ICCET 2022 - Proceedings Release v1

10th INTERNATIONAL CONFERENCE ON
“CONTEMPORARY ENGINEERING AND TECHNOLOGY –

2022”
(ICCET – 2022)
May 21st – 22nd, 2022
ORGANIZED BY
ORGANIZATION OF SCIENCE AND INNOVATIVE

ENGINEERING & TECHNOLOGY (OSIET).
Chennai, India.
Website: www.ijsiet.org
In association with
PRINCE SHRI VENKATESHWARA PADMAVATHY

ENGINEERING COLLEGE
PRINCE DR. K. VASUDEVAN COLLEGE OF ENGINEERING

AND TECHNOLOGY
Medavakkam - Mambakkam Road, Ponmar, Chennai – 600 127
Website: www.psvpec.in
www.princedrkvasudevan.com
ICCET 2022
Proceeding of International Conference on Contemporary Engineering and

Technology.
21st & 22nd May, 2022
Organized by:
Organization of Science and Innovation Engineering & Technology
4/3, Vallalar Street, Choolaimedu, Chennai – 600094
Venue:
Prince Shri Venkateshwara Padmavathy Engineering College
Prince Dr. K. Vasudevan college of Engineering and
Technology Medavakkam - Mambakkam Road, Ponmar,
Chennai – 600 127
ISBN 978-93-81288-22-1
Published by:
Sabanayagam Publications, Chennai.

Design & Printed by :
United Bind Graphics, Chennai.
[email protected], 9282102533
CHAIRMAN’S MESSAGE
I am delighted to welcome you all to the International Conference on
“Contemporary Engineering and Technology” (ICCET 2022) which brings
together experts and academics from around the world. The conference this year
has brought together a tremendous and rich diversity of authors and speakers from
universities, government and industry to share ideas and new perspectives on a
wide range of Communications, Electronics, Networking, Computer Science,
Mechanical and Electricals research and technologies. Addressing the new
technical and business issues are essential to advance in today's engineering and
technological environments. In order to provide an outstanding technical level for
the presentations at the conference, we have invited more than 15 distinguished
experts in the Engineering and Technology field to participate in the Advisory
Committee. Academic excellence has always been the hall mark of our institute
and we are committed to provide a comprehensive education which seeks to
develop the students into academically proficient, morally upright and socially well
integrated individuals. I would like to express my thanks to all authors for their
outstanding contributions and in particular the members of the program board for
their competent evaluation of the large number of submissions
We are very pleased with the quality, depth, and breadth of this year's
technical program. I wish you a most enjoyable experience at the Conference.
Dr.K.Vasudevan
Chairman
Prince Group of Educational Institutions
VICE CHAIRMAN’S MESSAGE

The words “Engineering” and “Technology” are not just words but are vital
disciplines that guide the world by bringing huge changes to the current way of
living. Modern Engineering and Technological innovations are making people
think that nothing is impossible in this world. Everything can be achieved with
an extremely powerful vision and a path to attain that vision. If a person is doing
a marvel, keep in mind that you too can outshine them with perseverance. Dr. A.
P. J Abdul Kalam said “Dream is not something that comes when you sleep, it is
something that doesn’t let you sleep”, so, dream a lot about how to bring changes
and try to make the changes come true. Remember earning money is not life; it is
just a part of life which can be easily earned when you do it with dedication.
Science and technology have become so closely intertwined, and I hope the
conference will help you to reinforce each other.
th
We are very much elated in welcoming you all to our college for the 10
International Conference on Contemporary Engineering and Technology and I wish
the students all the very best for their presentations.
Dr. V.Vishnu Karthick,
Vice Chairman,
Prince Shri Venkateshwara Padmavathy Engineering College.
ADMINISTRATIVE OFFICER’S MESSAGE

The world is growing fast in fact it grows in such a way that a thing invented
yesterday becomes obsolete today, which means there is a plethora of competition
in each and every discipline of engineering and technology. This scenario goes
well with a famous quote of a philosopher Heraclitus, who said, “Change is the
only constant in this world.” Now a days people need new inventions for their
daily works so that they can do it with ease. People not only need to do their
activities with ease but they also need it to be done within a period of time. As an
engineer we need to solve problems that are left unsolved. As a way of showcasing
th
the talents of young minds to find solutions we have organised this 10
International Conference on Contemporary Engineering and Technology.
I welcome you all to this event, my warm wishes to the students who are going to
present their papers and also hearty thanks for the staff who have made this event a
great success.
Er. K. Parthasarathy
Administrative Officer
Prince Group of Educational Institutions
PRINCIPAL’S MESSAGE
I express my happiness to meet you through this 10th International Conference on Contemporary
Engineering and Technology (ICCET 2022). I am much delighted to convey my warm greetings to
the Guests and the Host. We, at Prince Shri Venkateshwara Padmavathy Engineering College and
Prince Dr. K. Vasudevan College of Engineering and Technology, Ponmar, Chennai impart
futuristic technical education and instil high pattern of discipline. Education is sweeter and
valuable when it promotes the blossoming of natural talent that enriches students to be self reliant
individuals. The scenario of technical education in India is now in a transformational phase as
compared to earlier times. We are in a compelling era where global conditions require
conventional practices which make one to change and adapt to suit the present requirements and
also address futuristic needs. The college is committed in enhancing the quality of service delivery,
providing practical cum industrial-oriented programs and ensuring a valuable campus environment
for the faculty and students. With the support of the Management, faculty members, staff and
students are excelling in their work. The holistic and comprehensive education provided in the
College will enlighten the students and enable them to face the challenging world. Research is the
key parameter to promote the individuality to horizon. In order to create the best engineers we
have organized the international conference to enhance the research activities of the budding
engineers.
This proceeding of the conference has been documented with utmost care. I believe strongly that,
this will stand as a great source of knowledge and researchers. With immense pleasure and pride, I
welcome all the participants and convey my best wishes for ICCET 2022.
Dr. G Indira
Principal
PRINCIPAL’S MESSAGE
th
I am indeed most delighted to chair this 10 International Conference on
Contemporary Engineering and Technology (ICCET 2022). The sharp, clear sighted
vision and precise decision making powers of our management has benefited our
college to stay competitive. The Pedagogy at the Institute is Modern where a variety
of learning, behavioural tools are used in quality pursuance of knowledge,
development of skills, attitudes, and values complemented by academia-industry
interface imparting uniqueness to our programme. The Institute focuses on the
holistic development of its students through variety of methodologies and
extracurricular activities the whole year round. Today, we live in an era of incredibly
rapid technological change. Technology has dominated our lives and we now have
the ever evolving technology at our finger tips. Symposium, Seminars and
International Conferences and Workshops are organized in the institute for their
overall development. Teaching and Research are the two primary activities through
which we fulfil our Mission and Objective. The Institute takes pride in welcoming
th the National and
International participants for 10 International Conference.
With a firm foundation of the past and high hopes for a bright future, I wish everyone
good luck and prosperity ahead. May we grow many technological wings!
Dr. T.Sunder Selwyn

Principal
Prince Dr. K. Vasudevan College of Engineering and Technology.
ABOUT INSTITUTION
Prince Educational Society was established in 1978 by our Founder - Chairman Dr. K. Vasudevan, M.A.,
B.Ed., Ph.D. Going down memory lane in the seventies, our Chairman had realized the need for an
Institution which will serve as a role model and stand apart from other Educational Institutions.
VISION
The main aim of Prince Shri Venkateshwara Padmavathy Engineering College & Prince Dr. K. Vasudevan
College of Engineering & Technology is to meet the challenges and demand of the highly advanced scientific
and technological fields and to prepare to meet the man power needs of the world.
MISSION
To provide goal-oriented, quality-based and value-added education through state-of-the-art technology on
a par with international standards. To promote nation - building activities in science, technology,
humanities and management through research.To create and sustain a community of learning that sticks
on to social, ethical, ecological, cultural and economic.
OUR GROUP OF INSTITUTIONS
Prince Dr. K. Vasudevan College of Engineering & Technology
Prince Shri Venkateshwara Arts and Science College
Prince Matriculation Higher Secondary School,7,Kannagi Street, Madipakkam to Puzhuthivakkam, Chennai 091
Prince Sri Vari Vidyalaya CBSE School,93 ,College Road , Nanganallur, Chennai 114.
Prince Sri Vari Vidyalaya CBSE School, 12,Kannagi Street, Madipakkam to Puzhuthivakkam, Chennai 91 Shri
Venkateshwara Higher Secondary School, 9,Kannagi Street, Madipakkam to Puzhuthivakkam, Chennai 91
Prince Matriculation Higher Secondary School, 67&68, College Road, Nanganallur , Chennai 114.
ICCET 2022
PATRON-IN-CHIEF
Thiru. Dr. K. Vasudevan, M.A., B.Ed., Ph.D,

Chairman,Prince Group of Educational Institutions.
Thiru. Dr.V.VishnuKarthik, MD, Vice-Chairman, Prince Group

of Educational Institutions.
SECRETARY
Er.K.Parthasarathy, B.E,
Administrative Officer, Prince Group of Educational Institutions.
CONFERENCE CHAIR
Dr. V.Mahalakshmi, Principal,
Dr.Sunder Selwyn, Principal,

Prince Dr.K.Vasudevan College of Engineering
SPEAKERS
Dr. K. Aldrin Karunaharan, Assistant Professor, International Maritime College

Oman, Sohar, Oman
Dr. Vaithyasubramanian.S, Assistant Professor, PG and Research Department of

Mathematics, Dwaraka Doss Goverdhan Doss Vaishnav College, Chennai,
Tamil Nadu
Dr. Sivakumar Krishnan, Professor & Dean and Former Principal, Rathinam
Technical Campus, Coimbatore, Tamil Nadu
CONFERENCE SESSION CHAIR
Dr. S. Vijayakumar, IEEE Vice Chairman, PSES – Madras Chapter
Dr. Sivakumar Krishnan, Professor & Dean and Former Principal, Rathinam
Technical Campus, Coimbatore, Tamil Nadu
Dr. M.Jeyakarthic, Assistant Director, Tamil Virtual Academy, Anna

University, Chennai, Tamil Nadu
Dr. Vaithyasubramanian.S, Assistant Professor, PG and Research Department of

Mathematics, Dwaraka Doss Goverdhan Doss Vaishnav College, Chennai,
Tamil Nadu
Dr. G. N. R. PRASAD, Associate Professor, MCA department, Chaitanya

Bharathi Institute of Technology, Gandipet, Telangana
Dr. T Jayakumar, Assistant Professor, Department of Mechanical, AMET

University, Chennai, Tamil Nadu
Dr. H.S.Niranjana Murthy, Associate Professor, Electronics & Instrumentation,

Ramaiah Institute of Technology, Bangalore, Karnataka
Dr. Harshavardhan Awari, Senior Assistant Professor, Computer Science,

VNRVJIET, Hyderabad, Telangana
Dr. K. Sudhakar, Associate Professor, Computer Science and Engineering,

PSCMR College of Engineering & Technology, Vijayawada, AP
Dr. Velappagari Sekhar, Associate Professor & HOD, Kuppam Engineering

College, Kuppam, AP
Dr. C Sakthi Gokul Rajan, Assistant Professor (Senior), Kuppam Engineering

College, Kuppam, AP
Dr. G Rajesh, Teaching and Research Assistant, SENSE, VIT university,

Vellore, TN
Dr. C Kavitha, Associate Professor, Electronics and Communication

Engineering, SITAMS Autonomous, Chittoor, AP
Dr. Raviprakash ML, Associate Professor, Department of CSE, Kalpataru

Institute of Technology, Tiptur, Karnataka
Dr. Ananda Babu J, Associate Professor, Department of Information Science and
Engineering, Malnad College of Engineering, Hassan, Karnataka
Dr N Vinodhkumar, Associate Professor, Electronics and Communication

Engineering, Vel Tech Rangarajan, Dr Sagunthala R and D Institute of Science and
Technology, Chennai, Tamil Nadu
INDEX
S.NO PAPER ID PAPER TITLE
1 ICCET22002 UWB MICROSTRIP PATCH ANTENNA FOR WIRELESS SENSOR APPLICATIONS
MULTILAYER PERCEPTRON ANALYZED THE AUDIENCE PERCEPTION TOWARDS CITIZENSHIP
2 ICCET22004 AMENDMENT ACT-CAA (INDIA)
REVIEW OF RESEARCH ON AERODYNAMIC CHARACTERISTICS OF SUPERCRITICAL AIRFOIL 2414 SERIES
3 ICCET22001 ON TRANSONIC FLIGHT
TEXTILE HORIZONTAL T-SHAPED ULTRAWIDE BAND MICROSTRIP PATCH ANTENNA FOR WIRELESS BODY
4 ICCET22010 AREA NETWORK APPLICATIONS
5 ICCET22007 INCOMPLETE FUZZY SOFT SETS IN SANCHEZ'S METHOD OF DECISION MAKING
6 ICCET22008 ON SOLVING FUZZY TRANSPORTATION PROBLEM USING FUZZY ONE-POINT METHOD
7 ICCET22006 CRIME DATA FORECASTING USING MACHINE LEARNING AND BIG DATA ANALYTICS
8 ICCET22012 APPLICATION OF QUEUEING THEORY FOR THE IMPROVEMENT OF RAILWAY SERVICES
9 ICCET22011 APPLICATIONS OF GENERALIZED FUZZY SOFT SETS IN DECISION MAKING PROBLEM
DESIGN OF NOVEL T-FLIP FLOP FOR SYNCHRONOUS COUNTER USING QUANTUM-DOT CELLULAR
10 ICCET22020 AUTOMATA
INFLUENCE OF MANAGER’S SUPPORT IN CAREER PLANNING AND DEVELOPMENT ON CREATING
11 ICCET22028 EMPLOYEES’ POSITIVE ATTITUDE TOWARDS ORGANIZATION.
12 ICCET22035 AN APPLICATION SPECIFIC WEB-BASED DATABASE FOR INSTITUTIONAL DATA MANAGEMENT SYSTEM
13 ICCET22037 RANKING OF PARABOLIC TRAPEZOIDAL FUZZY NUMBER USING THE CENTROIDS AND FOCUS
14 ICCET22036 INSERTION OF TIE –BREAKING MECHANISM FOR FLOW SHOP SCHEDULING PROBLEM
15 ICCET22027 IOT BASED SOLDIERS MONITORING AND TRACKING IN THE WAR-FIELD
16 ICCET22032 ANN BASED IMAGE SEGMENTATION FOR PHOTOVOLTAIC HOT-SPOT DETECTION
17 ICCET22047 A DISTINCTIVE APPROACH FOR EARLY DETECTION OF PLANT DISEASE FOR SUSTAINABLE AGRICULTURE
A NEW MODEL BASED DESIGN AND PERFORMANCE ANALYSIS OF HIGH THROUGHPUT AND LOW
18 ICCET22039 LATENCY DA-FIR FILTER FOR EFFECTIVE NOISE REMOVAL IN SDR APPLICATIONS
TEACHING AID BASED ON CONTEXT ACTIVE CULTURAL BASED STUDY FOR LEARNING COMPILER DESIGN
19 ICCET22042 CONCEPT FOR STUDENTS
ANALYSIS OF KNN, LINEAR, RIDGE, LASSO, POLYNOMIAL REGRESSION AND SVM ALGORITHMS FOR
20 ICCET22034 MAXIMUM ACCURACY TO PREDICTING THE PRICE OF BITCOIN
21 ICCET22031 CHARMING SEWERAGE SPECIALIST CARE ARRANGEMENT USING EMBEDDED SYSTEM
HUMANA BRAIN-MOTIVATED MODEL FOR DISTINGUISHING EMOTIONAL STATES FOR AUSTIN
22 ICCET22054 DISORDER_CHATBOT
23 ICCET22033 SMART DELIVERY BOX BASED ON TENSORFLOW OBJECT DETECTION AND RECOGNITION API
24 ICCET22044 A NOVEL ENSEMBLED DEEP LEARNING MODEL FOR BIOMEDICAL QUESTION CLASSIFICATION
FORECASTING OF CROP YIELD IN VARIOUS CLIMATIC ZONES OF KARNATAKA USING PREDICTIVE
25 ICCET22046 ANALYTICS
26 ICCET22053 UN-STRUCTURED MEDICAL DATA DETECTION FOR AUTISM SPECTRUM DISORDER
27 ICCET22062 SIGN LANGUAGE CONVERTER USING FEATURE EXTRACTOR AND POSENET
SMART AND SUSTAINABLE HORTICULTURE USING INTERNET OF THINGS AND ARTIFICIAL INTELLIGENCE:
28 ICCET22068 A QUALITATIVE REVIEW
DENSITY BASED SUPPORT VECTOR MACHINE WITH OUTLIER FILTERS FOR REAL TIME ANALYSIS OF
29 ICCET22071 ROCKS AND STRUCTURES
30 ICCET22069 DESIGN OF 243 LEVEL TRINARY LADDER MULTILEVEL INVERTER USING VHDL
31 ICCET22065 A TRANSFER LEARNING-BASED DCNN FRAMEWORK FOR BRAIN TUMOUR MULTI-CLASSIFICATION
32 ICCET22067 PRE-SERVICE EDUCATION: A PRE-REQUISITE FOR THE TEACHERS OF NEW GENERATION ESL LEARNERS
33 ICCET22056 FACIAL RECOGNITION PROCESS USING MACHINE LEARNING ALGORITHMS: A REVIEW
PREDICTION OF RESPIRATORY DISEASES FROM AUDIO DATA CAPTURED THROUGH DIGITAL
34 ICCET22066 STETHOSCOPE AND SEVERITY ANALYSIS OF DISEASES
35 ICCET22063 A COMPARATIVE STUDY ON SELF-SOVEREIGN IDENTITY POWERED BY THE BLOCKCHAIN TECHNOLOGY
36 ICCET22079 CONDUCTION OF ONLINE EXAMINATIONS : A SURVEY
37 ICCET22083 DESIGN OF FUSION DIGITAL PULSE WIDTH MODULATION IN QUANTUM DOT CELLULAR AUTOMATA
EMOTIONAL INTELLIGENCE FOR SUSTAINABLE GROWTH OF ENTREPRENEURS: A REVIEW STUDY
38 ICCET22086 CONNECTED WITH WORK-LIFE BALANCE
39 ICCET220104 INFRA SOFT Α - OPEN(CLOSED) SETS ON INFRA SOFT TOPOLOGICAL SPACE

40 ICCET22090 PREDICTION ON FIELD CROPS YIELD BASED ON TIME SERIES ANALYSIS DEEP LEARNING MODEL
ENERGY-EFFICIENT AND BALANCED LOAD ROUTING APPROACH USING BALD EAGLE SEARCH
41 ICCET220119 ALGORITHM FOR PATH CONVERGENCE IN MANET
42 ICCET22091 SECURITY ANALYSIS AND MITIGATION IN AUTONOMOUS VEHICLE
USE OF CARBON NANO TUBE (CNT) AS A WRAPPING MATERIAL TO THE BUILDING STRUCTURES TO
43 ICCET220122 CONTROL SALT DAMAGE
TRANSIENT RESPONSE ANALYSIS OF PV BASED NON- ISOLATED TWO PHASE INTERLEAVED BOOST
44 ICCET220117 CONVERTER FOR STANDALONE DC MICROGIRD
45 ICCET220112 OPTIMIZATATION OF PARTIALLY REPLACEMENT OF NANO – SIO 2 WITH CEMENT
46 ICCET22072 PERSONALIZED RECOMMENDER SYSTEM USING SENTIMENT AND EMOTION ANALYSIS FROM TWEETS
47 ICCET220137 GROUP {1,-1,I,-I} CORDIAL LABELING SOME DUPLICATE GRAPHS
MODELING OF FILTERS FOR GRID CONNECTED VOLTAGE SOURCE INVERTER USED IN RENEWABLE
48 ICCET220110 ENERGY RESOURCES
49 ICCET220113 SEISMIC ANALYSIS STEEL TRUSS BRIDGE USING SPLICES CONNECTION
50 ICCET220140 RECURRENCE OF GENERALIZED LIMITING RATIO RELATIONS
51 ICCET220153 A RELIABLE DATA TRANSMISSION USING AN ADAPTIVE ROUTING PROTOCOL FOR FANETS
52 ICCET220154 STUDIES ON PREPARATION OF SQUASH FROM COCONUT SPROUT
ENHANCEMENT OF REJECTION CHARACTERISTICS OF FILTER USING AN ARRAY OF CSRR FOR NOTCHING
53 ICCET220111 THE ISM BAND IN UWB ANTENNAS
54 ICCET220223 WIRELESS POWER TRANSFER SYSTEM FOR IMPROVED COIL GEOMETRY
ARTIFICIAL NEURAL NETWORKS AND PI HYBRID CONTROLLERS FOR AN ELECTRIC VEHICLE'S ENERGY
55 ICCET220169 STORAGE SYSTEM
56 ICCET220118 A SHORT REVIEW ON BATTERY MANAGEMENT SYSTEM FOR ELECTRIC VEHICLE
IMPACT OF WORK FROM HOME ON IT EMPLOYEES CONSIDERING THE PRODUCTIVITY, WORK LIFE
57 ICCET220143 BALANCE AND COLLEAGUESHIP
58 ICCET220155 RECOGNITION OF TRAFFIC INFORMATION WITH THE HELP OF SOCIAL MEDIA TWEETS
SENTIMENT ANALYSIS WITH SARCASM DETECTION USING OPTIMAL BIDIRECTIONAL LONG SHORT TERM
59 ICCET220189 MEMORY ON TWITTER DATA
IOT BASED DECISION SUPPORT SYSTEM FOR TELEHEALTH MANAGEMENT THROUGH DATA MINING FOR
60 ICCET220149 BIOMEDICAL EQUIPMENT
EFFECT OF SURFACE ROUGHNESS ON THE AERODYNAMIC AND AEROACOUSTIC PERFORMANCE OF THE
61 ICCET220187 DARRIEUS WIND TURBINE
62 ICCET220141 PCOD PREDICT FOR OBESE PATIENT USING NAFLD AND T2D PROFILES
63 ICCET220205 NETWORK SECURITY ATTACKS & TECHNIQUES TO SECURE THE NETWORKS
64 ICCET220206 METHODS FOR RESTRICTING UNWANTED ACCESS ON THE NETWORKS USING FIREWALL
COMBINING DATA OWNER- SIDE AND CLOUD-SIDE ACCESS CONTROL IN ATTRIBUTE BASED ENCRYPTED
65 ICCET220107 CLOUD STORAGE
66 ICCET220191 TRIPLE BAND U-SLOT ULTRA-WIDEBAND ANTENNA AS A WIRELESS SENSOR FOR COMMUNICATION
67 ICCET220201 MESH MESSAGING SERVICE USING BLUETOOTH
ANALYSIS AND OPTIMIZATION OF MACHINABILITY OF AL2O3 AND FLY ASH REINFORCED ALUMINUM
68 ICCET220184 HYBRID COMPOSITES USING CARBIDE INSERTS
69 ICCET220185 ANALYSIS OF THE UPSHOT OF HEAT TREATMENT ON PROPERTIES OF MAGNESIUM HYBRID COMPOSITES
70 ICCET220197 IMPROVING THE LIQUID INSULATION PERFORMANCE WITH ANTIOXIDANTS
71 ICCET220135 MULTIPLE VENDOR E-PLATFORMS FOR FARMERS
72 ICCET220145 Q LEARNING ALGORITHMS :AN INTRODUCTION
ACHIEVING MORE SECURITY IN ELECTRONIC MONEY FOR E-COMMERCE WEBSITE THROUGH
73 ICCET220134 CRYPTOGRAPHY
74 ICCET220131 SPAM DETECTION USING LINGUISTICS AND SPAMMER BEHAVIORAL
BIDIRECTIONAL VISITOR COUNTER WITH NO-MASK AND NO-ENTRY ALONG WITH AUTOMATIC ROOM
75 ICCET220157 LIGHT FAN CONTROLLER
76 ICCET220159 THE VEHICLE SERVICE MANAGEMENT SYSTEM
77 ICCET220130 MULTI-SOURCE MEDICAL DATA INTEGRATION AND MINING FOR HEALTHCARE SERVICES
78 ICCET220133 FAKE NEWS DETECTION USING MACHINE LEARNING
79 ICCET220156 AN EFFICIENT WAY TO PREDICT THE DISEASE USING MACHINE LEARNING

80 ICCET220132 REAL TIME OBJECT DETECTION
81 ICCET220172 UNIFIED HEALTH CARE SYSTEM USING WEB FRAMEWORKS
82 ICCET220180 ALCOHOL DETECTION AND EMERGENCY ALERT SYSTEM
83 ICCET220181 DEVELOPMENT OF HUMAN RESOURCE: A CRITICAL ANALYSIS OF INDIA INNOVATION INDEX
AN OPTIMAL DEEP WAVELET AUTOENCODER -BASED DEEP NEURAL NETWORK WITH THE USE OF
84 ICCET220151 RIDERCUCKOO SEARCH ALGORITHM FOR CLASSIFICATION OF LUNG CANCER ON CT IMAGES
85 ICCET220207 A NOVEL METHOD OF APPROACHING TRANSPORTATION PROBLEM OF HEXAGONAL FUZZY NUMBER
86 ICCET220215 XGBOOST OPTIMIZED BY ADAPTIVE TREE PARZEN ESTIMATORS FOR CREDIT RISK ANALYSIS
A LIGHTWEIGHT RUBIK’S CUBE AND CHAOTIC-BASED IMAGE ENCRYPTION AND KEY GENERATION
87 ICCET220178 ALGORITHM FOR MULTIMEDIA INTERNET OF THINGS
88 ICCET220214 MUSIC RECOMMENDATION USING EMOTION RECOGNITION
89 ICCET220190 CLASSIFICATION OF YOGA HAND MUDRAS USING SIFT AND SURF FEATURES
90 ICCET220129 QUERY CACHING IN BIGDATA ANALYTICS
91 ICCET220211 DETECTION OF FACE MASK AND DISTANCE CALCULATION USING DEEP LEARNING
92 ICCET220218 PENETRATION TESTING TECHNIQUE WITH REVERSE SHELL ATTACK
93 ICCET220213 CONVERTING VOICE SIGNAL TO VISUAL INDIAN SIGN LANGUAGE
94 ICCET220273 MODIFIED MUSIC POPULARITY PREDICTION WITH DEEP LEARNING TECHNIQUES
95 ICCET220232 APPLICATION OF GIS TECHNIQUES IN CITY TRAFFIC
AN OVERVIEW ON ADVANCED GENETIC DISEASE DIAGNOSIS AND PREDICTION TECHNIQUES USING
96 ICCET220115 GENOME DATA
DEVELOPING A NEXT-GENERATION MULTITUDINOUS EPITOPE-BASED VACCINE AGAINST BIOTINIDASE
97 ICCET220230 DEFICIENCY; AN IMMUNOINFORMATICS BASED APPROACH
98 ICCET220194 PLANT DISEASE MANAGEMENT SYSTEM
99 ICCET220252 FACE MASK DETECTION USING VIOLA-JONES ALGORITHM
100 ICCET220250 ADVANCED DOORBELL TO PREVENT COVID-19 USING ARDUINO NANO
101 ICCET220234 UPPER LIMB ANATOMY LEARNING WITH VR
NEWS CLASSIFICATION FOR INDIAN LANGUAGES BASED ON THEIR HEADLINES USING MACHINE
102 ICCET220227 LEARNING AND DEEP LEARNING MODELS
103 ICCET220248 A/B TESTING FOR INDUSTRIAL COMAPANIES
104 ICCET220419 IMPROVED DIGITAL TWINNING TECHNIQUES FOR SMART SIMULATIONS
105 ICCET220418 MEDICAL CHATBOT USING MACHINE LEARNING AND NLP
106 ICCET220219 ASTHMA PREDICTION USING MACHINE LEARNING CLASSIFIERS
107 ICCET220248C A NOVAL APPROACH FOR WSN USING SHORTEST PATH (DIJKSTRAS ALGORITHM)
108 ICCET220248D IOT BASED SMART CHARGING SYSTEM
109 ICCET220248E A TOURISM MANAGEMENT WITH IMAGE SEARCH SYSTEM USING PIXABAY
110 ICCET220221 ML ASSISTED LOCALIZATION ALGORITHMS FOR INDOOR/OUTDOOR SETTINGS: AN INTROSPECTION
EVALUATION OF THE FUNCTIONAL ROLES PLAYED BY THE DORSOLATERAL PREFRONTAL CORTEX IN
111 ICCET220284 HUMAN COGNITION
RECENT TRENDS IN NEURAL NETWORK-BASED APPROACHES TO VISUAL PROCESSING VIS-À-VIS THEIR
112 ICCET220285 BIOLOGICAL CORRELATES IN THE BRAIN
EXPLORATIVE STUDY ON CHILDREN’S INTERNET USAGE AND PARENTAL CONTROL IN INDIA – USING
113 ICCET220257 CLASSIFICATION TREE MODEL ANALYSIS
114 ICCET220226 A REVIEW PAPER ON DIMENSIONALITY REDUCTION TECHNIQUES
115 ICCET220253 SOLAR PANEL DUAL MANAGEMENT SYSTEM
116 ICCET220242 SOCIAL IMPACT ASSESSMENT IN PROTECTED AREAS OF PANNA NATIONAL PARK, MADHYA PRADESH
117 ICCET220272 ATTRIBUTE BASED ACCESS CONTROL MODEL FOR EHR
IMPLEMENTING FUZZY LOGIC TO IMPROVE THE NETWORK SELECTION PROCESS FOR HETEROGENEOUS
118 ICCET220247 HANDOVER
119 ICCET220274 MEGAPSO: MOBILITY AWARE ENERGY EFFICIENT HYBRID GA & PSO BASED CLUSTERING FOR MANET
120 ICCET220238 WATER CONSUMPTION IN THE POWER SECTOR IN INDIA - PROBLEMS AND PROSPECTS
121 ICCET220283 RFID WIRELESS VEHICLES CHARGING STATION FOR ELECTRIC VEHICLES AND PAYMENT SYSTEM

122 ICCET220287 ASSESSMENT OF VARIOUS MILK PARAMETER MONITORING SYSTEM USING AN IOT
123 ICCET220253A ONLINE FOOD ORDERING SYSTEM
DATA AUGMENTATION TECHNIQUES FOR DIAGNOSIS OF KERATOCONUS USING CONVOLUTIONAL
124 ICCET220222 NEURAL NETWORKS
IMPLEMENTATION OF DEEP NEURAL NETWORK MODEL FOR DETECTION OF HUNTINGTON'S DISEASE IN
125 ICCET220225 HUMAN DNA SEQUENCE
126 ICCET220258 DELAY OF NETWORK ISSUES AND CHALLENGES
127 ICCET220266 IMPLEMENTATION OF DATA INTEGRITY EVALUATION PROCESS IN CLOUD COMPUTING
128 ICCET220313 A REVIEW ON IMAGE CLASSIFICATION USING DEEP LEARNING
GIS-BASED QUANTITATIVE ASSESSMENT OF PUBLIC GREEN SPACE DISTRIBUTION IN LUCKNOW CITY,
129 ICCET220224 INDIA
DESIGN AND IMPLEMENTATION OF U-SLOT MICROSTRIP PATCH ANTENNA USING CO-AXIAL FEEDING
130 ICCET220281 TECHNIQUE FOR WLAN APPLICATIONS
EVALUATION OF CORE POINT DESCRIPTORS ENFORCED IN THE PEDESTRIAN IDENTIFICATION IN LOWER
131 ICCET220317 QUALITY IMAGES
132 ICCET220298 IN SILICO INTERACTION STUDIES OF SALACINOL WITH INHIBITORS OF IAP PROTEINS
IN SILICO SCREENING OF SNP'S IN PRION PROTEIN AND COMPARATIVE STUDIES BETWEEN
PHYTOCHEMICAL COMPOUNDS THAT INHIBIT PRION PROTEIN CAUSING NEURO-DEGENERATIVE
133 ICCET220297 DISEASES.
MULTI-OBJECTIVE VIRTUAL MACHINE PLACEMENT USING ORDER EXCHANGE AND MIGRATION ANT
134 ICCET220307 COLONY SYSTEM ALGORITHM
135 ICCET220310 EFFICIENT DESIGN OF SRAM CELL USING 90 NM TECHNOLOGY
136 ICCET220293 THIOPHENE AS A GAS SENSOR FOR THE ADSORPTION OF CARBONYL SULFIDE: DFT CALCULATIONS
137 ICCET220304 RASPBERRY PI BASED READER FOR BLIND
138 ICCET220250A TEMPERATURE PREDICTION USING MACHINE LEARNING ALGORITHMS
139 ICCET220302 CLASSIFICATION OF PADDY AND TOMATO CROPS DISEASES DETECTION USING CNN
DEEP LEARNING AND SWARM INTELLIGENCE BASED SYSTEM FOR SMART AGRICULTURE USING
140 ICCET220437 WIRELESS SENSOR NETWORKS
141 ICCET220336 SIGN LANGUAGE RECOGNITION USING DEEP LEARNING
142 ICCET220409 DETECTION OF OBJECTS USING MACHINE LEARNING
CODE COVERAGE - STUDY ON APPLICATIONS OF CODE COVERAGE METRICS FOR IMPROVING TEST
143 ICCET220414 EFFECTIVENESS
144 ICCET220413 DETECTING SKIN CANCER USING DEEP LEARNING
145 ICCET220374 COVID-19 INDOOR SAFETY MONITORING SYSTEM USING MACHINE LEARNING
146 ICCET220327 SENTIMENT ANALYSIS ON USER REVIEWS AND TWITTER DATA TO PREDICT VEHICLE SALES
147 ICCET220325 CHILD ADOPTION BASED ON FACIAL FEATURES AND SKIN COLOR
148 ICCET220328 SOCIAL MEDIA ADVERTISEMENT RESULT ANALYSIS
DETECTION OF VIRULENT MESSAGES ON PROPOSED DATASET WRITTEN IN CODE-MIXED HINDI-ENGLISH
149 ICCET220429 LANGUAGE USING NLP TECHNIQUES
150 ICCET220330 INTELLIGENT TRAFFIC FLOW USING MACHINE LEARNING
151 ICCET220321 DESIGN AND DEVELOPMENT OF SOLAR WATER PUMP WITH DIFFERENT TIME SLOTS FOR POWER SAVING
152 ICCET220446 DETECTING PASSIVE ATTACKS IN FINGERPRINT SPOOFING USING CONVOLUTION NEURAL NETWORK
PREDICTION OF COVID-19 & PNEUMONIA BY USING CONVOLUTIONAL NEURAL NETWORK AND
153 ICCET220415 DEPLOYMENT BY USING THE WEB FRAMEWORK
154 ICCET220323 TEMPERATURE CONTROLLED DC FAN USING SOLAR ENERGY
SECURED DATA TRANSMISSION FOR SMART GRID SYSTEM USING BLOCKCHAIN TECHNOLOGY WITH A
155 ICCET220373 TRUSTED GAMING TOKEN HONEYPOT ALGORITHM
156 ICCET220375 FRAUD CLAIMS DETECTION IN INSURANCE USING MACHINE LEARNING
157 ICCET220430 PHISHING SCAM IN THE MIDST OF THE UKRAINE CRISIS
158 ICCET220447 CROP ANALYSIS AND PREDICTION USING ENSEMBLE LEARNING
159 ICCET220455 PREDICTING STUDENT'S MARKS USING MACHINE LEARNING
160 ICCET220451 MODELING OF 2DMULTILAYER MOLYBDENUM DISULFIDE MOS 2 TRANSISTOR
161 ICCET220424 A STUDY TO EXPLORE HINDRANCES IN IMPLEMENTATION OF DATA MINING IN BANKS

162 ICCET220393 DESIGN & IMPLEMENTATION OF BUS TRACKING SYSTEM UNDER IOT PLATFORM
163 ICCET220432 SENTIMENT ANALYSIS OF PRODUCT REVIEW
164 ICCET220262 VOICE BASED EMAIL FOR BLIND USING ARTIFICIAL INTELLIGENCE AND PYTHON
165 ICCET220427 PREDICTION OF DIABETES USING MACHINE LEARNING ALGORITHM
166 ICCET220420 RSSI BASED SAND THEFT PREVENTION
167 ICCET220422 A BRIEF REVIEW OF IMAGE CAPTIONING METHODS USING DEEP LEARNING APPROACH
168 ICCET220368 DETECTION OF CYBER ATTACK IN NETWORK USING DIFFERENT MACHINE LEARNING APPROACHES
169 ICCET220456 A CONTEMPLATION OF CLOUD SECURITY MENACES AND MITIGATION TECHNIQUES
170 ICCET220342 BLOOD BANK MOBILE APPLICATION OF IOT-BASED ANDROID STUDIO FOR COVID-19
171 ICCET220408 VIRTUALLY CONTROLLING COMPUTERS USING HAND GESTURE AND VOICE COMMANDS
A UNIFIED APPROACH FOR DETECTING TRAFFIC ACCIDENT AND ANALYSING THE ACCIDENT PRONE ZONE
172 ICCET220449 USING QUANVOLUTIONAL NEURAL NETWORKS
173 ICCET220309 PREDICTION OF HEART DISEASE USING MACHINE LEARNING ALGORITHMS
174 ICCET220410 CODE PINK A NEWBORN ABDUCTION DETECTION SYSTEM
175 ICCET220467 REAL TIME CRIME DETECTION USING GESTURE RECOGNITION
PERFORMANCE ENHANCEMENT OF FULLY HOMOMORPHIC ENCRYPTION AND DECRYPTION USING HIGH
176 ICCET220538 SPEED MULTIPLIER
177 ICCET220487 SMART ELECTRICITY DEMAND FORECASTING BY USING HYBRID ARIMA WITH LSTM ALGORITHM
178 ICCET220394 MODIFIED M-RCNN APPROACH (AOD-NET) FOR ABANDONED OBJECT DETECTION IN PUBLIC PLACES
179 ICCET220376 AUTONOMOUS HOME GARDENING SYSTEM USING IOT
180 ICCET220498 THERMAL IMPACT SENSITIVITY ANALYSIS OF FLASH POWDER MIXTURE FOR FIREWORKS APPLICATION
EVALUATION OF GROUNDWATER CHEMISTRY AND MULTIVARIATE STATISTICAL STUDIES IN PARTS OF
181 ICCET220388 TIRUPUR DISTRICT, TAMIL NADU, INDIA
182 ICCET220473 PREDICTION ON REAL ESTATE HOUSE PRICE USING REGRESSION MODELS
183 ICCET220445 DELAY NEURAL NETWORK ANALYSIS FOR LIP READING - NEURAL MACHINE TRANSLATION
184 ICCET220392 ON-ROAD VEHICLE BREAKDOWN ASSISTANCE
A SURVEY ON VARIOUS MACHINE LEARNING TECHNIQUES FOR AN EFFICIENT BRAIN TUMOR DETECTION
185 ICCET220595 FROM MRI IMAGES
186 ICCET220597 CLASSIFICATION OF DIABETIC RETINOPATHY DISEASE USING DEEP LEARNING
187 ICCET220377 ONLINE COLLABORATIVE DATA CACHING IN EDGE COMPUTING
MITIGATION OF FIRE AND EXPLOSION HAZARDS OF FLASH POWDER BY ADDING MONO AMMONIUM
188 ICCET220504 PHOSPHATE AS A SOLID INERTANT
A COMPREHENSIVE REVIEW ON TRACTION MOTORS COMPARISON,OVERVIEW OF OPTIMIZATION
189 ICCET220383 TECHNIQUES,RECENT AND FUTURE TRENDS IN ELECTRIC VEHICLE
190 ICCET220574 AI - BASED SMART IRRIGATION SYSTEM
191 ICCET220480 CUSTOMER SEGMENTATION USING NATURAL LANGUAGE PROCESSING
ENERGY EFFICIENT CH SELECTION & OPTIMAL RELAY NODE SELECTION ALGORITHM AND ITS
192 ICCET220530 PERFORMACE ANALYSIS FOR WIRELESS SENSOR NETWORKS
193 ICCET220432A ENERGY EFFICIENT STREET LIGHTING SYSTEM
194 ICCET220479 DEEP LEARNING BASED NEUROIMAGING ANALYSIS FOR PARKINSON’S DISEASE DETECTION
RECOMMENDATION SYSTEM FOR COLLEGE ELECTIVES USING ALS IMPLICIT COLLABORATIVE FILTERING
195 ICCET220513 AND LSTM NETWORK
196 ICCET220528 IOT BASED SMART SECURITY AND SURVEILLANCE SYSTEM
MITIGATING SHATTER BEHAVIOUR OF FIREWORKS BUILDING STRUCTURE USING REINFORCED HOLLOW
197 ICCET220500 BLOCK
198 ICCET220474 GESTURE RECOGNITION SYSTEM USING CONVOLUTED NEURAL NETWORKS AND TRANSFER LEARNING
199 ICCET220484 PRODUCT RECOMMENDATION ENGINE USING KNN AND K-MEAN
VISION AND INFRARED SENSOR BASED DISTANCE ESTIMATION FOR MICRO AERIAL VEHICLE
200 ICCET220337
NAVIGATION
201 ICCET220340 INDOOR MICRO AERIAL VEHICLE NAVIGATION USING OPTICAL FLOW
202 ICCET220493 BRAIN TUMOR DETECTION USING DEEP LEARNING

203 ICCET220407 SEISMIC RESPONSE CONTROL OF IRREGULAR RCC BUILDING USING BRACING SYSTEM
204 ICCET220570 PERFORMANCE ANALYSIS OF SPATIO- TEMPORAL HUMAN DETECTED KEYFRAME EXTRACTION
205 ICCET220378 AN INTERACTIVE PHYSIOTHERAPIC EQUIPMENTS FOR CEREBRAL PALSIC CHILDREN
206 ICCET220582 FLY ASH AS MINERAL ADMIXTURE IN CONCRETE MATRIX – A STATE OF THE ART
207 ICCET220395 A REVIEW: BLOCKCHAIN & CRYPTOGRAPHY USED TO SECURE DATA
BIOMETRIC SYSTEM FOR FACE RECOGNITION USING TENSOR FLOW ON RASPBERRYPI FOR EDUCATIONAL
208 ICCET220533 INSTITUTIONS
209 ICCET220382 EFFECT OF SURFACE ROUGHNESS ON LEADING EDGE PROTUBERANCED WING
HYBRID RECOMMENDATION SYSTEM WITH GRAPH BASED AND COLLABORATIVE FILTERING
210 ICCET220577 RECOMMENDATION SYSTEMS
DRAIN CURRENT ANALYSIS OF MG0.25ZN0.75O/ MG0.6ZN0.4O/ZNO DOUBLE- CAVITY MOSHEMT BASED
211 ICCET220516 BIOSENSOR
DECENTRALIZED PAYMENT ARCHITECTURE FOR E-COMMERCE AND UTILITY TRANSACTIONS WITH
212 ICCET220532 GOVERNMENT VERIFIED IDENTITIES
213 ICCET220611 A REVIEW ON SECURE TECHNIQUES FOR KEYWORD-BASED SEARCH AND DATA SHARING
214 ICCET220641 DETECTING MALIGN REQUESTS USING HYBRID DEEP LEARNING MODEL WITH SEQUENCE-BASED LSTM
215 ICCET220679 LOW WEIGHTED HUMAN ACTIVITY RECOGNITION USING DEEP LEARNING
216 ICCET220681 CARDIAC DISEASE ANALYSIS USING MACHINE LEARNING
217 ICCET220486 REAL TIME CROWD ANALYSIS
218 ICCET220494 SMART BLIND STICK WITH IMAGE TO SPEECH CONVERSION
219 ICCET220529 IMPLICATIONS OF THE EXTERNAL FLOW AND ACOUSTICS ON LATERAL UPWARD SMOLDERING
220 ICCET220651 DISEASE PREDICTION USING MACHINE LEARNING ALGORTIHMS
221 ICCET220636 STUDENTS LIVE BEHAVIOUR MONITORING IN ONLINE CLASSES USING ARTIFICIAL INTELLIGENCE
222 ICCET220698 DETECTION OF POTHOLES USING IOT
223 ICCET220550 STEGANALYSIS TECHNIQUES: A SYSTEMATIC REVIEW
224 ICCET220542 ONLINE SECURE VOTING SYSTEM USING SMART CONTRACTS
225 ICCET220546 SUICIDAL IDEATION DETECTION USING MACHINE LEARNING TECHNIQUES
AN ADAPTIVE SEMANTIC APPROACH FOR SOCIAL RECOMMENDATION USING CONTENT BASED
226 ICCET220426 ALGORITHM
227 ICCET220572 AUTONOMOUS FIRE DETECTING AND EXTINGUISHING ROBOT
228 ICCET220523 AN IMPROVED ECOURT SYSTEM WITH SUSPECT PREDICTION USING HAARCASCADE ALGORITHM
229 ICCET220685 MODIFIED PARABOLIC SOLAR COLLECTOR HOT-AIR TRAY DRYER
230 ICCET220366 CONTINUOUS SIGN LANGUAGE ESTIMATION USING LSTM
231 ICCET220604 LARGE SCALE PROCESSING OF SIGNALS IN NETWORK ENVIRONMENT
232 ICCET220658 SECURED BLOCKCHAIN TECHNOLOGY FOR AGRICULTURE FOOD SUPPLY CHAIN
233 ICCET220657 SMART HEALTH PREDICTION USING MACHINE LEARNING ALGORITHMS
234 ICCET220630 ANALYZING IMAGE DENOISING USING GENERATIVE ADVERSARIAL NETWORK
ENGINEERING A NOVEL RECRUITMENT SYSTEM USING GRADIENT BOOSTED DECISION TREE ALGORITHM
235 ICCET220713 AND BUSINESS INTELLIGENCE PRINCIPLES
A NOVEL APPROACH FOR NEXT GENERATION BUSINESS BY USING PARALLEL AND DISTRIBUTED DEEP
236 ICCET220627 LEARNING
237 ICCET220701 DECENTRALIZING ELECTRONIC MEDICAL RECORDS ON THE BLOCKCHAIN USING SMART COBTRACTS
238 ICCET220596 PRESERVING DATA PRIVACY IN WIRELESS COMMUNICATION IN IOT
239 ICCET220625 MUSIC GENRE CLASSIFICATION
240 ICCET220732 THE UNUSED POTENTIAL OF PROGRESSIVE WEB APPLICATIONS (PWA)
241 ICCET220571 DEDUCTION MANAGEMENTSOFTWARE
242 ICCET220704 GROUPING OF CUSTOMERS USING CLASSFICATION ALGORITHS
243 ICCET220667 WHEEL RIM RADIAL FATIGUE LIFE ANALYSIS CONSIDERING STEERING GEOMETRY
244 ICCET220676 A SURVEY ON KEY TECHNOLOGY AND DECISION-MAKING IN SELF-DRIVING CARS

245 ICCET220677 FEATURE EXTRACTION FOR ENGLISH LANGUAGE USING NLP
246 ICCET220734 STOCK ANALYSIS AND AUTOMATION USING MACHINE LEARNING
247 ICCET220629 TO DETECT SIGNATURE-BASED MALWARE USING MACHINE LEARNING
248 ICCET220620 DESIGN AND DEVELOPMENT OF ELECTRONIC VOTING MACHINE
249 ICCET220560 A SURVEY ON LEARNING MODELS OF REINFORCEMENT LEARNING
250 ICCET220471 AN ENSEMBLE APPROACH FOR DETECTION AND CLASSIFICATION OF FAKE NEWS
251 ICCET220602 ML-BASED OFFENSIVE TWEET ACCURACY DETECTOR ON SOCIAL MEDIA
COMPARATIVE ANALYSIS OF DIFFERENT MACHINE LEARNING TECHNIQUES FOR PREDICTING DIABETICS
252 ICCET220724 DISEASE
253 ICCET220750 AN INTELLIGENT TRAFFIC LIGHT CONTROL SYSTEM USING CONVOLUTIONAL NEURAL NETWORK
254 ICCET220770 ONLINE STUDENT CAREER GUIDE PORTAL
255 ICCET220499 AUTOMATION ON MAINTAINING OF SOIL MOISTURE LEVEL IN EARTHING PIT
256 ICCET220647 MACHINE LEARNING AND STOCK MARKET ANALYSIS
257 ICCET220752 PREDICTION AND DETECTION OF FOREST FIRE BASED ON DEEP-LEARNING APPROACH
258 ICCET220331 IMPROVISING AIR QUALITY FORECASTING USING MACHINE LEARNING
259 ICCET220621 INHIBITORY POTENTIAL OF NATURAL PLANT PRODUCTS AGAINST INFLUENZA VIRUS
260 ICCET220689 BIOMETRIC AUTHENTICATION ON FINGER VEIN RECOGNITION
261 ICCET220720 THERMAL WRAPPER USING T-CALL ESP32 SIM800L
262 ICCET220616 IOT BASED ENERGY MONITORING SYSTEM
A UNIFIED, SIMPLIFIED, AND INNOVATIVE TACTIC PREDICTION OF CUSTOMER CHURN FOR BANK
263 ICCET220765 CUSTOMERS
264 ICCET220649 BLOCKCHAIN APPLICATION IN INCENTIVIZING STUDENTS ON PARTICIPATION IN ONLINE CLASSES
265 ICCET220718 YIELD FORECASTING OF OILSEEDS USING MACHINE LEARNING TECHNIQUES
266 ICCET220886 COMMUNITY MONITORING SYSTEM FOR COVID-19 PREVENTION
267 ICCET220603 DECENTRALIZED SOCIALMEDIA APPLICATION WITH CHAT AND NFT MARKETPLACE
268 ICCET220556 NLP BASED QUESTION & ANSWERING SYSTEM
A NOVEL DATA-DRIVEN OPTIMAL METHODOLOGY FOR DETECTING SHIP FROM SAR IMAGES BASED ON
269 ICCET220839 ARTIFICIAL INTELLIGENCE
270 ICCET220318 METHODOLOGY TO IMPLEMENT HOME AUTOMATION SYSTEM USING FPGA BASYS 3 BOARD
271 ICCET220705 VIRTUAL CLASSROOM
272 ICCET220746 A SURVEY OF IMAGE PROCESSING TOOLS PACKAGE
273 ICCET220673 INTERNET OF THINGS(IOT) AND ITS APPLICATIONS: A REVIEW
274 ICCET220803 MODELING AND CONTROL OF DC-DC CONVERTERS
275 ICCET220631 AI BASED E-COMMERCE WEB APPLICATION FOR EVENT MANAGEMENT
CORRELATION BETWEEN THE DEPENDENT AND INDEPENDENT PARAMETERS IN THE EXTENDED
276 ICCET220585 HODGKIN AND HUXLEY NEURON MODEL USING PARTIAL LEAST SQUARE REGRESSION TECHNIQUE
277 ICCET220586 ONLINE ELECTRICITY BILLING SYSTEM
278 ICCET220607 MODELLING A VOTING-BASED MODEL FOR DIABETES PREDICTION USING LEARNING MODELS
STUDY OF ELECTRO-MECHANICAL PROPERTIES OF FDM MATERIAL USING DIFFERENT COATING
279 ICCET220690 COMPOSITIONS
280 ICCET220591 FOORPTINTING AND RECONNAISSANCE TOOL
281 ICCET220632 DESIGN AND ANALYSIS OF SOLAR AND WIND BASED CHARGING STATION FOR ELECTRIC VEHICLES
REMOVAL OF CHROMIUM VI IONS FROM TANNERY EFFLUENT USING EMULSION LIQUID MEMBRANE
282 ICCET220683 AND OPTIMIZATION OF PARAMETERS USING RESPONSE SURFACE METHADOLOGY
283 ICCET220810 WASTE MANAGEMENT AND WEALTH GENERATION THROUGH WASTE ELIMINATION
284 ICCET220693 IMPLEMENTATION OF ADVANCED ENCRYPTION STANDARD (AES) ON FPGA
285 ICCET220610 REGULARIZATION IN NEURAL NETWORKS
INVESTIGATION OF VARIOUS SUPERVISED MACHINE LEARNING ALGORITHM TO CHARACTERIZE
286 ICCET220794 MAMMOGRAM BREAST IMAGES ELECTRONIC AND COMMUNICATION ENGINEERING

A LITERATURE REVIEW OF BLOCKCHAIN AND GAMIFICATION BASED INCENTIVE MECHANISM IN
287 ICCET220702 EDCUACTION SECTOR
288 ICCET220592 STOCK MARKET PREDICTION USING MACHINE LEARNING
A WEIGHTED MULTIVARIABLE ADAPTIVE REGRESSION MODEL FOR DELIVERY TIME PREDICTION IN
289 ICCET220638 ECOMMERCE SUPPLY CHAIN
290 ICCET220699 A MACHINE LEARNING BASED FAULT DETECTION FOR PREDICTIVE ANALYSIS OF MACHINE
291 ICCET220545 NON-INVASIVE METHOD TO DETECT DIABETICS
SUPLYING ADJUSTABLE POWER SUPPLY TO THE BIOMEDICAL DEVICES BY USING INTERNET OF THINGS
292 ICCET220579 (IOT)
293 ICCET220731 E-COMMERCE: CONCEPT AND DESIGN
294 ICCET220558 BITCOIN PRICE PREDICTION USING MACHINE LEARNING ALGORITHMS
295 ICCET220766 A STUDY ON IMPROVING BUSINESS EXCELLENCE IN SUPERSTORES
296 ICCET220762 ANALYZING FACTORS AFFECTING YOUTUBE VIDEOS TO BECOME TRENDING VIDEO
297 ICCET220736 COVID-19 PREDICTION USING MACHINE LEARNING USING REAL WORLD DATA
COST EFFECTIVE, 3D PRINTED BINIC ARM MECHANICAL DESIGN CONTROLLED BY ELECTROMYOGRAPHIC
298 ICCET220561 SIGNALS
299 ICCET220656 FINGER VEIN DETECTION USING DEEP LEARNING
300 ICCET220749 MACHINE LEARNING APPROACH TO CLASSIFY AND PREDICT OSTEOSARCOMA GRADING
301 ICCET220767 ROAD SURFACE CLASSIFICATION THROUGH VEHICLE VIBRATION SINGAL PROCESSING
PREPARATION OF HEAT RESISTANCE AND FIRE-RETARDANT COATING USING SILICONE POLYESTER RESIN
302 ICCET220789 DOPED WITH AL2O3
303 ICCET220922 INTELLIGENT DIGITAL IMAGE WATERMARKING USING WAVELET TRANSFORM
304 ICCET220941 AI BASED ALGORITHMS FOR COMPUTER AIDED DIAGNOSIS
EXAMINING FAKE NEWS OBSERVATION AND DETECTION IN SOCIAL MEDIA: USING MACHINE LEARNING
305 ICCET220819 AND DEEP LEARNING ALGORITHMS
306 ICCET220267 ROLE OF INTELLIGENT CAD SYSTEMS FOR REAL WORD APPLICATIONS - A REVIEW
307 ICCET220664 AN APPROACH TO DETECT BRAIN MALIGNANCY USING BAT ALGORITHM
308 ICCET220903 SINGLE IMAGE DEHAZING USING MULTI-SCALE DCP-BCP FUSION
309 ICCET220798 PHOTOLUMINESCENCE OF ER-DOPED KNN CERAMICS
310 ICCET220578 ONLINE PROCTORING SYSTEM USING IMAGE PROCESSING AND MACHINE LEARNING
311 ICCET220813 WEATHER FORECASTING USING MACHINE LEARNING
312 ICCET220937 DRIVER DROWSINESS DETECTION
313 ICCET220913 AUGMENTED ANALYTICS: A REVIEW OF CURRENT TRENDS AND FUTURE CHALLENGES
314 ICCET220719 AUTOGRICULTURE – AN HI-TECH AGRICULTURAL FRAMEWORK
315 ICCET220740 DETECTION OF PHISHING WEBSITES USING MACHINE LEARNING
316 ICCET220899 COMPUTER SECURITY RISK ANALYSIS IN DISTANCE LEARNING
CUSTOMER CHURN PREDICTION - A COMPARISON BETWEEN STOCHASTIC GRADIENT DESCENT AND
317 ICCET220741 NEURO-EVOLUTION
318 ICCET220717 HITECH AGRI - CROP RECOMMENDATION SYSTEM
319 ICCET220743 BIOACTIVE PHYTOCHEMICALS: A PROMISING TREATMENT FOR CANCER
320 ICCET220911 COMPASSION:ONE MORE STEP AGRICULTURE MANAGEMENT SYSTEM
321 ICCET220581 DESIGN OF HEART RATE EQUIPMENT BASED ON IOT WHILE RIDING VEHICLE
322 ICCET220748 SEWAGE BLOCK DETECTION SYSTEM
323 ICCET220773 DATA SCIENCE AND AI/ML IN CONSTRAINING TAX EVASION
324 ICCET220962 A REVIEW AND PROPOSAL OF AN INNOVATIVE APPROACH FOR E-LEARNING PLATFORM
325 ICCET220935 CONVERSION OF SIGN LANGUAGE INTO TEXT AND SPEECH USING DEEP LEARNING
THERMAL DEGRADATION OF POLYVINYL CHLORIDE (PVC) MODIFIED BY HYDROXYAPATITE NANO
PARTICLES
326 ICCET220838
BRAIN DISEASE CLASSIFICATION & BRAIN TUMOR ESTIMATION USING CNN
327 ICCET220900

QUANTIFICATION OF MODIFIER IN POLYMER MODIFIED BITUMEN USING FOURIER TRANSFORM
INFRARED RADIATION (FTIR) SPECTROMETER AN INVESTIGATIVE STUDY TO CURB ADULTERATION IN
328 ICCET220758 MODIFIED BITUMEN
329 ICCET220745 CONTACTLESS COVID TESTING BOOTH WITH PATIENT TRACKING AND MONITORING
330 ICCET220454 CROP YIELD PREDICTION IN SMART FARMING USING MACHINE LEARNING
331 ICCET220918 APPOINT US
SENTIMENTAL CLASSIFICATION ON ONLINE CUSTOMER REVIEWS BY USING NATURAL LANGUAGE
332 ICCET220721 PROCESSING
333 ICCET220663 FLOOD INTENSITY PREDICTION VIA ENSEMBLE MACHINE CODE MODEL
334 ICCET220659 CONTROL OF FUEL CELL BASED SUPERLIFT LUO CONVERTER
PREDICTION OF STUDENT PERFORMANCE BASED ON TEXT MINING USING NEURAL NETWORKS - A
335 ICCET220888 REVIEW
A CHAOTIC SALP SWARM FEATURE SELECTION ALGORITHM FOR ORANGE AND TOMATO PLANT LEAF
336 ICCET220979 DISEASE DETECTION
337 ICCET220792 BIRD WATCHER USING DEEP LEARNING
338 ICCET221022 FACE RECOGNITION BASED ATTENDANCE MANAGEMENT SYSTEM
339 ICCET221045 CHALLENGES IN HYBRID APPLICATION DEVELOPMENT: A CASE STUDY
DESIGN AND DEVELOPMENT OF MULTI-PURPOSE 3D MODELLED PORTABLE POWER BANK USING 18650
340 ICCET220774 LITHIUM-ION BATTERY
EFFECTS OF PROCESS PARAMETERS THAT AFFECT THE MICROSTRUCTURE OF ALUMINIUM ALLOY: A
342 ICCET220957 IOT-BASED SYSTEM FOR DETECTING AND FORECASTING INDUSTRIAL FIRE
343 ICCET220799 FRICTION STIR WELDING OF ALUMINIUM ALLOY PLATES WITH THREADED PIN TOOL: A REVIEW
344 ICCET220997 FILTERING THEORY FOR QUANTUM STRING IN NOISE
345 ICCET220996 NEW MOS SQUARE-LAW TRANSCONDUCTOR
346 ICCET220726 DATA SECURITY AND PRIVACY PROTECTION FOR CLOUD STORAGE
347 ICCET220949 DRIVER'S DROWSINESS DETECTION USING DEEP LEARNING
348 ICCET220710 HEART DISEASE PREDICTION AND DETECTION SYSTEM
349 ICCET220754 BLIND IMAGE QUALITY ASSESSMENT WITH IMAGE DENOISING: A SURVEY
350 ICCET220982 HOUSEHOLD MANAGEMENT WEBSITE FOR RURAL AREAS
351 ICCET220928 SPEAKER IDENTIFICATION FROM VOICE
352 ICCET220778 BIOCATALYTIC POTENTIAL OF PROTEASE AND AMYLASE ENZYME ON WASTE TANNERY EFFLUENT
353 ICCET221058 MANAGERIAL EFFICIENCY WITH PRICE SENSITIVE DEMAND
354 ICCET220626 EFFECTIVE TAMIL HANDWRITTEN CHARACTER RECOGNITION USING DEEP LEARNING TECHNIQUE
ADAPTIVE AND INTEGRATED TEXT MINING FOR ASSESSING MASSIVE OPEN ONLINE COURSES (MOOC)
355 ICCET220782 USING VECTOR SPACE MODEL
356 ICCET220152 ATTENDANCE MONITORING SYSTEM USING OPENCV
357 ICCET220158 CLOUD NETWORK MANAGEMENT SYSTEM
358 ICCET220133 FAKE JOB FORECAST USING DATA MINING TECHNIQUES
359 ICCET220161 CYBER ATTACK DETECTION SYSTEM
360 ICCET220126 MUSIC CLASSIFICATION MANAGEMENT SYSTEM
361 ICCET220756 SECURITY LEVEL DETECTION OF VARIOUS CRYPTOSYSTEMS USING MACHINE LEARNING MODEL
362 ICCET220639 DETECTION OF PARKINSON’S DISEASE USING DEEP LEARNING ALGORITHMS
363 ICCET220742 TEA LEAF DISEASE DETECTION USING DEEP LEARNING
364 ICCET220367 REAL AND FAKE NEWS CLASSIFICATION USING NATURAL LANGUAGE PROCESSING
365 ICCET220923 A SURVEY OF CYBER CRIMES AND CYBER SECURITY
366 ICCET220946 MODELLING A NOVEL LEARNING-BASED APPROACH FOR COVID-19 PREDICTION
367 ICCET220682 IOT BASED REAL-TIME HEALTH CARE MONITORING SYSTEM FOR COMPLEX EVENT PROCESSING
368 ICCET220917 DESIGN AND IMPLEMENTATION OF INTEGRATED EVACCINATION CHIP FOR COVID-19
369 ICCET221061 CHARACTERISTIC ANALYSIS OF OFDM-OQAM SYSTEM IN BELOW DECK SHIP ENVIRONMENT
A XTREME FEATURE-BASED BOOSTING MODEL FOR HEART DISEASE PREDICTION USING FEATURE
370 ICCET220920 REPRESENTATION
371 ICCET221062 HEART RATE MONITORING , HEART ATTACK DETECTION AND PREDICTION
DEVELOPMENT OF NON-INVASIVE CONTINUOUS PATIENT GLUCOSE BOTTLE LEVEL AND HEALTH CARE
372 ICCET220992 MONITORING SYSTEM USING IOT
373 ICCET220940 DESIGN AND DEVELOPMENT OF ANESTHESIA CONTROLLER
374 ICCET220829 TEXTURE BASED CLASSIFICATION OF LIVER CANCER USING ULTRASOUND IMAGES
375 ICCET221066 A LANGUAGE BASED SYSTEM ARCHITECTURE FOR MULTIPLANAR CRYPTOSYSTEMS
376 ICCET221090 PROTECTION OF IOT USING BLOCKCHAIN: A SURVEY
377 ICCET220944 CYBERBULLYING DETECTION IN SOCIAL MEDIA USING MACHINE LEARNING
PREPARATION AND CHARACTERIZATION OF PROSOPIS JULIFLORA STEM ACTIVATED CARBON FOR
378 ICCET221109 METHYLENE BLUE DYE ADSORPTION
379 ICCET221048 FINITE ELEMENT MODELING FOR RUTTING IN FLEXIBLE PAVEMENT
380 ICCET221069 A TRIFOLD ELECTRONICS SURVEILLANCE SYSTEM USING IMAGE PROCESSING
381 ICCET221071 ANALYSIS OF PILED RING FOUNDATIONS IN MEDIUM DENSE SAND USING EXPERIMENTAL APPROACH
ROLLER COMPACTOR CUM RUT ANALYSER, AN EQUIPMENT FOR PAVEMENT TESTING AND BITUMINOUS
382 ICCET221043 MIX DESIGN
383 ICCET221018 PROSTATE CANCER DETECTION USING DEEP LEARNING & TRADITIONAL TECHNIQUES
384 ICCET220971 ROAD SIGNS CLASSIFICATION USING TRANSFER LEARNING
385 ICCET220889 DETECTION OF DDOS ATTACK USING MACHINE LEARNING
ADDITIVE MANUFACTURING ASPECTS FOR DESIGN AND DEVELOPMENT OF GARBAGE PICKING MACHINE
386 ICCET221056 COMPONENTS FOR COMPACT DESIGN
387 ICCET220828 MORPHOLOGICAL DESCRIPTORS BASED URBAN CHANGE DETECTION USING REMOTE SENSING IMAGES
NUMERICAL INVESTIGATION OF FLOW OVER AN OSCILLATING CIRCULAR CYLINDER AT LOW REYNOLDS
388 ICCET220939 NUMBER
389 ICCET221072 INVESTIGATION OF FIBRE-BASED BAG FILTER COATED WITH METAL OXIDES FOR DUST ADSORPTION
NUMERICAL ANALYSIS OF FLUID FLOW OVER PLUNGING NACA0012 AIRFOIL AT LOW REYNOLDS NUMBER
390 ICCET220952
391 ICCET221028 DESIGN OF EFFICIENT TOPOLOGY IN TIME EVOLVING AND ENERGY HARVESTING
392 ICCET220805 CRYPTOCURRENCY PREDICTION:METHODS,TECHNIQUES AND TOOLS
NUMERICAL STUDIES ON THE INFLUENCE OF SURFACE HEATING ON THE FLOW CHARACTERISTICS OF A
393 ICCET220912 CONVENTIONAL TESLA VALVE
ENHANCING OBJECT DETECTION IN AUTONOMOUS DRIVING USING YOLOV4 AND DEHAZING
394 ICCET220815 ALGORITHMS
395 ICCET220965 FACE RECOGNITION AND ATTENDANCE MARKING USING MACHINE LEARNING
A NEW OTA-BASED SHADOW FILTER PROVIDING INDEPENDENT ELECTRONIC CONTROLS OF Ω0 AND
396 ICCET221057 BANDWIDTH
397 ICCET220820 BLOOD VESSEL SEGMENTATION AND CLASSIFICATION IN FUNDUS IMAGE
DETECTION OF CYBERBULLYING IN SOCIAL MEDIA USING RECURRENT NEURAL NETWORK
398 ICCET221044 ARCHITECTURES
APPLICATION OF DEEP LEARNING CONCEPT IN DERMOSCOPIC IMAGES FOR HAIR SEGMENTATION AND
399 ICCET221044A REMOVAL
400 ICCET220921 VISUAL CUES TO VOICE CUES
401 ICCET221015 A NEW COMPUTATIONAL INTELLIGENCE APPROACH TO UNDERSTAND AUTISM SPECTRUM DISORDERS
402 ICCET220827 IOT BASED CONSERVATION AND MONITORING OF NATURAL RESOURCES IN DAY-TO-DAY LIFE
403 ICCET220894 PREVENTION OF SQL INJECTION ATTACKS IN WEB APPLICATIONS
404 ICCET220823 LUNG CANCER DETECTION USING DEEP LEARNING TECHNIQUE
405 ICCET220931 MRI SEGMENTATION OF LUNGS USING LEARNED SEMANTIC KNOWLEDGE AND GRAPH CUTS
406 ICCET221070 DESIGN OF D FLIP – FLOP USING KISAN GATE
407 ICCET221093 FISHING VESSELS MONITOIRNG SYSTEM
PERFORMANCE EVALUATION OF CNN-SVM CLASSIFIER FOR DIAGNOSING CORONARY ARTERY
408 ICCET221037 BLOCKAGE
409 ICCET221021 INSECT CLASSIFICATION USING DEEP LEARNING

ADVANCED MULTI SWARM INTELLIGENT ANT COLONY OPTIMIZATION ALGORITHM FOR VIRTUAL
410 ICCET221039 MACHINE PLACEMENT PROBLEM IN CLOUD PLATFORM
PERFORMANCE ANALYSIS OF MACHINE LEARNING AND DEEP LEARNING ALGORITHMS IN CREDIT CARD
411 ICCET221009 FRAUD DETECTION
412 ICCET221031 DESIGN AND IMPLEMENTATION OF AN AUTOMATED AGRICULTURE MONITORING AND CONTROL
413 ICCET220219A SURVEY ON AI BASED CAR PARKING SYSTEM
FRAMEWORK FOR NETWORK THROUGHPUT MAXIMIZATION TO PROLONG LIFE TIME OF WIRELESS
414 ICCET220990 SENSOR NETWORKS
415 ICCET221124 SOLAR IRRADIANCE VALUE BASED OPTIMAL OPERATING ELECTRIC ASSEMBLY DESIGN
416 ICCET221065 WEB BASED INTERNSHIP PORTAL
417 ICCET221042 A VANET BASED ACCIDENT DETECTION AND MANAGEMENT SYSTEM IN A SMART CITY
418 ICCET221118 SENTIMENT ANALYSIS OF TWITTER DATA REGARDING RUSSIA UKRAINE WAR
419 ICCET221105 IOT BASED ACCIDENT PREVENTION AND INTIMATION SYSTEM
420 ICCET221080 ANALYZING THE EFFECTS OF LOTTERY SCHEDULER
PREDICTION OF CROP YIELD PRODUCTION USING HYBRID RANDOM FOREST MODEL IN MACHINE
421 ICCET221088 LEARNING
422 ICCET221049 FOOD ALLERGY PREDICTION SYSTEM
423 ICCET221053 STOCK MARKET PREDICTION USING LSTM
424 ICCET221078 FRICTION STIR WELDING OF ALUMINIUM ALLOY PLATES USING CYLINDRICAL TOOL: A REVIEW
425 ICCET221055 DEEP LEARNING BASED DEFORESTATION CLASSIFICATION USING CNN
426 ICCET220804 METHODOLOGY TO IMPLEMENT IMAGING APPLICATIONS ON FPGA BASYS 3 BOARD
CHARACTERIZATION OF SOIL BY SCANNING ELECTRON MICROSCOPY WITH ENERGY DISPERSIVE X-RAY
427 ICCET221126 SPECTROSCOPY
428 ICCET221003 SMART CARD BASED E-HEALTH RECORD
429 ICCET221029 SMART MEDIC ASSIST IN TRAIN USING IOT
430 ICCET221148 EAR DETECTION IN UNCONSTRAINED ENVIRONMENT USING FASTER R-CNN
431 ICCET221216 IMPROVEMENT IN ENGINEERING BEHAVIOUR OF BLACK COTTON SOIL USING GGBS
432 ICCET221110 AGRICULTURE OPTIMIZATION AND LIFE ANALYSIS OF A PLANT USING IOT
433 ICCET221150 REAL TIME IMPLEMENTATION OF FACIAL RECOGNITION
434 ICCET221142 A STUDY ON IOT BOTNET DETECTION TECHNIQUES
435 ICCET221165 MOVABLE PLATFORM FOR SOLAR POWER GENERATION
436 ICCET221159 A SYSTEMATIC LITERATURE REVIEW ON DEEP LEARNING BASED MEDICAL IMAGE SEGMENTATION
437 ICCET21176 IOT BASED ASSIST DEVICE FOR MONITORING STROKE PATIENT
438 ICCET221103 INSTANT VEHICLE INSURANCE RENEWAL SYSTEM
439 ICCET221167 DATA SCIENCE APROACH TO STOCK PRICE FORECASTING IN INDIA
440 ICCET221201 IOT CONTROLLED LAWNMOWER
441 ICCET221190 SMART APPLICATION BASED PARKING SYSTEM USING INTERNET OF THINGS
442 ICCET221127 DESIGN AND DEVELOPMENT OF A NOVEL ACCIDENT DETECTION AND EMERGENCY REPORTING SYSTEM
BIO-CERAMIC COATING FOR KNEE IMPLANTS FOR BETTER INTERFACE BETWEEN BONE TISSUES AND
443 ICCET221166 TITANIUM METALS
444 ICCET221083 SMART SURVEILLANCE USING MOBILENETV2
445 ICCET221164 ANOMALY BASED INTRUSION DETECTION SYSTEM USING HYBRID MACHINE LEARNING ALGORITHMS
446 ICCET221084 IOT BASED INDUSTRIAL ADOPTATION CONTROL FOR USING WIND ENERGY CONVERSION SYSTEM
447 ICCET221149 IOT BASED ENVIRONMENT MONITORING SYSTEM
448 ICCET221225 PROTECTING CROPS FROM WILD AND DOMESTIC ANIMALS INTRUSION USING DEEP LEARNING
AUTOMATIC NUMBER PLATE RECOGNITION SYSTEM USING DEEP LEARNING ALGORITHM AND IMAGE
449 ICCET221004 PROCESSING TECHNIQUES
450 ICCET221193 SYSTEMATIC INVESTIGATION ON CLOUD ASSISTED IOT DEVICES, CHALLENGES AND APPLICATIONS
CARBON EMISSION CONSIDERATION IN A FUZZY INVENTORY MODEL UNDER HYBRID PAYMENT POLICY
451 ICCET221157 AND TRADE CREDIT FINANCING

452 ICCET221131 BUS TICKET BOOKING SYSTEM USING CLOUD COMPUTING
453 ICCET220916 PERPETUAL PRESERVATION OF PESTICIDE FREE PRODUCTS USING IOT
454 ICCET221129 YDR'S BROWSER
455 ICCET221137 MODERN DAY SMART HOME WITH E-BILLING
456 ICCET221074 ETHEREUM BASED BLOCKCHAIN FOR SECURED TRANSACTION IN WSN
DETECTION AND CLASSIFICATION OF CERVICAL CANCER BASED ON ELEPHANT HERDING OPTIMIZATION
457 ICCET221114 ON CONVOLUTIONAL NEURAL NETWORK
458 ICCET221231 STUDY ON EFFECT OF FOUNDATION SOIL ON MASS IRREGULAR BUILDINGS UNDER SEISMIC LOADS
459 ICCET220757 CUSTOMER CHURN PREDICTION USING MACHINE LEARNING
460 ICCET220841 FAKE REVIEW DETECTION SYSTEM
461 ICCET221177 SMART PORTABLE DUSTBIN WITH METHANE DETECTOR USING NODE MCU
A SMART CONTRACT BASED ACCESS CONTROL FRAMEWORK FOR CLOUD BASED SMART HEALTHCARE
462 ICCET220775 SYSTEM
463 ICCET221233 A BRIEF REVIEW OF E-NOSE FOR QUALITY ANALYSIS OF MATERIALS IN DIFFERENT CATEGORIES
SIMULATING SUBJECTIVE OPINIONS ON WOMEN’S SAFETY IN HIGHER EDUCATION CAMPUSES USING AN
464 ICCET221079 ARTIFICIAL INTELLIGENCE-DRIVEN DESIGN FRAMEWORK
465 ICCET221198 SYSTEMATIZED VLSI ARCHITECTURE FOR AES ALGORITHM
466 ICCET221161 AUTO ATTENDANCE SYSTEM FROM SINGLE IMAGE OF THE CLASSROOM USING FACE RECOGNITION
A USER EXPERIENCE STUDY ON RESPONSIVE MOBILE TOURISM WEBSITES' PERCEIVED USABILITY,
467 ICCET221133 ATTRACTIVENESS, AND INTUITIVENESS
468 ICCET221237 REAL TIME MONITORING OF CONVEYOR GUARDS USING IOT
469 ICCET221168 AGRO DIRECT SELLING APP WITH VOICE BASED CHATBOTS
470 ICCET221151 SIGN LANGUAGE INTO TEXT CONVERTER
471 ICCET220980 E-WASTE CLASSIFICATION AND SORTING USING IOT
472 ICCET221119 SOLAR POWERED POND CLEANING MACHINE
A STUDY ON METHODS AND DEVELOPMENT OF SELF LEARNING PORTAL FOR MATHEMATICS
473 ICCET221081 ENTHUSIASTS
474 ICCET221153 HUMAN-COMPUTER INTEACTION: CONTROLLING MOUSE FUNCTIONALITIES USING HAND GESTURES
475 ICCET221140 RECOGNITION OF WHEAT LEAF DISEASE USING TENSOR FLOW AND KERAS
476 ICCET221247 SMART ONLINE BOOKING SYSTEM THROUGH HAND GESTURE RECOGNITION
477 ICCET221098 ANALYSIS ON ELECTRONIC SKIN
LUNG NODULE MALIGNANCY CLASSIFICATION USING MACHINE LEARNING TECHNIQUE CNN ALGORITHM
478 ICCET221143
479 ICCET221106 DEVELOPMENT OF AUTOMATIC POSTAL PARCEL COLLECTOR WITH PINCODE BASED SEGREGATION
480 ICCET221234 SEGMENTATION AND CLASSIFICATION APPROACH TO IMPROVE BREAST CANCER SCREENING USING
HISTOPATHOLOGICAL IMAGES
481 ICCET221154 SMART SOLAR PANEL CLEANING SYSTEM USING IOT
MUSIC RECOMMENDATION SYSTEM USING SENTIMENT ANALYSIS THROUGH HAAR CASCADE IN
482 ICCET221245 MACHINE LEARNING
483 ICCET221107 FINGER VEIN RECOGNITION BASED ON DEEP NEURAL NETWORK
484 ICCET220808 EFFECTIVE UTILIZATION OF FOOD WASTE AS CONSUMABLE RESOURCES
485 ICCET221226 AN ALTERNATIVE METHOD FOR NON-INVASIVE GLUCOSE DETECTION USING FTIR SPECTROSCOPY
USABILITY STUDY ON VIRTUAL FITNESS APPLICATION USING ARTIFICIAL INTELLIGENCE AND DEEP
486 ICCET221237A LEARNING
487 ICCET221122 ANALYTICAL DEVELOPMENT OF MICRO COMBUSTION CHAMBER
488 ICCET220898 CUSTOMER SALES PREDICTION USING MACHINE LEARNING
489 ICCET221128 A SYSTEMATIC ASSISTANCE FOR FOOD WASTAGE REDUCTION USING MOBILE APPLICATION
490 ICCET221162 LOCALIZATION APPROACH USING BLE AND GSM
491 ICCET221100 A SURVEY ON CHANNEL ESTIMATION IN MMWAVE MASSIVE MIMO SYSTEMS
492 ICCET221076 PERFORMANCE ANALYSIS OF IMAGE RESTORATION FILTERS FOR TEXTUAL IMAGES
493 ICCET221172 PERKSUITY - GET CONTROL AND BENEFITS FOR YOUR DATA

494 ICCET221156 TRAFFIC SIGN RECOGNITION AND ACCIDENT PRONE ZONE INDICATION IN AUTONOMOUS CAR
495 ICCET221207 INVESTIGATION OF FAULTLESS STRESS LEVEL MONITORING SYSTEM USING ARTIFICIAL ALGORITHMS
496 ICCET221227 EYE MOTION ANALYSIS USING COMPUTATIONAL FLUID DYNAMICS
497 ICCET221210 SMART HOME CONTROL SYSTEM USING MACHINE LEARNING ALGORITHM.
498 ICCET221221 BED FALL DETECTION AND PREDICTION FOR ELDERLY PEOPLE
499 ICCET221238 DEVELOPMENT OF AN AUTOMATIC PATHWAY DIVERTER FOR THE COUNT BASED CONVEYOR BELTS
4D PRINTING USING FUSED DEPOSITION MODELLING MECHANISM FOR SHAPE MEMORY POLYMER:
USING 3D PRINTING TECHNOLOGY FOR PRE-OPERATIVE SURGICAL PLANNING AND TRAINING FOR
501 ICCET221220 OPTICAL NERVE DISEASES
502 ICCET221097 HUMAN ACTIVITY RECOGNITION USING MACHINE LEARNING
503 ICCET220793 ADVANCED HYBRID SOLID PROPELLANT COMPOSITION(S) FOR FUTURISTIC MISSILE TECHNOLOGY
504 ICCET221087 AUTOMATIC PATHWAY DIVERTER FOR CONVEYOR IN PACKING SECTION OF CEMENT FACTORY
TRANSFER LEARNING WITH VGG16 AND RESNET50 FOR LUNG MALIGNANCY CELLS CLASSIFICATION AND
505 ICCET221212 COMPARING ITS PERFORMANCE VIA EVALUATION METRICS.
506 ICCET221205 A REVIEW ON ROBOTIC TECHNOLOGY AND ITS APPLICATIONS
507 ICCET221246 ASSESSMENT OF POST-FIRE CHANGES IN FOREST COVER USING SATELLITE IMAGE PROCESSING
508 ICCET221054 AN ALGORITHM FOR CRIME DETECTION IN DIGITAL FORENSICS
509 ICCET221184 PERSON FALL DETECTION AND HEALTH MONITORING SYSTEM
510 ICCET221009A ATTEMPT TO CLASSIFY TREE SPECIES USING MACHINE LEARNING ALGORITHM WITH SPATIAL DATA
511 ICCET221197 A NOVEL METHOD FOR FAULT DETECTION AND PROTECTION IN SOLAR PHOTO VOLTAIC ARRAYS
512 ICCET221196 BIG DATA ANALYSIS FOR HEART FAILURE: DETECTION, PREVENTION AND PREDICTION
REVIEW: COMPARISON OF PARAMETER ANALYSIS OF RING OSCILLATORS FOR MEASUREMENT OF DELAY
513 ICCET221163 TIME
DIVERSE OBSERVER'S STANPONIT ON THE EXTRATERRESTRIAL INTELLIGENCE - A HYPOTHETICAL
514 ICCET220781 NOTION
515 ICCET221239 AUTOMATIC LUNG SEGMENTATION USING U-NET
516 ICCET221248 BATTERY MANAGEMENT SYSTEM IN ELECTRIC CYCLE
517 ICCET22131 A SURVEY ON EMOTIONAL DETECTION IN DEEP LEARNING
518 ICCET221243 DETECTION OF MAN IN THE MIDDLE ATTACK IN 5G IOT USING MACHINE LEARNING
DIFFERENT LEVEL MULTI MEDIA CONTENT DELIVERY SYSTEM BASED ON ADVANCE WATERMARK
519 ICCET221204 ALGORITHMS
520 ICCET221351 CREDIT CARD FRAUD DETECTION USING DYNAMIC BAYESIAN NETWORK
521 ICCET221195 SIGNATURE AND FORGERY VERIFICATION USING CONVOLUTIONAL NEURAL NETWORK
3D DENTAL PROSTHETIC TOOTH CROWN: DESIGNING AND RE-MODELLING OF CUSTOMIZED TOOTH
522 ICCET221181 CROWN USING ADDITIVE MANUFACTUREING
523 ICCET221291 MONITORING THE TEXTILE WASTE WATER TREATMENT BY USING NATURAL RESOURCES
DEEP LEARNING APPROACH FOR ECG –.BASED HEARTBEAT CLASSIFICATION FOR ARRHYTHMIA
524 ICCET221285 DETECTION
525 ICCET221352 DOCTOR PAITENT APPOINTMENT SYSTEM USING KNN
526 ICCET221263 A PREDICTIVE APPROACH ON STOCKPILE SYSTEM USING ABC ANALYSIS
527 ICCET221213 DEVELOPMENT OF EMBEDDED CUM AI BASED SMART HELMET FOR UNDERGROUND MINERS
528 ICCET221354 CATARACT DETECTION USING DEEP LEARNING
529 ICCET221217 COMPUTER BASED DIAGNOSIS AND CLASSIFICATION OF SKIN DISEASES IN DERMOSCOPIC IMAGES
530 ICCET221158 WOMAN SAFTEY APPLICATION USING GEO- FENCING
531 ICCET221241 CONTROLLING AND MONITORING OF BOILER PARAMETRS USING SCADA
532 ICCET221257 LUNG CANCER DETECTION USING MACHINE LEARNING TECHNIQUES
533 ICCET221358 PRIVACY PRESERVING HEALTHCARE ADMINISTRATION USING BLOCKCHAIN TECHNOLOGY
534 ICCET221277 DESIGN AND ANALYSIS OF COMPACT MICROSTRIP PATCH ANTENNA FOR 5G APPLICATIONS
535 ICCET221151A AUTOMATIC TEXT SUMMARIZER

536 ICCET221314 SURVEY OF COIN RECOGNITION AND DISCRIMINATORS
537 ICCET221355 A STATE OF ART TECHNOLOGIES IN SMART MASK
REAL-TIME TOMATO LEAF DISEASE CLASSIFICATION USING OPTIMIZED PRETRAINED CNN MODEL BASED
538 ICCET221287 ON TRANSFER LEARNING IN DEEP LEARNING AND PREVENTION STRATEGIES
539 ICCET221178 SMART BAG FOR WOMEN'S SAFETY USING EMBEDDED SYSTEM
540 ICCET221368 PERFORMANCE ANALYSIS OF ENHANCED LIGHTWEIGHT ALGORITHM FOR IOT DEVICES
541 ICCET221276 ECG–BASED HEARTBEAT CLASSIFICATION FOR ARRHYTHMIA DETECTION USING TRANSFORMER MODEL
ON-SITE WASTE MINIMIZATION PRACTICES AND EFFECTIVE QUALITY MANAGEMENT AS A WAY OF
542 ICCET221289 REDUCING ENVIRONMENTAL FOOTPRINT OF THE CONSTRUCTION INDUSTRY: A REVIEW
543 ICCET221269 OBSTACLE AVOIDING ROBOT USING ARDUINO
544 ICCET221284 ARTIFICIAL INTELLIGENCE BASED VOICE ASSISTANT
545 ICCET220670 PRAGMATIC ANALYSIS OF E-COMMERCE SUCCESS FACTORS AND DESIGN ATTRIBUTES
546 ICCET221362 PNEUMONIA DETECTION FROM CHEST X-RAY IMAGES BASED ON CONVOLUTIONAL NEURAL NETWORK
GRID AND RENEWABLE ENERGY SOURCE DRIVEN INDUSTRIAL MOTOR WITH POWER FLOW
547 ICCET221293 MANAGEMENT
548 ICCET221179 CLOUD COMPUTING
549 ICCET221295 ANALYSIS OF A PROJECT COST OVERRUNS AND DELAYS USING RII METHOD
FINITE ELEMENT MODELLING AND SIMULATION OF GLASS FIBRE REINFORCED INDUSTRIAL SAFETY
550 ICCET221357 SHOES USING ANSYS
551 ICCET221363 IDENTIFYING ELECTRICAL FAULT USING MACHINE LEARNING TECHNIQUE
552 ICCET221271 SKIN CANCER DIAGNOSIS USING AI COUPLED WITH IMAGE PROCESSING
553 ICCET221363A ANALYSIS OF ACCURACY RATE IN CREDIT CARD FORGERY
554 ICCET221298 AUTOMATIC CLASSIFICATION OF EVENT RELATED POTENTIAL USING NEURAL NETWORK
555 ICCET221274 STUDY AND ANALYSIS OF VARIOUS CONTROL ALGORITHMS FOR COUPLED TANK SYSTEM
556 ICCET221202 DESIGN OF CRANE BEAM AND GANTRY GIRDER
557 ICCET221199 DESIGN OF CONNECTIONS AND DEVELOPING DESIGN GUIDELINES
558 ICCET221367 ENHANCE DETECTION AND SECURITY IN ATM USING HAAR CASCADE ALGORITHM
559 ICCET221362A A SURVEY ON FLIPPING AND PICK-AND-PLACE MECHANISMS USED IN THEFIELD OF ROBOTICS
EVALUATING THE THERAPEUTIC POTENTIAL AND EFFECTIVENESS OF EXTENDED REALITY EXPOSURE
560 ICCET221251 THERAPY IN THE TREATMENT OF PHOBIAS AND EMOTION REGULATION
INVESTIGATIONS ON THE COUPLED EFFECT OF SPRAY BURNER AND FLUIDIC PARAMETERS ON THE
561 ICCET221364 ONSET OF COMBUSTION INSTABILITY IN SPRACE PROPULSION
562 ICCET221229 DESIGN OF COMPOSITE SLAB IN MULTISTORIED BUILDING
563 ICCET221306 PHOTO VOLTAIC POWER FORECASTING USING REGRESSION ANALYSIS FOR POWER SYSTEMS
564 ICCET221258 ADVANCE HYBRID SOLID PROPELLANT FOR FUTURISTIC MISSILE TECHNOLOGY
565 ICCET221174 DATA MINING AND NEW AGE ELECTIONS
566 ICCET221364A INTELLIGENT CRUTCH TOOL FOR BODY POSTURE MANAGAEMENT SYSTEM USING IOT
567 ICCET221369 PREDICTION OF HOUSE PRICE USING SML TECHNIQUE
568 ICCET221242 MULTI OUTPUT FLY BACK CONVERTER USING SOLAR ENERGY
569 ICCET221367A EOG SIGNALS - A BIOMARKER FOR NEURODEGENERATIVE DISEASES
ON THE STRUCTURE AND DYNAMICS OF LINEAR TIME VARIANT AND NON LINEAR TIME VARIANT
570 ICCET221249 FLAMES
571 ICCET221354A FACE RECOGNITION BASED ATTENDANCE MANAGEMENT SYSTEM IMAGE PROCESSING
572 ICCET221367B COMPARATIVE STUDY OF ENSEMBLE METHODS FOR CLASSIFICATION IN HEALTHCARE DATA
573 ICCET221369A OPTIMIZATION OF VISUAL AND THERMALCOMFORT IN GREEN BUILDING : A REVIEW
574 ICCET221376 ASSISTIVE SYSTEM FOR BLIND PERSON
DENSE RECURRENT NEURAL NETWORK ARCHITECTURE TOWARDS PARTICIPANT PREDICTION FOR
575 ICCET221261 EVOLVING EVENTS IN EVENT BASED SOCIAL NETWORK
576 ICCET221356 ASSESSMENT OF SOIL MOISTURE PROFILE USING SMART REMOTE SENSING AND IOT TECHNIQUES
DESIGN OF VEDIC MULTIPLIER USING QUANTUM DOT CELLULAR AUTOMATA FOR CRYPTOGRAPHIC
577 ICCET221256 APPLICATIONS
578 ICCET221379 ANALYSIS OF FORENSIC IMAGES DETECTION FOR SUPERIMPOSED OF AN IMAGE AND OTHER IMAGES
579 ICCET221375 ADVERSE DRUG REACTION ANALYSIS USING SUPPORT VECTOR MACHINE MODEL
580 ICCET221382 ARTIFICIAL INTELLIGENCE AND TELEGRAM CHAT-BOT BASED INTELLIGENT AUTHENTICATION SYSTEM
CLOCK BASED TRANSMITTER IDENTIFICATION AND ATTACK RECOGNITION USING INTRUSION
581 ICCET221385 DETECTION
582 ICCET221389 DETECTION OF PROSTATE CANCER USING J48 ALGORITHM
583 ICCET220144 CUSTOMER PERCEPTION TOWARDS ONLINE FOOD BUSNIESS APPS.
584 ICCET221389A SPEECH EMOTION RECOGNITION USING MACHINE LEARNING
585 ICCET221147 FACE MASK DETECTION
586 ICCET221282 QOS ENABLED LOAD BALANCING IN CLOUD USING GREY WOLF OPTIMIZATION ALGORITHM
1 Proceeding of International Conference on Contemporary Engineering and Technology
May 21st – 22nd, 2022
1. UWB MICROSTRIP PATCH ANTENNA FOR WIRELESS SENSOR

APPLICATIONS
K.Sugapriya
Research scholar,ECE
Sri Chandrasekharendra
Saraswathi Viswa Maha Vidyalaya
Kanchipuram, India
S.Omkumar
Associate Professor,ECE
Sri Chandrasekharendra
Saraswathi Viswa Maha
Vidyalaya
Kanchipuram, India
In the modern world antenna plays a very vital role in the field of communication, the Ultrawide
band antenna uses many types of wireless sensor applications. The proposed design consist of a
circular microstrip patch antenna with diamond shaped slot is designed using jeans substrate
material, it gives the good sensing in wireless communication at the resonance frequency of 9.5 GHz
in the UWB range 3.1GHz to 10.6GHz. The Specific Absorption Rate SAR value of an antenna in
the acceptable range, it can be calculated with the Human Phantom. Hence the proposed UWB
antenna gives the good VSWR, return loss and gain value compared with various UWB designs. the
antenna gives good candidature for wireless sensor applications.
2. MULTILAYER PERCEPTRON ANALYZED THE AUDIENCE PERCEPTION

TOWARDS CITIZENSHIP AMENDMENT ACT-CAA (INDIA)
Mr. M. Shriram1; Dr. S. Arulchelvan2
Department of Media Sciences,
Anna University, Chennai - 600025, Tamil Nadu, India.
This study provides an in-depth analysis of the public perceptions on Citizenship Amendment Act-
(CAA) portrayed in the media. In a multicultural context, the media and their activities are generally
seen as the most important and strategic tool for information flow far the government and
policymakers and their actions and activities. Although the media has a certain influence on public
policy, it is often hard to see how much it magnifies or restricts / censors messages sent to the public
regarding specific policies. The main purpose of this research is to assess the extent of media
influence on various public policies on audiences. This research not only examines the specific cases
of the CAA’s Media Coverage but also seeks to gain publics feedback on the coverage. Furthermore,
this research examines how observers perceived the way media responded and acted on the coverage
the controversial bill -CAA while it was passed in the Parliament. Here this study analyzes the
CAA's media coverage and its impact on audiences. The research further assesses how multilayer
perception is measured and thereby how the media responds to messages that are of public interest to
the audience. Although this study raises a particular issue of audience perception of media coverage
in the CAA and its aftermath which have caused widespread debate in the public sphere. But a
limited amount of research has been conducted on this subject.
ISBN 978-93-81288-22-1 Prince Shri Venkateshwara Padmavathy Engineering College

Prince Dr. K. Vasudevan college of Engineering and Technology
May 21st – 22nd, 2022
3. REVIEW OF RESEARCH ON AERODYNAMIC CHARACTERISTICS OF

SUPERCRITICAL AIRFOIL 2414 SERIES FOR TRANSONIC FLIGHT
Dr. P. Booma Devi1, Dr. D. Piriadarshani2, Ms. P. Catherine Victoria3
1Professor, Department of Aeronautical Engineering, Sathyabama University, Tamil Nadu, India
2 Professor, Department of Mathematics, Hindustan University, Tamil Nadu, India
3PhD Scholar Student, Department of Aeronautical Engineering, Sathyabama University, Tamil Nadu, India
Decades of research in shocks and boundary layer interaction of conventional airfoil has yielded
many technical solutions. The primary objective of this review paper is to focus on the early design
and development of supercritical airfoil and its influence on improving lift coefficient, drag
reduction and pressure distribution in high subsonic speed aircraft. The stall effects and buffeting
phenomena due to unsteady flow field are analyzed using the Reynolds average Navier–Stokes
equation (RANS). This paper reviews studies that optimized airfoil shape with increased lift and
reduced drag to maximize its applications in aeronautical and aerospace industries.
4. TEXTILE HORIZONTAL T-SHAPED ULTRAWIDE BAND MICROSTRIP PATCH

ANTENNA FOR WIRELESS BODY AREA NETWORK APPLICATIONS
1K.Sugapriya, 2 S. Omkumar
Research Scholar, Sri Chandrasekharendra Saraswati Viswa Maha Vidyalaya, Kanchipuram,
Tamilnadu, India1,
Asso. Prof. /ECE, Sri Chandrasekharendra Saraswati Viswa Maha Vidyalaya, Kanchipuram,
Tamilnadu, India2,
In wireless body area network WBAN technology, the most common and difficult task is continuous
health monitoring. The proposed design's major goal is to achieve low specific absorption rate SAR
value and to get excellent communication in wireless sensor environment. an antenna is a most
common communicating equipment in the digital era with its unique designs and excellent findings.
The design approach is appropriate to create a new horizontal T-shape slot in a rectangular patch
UWB microstrip patch antenna for wearable applications and the antenna was simulated using a
highfrequency structure simulator utilising a textile material, denim substrate. The designed antenna
resonates at 7.7GHz with acceptable gain of 4dB. Radiation pattern was measured using anechoic
chamber and the antenna gives the good return loss, better bandwidth, a VSWR of 1.05 also the
specific absorption rate SAR value of the UWB antenna is below 2W. The result shows that
proposed design satisfies the requirements for a textile wearable antenna used for continuous health
monitoring in wireless body area network, sports and military applications.
5. INCOMPLETE FUZZY SOFT SETS IN SANCHEZ’S METHOD OF DECISION

MAKING
K. Megala1 and K. Selvakumari2*
1Research Scholar, Department of Mathematics,
2Professor, Department of Mathematics,
1,2Vels Institute of Science, Technology and Advanced Studies, Chennai, Tamilnadu, India .

May 21st – 22nd, 2022
Incomplete fuzzy soft sets is of paramount importance in fuzzy soft sets, where the combination of
incomplete fuzzy soft set and decision-making problem is of great significance. The focus of this
paper to propose an algorithm of fuzzy soft set-based decision- making problems under incomplete
information. To obtain the best feasible solution of these problems, we have to consider various
parameters related to the solution. we apply incomplete fuzzy soft set to deal with decision-making
and propose a Sanchez’s algorithm to make it more accurate and simpler.
6. ON SOLVING FUZZY TRANSPORTATION PROBLEM USING FUZZY ONE

POINT METHOD
T. Benith Sowmya1 and K. Selvakumari2*
1,2Vels Institute of Science, Technology and Advanced Studies, Chennai, Tamilnadu, India.
In this paper a procedure based on fuzzy one point method is proposed to solve fuzzy transportation
problems where the transportation cost, supply and demand are taken s Pentagonal fuzzy numbers.
Numerical Examples are illustrated to demonstrate the proposed approach in detail.
7. CRIME DATA FORECASTING USING MACHINE LEARNING AND BIG DATA

ANALYTICS
R.Brindha1, Dr.M.Thillaikarasi2
1Research Scholar,Dept of CSE,Annamalai University,Chidambaram,Tamilnadu.
2Assistant Professor,Dept of CSE,Annamalai University,Chidambaram,Tamilnadu.
Big data analytics (BDA)” is a system based method with an aim to recognize and examine different
designs, patterns and trends under the big dataset. In this paper, BDA is used to visualize and trends
the prediction where exploratory data analysis examines the crime data.”A successive facts and
patterns have been taken in following cities of California, Washington and Florida by using
statistical analysis and visualization”. The predictive result gives the performance using Keras
Prophet Model, LSTM and neural network models followed by prophet model which are the existing
methods used to find the crime data under BDA technique. But the crime actions increases day by
day which is greater task for the people to overcome the challenging crime activities. Some ignored
the essential rate of influential aspects. To overcome these challenging problems of big data, many
studies have been developed with limited one or two features. “This paper introduces a big data
introduces to analyze the influential aspects about the crime incidents, and examine it on New York
City. The proposed structure relates the dynamic machine learning algorithms and geographical
information system (GIS) to consider the contiguous reasons of crime data. Recursive feature
elimination (RFE) is used to select the optimum characteristic data. Exploitation of gradient boost
decision tree (GBDT), logistic regression (LR), support vector machine (SVM) and artificial neural
network (ANN) are related to develop the optimum data model. Significant impact features were
then reviewed by applying GBDT and GIS”. The experimental results illustrates that GBDT along
with GIS model combination can identify the crime ranking with high performance and accuracy
compared to existing method.”

May 21st – 22nd, 2022
8. APPLICATION OF QUEUEING THEORY FOR THE IMPROVEMENT OF

RAILWAY SERVICES
J.Vasanthi1,S. Santhi2 and K. Selvakumari3
Research Scholar1 , Department of Mathematics
Assistant Professor2 , Department of Mathematics,
Professor3, Department of Mathematics,
1,2,3Vels Institute of Science, Technology and Advanced Studies, Chennai, Tamilnadu.
In most railway stations, the lines of waiting passengers are always very long. The M/M/Z/: FCFS
model is changed into M/M/1/: FCFS in this study to determine which is more efficient, a single line
or several lines. Here, we create an optimization model for queuing and then determine the best
model for queuing. The outcome of analysis was effective and realistic of these factors.
9. APPLICATION OF GENERALIZED FUZZY SOFT SETS IN DECISION MAKING

PROBLEM
V. Renuka1 S. Sandhiya2 and K. Selvakumari3
Vels Institute of Science, Technology and Advanced Studies, Chennai, Tamilnadu, India.
2Assistant Professor, Department of Mathematics,
Majumder and Samanta initiated the notion of generalized fuzzy soft set and in recent times
application of generalized fuzzy soft sets in solving decision problems is a catching momentum.
Generalized fuzzy soft set theory is a powerful tool for decision making. In our work, an attempt has
been made to apply the concept of generalized fuzzy soft set in student ranking process. Student’s
ranking provides a clear picture to students, parents, teachers and other concerned persons. It creates
a healthy competition among the students. A new technique for students ranking is introduced in this
paper based on generalized fuzzy soft matrix theory.
10. DESIGN OF NOVEL T-FLIP FLOP FOR SYNCHRONOUS COUNTER USING

QUANTUM-DOT CELLULAR AUTOMATA
Dr. G.A.E. Satish Kumar1*, Rafi Naquib2, Pranav Bhargavi3, Dr.T.S.Anandhi4,
1*Professor, Department of E.C.E, Vardhaman College of Engineering, Hyderabad, India,
2Team Lead, Plintron Mobility Solutions, Chennai, Tamil Nadu, India
3Assistant Professor, Sri Venkateswara College of Engineering & Technology, Chittoor, Andhra Pradesh
4Professor, Department of E.I.E, Annamalai University, Chidambaram, Tamil Nadu
The alternate method for the CMOS technology based electronics circuit design with significant
reduction in size is the novel nano scale technology referred as Quantum-dot Cellular Automata
(QCA). Though there are many manipulation circuits, the Arithmetic Logic Unit is a integral
component of the Central Processing Unit to perform the arithmetic and logical calculations. In this
May 21st – 22nd, 2022
work, a novel T-Flip Flop circuitry is proposed for the design of the synchronous counter with 3 bit
resolutions. The QCAD Designer tool is used to develop the novel T-FF and its utilization in the
counter design. The simulation results exhibits that the proposed QCA design performs effectively.
Keywords: Quantum-dot Cellular Automata (QCA), T-Flip Flop, Digital Counter Circuit.
11. INFLUENCE OF MANAGER’S SUPPORT IN CAREER PLANNING AND

DEVELOPMENT ON CREATING EMPLOYEES’ POSITIVE ATTITUDE TOWARDS
ORGANIZATION.
H Jyothilakshmy; Amrita Vishwa Vidyapeetham, Amritapuri, India
Dr. N Ajith Kumar, Amrita Vishwa Vidyapeetham, Kochi, India
McKinsey’s hot pursuit for talent has triggered diverse studies in talent management. Managing and
retaining talented people have become the greatest challenge for employers. Managers, seen as
superiors as well as the organization’s representative, are the closest link to human assets. As change
makers, managers who nurture a positive culture in an organization, benefit the organization in the
long-term. To bring about positive changes in the organization, we propose that managers’ positive
involvement in career planning and skills training would inculcate a positive attitude towards the
organization and enhance employee development. A systematic probe of IT professionals, using
structural equation model, shows that managers involvement in career planning and employee
development have a sustained effect on fostering positive attitude towards an organization.
12.AN APPLICATION SPECIFIC WEB-BASED DATABASE FOR INSTITUTIONAL

DATA MANAGEMENT SYSTEM
Sini Anna Alex

Computer Science and Engineering
M S Ramaiah Institute Of Technology
Bengaluru, India
Shradha Ravi
Bengaluru, India
Akash Hegde
Bengaluru, India
Anita Kanavalli
Bengaluru, India
D Sanath Kumar
Bengaluru, India

May 21st – 22nd, 2022
This Most educational institutes admit thousands of students, yearly, into various programs and
employ faculty who guide and mentor these students. Such institutions require a well-organized and
efficient solution for the management and maintenance of all the information pertaining to them.
This paper elaborates on the database system design that allows for the storage and retrieval of
information of various departments, faculty and students of the college and it also highlights the
details of the web portal created as a user interface for the database. With this system, one can enter
large amounts of data into the database, update it regularly, retrieve and download the data all
through the easy to use and understand web portal. This way all information of the college is
digitized, secure and centralized which enables ease of access to anyone in the college. The database
also has various levels of security so the users can access information only if the related permissions
are granted. Such a system can do away with the unreliable and traditional method of maintaining
information on paper while also keeping the process just as simplified so the burden on the workers
are lessened but the quality of the work remains unhindered.
13. RANKING OF PARABOLIC TRAPEZOIDAL FUZZY NUMBER USING THE

CENTROIDS AND FOCUS
A. Thiruppathi1, C. K. Kirubhashankar2, E.Janaki3
1 Associate professor, Department of Mathematics, Panimalar Institute of Technology
2 Associate professor, Department of Mathematics, Sathyabama Institute of Science and Technology
3 Assistant professor, Department of Mathematics, Panimalar Institute of Technology
Fuzzy set theory has a wide range of application in all fields. Many researchers have developed a
different type of fuzzy numbers and its membership function. The fuzzy membership function
attains the highest value only between the intervals. The fuzzy numbers are parabolic once they
obtain the highest value at the midpoint of an interval and is referred as parabolic fuzzy number.
Here a new ranking of centre of centroids and focus of the parabolic ranking trapezoidal fuzzy
numbers has been developed. So, the parabolic fuzzy number is transformed into a crisp number
using new ranking methods.
14. INSERTION OF TIE –BREAKING MECHANISM FOR FLOW SHOP

SCHEDULING PROBLEM
E.Janaki1 A. Mohamed Ismail2 A.Thiruppathi 3

1 Assistant Professor, Panimalar Institute of Technology, Chennai, India
2 Associate Professor,Sathyabama Institute of Science and Technology, Chennai, 3Associate Professor,
Panimalar Institute of Technology, Chennai, India
This paper deals with Tie –Breaking mechanism applied in NEH technique for four machines and
four job problem in a Flow Shop Scheduling. To obtain a optimal sequence in NEH technique
TieBreaking mechanism plays an vital role. Job sequential order is same for all machines in
Permutation flow-shop scheduling problem(PFSP). The possible number of sequence is (n!)m. Since
the number of machine is four, the problem can be taken as NP-complete. Numerical illustrations are
provided to understand the NEH algorithm easily
15. IOT BASED SOLDIERS MONITORING AND TRACKING IN THE WAR-FIELD

Mrs.G.Shobana, Srivarshinee S T, Tamilselvi P,

May 21st – 22nd, 2022
Department Of Electronics & Communication Engineering,

Mepco Schlenk Engineering College, Sivakasi – 626005, Tamil Nadu, India.
Soldiers are the backbone of our Nation’s protection. They guard us despite any worse situations.
But, nowadays the life of the ones who are guarding us day and night is at stake. One of the crucial
challenges in the military operations is that the soldiers are not able to communicate with the control
station. Herein, a soldiers monitoring and tracking system based on Internet of Things (IoT) is
proposed to monitor and track the soldiers in the war-field. In this proposed system design, PIC
controller (16F877A) is used for sensor interfacing and for communicating with the server unit
using IoT. The main aim of this system is to track the health status like body temperature, heart and
blood oxygen rate of the soldiers and also to monitor the environmental parameters like air quality
of the war-field using wireless sensor network. The proposed system also comprises of a LCD which
displays the normal/abnormal results of the sensors and a keypad which is programmed to send
instant commands like mentioning the necessity of man power and weapons or rescue them in case
of emergency in the war-field. It sends GPS location of the soldiers using IoT Module from soldier
unit to the server unit when there is a variation in the sensors result. The GPS location and the
instant commands are received and displayed in the PC in the server unit. With the utilization of this
proposed hardware, it is possible to execute a system that ensures the safety of soldiers on the war
zone.
16. ANN BASED IMAGE SEGMENTATION FOR PHOTOVOLTAIC HOT-SPOT

DETECTION
S.Gayathri Monicka
Professor, Department of EEE
SRM Institute of Science and Technology
Chennai, India
D.Manimegalai
Research Scholar, Department of EEE
Chennai, India
M.Karthikeyan
Assistant Professor, Department of EEE
Vel Tech Multi Tech Dr.Rangarajan Dr.Sakunthala Engineering College
Chennai, India
Photovoltaic systems have become more popular as people become more interested in developing
energy from renewable resources. Even after the installations, however, there is still a lack of
understanding about the importance of inspecting the condition of the PV modules. To keep the PV
running, early hot-spot detection is required. For detecting hot-spots, thermal imaging is still a
popular technique. This research proposes to develop a method for detecting hot-spots in thermal
images of photovoltaic modules using artificial intelligence techniques. Pre-processing,
segmentation with an Artificial Neural Network and identification are the three main processes. The
proposed method appears to be a good choice for improving hot-spot detection in PV monitoring
systems, according to the results of the experiments.
17. A DISTINCTIVE APPROACH FOR EARLY DETECTION OF PLANT DISEASE

FOR SUSTAINABLE AGRICULTURE

May 21st – 22nd, 2022
Sivasangari Ayyappan 1, Gnanasekaran Sasikumar 2

1 Department of Electronics and Communication Engineering, GMR Institute of Technology,
Srikakulam-532127, Andhra Pradesh, India.
2 Department of Mechanical Engineering, GMR Institute of Technology, Srikakulam-532127,
Andhra Pradesh, India
Agriculture has always been the primary resource in producing food. Plant disease has been a major
factor influencing the food production. If there are low yields and less stock of food, and a disease
epidemic occurs then food shortage can easily develop resulting in worst effects on human society.
So, it is essential to deal with the identification of the plant diseases as early as possible to avoid low
crop yield because of diseases. Because of the rapid devel-opment of the Smart Farming, the
identification of the plant disease becomes data driven enabling smart analyses. In this context, Deep
Learning algorithms is used for disease identifi-cation. So, a mathematical model of plant disease
classification and its findings are proposed. The proposed work consist of two layers namely
Preprocessing Layer and Fully Connected Layer (FC). The Pre-processing Layer is for Feature Map
extraction and the Fully Connected Layer is Neural Network. The model is trained using larger
dataset of diseased leaves and healthy leaves and it is tested against random samples. The proposed
model obtained 97% of accuracy in classification and recognition of various plants.
18. A NEW MODEL BASED DESIGN AND PERFORMANCE ANALYSIS OF HIGH

THROUGHPUT AND LOW LATENCY DA-FIR FILTER FOR EFFECTIVE NOISE
REMOVAL IN SDR APPLICATIONS
S. Hari Krishnan, PhD Scholar/ECE, Saveetha School Of Engineering, Chennai, Tamilnadu, India.
Dr.K.Vidhya, Professor/ECE, Saveetha School Of Engineering, Chennai,Tamilnadu,India.
A software-defined radio (SDR) allows for digital communication systems to easily adopt more
sophisticated coding and modulation technologies, which is extremely important in meeting the
everincreasing demands of the wireless communication industry. In this paper, we have analyzed the
Design of Efficiency in FIR filter using faster multiplier and faster adder and verification on Artix-7
FPGA. Design of channel equalizer which is part of the SDR application is applied to FIR for
validation to minimize the BER and Latency, and to improve the throughput. Here, through
decimation factor, frequency response for filter output is changed dynamically without changing
filter coefficients. The High configurable parallel prefix adder can be used to narrow down the worst
case critical path delay during partial product accumulation. The number of LUT’s can be reduced
here based on FIR. The number of coefficients that can access parallel for performing FIR filter
operations can be reduced to reduce delay and memory.
19. TEACHING AID BASED ON CONTEXT ACTIVE CULTURAL BASED STUDY

FOR LEARNING COMPILER DESIGN CONCEPT FOR STUDENTS
Sini Anna Alex
Bengaluru, India
Anita Kanavalli
May 21st – 22nd, 2022

Bengaluru, India
Parkavi A
Bengaluru, India
Context awareness (CA) alludes to frameworks can both sense and respond based on their
environment. We present a complete description of a project for a compiler called Compiler
Evaluator Module (CEM) that generates intermediate representations for each of the phases in the
design of a compiler. A Table driven parsing model for Lexical and Syntax Analysis phase. An
annotated parse tree for a given Syntax Directed Definition using suitable data structures and
algorithms. The syntaxdirected interpretation procedures will be connected to sort checking and
intermediate-code era. The most application is the development of sentence structure trees. Since a
few compilers utilize language structure trees as a middle representation, a common frame of Syntax
Directed Definition turns its input string into a tree. To complete the translation to intermediate
code, the compiler may then make the syntax tree, using another set of rules that are in effect a
Syntax Directed Definition on the syntax tree rather than the parse tree. All these translations are
designed through game based learning [3] for the students with Learning Disabilities like slow
learners, hyper active students who can’t concentrate.
20.ANALYSIS OF KNN, LINEAR, RIDGE, LASSO, POLYNOMIAL REGRESSION

AND SVM ALGORITHMS FOR MAXIMUM ACCURACY TO PREDICTING THE
PRICE OF BITCOIN
Dr.P.A.Selvaraj 1
Associate Professor
Department of Computer Applications
Kongu Engineering College
Perundurai, Erode,
Tamilnadu, India.
Dr. M.Jagadeesan2
Associate Professor
Perundurai, Erode,
Tamilnadu, India.
Mr.K.Jayasurya 3
PG Scholar
Perundurai, Erode,
Tamilnadu, India.
Ms.K.Karthika 4
PG Scholar
Perundurai, Erode,
Tamilnadu, India,
May 21st – 22nd, 2022
Mr.K.P.Rajkumar 5
PG Scholar
Perundurai, Erode,
Tamilnadu, India.
Bitcoin has become the most valuable crypto currency in recent years. Bitcoin values, on the other
hand, have been extremely volatile, making them extremely difficult to forecast. As a result, the goal
of this study is to use machine learning linear regression models to train the dataset and then use
Grid Search to find the best scaling parameter and discover the most efficient and accurate model to
predict Bitcoin prices from various machine learning algorithms such as KNN regression, linear
regression, Ridge, Lasso, polynomial regression, and SVM, both simple and with kernels, for
predicting the accuracy of 94 percent of the bitcoin market price.
21. CHARMING SEWERAGE SPECIALIST CARE ARRANGEMENT USING

EMBEDDED SYSTEM
ANITHA MARY.M1, BHARATHY R2, JHANANI SHREE U3, SURENDAR R4

1, 2, 3, 4Assistant Professor (SS)
Department of Electronics and Communication Engineering,
Rajalakshmi Engineering College, Chennai, India
Sewerage system plays an important role to keep the city clean. The traditional Drainage Monitoring
System failed to acknowledge in the field of alerting the people about the gas explosion, increase in
the water level and the opened lid. This system directly impacts on health issues of citizens and
worker who cleans the underground drainage. The proposed system is low cost, less maintenance,
long life and web based real time system, which update the municipal officer by text communication
when any manhole crosses the threshold value. Thus, the proposed system predicting the dangerous
situations in drainage system.
22. HUMAN BRAIN-MOTIVATED MODEL FOR DISTINGUISHING EMOTIONAL

STATES FOR AUTISM DISORDER
Shushma GB1, I Jeena Jacob2
1PhD Scholar, Department of Computer Science, GITAM University, Bengaluru, India
2Department of Computer Science, GITAM University, Bengaluru, India
The machine learning has immense potential to amplify diagnostic and interference research in the
behavioral sciences. It may be instrumental in research involving the highly frequent and varying
syndrome of autism spectrum disorder. However, ML chatbots in the absence of clinical domain
specialists or physicians are questionable and lead to fabricated closures. We conclude with
proposed better practices while using ML in researches related to autism and highlight some up-and-
coming areas for collective work at the intersection of behavioral and computational sciences. As in
the recent years show the growing popularity of chatbots in the absence of the physician in health
care industries, with latest efforts aiming to make them more empathic and humanlike, finding

May 21st – 22nd, 2022
applications, for example, in health care service or in treating mental illnesses, this proposed chatbot
is a human brain motivated emotional Intelligence.
23. SMART DELIVERY BOX BASED ON TENSORFLOW OBJECT DETECTION

AND RECOGNITION API
Dr.B.Muthusenthil, Dr.V.Dhanakoti , Dr.M.Mayuranathan, T.S.Ramachandran
Associate Professor, CSE
SRM Valliammai Engineering College
Chennai, Tamil Nadu, India
The Object Detection Smart Delivery box operates on raspberry pi with two cameras, one camera
that takes a picture of the delivery person and sends it to the user so that the user can remotely open
the delivery box with the help of an android application. Another camera which is placed inside the
box scans the package, recognizes it with help of “TensorFlow object detection API” and mails the
picture of the package to the user. Receiver module displays the name of the package and tells what
package is delivered. The user can also use Google assistant voice command to open or close the
Object Detection Smart Delivery box.
24. A NOVEL ENSEMBLED DEEP LEARNING MODEL FOR BIOMEDICAL

QUESTION CLASSIFICATION
Tanu Gupta, Ela Kumar
Indira Gandhi Delhi Technical University for Women
The medical question classification is one of the essential steps in building an outstanding ques-tion
answering system for the medical domain. The accurate question classification plays a crucial role in
selecting appropriate documents for answering questions. Due to the competitive nature of deep
learning models in mining deep hidden features, they are popular in accomplishing various NLP
tasks. In this study, we examine the importance of the Temporal CNN (TCN) model in extracting
deep insight features from biomedical questions. A novel deep learning model, Bi-GRU-TCN is
proposed and implemented it for classifying medical question datasets, that blend together the
benefits of BiGRU and TCN. This model not only ex-tract contextual features from the Bi-GRU
model, but also learn spatial features using TCN layers. Comparing our work with the finest deep
learning models with series of experiments, the results show that the proposed model attains an
outstanding performance on biomedical question classification datasets.
25. FORECASTING OF CROP YIELD IN VARIOUS CLIMATIC ZONES OF

KARNATAKA USING PREDICTIVE ANALYTICS
Parkavi A1, Pushpalatha M N2, Sini Anna Alex3, Manasa Manjunatha4

1,3Department of Computer Science and Engineering, M S Ramaiah Institute of Technology, Bengaluru
560054
2 Department of Information Science and Engineering, M S Ramaiah Institute of Technology, Bengaluru -
560054
May 21st – 22nd, 2022
4Department of Information Science and Engineering, National Institute of Engineering, Mysuru-570008
Agriculture is the most widely practiced occupation in India. It contributes its major share in the
country’s GDP. Most of the times, agriculturists in India do not get the desired crop yield. This is
due to various reasons like government policies, international market policies, limitations of
development and usages of technologies. These reasons influence the results of lesser economic
value of the crop. The advancement of several engineering techniques provide solutions to combat
this problem of obtaining lesser yield. Data analytics is one among such techniques, which is being
used as a part of meticulousness in agriculture; Data analytics consists of powerful set of machine
learning algorithms. It helps in deriving useful information out of the huge data. These enormous
data are collected which aids in making smart decisions. We can use data analytics in the field of
Arecanut crop yield prediction by analyzing the effect of distinct environmental parameters. Some of
these parameters are like rainfall, temperature, humidity, precipitation, soil quality, etc. During
various stages of crop growth, these analysis provides insight on the crop yield. Such insights
obtained can provide suggestions to the agricultural community. This makes them to think
scientifically in the direction of crop cultivation. The current study uses several data analytical
techniques in analyzing the Arecanut crop yield in various climatic zones of Karnataka state in India.
In this study the effect of rainfall, temperature, humidity and soil type parameters are used to predict
the Arecanut crop yield by using Multiple Linear Regression and Random Forest. The prediction
was followed by classification in order to analyze the yield patterns in different climatic zones. The
districts of Karnataka are classified into different climatic zones by using Random Forest and Fuzzy
Logic classification algorithms. In this study, Random Forest gives good accuracy compared to other
techniques for predicting Arecanut crop yield as well as classification of districts into different
climatic zones.
26. UN-STRUCTURED MEDICAL DATA DETECTION FOR AUTISM SPECTRUM

DISORDER
Shushma GB1, I Jeena Jacob2
1PhD Scholar, Department of Computer Science, GITAM University, Bengaluru, India
2Department of Computer Science, GITAM University, Bengaluru, India
The Autism spectrum disorder (ASD) is a neural developmental disorder that impairs patients'
capability to perform regular activities like social interaction and communication significantly.
Moreover, the procedure of diagnosing ASD is highly time-consuming, labor-intensive, and
demands extensive expertise. Even though there is no such known cure for ASD, physicians have
consent regarding the importance of early intervention for the diagnosis and recovery of ASD
patients. Therefore, for the benefit of autism patients by improving their access to treatments such as
early intervention, we aim to develop a robust machine learning-based system for autism detection
using Natural Language Processing techniques based on information extracted from medical forms
of potential ASD patients. Our detecting framework includes converting semi-structured and
unstructured medical records data into digital formats representation, preprocessing, learning
document representation, and finally performing appropriate clustering so that it could help the
enduser (Physicians)

May 21st – 22nd, 2022
27. SIGN LANGUAGE CONVERTER USING FEATURE EXTRACTOR AND

POSENET
Pushpalatha M N1, Parkavi A2, Sachin RS3, Amy S Vadakkan4, Aakifha Khateeb5, Deeksha KP6
1 Assistant Professor, Ramaiah Institute of Technology, Department of Information Science and Engineering,
Bangalore, Karnataka
Associate Professor, Ramaiah Institute of Technology, Department of Computer Science and Engineering,
Bangalore, Karnataka
2Student, Ramaiah Institute of Technology, Department of Information Science and Engineering, Bangalore,
Karnataka
Karnataka
Karnataka
KarnatakaAbstract
Human beings commune with each other to convey thoughts, expressions, experiences and so on to
the people around them. But this case is not the same when it comes to the deaf and mute people.
Sign language paves the way for people with such difficulties to communicate with one another. The
aim of this paper is to facilitate easy and accurate communication between people who have hearing
and speaking disabilities, and those who do not. This paper shows how the communication gap can
be bridged using sign language to text and audio converter with the help of Feature Extractor and
Posenet with an accuracy of 92%. A webcam is used to capture the sign language shown by a
person. Posenet with Artificial Neural Network is used to classify essential words used in day to day
life. Various parts of the body are tracked by the webcam and then converted to text and audio to
convey what the person is trying to express in real time.
28. SMART AND SUSTAINABLE HORTICULTURE USING INTERNET OF THINGS

AND ARTIFICIAL INTELLIGENCE: A QUALITATIVE REVIEW
Prof. Subramanya S. G1, Dr. Parkavi A2
Research Scholar, Visvesvaraya Technological University, Belagavi and Assistant Professor, Dept. of C.S.E1
Nagarjuna College of Engineering & Technology (Affiliated to VTU, Belagavi), Bengaluru, Karnataka,
India1
Associate Professor, Dept. of C.S.E, M S Ramaiah Institute of Technology, Bengaluru, Karnataka, India2
Horticulture is a branch of agriculture which deals particularly about growing fruits, nuts,
vegetables, flowers and ornamental plants. It is one of the fastest growing economic sectors and 30%
of Indian Gross Domestic Product (GDP) accounts from horticulture alone making India a second
largest horticulture producer in the world. However, Horticulture crops face many challenges like
effective utilization of land, shortage of labour, shortage of water, low soil fertility, early detection
and treatment of plant disease, pest control, crop monitoring, timely harvesting and yield prediction.
In this paper, we present a qualitative survey to address before mentioned challenges in agriculture
as well as horticulture by using technologies like IoT and AI. We discuss about the application of
IoT at various stages of farming starting from irrigation to crop harvesting and also, about the
application of AI techniques for disease detection, yield prediction etc.
May 21st – 22nd, 2022
29. DENSITY BASED SUPPORT VECTOR MACHINE WITH OUTLIER FILTERS

FOR REAL TIME ANALYSIS OF ROCKS AND STRUCTURES
N. LEO BRIGHT TENNISSON,
Assistant Professor [Sr.G], Department of Computer Science and Engineering,
SRM Valliammai Engineering College, Kattankulathur, Chengalpattu District-603203,
Artificial Intelligence provide effective solution to many real time problems and challenges in the
world. Real Time Analysis of rocks and structures is one of such challenges. Many situations require
gaining the knowledge of rocks or building or iceberg structures. Examples where such analysis
required are mining, rescue operations, sailing of ships, etc…. Structure of rocks / debris / icebergs
are got through advanced imaging system and sensors. This paper introduces application of
Artificial Intelligence algorithm for analyzing and predicting the structure of earth or rock.
30. DESIGN OF 243 LEVEL TRINARY LADDER MULTILEVEL INVERTER USING

VHDL
R.Nithyaa*, Research Scholar, Department of Electronics and Instrumentation Engineering, Annamalai
University, Chidambaram, Tamil Nadu, India.
Dr. T. Anithab, Associate Professor, Department of Electronics and Instrumentation Engineering, Annamalai
University, Chidambaram, Tamil Nadu, India
Dr.M.Arulaalan c,Professor, Department of Electronics and Communication Engineering, C K College of
Engineering & Technology, Cuddalore, Tamil Nadu, India
In this paper, a 243 level Trinary Switched Ladder Multi-level Inverter based on a non-carrier
switching angles algorithm is proposed. The switching angle algorithms generate the triggering
angles for the Digital Pulse Width Modulation signals. The advantage of the non-carries switching
angle algorithm is the precision of the switching signals. The Trinary Switched Ladder Multi-Level
Inverter is a topology that utilizes two wings of DC sources with switch controls for the generation
of the Multi-Level Inverter levels. The 243 –level Trinary Switched Multi-Level Inverter requires 10
switches that split 5 each between the positive wing and negative wing respectively. The 10
switching patterns are developed using the VHDL code to control the proposed switched ladder
Inverter. The validation of the proposed method is attained by cross-compiling the developed VHDL
code in the MATLAB SIMULINK tool. The parametric analysis namely %THD, Vpeak, Vrms are
manipulated for the proposed 243-level Trinary Switched Ladder Multi-Level Inverter.
31. A TRANSFER LEARNING BASED DCNN FRAMEWORK FOR BRAIN TUMOUR

MULTI-CLASSIFICATION
R. Tamilarasi
Research Scholar
Department of Computer Science,
University of Madras, Chennai-600025
Dr.S. Gopinathan
University of Madras, Chennai-600025.

May 21st – 22nd, 2022
One of the most frequent diagnoses in neuroradiology is brain tumours. The brain tumours are the
most frequent and most aggressive diseases and their greatest quality of life is quite short. Thus, the
planning of treatments is a vital step in improving patients' quality of life. Improved technology and
machine learning can enhance tumour diagnosis by radiologists without intrusive methods.
Nowadays, Convolutional Neural Networks (CNN) is the most often used algorithm for visual
learning and image recognition. This research seeks to multi-classify brain tumours for early
detection by using CNN. Two distinct CNN models are presented for two classification tasks. The
first model of CNN is used to identify brain tumours with 97.60% accuracy. The other model can
classify the brain tumour with an accuracy of 98% into four tumour kinds such as glioma, normal,
meningioma and pituitary. In addition, the performance of our CNN model was evaluated using 5-
fold cross validation technique and compared with other state-of-the-art models of CNN.
32. PRE-SERVICE EDUCATION: A PRE-REQUISITE FOR THE TEACHERS OF NEW

GENERATION ESL LEARNERS
Dr Jayanthi Rajendran
Professor - Language Training Centre
Saveetha University
Teachers must be equipped with the skills to integrate technology seamlessly into their teaching in
ways that move beyond mere presentation and communication. It is very important that concrete
preparation program for teachers with appropriate technologies and resources would generate an
effective and creative student engagement and learning. The key challenges and solutions to the
effective implementation and integration of technology in teacher preparation program would widen
the arena of opportunities for students learning. Unless and until teachers implement or impart a new
learning methodology towards the competencies of technology education in colleges and
universities, active learning cannot be achieved. Pre-service teaching is an vital acquaintance in the
professional education program. Although college courses, activities, and practices contribute
knowledge and experience to the prospective teacher, pre-service teaching provides the opportunity
to experience the challenging and gratifying task of assuming major teaching responsibilities. This
paper will focus on how pre-service education would enable the teachers to equip themselves to
teach the new generation ESL learners with ideas and technologies that has become the part of our
daily life. Digital pedagogy permits teachers to develop a plan to successfully amalgamate
technology into their curriculum of teaching that would enable learning through formation,
innovation and problem-solving. pre-service educators can create expert teachers, they plan and
prepare programs that would be helpful to assimilate a combination of skills and knowledge
mentioned to as (Technological Pedagogical Content Knowledge). Pre-service teachers with no
doubt be able to subsume a firm comprehension and knowledge of the content matter, a deep
understanding of how students learn, and a practice facility with technology for learning. The
preservice teaching experience is designed to provide a controlled learning situation in which the
prospective teacher can put into practice the principles and methods learned in the pre-service
teacher education Program. The eventual goal of the pre-service teacher is to achieve competency
and proficiency in the entry level skills in the teaching profession.
“There’s no doubt technology can enhance education, but only it is in the hands of well-prepared
teachers.” (AT&T, October 31, 1995)

May 21st – 22nd, 2022
33. PREDICTION OF REAL TIME SOUND USING MACHINE LEARNING FOR THE
HEARING IMPAIRED PEOPLE
S.Veena1, D.John Aravindhar2

1SRM Institute of Science and Technology, Ramapuram, India
2Hindustan Institute of Technology and Science, Padur, India
This paper focuses on aiding the hearing impaired people for their development in the society. Most
of the hearing impaired is caused due to the inner ear or nerve damage. Only a few cases can be
resolved using the Hearing Aid. Hence, we have designed a system that could work for all the sound
impaired people regardless of the case they hold. The main reason behind this is that sound plays a
very important role in everyone’s life. It tells us about the environment, the characteristics of it,
about people, place and move in a way where visuals can’t. To make sure that the Hearing Impaired
people are not isolated with the rest of the world we have proposed to design an IoT based Machine
Learning Sound Classification System that captures sound in Real-Time, processes it and then
classifies it using a trained machine. The machine is trained using the Feature Extraction Techniques
and the final result is given as a push notification to the user. Observations made from the results
state that this proposed system can be used in a deaf school or in a place where there are a group of
Hearing Impaired people so that they are also connected to the happenings of the environment.
34. PREDICTION OF RESPIRATORY DISEASES FROM AUDIO DATA

CAPTURED THROUGH DIGITAL STETHOSCOPE AND SEVERITY ANALYSIS OF
DISEASES.
Soumyadeep Ganguly
University of Calcutta
Kolkata, India
Diptakshi Sen
University of Calcutta
Kolkata, India
Anil Kumar Gupta
HPC E &I group, C-DAC
C-DAC, Pune
Pune, India
Respiratory diseases are diseases that affect the lungs and other parts of the respiratory system.
They can be severe and can be deadly. In this paper, we have used deep learning models to predict
the disease and analyse the severity of diseases.Thus giving faster detection of disease and its
intensity. The detection of the respiratory diseases was performed on the electronically recorded
lung sounds using a digital stethoscope. The data consist of the respiratory database of 126 patients.
The dataset contains clean respiratory sounds as well as noisy data. The dataset also has 920
annotated recordings of varying lengths of 10 sec to 90 sec, 6898 respiratory cycles where 1864
crackles,886 wheezes, 506 containing both wheezes and crackles. Several preprocessing steps were
introduced to achieve smoother and less noisy signals. First, the data has been filtered of outliers,
and then feature is extracted from the audio clips namely MFCC, chroma, and mel spectrogram
May 21st – 22nd, 2022
vector. The deep learning network architecture consists of a Convolutional neural network and
recurrent neural network models. Evaluation of the model was done on the accuracy of the train and
test. The developed algorithm has achieved the highest average accuracy of 94% on tests. This study
helps in the detection of diseases in clinical settings to assist doctors or health workers in decision-
making related to the recognition of respiratory diseases.
35. A Comparative Study on Self-Sovereign Identity powered by the

Blockchain Technology
Sangeetha V1, Shilpa H2, Darshana A Naik3, Shashank M G4
1,2,3,4 Department of Computer Science and Engineering,
M S Ramaiah Institute of Technology, Bangalore, India
The current state of Digital Identity Systems is fractured among service providers. Users must
duplicate their identity information across services, which reduces overall accessibility and increases
the likelihood of privacy breaches. Users have no knowledge of how their data is being misused by
providers and they have little real influence of it. The concept of Self Sovereign Identity (SSI) has
emerged, promising to usher in a new era in which the individual, and only the individual, has
complete autonomy over their identity records, with clear support for a user - controlled data storage
facility. With the introduction of Blockchain technology, the concept of Self – Sovereign Identity
has gained traction, and it is expected to have a significant impact on how internet users
communicate in the future. This paper primarily focuses on several existing Identity Management
Systems and shows how Self Sovereign Identity is better in many aspects when compared to them.
36. CONDUCTION OF ONLINE EXAMINATIONS : A SURVEY

Evangeline D and Parkavi A
M S Ramaiah Institute of Technology, Bangalore,India
Owing to the Covid-19 pandemic, online examinations have been widely conducted across the globe
by educational institutions. Such examinations are proctored manually in many cases. But,
eproctoring tools are also available. Such tools employ Artificial Intelligence, soft biometric traits,
emotion analysis, etc., for monitoring the candidates. There are certain limitations with the use of
these e-proctoring tools. These loopholes permit the candidates taking up the examination to indulge
in impersonation or malpractice of any kind. There are several aspects to be considered during
online exam conduction including authentication, security and evaluation. In this paper, a survey is
conducted on online examination conduction techniques. A comparative study on the advantages,
limitations and scope of such techniques has been discussed in detail.
37. DESIGN OF FUSION DIGITAL PULSE WIDTH MODULATION IN QUANTUM

DOT CELLULAR AUTOMATA
P. Jegadeeshwaria*, Research Scholar, Department of Electronics and Instrumentation Engineering,
Annamalai
University, Chidambaram, Tamil Nadu, India

May 21st – 22nd, 2022
Dr. T. Anithab, Associate Professor, Department of Electronics and Instrumentation Engineering, Annamalai
University, Chidambaram, Tamil Nadu, India.
Dr. M.Arulaalanc, Professor, Department of Electronics and Communication Engineering, CK College of
Engineering and Technology, Cuddalore, Tamil Nadu, India
The Quantum Dot Cellular Automata (QCA) is a novel technique in Nano-Electronics based digital
design. The QCA is the best alternate for the CMOS design for its less area occupancy, high
performance and low power consumption. This brief presents the Quantum dot Cellular Automata
design for the fusion Digital Pulse Width Modulation. The proposed fusion Digital Pulse Width
Modulation includes the counter and comparator circuits for the Counter based Digital Pulse Width
Modulation, the ring counter and multiplexer circuits for the Delayline based Digital Pulse Width
Modulation and the combination of CDPWM and DDWPM for the Hybrid based Digital Pulse
Width Modulation. In this work, the resolution is fixed at 23 bits for CDPWM, 22 bits for the
DDPWM and 25 for HDPWM. The QCA based fusion Digital Pulse Width Modulation is designed
and evaluated using the QCADesigner Tool. The parametric analysis of the proposed QCA based
Fusion DPWM is compared with the CMOS based design.
38. EMOTIONAL INTELLIGENCE AS A REINFORCER TO ENHANCE

PSYCHOLOGICAL WELLBEING IN THE COVID 19 PANDEMIC.
Dr Rethy B Menon, Vidya Avadhani
Department of Management and Commerce,
Amrita School of Arts and Sciences, Mysuru,
Amrita Vishwa Vidyapeetham, India
COVID-19 pandemic is a severe health disaster affecting multiple countries. Widespread breakouts
have been linked to negative psychological health repercussions. COVID has brought many changes
in the way human beings think and act. Most of the people in the country suffer due to physical
health and mental health issues. It has affected all the age groups and all the sectors. Schools and
colleges have been affected and led to shutting down. Human resources those who are working in
the service sector facing challenges. Due to the mode of work where face-to-face interaction is
needed. Employee-customer contact is the most critical aspect for success in the service industry.
Organizations in the service sector exert ongoing pressure on their staff to perform since front-line
employees are frequently the first point of contact for customers; thus, they must be active,
thorough, analytical, and pleasant. Service organizations like Insurance companies play a significant
role in building up a nation. It contributes in terms of creating huge employment opportunities. This
sector has undergone a lot of changes after globalization. Sales department employees go through a
lot of ups and downs due to heavy targets on them. These Employees in the Insurance sector are
under extreme time pressure to complete tasks. As a result, employees are under much stress and
have many emotional problems. Performance is influenced directly or indirectly. Employees are
expected to deliver exceptional results. Employees must be notable performers to provide these
outstanding results in their job more than usual, resulting in excellent productivity. So, Investigator
has attempted to find out the level of emotional intelligence among the sales department employees
of the Insurance sector and offer suggestions to enhance EI among employees. Data was collected
from 612 employees of the private insurance sector through a structured questionnaire. Data was
analyzed using descriptive statistics-test, ANOVA, and cluster analysis.

May 21st – 22nd, 2022
39. INFRA SOFT Α-OPEN (CLOSED) SETS ON INFRA SOFT TOPOLOGICAL

SPACE
M John Peter1,2,a and R Manoharan3,b

1Assistant Professor, Panimalar Institute of Technology, Chennai-600123, India.
2Research Scholar, Sathyabama Institute of Science and Technology, Chennai-600119, India.
3Associate Professor, Sathyabama Institute of Science and Technology, Chennai-600119, India.
In this article, we define a new soft open (closed) set called infra soft α-open (closed) set and
establish some fundamental properties of this soft open (closed) set. The relation between infra soft
αopen (closed) set and other topological sets are studied. The findings mentioned in this study are
preliminary and serve as an introduction to more advanced research in theoretical and practical
areas. AMS Subject Classifications: 54A40, 54C08, 54C10
40. PREDICTION ON FIELD CROPS YIELD BASED ON TIME SERIES ANALYSIS

DEEP LEARNING MODEL
*S.Iniyan1, R.Jebakumar2, M.Gayathri3
1,2,3
SRM Institute of Science and Technology , Kattankuathur, Chennai 603203.
The agriculture plays a very crucial role within the overall development of the country’s economy.
Climate change, irregular rainfall, change in the nutritional content of the soil and some other
environmental changes have been considered as a serious issue within the agriculture. Using deep
learning models which takes on the various parameters can be considered as an important approach
for achieving an accurate and effective solutions for this drawback. Prediction of the crop yield can
be done by using the yield data of the crop which can be obtain from a historical source which
should have the information like weather, soil nutrient content, soil type, season in which the crop
was grown and its yield. A large dataset containing various parameters would be required in order to
train the model and attain a good accuracy. This paper targets on predicting the yield of a particular
crop which will be supported by the present information using LSTM time series analysis. The
information that is used for building the models was taken from a reliable source thus containing the
accurate values. Also, a lot of models were tested with sample dataset. The prediction of the yield
using the advanced methods can help the farmers to forecast the yield of a particular crop even
before cultivating the crop sown on a piece of agricultural land.
41. ENERGY-EFFICIENT AND BALANCED LOAD ROUTING APPROACH USING

BALD EAGLE SEARCH ALGORITHM FOR PATH CONVERGENCE IN MANET
Venkatasubramanian S
Saranathan college of Engineering
Trichy, India
Suhasini A S

May 21st – 22nd, 2022
Annamalai University
Chidambaram, India
Vennila C
Electronics and Communication Engineering
Trichy, India
Gomathi K
Trichy, India
Mobile Ad hoc Networks (MANETs) are overburdened with stagnant traffic. As a result, the routing
protocol for energy-efficient with load balancing has become an urgent need for MANET, especially
given the limited battery resources in the nodes. Most of the existing routing protocols are
timeconsuming because they consider power at the expense of time and over-routing. An energy-
saving, meta-inference-based load balancing protocol is an alternative model that is more
challenging than traditional routing schemes. Therefore, in this research activity, a meta-inference-
based Bald Eagle Search (BES) approach is developed for an optimal routing protocol for balancing
the load in energy usage. There are three major steps used in the BES, while hunting the fish.
Initially, the location of the vast number of prey is selected by an eagle in the space selection stage;
the searching process of prey is carried out by moving into the selected space during the space
search stage. Finally, during the dive stage, the best point is identified for hunting by swings from
the best position that are determined in the second stage. Therefore, the optimal path is identified by
the proposed BES, while considering the parameters of energy, bandwidth, delay, reliability, and
quality. Finally, the proposed algorithm supports Quality of Services (QoS) and finds the shortest
path from one source to the destination with the lowest power consumption. When analyzing the
results based on simulation, the proposed BES scheme showed a significant improvement in better
battery life, less energy consumption, and high throughput.
42. DDOS ATTACK IN VANET –DETECTION AND PREVENTION

Kalki J
Fourth Year
B.Tech
Vijeta Sarkar
Fourth Year
B.Tech
Dr. K. Kalaiselvi
Assistant Professor
Department Of Networking and Communications
Vehicular conversation networks is a effective device that permits several vehicular information
offerings and programs. The fast boom in automobiles has additionally resulted withinside the
vehicular community turning into heterogeneous, dynamic, and large-scale, making it difficult to
satisfy the stern requirements, along with extraordinarily latency, excessive mobility, pinnacle

May 21st – 22nd, 2022
safety, and sizable connections of the fifth-era community. Previous research have proven that with
the growth withinside the utility of Software-Defined Networking (SDN) on Vehicular Ad-hoc
Network (VANET) in industries, researchers have exerted widespread efforts to enhance vehicular
communications. This observe offers an exhaustive overview of preceding works with the aid of
using classifying them primarily based totally on primarily based totally on wi-fi conversation,
mainly VANET. First, a concise precis of the VANET shape and SDN controller with layers and
information in their infrastructure is provided. Second, an outline of SDN-VANET programs in
distinctive wi-fi communications, along with the Internet of Things (IoT) and VANET is supplied
with attention at the exam and contrast of SDNVANET works on numerous parameters. This paper
additionally highlights the present day and rising technology with use instances in vehicular
networks to deal with the numerous demanding situations withinside the VANET infrastructure.
This survey acts as a catalyst in elevating the emergent robustness routing protocol, latency,
connectivity and safety problems of destiny SDN-VANET architectures.
43. USE OF CARBON NANO TUBE (CNT) AS A WRAPPING MATERIAL TO THE

BUILDING STRUCTURES TO CONTROL SALT DAMAGE
R.Lilly1 S.Prabhakaran2
1Assistant professor, Department of Naval Architecture and Offshore Engineering, Academy of Maritime
Education and Training (AMET), Chennai, India.
2Professor, Department of Marine Engineering, Academy of Maritime Education and Training (AMET),
Chennai, India.
Salt damage affects the life of the structure significantly. Huge Constructions with higher
investments needs larger life. The Underground metro rail corridor of Chennai constructed along sea
coast was taken as study area which had a greater chance of affected with salts. As a protective
measure from the salt damage, Carbon Nano Tube the Nano material was tested as wrapping
material and its effectiveness in reducing the salt contaminants was found in the laboratory. The
sample points are located on either side of the corridor and sampling was done .The salinity of the
water was examined in the laboratory for the sample points and the water was allowed to pass
through the Carbon Nano Tube. The passed through water was tested again for the salinity levels.
Significant reduction of the salt contaminants was observed during this laboratory investigation. The
salinity was identified as a measure of Electrical Conductivity. Cost of the Nano materials are
highly reduced in recent years. Hence the Nano material founds to be a highly effective solution for
the future planet fresh water.

May 21st – 22nd, 2022
44. TRANSIENT RESPONSE ANALYSIS OF PV BASED NON- ISOLATED TWO

PHASE INTERLEAVED BOOST CONVERTER FOR STANDALONE DC
MICROGIRD
M. Kavitha1,V. Meenakshi2,W.Abitha Memala3,C.Bhuavneswari4, M.Pushpavalli5, P.Sivagami6,
S.Nirmalraj7 1-7Department of Electrical and Electronics Engineering, Sathyabama Institute of Science and
Technology, Rajiv Gandhi Salai,, Chennai, Tamil Nadu, India
The renewable energy sources plays a major role in village electrification. PV fed standalone micro
grid suits well for village electrification as grid connected micro grid needs extension of
infrastructure and takes nearly a year to reach the villages. This paper focuses on designing a non-
isolated two phase interleaved boost converter(NI-TPIBC) with Voltage Multiplier Module (VMM)
which acts as an intermediate and promising candidate that connects PV source to the DC load or
DC bus. Since PV is intermittent in nature and produces low voltage, the proposed converter is
designed to achieve high voltage gain with moderate duty cycle. The proposed converter is built to
cut down the voltage stress of the switches and diodes. The leakage energy of the coupled inductor is
recycled and the clamp circuit reduces voltage spikes of the power devices during turn off operation.
Further PID controller and Fuzzy Logic Controller are also used to supply regulated DC bus voltage
during different irradiance level of photovoltaic array and different load variation conditions. The
transient response parameters such as settling time, peak overshoot of the converter is compared for
both the controller and FLC which gives better results.
45. OPTIMIZATATION OF PARTIALLY REPLACEMENT OF NANO – SiO2 WITH

CEMENT
A Jayaraman1, N Sathyakumar2 , S Dhivyabharathi3and J Jeba Jenkin3

1,2,3 Assistant professor, Department of Civil Engineering, Bannari Amman Institute of Technology,
Sathyamangalam, Tamilnadu
4Assistant professor, Department of Civil Engineering, DMI Engineering College, Aralvaimozhy, Tamilnadu
Nowadays, the high strength concrete is develop using nano- materials incorporated by cement. The
nano - materials fill the voids or porous in concrete so that strength and durability are easily has been
achieved. In this paper, the nano-SiO2 materials are partially incorporated into cement and enhance
the strength. The various percentages of nano - SiO2 materials partially incorporated such as 1%,
2%, 3%, 4% and 5% . In this paper, briefly introduced the influence of nano-SiO2 on mechanical
and durability performance of concrete. In addition, this review also includes the micro structure
measured by scanning electron microscope (SEM). The results show that the enhanced compressive
strength and durability of concrete has been achieved, when nano - SiO2 incorporated up to 2% by
cement. The nano - material increased above 2 %, the strength and durability of concrete also
reduced. Whan up to 2 % nano - SiO2 incorporate with cement, the 28 days compressive and tensile
strength has been increased, and also durability performance also good.

May 21st – 22nd, 2022
46. PERSONALIZED RECOMMENDER SYSTEM USING SENTIMENT AND

EMOTION ANALYSIS FROM TWEETS
First A. Swetha Sri T S, Second B. Lekshmi Kalinathan, Third C. Sharon Julia S, Fourth D. Yamuna M
Sri Sivasubramaniya Nadar College of Engineering, Kalvakkam, Tamil Nadu, India
Online social networks provide an arena for people to share their views and perspectives on different
issues and subjects with their friends, family, relatives, etc. It is planned to detect and analyze
sentiment and emotion expressed by people from text in their twitter posts and use them for
generating recommendations. The tweets are collected on a few specific topics and created a data set
which is used to detect sentiment and emotion from them and measure the influence scores of users
based on various user-based parameters. Then the latter information is used to generate generalized
and personalized recommendations for users based on their twitter activity. These recommendations
have a list of users who agreed on the same topic and expressed similar emotions and sentiments
towards that particular topic. Among the three supervised classifiers experimented, SVM provides
the best results for the full text by considering sentiments and emotions.
47. GROUP{1,-1,i,-i}CORDIAL LABELING IN SOME DUPLICATE GRAPHS

R. Jegan1, P. Vijayakumar2
1Associate Professor, Panimalar Engineering College, Chennai
2COE, GIET University, Gunupur, Odisha
We present algorithms in this article and demonstrate the existence of Group{1,-1,i,-i} Cordial
Labeling in the extended duplicate graphs of complete bipartite, quadrilateral snake, alternate
quadrilateral snake and duplicate graph of complete graph with even vertices, triangular snake,
alternate triangular snake .
48. MODELING OF FILTERS FOR GRID CONNECTED VOLTAGE SOURCE

INVERTER USED IN RENEWABLE ENERGY RESOURCES
V.Meenakshi
Department of EEE
Sathyabama Institute of Science and
Technology
Chennai ,India
C.Bhuvaneswari
Department of EEE
Technology
Chennai ,India
M.Kavitha
Department of EEE
Technology
May 21st – 22nd, 2022
Chennai ,India
W.Abitha Memala
Department of EEE
Technology
Chennai ,India
V.Vijeya kaveri
Department of CSE
Sri Krishna college of Engineering and
Technology
Coimbatore- INDIA
D.Godwin Immanuel
Department of EEE
Technology
Chennai ,India
Among the non-conventional energy sources, it is much beneficiary and mandate to utilize wind
energy for the development of decentralized integrated energy systems in conjunction with solar
energy, hydel energy and biomass etc., Wind turbines do not launch emission which can pollute air
or water and water is not at all needed for the cooling system used in it. The quantity of carbon
dioxide emissions, amount of green-house gases is remarkably reduced. Wind power is suggested
for a prolonged usage while problems like oil shortages and environmental effects etc., arises in the
power market. In view of small scale energy manufacturing system, Permanent Magnet
Synchronous Generator (PMSG) is appropriate considering its cost and simple operation, while for
large scale, Doubly Fed Induction Generator (DFIG) is preferred . In this paper the filter design
along with its mathematical expressions to validate the design details both in analytical and
experimental manner is discussed. The transfer function of LCL filter in grid side is found out and
the consequent frequency domain approach using Bode plot has been presented. The total harmonic
distortion with reduced ripple component is observed in the grid current for LCL filter and design
parameter for LCL filter is done.
49. SEISMIC ANALYSIS STEEL TRUSS BRIDGE USING SPLICES CONNECTION

A Jayaraman 1, J Jeba Jenkin2, M Saravaman3, K Sreelaksmi4
1 Assistant professor, Department of Civil Engineering, Bannari Amman Institute of Technology,
Sathyamangalam, India
2Assistant professor, Department of Civil Engineering, DMI Engineering College, Aralvaimozhy, Tamilnadu,
India
3Associate professor, Department of Civil Engineering, Marri Laxman Reddy Institute of technology and
management, Dundigal, Hyderabad, India
4 Students, Department of Civil Engineering, Bannari Amman Institute of Technology, Sathyamangalam, India
The aim of the research project is to resist the seismic force/ vibration force in railway steel truss
bridges using splice connection. Using the Warren type of railway truss bridges Analysis and
designed by Indian standard railway code (IRC) and IS 800 -2007. The connection of the railway
truss bridge is bolted with splice connection. Same cross sectional area has been carried for both

May 21st – 22nd, 2022
theoretical and experimental investigation. Observed from the results, the splice connection has high
load carrying capacity, low deflection and high level seismic resistance.
50. RECURRENCE OF GENERALIZED LIMITING RATIO RELATIONS

R. Balaji 1,* , B. Antline Nisha 2, M. John Peter 3, SP. Ranganathan 4 , V. Kamalakannan 5
1. Department of Mathematics, Panimalar Institute of Technology,
Chennai 600123, Tamil Nadu, India. E. Mail: [email protected]
2. Department of Mathematics, St. Joseph’s Institute of Technology,
Chennai-600119, Tamil Nadu, India. E. Mail: [email protected]
3. Department of Mathematics, Panimalar Institute of Technology,
Chennai 600123, Tamil Nadu, India. E. Mail: [email protected].
4. Department of Mathematics, Surya Group of Institutions,
Vikravandi-605652, Tamil Nadu, India. E. Mail:[email protected]
5. Department of Mathematics, Panimalar Engineering College,
Chennai-600123, Tamil Nadu, India. E. Mail: [email protected]
*Corresponding author: R. Balaji
Progression of a distinct behavioral examination is studied using continued process. Continued

process is mostly multidimensional array to previous terms using recursion, in this continued
process with the nth term is straight line. Among them, in a sequence or a multidimensional array to
previous terms using recursion with constant coefficients and linear recurrences with polynomial
coefficients are very much important. Continued process results are generally straight-line
recurrences, in this nth term would be straight line regarding to its previous terms. Objective of this
paper is finding the fraction of f term to term and it’s known as mean ratio. Continued process is
applied step-by-step for a common sequence to determining a précised iterative relation and mean
ratio. By characterizing graphs, we can easily understand exact results obtained which is nearest to 2
on either side of progression. Closest point of progression when, m=18 the result obtained is 1.9999
and then when, m=19 the result obtained is 2.0001.
51. A RELIABLE DATA TRANSMISSION USING AN ADAPTIVE ROUTING

PROTOCOL FOR FANETS
Orchu Aruna1, Dr. Amit Sharma2
1Assistant Professor, Department of Computer Science and Engineering, Narasaraopeta Engineering

College, AP, India.
2Associate Professor, Department of Computer Science and Engineering, Lovely Professional

University, Punjab, India.
This paper converses instigating UAVs for several civilian and armed zones by generating FANETs
as modern solicitations comprise slightly floating nodes. Flexible, consistent, and delay-bounded
infrastructures among UAVs are essential in FANETs due to the quickly altering topology and
augmented quantity of UAVs by continuing the anticipated system with eminence facilities. Data
transmitting between UAVs is a key challenge in FANET, because FANET is not similar to the

May 21st – 22nd, 2022
MANETs with the limitations of less convenience and high mobility. The dynamic updating of
routing tables should achieve as indicated by topology changes. To afford a reliable communication
between UAVs, metric controls have not measured by maximum number of current routing
procedures in FANET. Due to the sophisticated movement of FANETs, the actual challenge is the
routes consistency. In this research work, we have introduced a new routing system, a predicted
future weight-based routing scheme for FANETs that considers the node's present and upcoming
values of detailed parameters for the reliable routing. This proposed routing procedure allocates
weight to every node in the network by calculating the node's future. Depending on this process, the
connecting nodes can found a reliable path that continues for a long time. This dynamic future
prediction-based routing scheme ensures better data delivery with minimum overhead and optimized
energy consumption in all conditions than traditional routing protocols. The proposed routing
protocol could achieve better results than other routing protocols based on the valuation of
simulation results.
52. STUDIES ON PREPARATION OF SQUASH FROM COCONUT SPROUT
Nithiya priya.B.T*, Jei Subhashini V, Vidhya Lakshmi J, Kundavaidevi K, Ragavi V J

Dept. of Food processing and preservation technology, School of engineering,
Avinashilingam Institute for Home Science and Higher Education for women, Coimbatore,
Tamil Nadu, India.
Squash is a non-alcoholic beverage that has 7g of fiber, 89% saturated fat, 150 ml of coconut sprout
extract, and 150 ml of tender coconut water it also includes two ml of ginger, 2ml of lemon, and 0.1
ml of sodium benzoate. Before serving, it is diluted. Coconut sprout squash is indeed a naturally
highenergy drink that is quite nutritious. Coconuts are still one of the world's most significant crops,
they are grown in tropical countries like India. Tender coconut is a plentiful and naturally occurring
resource that is utilized as a meal and beverage. The goal of this study was to construct a squash
with coconut sprout extract, tender coconut water, lemon juice, and ginger juice. To determine the
most appropriate boiling methodology for generating coconut sprout squash and to assess the overall
acceptability of squash in prepared sample A. In comparison to sweet orange squash, sample A had
the highest squash score. In comparison to sweet orange squash. Sample A had the greatest protein
and carbohydrate score. The sample had the highest overall acceptance. The prepared sample is
placed in a glass bottle and air-cooled separately on wooden planks. In physicochemical and sensory
evaluation, sample A was preferable. The samples are tested for proximity and microbiological
contamination. The formulated coconut sprout squash contains a significant level of Omega 3 fatty
acid, making it ideal for people with colorectal cancer and high blood pressure.
53. ENHANCEMENT OF REJECTION CHARACTERISTICS OF FILTER USING AN

ARRAY OF CSRR FOR NOTCHING THE ISM BAND IN UWB ANTENNAS
M. SATHISH1 S. ANITHA2 R. BHARATHY3 U. JHANANI SHREE3 M. ANITHAMARY3

1ASSISTANT PROFESSOR (SG), 3ASSISTANT PROFESSOR (SS)
13 DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING
13RAJALAKSHMI ENGINEERING COLLEGE, CHENNAI, INDIA

May 21st – 22nd, 2022
This proposed work uses Complementary Split Ring Resonator (CSRR) to filter out the frequency
bands. It is evident that any planar antenna will radiate in its resonant frequency as well as other
frequencies. Those harmonics need to be eliminated from the radiation characteristics to have
effective radiation. This filter is designed for 2.5GHz to remove those undesirable harmonics and
hence a band elimination filter is approached using square and hexagonal ring configuration.
54. WIRELESS POWER TRANSFER SYSTEM WITH IMPROVED COIL GEOMETRY

Anbuselvi Mathivanan and Saravanan Palani
Sri Sivasubramaniya Nadar college of Engineering, Tamilnadu, India
This research work focuses on the design and development of the wireless charging system for
electric vehicles. The research components of the work include the design of a compensation circuit
and coupling coil geometrical analysis for maximizing the power transfer efficiency. The
preliminary work on the compensation circuit, involved the design of LCCC-LCC, unbalanced
series-series topology. The parameters of the circuit components are optimized under simulation
environment and it is experimentally validated with the real-time implementation of the proposed
circuit. The coupling coil used in the wireless charging system has a prevalent impact on the
efficiency of power transfer. The work spans across the three different geometrical designs of
circular coils and analysis of the same under the developed experimental design. The analysis of the
proposed system is extended towards the understanding of varied coupling factor and coil
dimensions. The outcome of the analysis attains an optimized wireless charging system with
improved efficiency and reduced design complications
55. ARTIFICIAL NEURAL NETWORKS AND PI HYBRID CONTROLLERS FOR AN

ELECTRIC VEHICLE'S ENERGY STORAGE SYSTEM
Swagata Banerjee1, S Shiva Kumar2. Bit Mesra,
Techniques for designing an Electric Vehicle Hybrid Energy Storage System are being considered.
As a result of the increased demand for electricity, there must be greater monitoring and more
frequent regulation, and there must be more difficulties overcome in the creation and refining of
specialised techniques to meet it. Projects to reduce energy consumption in the construction,
transportation, commerce, and other activities are examples of such initiatives. Computer and
smartphone-based entertainment, basically portable PCs, is becoming more popular. This difference
must be made between the capacity to store energy utilising renewable resources for lengthy periods
of time and the technology for minimising energy use. A great amount of study and investigation is
required as a consequence of the nature and scope of this subject matter. An in-depth grasp of the
significance of electric automobiles for hybrid energy storage systems would be incorporated in a
more focused comprehension of their respective points of view. PI and neural networks are
employed in this thesis to create improved distortion-free outputs using a unique technique based on
neural networks and PI.

May 21st – 22nd, 2022
56. A SHORT REVIEW ON BATTERY MANAGEMENT SYSTEM FOR ELECTRIC

VEHICLE
Yasvanthkumar Sundaram1, Dr.Sivanandan Nanjan2, *
1ResearchScholar, Department of Electronics, PSG College of arts and Science, Coimbatore-641014,
Tamilnadu, India.
2 Assistant Professor, Department of Electronics, PSG College of arts and Science, Coimbatore-641014,
Tamilnadu, India .
Background: Rechargeable batteries have become a key renewable energy source for various
applications such as transport, grid storage and mobile systems in recent years. Much more powerful
batteries and the introduction of electric cars with a greater performance than fossil fuel cars were
accomplished by the advancement of power storage technology. Despite of the relative maturity of
batteries, the academic industry and businesses are working hard to make vehicle electrification
batteries safe and affordable. Methods: A battery test commonly used for charge and discharge
constant voltage, Current and constant exile electric mode, record the process of testing data, such as
testing time, voltage and current. The electrochemical performance parameters characterization of
material capacity, columbic efficiency, charging and discharging platform and battery internal
parameters. The battery management system (BMS) is a critical component of electric and hybrid
electric vehicles. Findings: The purpose of the BMS is to guarantee safe and reliable battery
operation. It is used to improve the battery performance with proper safety measures within a
system. Therefore, a best BMS is that must need for operating an electrical system. Novelty: This
paper reviews the characteristic of battery management system and its technology with advantages
and disadvantages. Also, aims to analyzing this battery management system in detail
Keyword: Battery, Battery Management System, energy storage safety, electric vehicle.
57. CUSTOMER PERCEPTION TOWARDS ONLINE FOOD BUSINESS APPS

1Dr.P. Vanitha2Priya R 3Gokul R4Arunika5Naveen C6Arulmurugan
1Associate Professor, Department of Management Studies, M. Kumarasamy College of Engineering, Karur,
Tamil Nadu, India.
E-Mail: [email protected]
2 Students, First Year MBA, M. Kumarasamy College of Engineering, Karur, Tamil Nadu, India.
3,4,5,6 Student Second Year MBA, M. Kumarasamy College of Engineering, Karur, Tamil Nadu, India.
The main aim of this research is to examine the consumer perception towards online food business
apps and to analyze the perception towards quality of apps and websites of electronic food ordering
companies among the respondents in Karur district .Both primary and secondary data has been used
to collect the information. Technology Acceptance Model questionnaire model has been used to
collect the data. Sample Size is 120.Kruskal Wallis test has been used to find the relationship
between variables.

May 21st – 22nd, 2022
58. RECOGNITION OF TRAFFIC INFORMATION WITH THE HELP OF SOCIAL

MEDIA TWEETS
Dr. Cuddapah Anitha, Associate Professor, Department of Computer Science and Systems
Engineering, SreeVidyanikethan Engineering College (Autonomous), Andhra Pradesh, India
Dr. C. Sivakumar, Asst.Professor, Department of Computer Science and Systems Engineering,

SreeVidyanikethan Engineering College (Autonomous), Andhra Pradesh, India
In present era, social media is been evolving widely in terms of using for connecting to their friends
and beloved one’s, posting and sharing the information. Few social Medias like, Face book,
LinkedIn, twitter was also using for sharing the ongoing daily trends in the real world. In metro
cities like Bangalore, Chennai, Delhi Mumbai and Hyderabad are well known to their traffics,
people are struggling to reach their destination on time. In our paper, we were focusing on detecting
the traffic information from social media text to make sure traveler reach their destination or places
with a short time or quick. Also, we were focused to provide the necessary safely precaution on the
road accident on their go. The main advantage is to increase the accuracy of traffic event and to
display the location information of huge traffic. In this paper, we used the Logistic Regression
method is used to extract the real-time traffic information which is also compared with combination
of CNN-LSTM Method.
59. OPTIMAL BIDIRECTIONAL LONG SHORT TERM MEMORY BASED

SENTIMENT ANALYSIS WITH SARCASM DETECTION AND CLASSIFICATION
ON TWITTER DATA
M. Jeyakarthic1, Assistant Professor, Department of Computer and Information science, Annamalai
University, Chidambaram.
J. Senthilkumar2, Assistant Professor, Department of Computer and Information science, Annamalai

University, Chidambaram
Sentiment analysis (SA) and sarcasm detection on social networking platforms have gained
significant attention in recent times. The existence of sarcasm in social networking data like Twitter
is a major cause of wrong categorization of sentiments. It remains a challenging issue in natural
language processing (NLP) as it restricts the way of identifying the original sentiment of the people.
Different feature engineering approaches are available in the literature to detect the presence of
sarcasm in Twitter data. This study develops an Optimal Bidirectional Long Short Term Memory
based Sarcasm Detection and Classification (OBiLSTM-SDC) using Twitter Data. The goal of the
OBiLSTM-SDC technique is to identify and classify the existence of sarcasm in social networking
data. The OBiLSTM-SDC technique involves term frequency-inverse document frequency (TF-IDF)
model for feature vector generation. In addition, the BiLSTM model is employed for the
classification of sarcasm and the hyperparameter tuning of the BiLSTM model takes place by the
use of chicken swarm optimization (CSO) algorithm. The performance validation of the OBiLSTM-
SDC technique takes place on benchmark dataset and the results are inspected under various aspects.
The simulation results highlighted the supremacy of the OBiLSTM-SDC technique over the recent
approaches.

May 21st – 22nd, 2022
60. IoT BASED DECISION SUPPORT SYSTEM FOR TELEHEALTH MANAGEMENT

THROUGH DATA MINING FOR BIOMEDICAL EQUIPMENT
1st Dr.S.Rajalaxmi
Dept.of Biomedical Engineering
Mahendra College of Engineering
Salem, India
2nd G.Shyamala
Salem , India
3rd R.Priyadharshini
Salem, India
Often, small clinics and hospitals using previous generation of Medical equipment do not have
facility to transfer patient data to a remote location for analysis. As the ratio of doctors to patients is
declining due to rise in health care issues, it is necessary to monitor patient data from a remote
location, where qualified medical professionals are available. India counts to 1,85,505 Primary
Health Centres in rural villages alone with approximately 371 crore worth medical equipment. Due
to the cost ratio, the entire equipment replacement with modern technology would be myth. Rather,
implementation of add-on equipment with the existing equipment will help provoke of a decision
support system for telehealth enhancement. The proposed work includes a clip on camera supported
with IoT to capture the screen continuously and transfer the data wirelessly to remote location. This
will be a small add-on camera to every equipment. It is a portable, small clip on device which can be
added to any biomedical equipment by clipping on it to the display monitors. The work proposes a
system that consists of a camera that can be clipped onto the screen of the existing equipment and a
data processing module and a data transmission module. The receiving end will be a wireless
receiver module and a translator and display capable of receiving from multiple equipment. The
development of camera based medical equipment is very helpful in the hospitals located in the rural
areas. It includes data extraction, data storage and data visualization. This works under the basis of
Optical Character Recognition and Application Programming Interface in Google Cloud Platform.
This proposed work would be useful for patients in need of diagnosis and treatment who are remote
from the doctors and using the existing older generation biomedical equipment itself.
61. EFFECT OF SURFACE ROUGHNESS ON THE AERODYNAMIC AND

AEROACOUSTIC PERFORMANCE OF THE DARRIEUS WIND TURBINE
A.Sasi1, J.Gunnal2, T.J.Ramkumar3, K.Karthik4,J. Sarathkumar Sebastin5
1,2,3,4,5Department of Aeronautical Engineering, Kalasalingam Academy of Research and Education,

Sirivilliputhur, Tamil Nadu.
. The recent advancements in small wind turbines show an increased demand for the Darrieus wind
turbine in the urban environment. The present work is devoted to the numerical analysis of the
aeroacoustic sound emission from a straight bladed Darrieus wind turbine with NACA0018.
Computations are performed at Reynolds number of 28,000 with a tip speed ratio of 0.4 using the
unsteady Reynolds averaged Navier Stokes equations with Ffowcs Williams- Hawking's (FW-H)

May 21st – 22nd, 2022
acoustic analogy. The FW-H method is based on a free field Green's function where the scheme uses
a porous integration surface and implements an advanced time formulation. The study aims to obtain
a numerical methodology to predict the effect of surface roughness on wind turbines' aerodynamic
power and sound emission.
62. PCOD PREDICT FOR OBESE PATIENT USING NAFLD AND T2D PROFILES
H.Kala, K.Sharmila, A.T.Priyeshkumar

Salem, India
PCOD is the major problem faced by women in now-a days which is caused by increasing the male
androgen level and cause heavy bleeding, infertility and cause severe effects like diabetes also.. For
that an indigenous device has been designed for detection of Polycystic ovarian disease based on
measurement of the volatiles in the Exhaled Breath. This device useful in clinical field. The sensing
unit is based on the two commercial sensors of MQ136-Hydrogen sulphide and MQ3-Alcohol
sensor which are sense the VOCs of Hydrogen sulphide and Ethanol contained in the human
Exhaled Breath. The patient samples of Exhaled breath to be blowing through inhalation tube, a
mouthpiece is to be linked with patient mouth for blowing. We have designed and tested this device
with two different types of patients between obese patient with PCOD and Lean patient without
PCOD. This device can differentiate the with PCOD and without PCOD patients for predicting the
PCOD patient correctly by analyzing the VOCs of NAFLD and T2D because the NAFLD and T2D
are the major influences for getting PCOD, particularly in obese patient. With further developments
and measurements gives the hope which this system will become as useful device in clinical purpose
for screening the PCOD problem at easiest way, in Non-invasive method than invasive method in
future. Such that this device will be reduce the usage of invasive method, patient discomfortness and
cost.
63. NETWORK SECURITY ATTACKS & TECHNIQUES TO SECURE THE

NETWORKS.
Bhagwant Singh1 Dr. Sikander Singh Cheema2
Punjabi University Patiala
Department of Computer Science and Engineering
Network safety is a procedure utilized to ensure the client’s classified Intel and shield from online
cheats and frauds. Network protection gives a safe space for the clients to get to their Intel safely. In
this paper, the author breaks down bunches of escape clauses that threaten digital protection,
prompting network assaults. This paper additionally discusses distinctive organization security
methods. The network safety strikethrough has sought me to explore this space. The information and
secret data of different organizations are in grave danger, so this exploration is an endeavour to give
a rampart with the goal that no intruder can encroach and strikethrough data. It is an endeavour to
plug all escape clauses to give a superior and dependable organization. While more businesses suffer
severe security breaches, the amount of data that has been exposed is increasing. According to data

May 21st – 22nd, 2022
breach stats, hackers are motivated by money to obtain data, and private information is valuable to
infiltrate. It’s also clear that businesses are still unprepared even though data breaches are becoming
more regular. In today’s environment, network security flaws are the most frequent dangers and
difficulties. The justification for this is that everything and everyone is connected to the Internet.
That implies there’s a chance the technology may be hacked, but there’s also a chance it won’t.
There are some manufacturing deficiencies, and sometimes, while designing a product, the
developer is left with a loophole or backdoor in the product; the product can be anything; for
illustration, when an ISO image is configured in hardware, there is a possibility that certain
backdoors and loopholes stay, which may impact clients or associations in the long term. There are
numerous methods for this sort of incident, and when we implement specific strategies. Some of the
strategies provided by the author in this paper demonstrate how to discover issues and implement
specific techniques to make their network more secure and robust.
64. METHODS FOR RESTRICTING UNWANTED ACCESS ON THE NETWORKS

USING FIREWALL
Bhagwant Singh1 Dr. Sikander Singh Cheema2
Punjabi University Patiala
This paper examines firewall mechanisms routinely implemented for network security in depth. A
firewall can't protect you against all the hazards of unauthorized networks. Consequently, many
kinds of infrastructure are employed to establish a secure network. Firewall strategies have already
been the subject of significant analysis. This study's primary purpose is to avoid unnecessary
connections by combining capability of the firewall of the use of additional firewall mechanisms
which including packet filtering and NAT, VPNs, and backdoor solutions. There are insufficient
studies on firewall potential and combined approaches, but there aren't many. The research team's
goal is to build a safe network by integrating firewall strength and firewall methods. The study's
findings indicate that the recommended concept can form a reliable network. This study examines
the characteristics of network security and the primary danger, synthesises existing domestic and
foreign firewall technologies, and discusses the theories, benefits, and disadvantages of different
firewalls. Through a synthesis and comparison of various techniques, as well as an in-depth
examination of the primary factors that affect firewall effectiveness, this study investigated firewall
technology's current application in computer network security, then introduced a new technique
named "tight coupling firewall." Eventually, the article discusses the current state of firewall
technology as well as the direction in which it is developing.
65. COMBINING DATA OWNER- SIDE AND CLOUD-SIDE ACCESS CONTROL IN

ATTRIBUTE BASED ENCRYPTED CLOUD STORAGE
Balasundar V1* , Zeus Kevin Anderson S2* , Dinesh S 3, M.Manimegalai 4 , K.

Sebasthirani5 1Under Graduate, Information Technology, National Engineering College,
Kovilpatti. 2Under Graduate, Information Technology, National Engineering College,

May 21st – 22nd, 2022
Kovilpatti. 3Under Graduate, Information Technology, National Engineering College,

Kovilpatti.
4Assistant Professor,Information Technology,National Engineering College, kovilpatti,
5Associate Professor, Electrical and Electronics Engineering, Sri Ramakrishna Engineering College,
Coimbatore
Due to the lack of user-to-cloud controllability, people endorse the benefits of cloud computing, but
are hesitant to trust cloud providers with private information. Owners of data outsource encrypted
data rather than plaintext in order to ensure confidentiality. CP-ABE (cypher text-policy attribute-
based encryption) can be used to conduct fine-grained and user-centric access control when sharing
encrypted data with others. However, this isn't sufficiently secure from other attacks. Previous
schemes did not allow cloud providers to verify whether a downloader could decrypt data. As a
result, all users should have access to the files. To launch economic denial of service (EDoS)
attacks, malicious attackers can download thousands of files, consuming much of the cloud's
resources. Users pay for these resources. A cloud provider is both an accountant and a payee for
resource consumption, which leaves data owners in the dark. These concerns must be supervised in
cloud applications in the real world. The project proposed a method for securing encrypted cloud
storage from EDoS attacks and providing resource usage reporting. By using CP-ABE schemes in a
black-box manner, it complies with the arbitrary access policy of the CP-ABE. Two protocols are
presented for different settings, and performance and security analyses are presented.
66. TRIPLE BAND U-SLOT ULTRA-WIDEBAND ANTENNA AS A WIRELESS

SENSOR FOR COMMUNICATION
1K.Sugapriya, 2 S. Omkumar
Research Scholar, SCSVMV, Kanchipuram, Tamilnadu, India1,
Asso. Prof. /ECE, SCSVMV, Kanchipuram, Tamilnadu, India2,
The communication industry involves in all the digital environment in the world, the Ultra-wide
band Microstrip patch antenna plays a vital role, because the patch antenna consists of compact size,
less weight, better return loss and good radiation pattern. The proposed U-slot ultra-wide band
antenna was designed for wireless sensor applications and it operates in triple band frequencies of
spectrum sensing function for three different applications. The antenna is designed using the FR4
substrate material with reference to the ground plane, the U-slot is etched from the rectangular patch
to get the good wireless sensing, triple frequency band operation for spectrum sensing and it gives
the good return loss i.e., below 10dB in negative, gives the Omni directional radiation pattern in the
ultra-wide band range 3.1GHz to 10.6GHz also to achieve a good gain, hence the ultra-wide band
antenna is suitable for triple band application like cognitive radio, satellite, radio navigation etc.,
67. MESH MESSAGING SERVICE USING BLUETOOTH

Aswini.J
Sree Vidyanikethan Engineering College
Tirupati, India
P. Ruchitha
Tirupati, India
May 21st – 22nd, 2022
P. Hitesh Reddy
Tirupati, India
P. Ashish
Tirupati, India
R.B.Moulya
Tirupati, India
S.Amisha
Tirupati, India
The idea is to create a service that enables users to communicate with other users using a Mesh
Network established between their devices. The Mesh network is to be built using the Bluetooth
service available in the user devices. There are apps built using this technology such as Bridgefy,
Mesh Messaging, Fire chat, etc. But the proposed application enables users to send messages to
other users through the Bluetooth network established between the user devices. The application
does not require any Internet connection. The idea is to transfer data from the sender to the receiver
where it finds its way using a Dynamic routing algorithm. The middle devices (user devices that are
not participating in the communication) act as routers in the network. The app encrypts the messages
which are sent from the sender so that the middle devices cannot see them.
68. ANALYSIS AND OPTIMIZATION OF MACHINABILITY OF AL2O3 AND FLY

ASH REINFORCED ALUMINUM HYBRID COMPOSITES USING CARBIDE
INSERTS
V Veeranaath1*, Mobashir2, Ravi Shankar3
1,2,3 Department of Mechanical Engineering, SRM Institute of Science and Technology,

Kattankulathur, Chennai, India, 603203
Aluminum matrix hybrid composites reveal excellent mechanical behavior compared to traditional
materials for a wide range of conditions and applications. However, the machining of this kind of
composites is a challenging job due to its improved mechanical hardness. The problems faced during
cutting of these composites increased machining forces, poor surface quality, and increased tool
wear. Many types of research are still in progress across the world to overcome these issues by using
a variety of inserts and processing constraints. So, this paper is mainly focused on the study of the
machinability of aluminum hybrid composites. The composites are synthesized by employing stir
casting via varying the filler content, stirring duration, and speed. The fabricated composites are
machined using coated and uncoated carbide inserts. Machining forces, surface roughness, and insert
wear are assessed to learn the machinability of these materials by varying the machining constraints.
Taguchi analysis was employed to assess the effect of the constraints on the retorts.

May 21st – 22nd, 2022
69. ANALYSIS OF THE UPSHOT OF HEAT TREATMENT ON PROPERTIES OF

MAGNESIUM HYBRID COMPOSITES
V Veeranaath1*, Uddhav Daga2, Yuvraj Kumar3
1,2,3 Department of Mechanical Engineering, SRM Institute of Science and Technology, Kattankulathur,
Chennai, India, 603203
Magnesium-based materials are in superior demand in auto industries due to their improved
properties and behavior. However, processing and further post-treatment of this kind of magnesium
composites is a stimulating chore because of its characteristics. In particular, heat treatment of this
kind of magnesium composites may induce a change in grain structure which in turn alters the
properties according to the need. So, this article is chiefly focused on the fabrication of magnesium
composites by filling with SiC and graphite. According to the Taguchi array, the composites were
developed by altering the filler content, stirring duration, and speed. Four samples were fabricated
accordingly by employing the stir casting technique. The fabricated composites were exposed to
different types of heat treatment processes such as annealing, normalizing, tempering, and
quenching to learn the upshot of the same on its behavior. The morphology of the dispersions of the
developed composites was analyzed using an optical microscope, the mechanical and physical
behavior like hardness, density, wear, and tensile tests were performed on the developed samples.
The morphology and behavior of the fabricated composites were compared and analyzed pre and
post-heat treatment.
70. IMPROVING THE LIQUID INSULATION PERFORMANCE WITH

ANTIOXIDANTS
HelanGethse J
EEE (UG)
National Engineering College
Kovilpatti
Diana Hyden N
Dhanya K
EEE (UG)
Kovilpatti
Samuel Pakianathan P
Muthuselvi G
EEE (UG)
Kovilpatti
Transformer oil is mostly used to keep the transformer cool. It functions as a cooling agent. Mineral
oil has long been used in transformers. Mineral oil has a high dielectric strength, which allows it to
withstand high temperatures. Mineral oil's main disadvantage is that it is not environmentally
friendly and can be dangerous to the environment. The features of breakdown voltage (BDV),
May 21st – 22nd, 2022
viscosity, flash point, and fire point are measured and reported in this study, and the characteristics
of olive oil are compared to the characteristics of mineral oil.
71. MULTIPLE VENDOR E-PLATFORM FOR FARMERS

Dr. C. K. Gomathy, V. Jaswanth Reddy, P. Venkatesh
SCSVMV
This study motivates the farmer to utilize an e-commerce platform to shape their revenue.
Ecommerce is one of the fast-moving industries in the world. Many businesses were already shifted
to e-commerce and generating huge revenue within a short period of time. The major problem is
"The farmer who produces food for the nation is not benefitting with the income". Because there is
no such dedicated platform for farmers to generate profitable income for their cultivated products.
Existing platforms are combined with all categories (home, kitchen & electrical appliances) of
products that might muddle up farmers, and also the lack of internet knowledge is one of the reasons
that a farmer may not be able to sell their products online. To overcome this, The online multivendor
platform will be helpful where a farmer can able to sell or purchase anything related to the
agriculture & farming category. A simple user-friendly web page with multiple farmer vendors and
multiple agricultural products along with detailed studies (blogs) about the latest farming
technologies, will surely shape the farming sector revenue. This will benefit farmers to get internet
knowledge and save both time and energy. Also, new connections & communications will be
established and the product stock will sell easily without any labor help. In a nutshell, It is a
dedicated multivendor platform for farmers with all farming products like pesticides, seeds,
fertilizers, crops, etc along with a knowledge base. So, this initiative will surely play a vital role in
uplifting & benefitting framers.
72. Q Learning Algorithms: An introduction

R.Raja Rajeswari
Research Scholar,
Department of Computer Science, Mother Teresa Women’s University,
Kodaikanal.
Dr.A.Pethalakshmi
Principal,
Alagappa Government Arts College,
Karaikudi
Reinforcement Learning is a combination of supervised learning and unsupervised learning, the two
main streams of Machine Learning and has many applications in Artificial Intelligence arena.
Reinforcement learning algorithms range from Q learning algorithms to Deep Q Network.
This research article analyses the elements of Reinforcement learning, discusses basics of Q
Learning algorithm ,with a simple implementation and concludes with the findings .

May 21st – 22nd, 2022
73. ACHIEVING MORE SECURITY IN ELECTRONIC MONEY FOR ECOMMERCE

WEBSITE THROUGH CRYPTOGRAPHY
Dr. V. Geetha1, K.Hemanth2, M.Chaitanya3 ,Dr. C.K.Gomathy4

1,4 Assistant Professor, Dept. of CSE, SCSVMV(Deemed to be University), Kancheepuram ,TamilNadu, India
2,3 Student,Dept. of CSE, SCSVMV(Deemed to be University), Kancheepuram ,TamilNadu, India
E-Commerce is the platform that merchant publish their products and makes online marketing
process online by using web browsers. Now days this online marketing is growing quickly and
demand for it is high. To purchase these products from the merchants, user needs to do an online
transaction. Most Internet payment gateways in e-commerce provide financial information to
customers using believed third parties directly to a payment gateway. Either the payment gateways
follow the strict policies and its there might be a chance of fraud that allows stealing the monetary
information of users. So, this project main role is to provide an effective and secured electronic
payment gateway where users can connect immediately to the merchants directly. In this proposed
system the user would do the transaction under a secret key that is generated by using encrypted
algorithms, and the details of user will be in secured mode. So, we can say that our protocol has
much improved security analysis that user can do their transactions without any doubt. Any payment
gateway should have some entities are confidentiality, integrity, non-repudiation, anonymity
availability, authentication, and authorization. We can say that these would give trust to the
customers that their information will be in safe and secured. And the proposed system of this project
has these entities and maintains the customer details or information secured.
74. A STUDY ON SPAM REVIEW DETECTION USING LINGUISTICS

Dr. V. Geetha1, Dr. C.K.Gomathy2
1, 2 Assistant Professor, Dept. of CSE, SCSVMV (Deemed to be University), Kancheepuram, TamilNadu,

India
Social Media Services are emerging communication mediums for people to establish and manage
social relationships. On social media websites, consistently, billions of users are engaged in social
interaction, content and opinion dissemination, networking, recommendations, advertising, scouting,
alerting, and social campaigns. The hype of social media services open-up a new opinions and
hurdles to the study of social media networks, being of interest to many fields. Social media is a
place where social activities, business-oriented activities, entertainment, and information are traded.
It inaugurated a worldwide connectivity environment where communities of people share their
interests, ideas, talents and activities, or who are interested in interests and activities of others
Although social network has given enormous benefits to the people at the same time harming people
with various devious activities that take place on social platforms. This causes notable economic
losses to our society and even endangers national security. Platforms like Facebook, Twitter,
LinkedIn, etc. are highly receptive to malware activities. Twitter is one of the biggest microblogging
networking platforms; it has more than half a billion tweets posted every day on average by millions
of users on Twitter. Such adaptability and wide usage of Twitter and other social media networks
makes it the target. They can be easily invaded with poisonous activities such as malware intrusion,
spam distribution, social attacks, data breaches etc. Ad mailer/Spammers use social planning attack
strategy to send junk tweets, spam URLs, etc. This made twitter an exemplary field for the
expansion of unidentified spam accounts. The impact stimulates researchers to prepare a model that

May 21st – 22nd, 2022
analyses, detects and recovers from scandalous actions on Twitter. Twitter’s system is besieged with
tens of millions of fake spam profiles which may cause trouble for the normal user's security and
privacy. To improve real customer safety and recognition of spam profiles became key parts of the
research.
75. SMART ROOM CONTROLLER FOR VISITORS USING INTERNET OF THINGS

Dr. V. Geetha1, Dr.C.K.Gomathy2

India
The user will be given a set of hardware with pre-installed software which needs to be placed at the
entrance the hardware will detect whether the person entering the premises is wearing a mask or not
and the body temperature using no contact temperature sensors after those two things have been
checked if the person’s body temp. is normal and he/she is wearing a mask than the gate will open or
vice versa automatic Room light and fan controls are mostly implemented in theatres, shopping
malls.
But this is the case where we can control the lights and fans of the room as well as we can also count
how many persons are entering into the room and also shows how many are exiting the room. here
in this we have used the Arduino UNO board, IR sensors and LCD display screen.
76. A WEB BASED MANAGEMENT SYSTEM USING TRANSPORT SERVICE

Dr. C.K. Gomathy1, Dr.V Geetha2
1, 2 Assistant Professor, Dept. Of CSE, SCSVMV (Deemed to be University), Kanchipuram,
TamilNadu, India
Regular vehicle servicing helps vehicle owners prevent unexpected vehicular malfunctions.
Identifying vehicle problems at an early stage is essential before they’ll turn into significant issues.
Vehicle owners tend to visit vehicle garages to enquire about vehicle servicing. Vehicle mechanics
check the vehicle’s vital components and repair issues to ensure that they are in good shape and will
not fail unexpectedly. The researchers conducted the study to assess how vehicle service shops or
garages perform daily operations and transactions. The researchers gather initial data and found out
that most vehicle shops or garage still uses manual processes in handling their day-to-day
transactions. Recording of customer records, monitoring of vehicle repair progress, vehicle service
updates, transactions with clients, and billing are still done manually. Continuous use of manual
methods will slow the operation and response time of the garage to the inquiries of the customers.
The researchers have seen the need to upgrade the way how vehicle servicing garage operates. The
implementation of the system will both benefit vehicle garages and customers. For vehicle garages,
utilizing the system to operate the business will help them grow and increase business efficiency.
For customers, they can easily and conveniently make transactions for vehicle servicing.

May 21st – 22nd, 2022
77. MULTI-SOURCE MEDICAL DATA INTEGRATION AND MINING FOR

HEALTHCARE SERVICES
Dr. C.K. Gomathy1, Mr. K.Nishanth Reddy2, Mr. K.Sai Abhishek3, Dr.V Geetha4
1, 4 - Assistant Professor, Dept. Of CSE, SCSVMV (Deemed to be University), Kanchipuram, TamilNadu,
India
2, 3 - Student, Dept. Of CSE, SCSVMV (Deemed to be University), Kanchipuram, TamilNadu, India
Standards with the advent of the Internet of Health (IoH) age, traditional medical or healthcare
resources are slowly moving to the web or the internet and generating a large amount of medical
data associated with patients, doctors, medications, medical infrastructure, and so on. The effective
integration and analysis of this IoH data are excellent indicators for disaster diagnostics and medical
care services. However, IoH dthe is usually distributed to different departments and contains the
privacy of the users. Therefore, it is often a challenging task to compile or extract sensitive IoH data,
where user privacy can be disclosed. To overcome the above difficulties, we prioritize a multi-
source medical data collection and mining solution for better health care services, called PDFM
(Data Fusion and Private Mining). With PDFM, we can search for similar medical records in a time-
saving and confidential manner, in order to provide patients with better medical and health care
services. A test team is developed and used to demonstrate the feasibility of a proposal for this task.
78.A STUDY ON REAL TIME OBJECT DETECTION AND TRACKING SYSTEM
Dr. V. Geetha 1Dr. C.K.Gomathy2

1,2Assistant Professor, , Department of Computer Science and Engineering,
Sri Chandrasekharendra Saraswathi Viswa Mahavidyalaya [SCSVMV Deemed to be University],
Kanchipuram, Tamil Nadu, India
Globally the amount of individuals of all ages visually impaired is calculable to be 285 million, of
whom thirty-nine million are blind. For these people knowing the surroundings learning about
anything is difficult compared to normal people. Some of them have personal assistants/ helpers to
help with this problem but a lot of them don’t. The main aim is to assist & support visually
challenged persons. Machine learning algorithms are used in a wide variety of applications, such as
email filtering and computer vision, where it is difficult or unfeasible to develop conventional
algorithms to perform the needed tasks. Deep learning has become the dominant approach for much
ongoing work in the field of machine learning. Deep learning (also known as deep structured
learning) is part of a broader family of machine learning methods based on artificial neural networks
with representation learning. Learning can be supervised, semi-supervised or unsupervised.
79. SMART HEALTH CARE DIAGNOSIS MANAGEMENT SYSTEM

Dr. V. Geetha1, Dr. C.K.Gomathy2
In the health-care industry, a system is critical for managing the administrative, financial, and
clinical components of a hospital. Accurate analysis of medical data aids early illness detection,
patient treatment, and community services, thanks to advances in data mining in the biomedical and
healthcare communities. When medical data is of poor quality, the precision of the study suffers.
May 21st – 22nd, 2022
Furthermore, different regions have distinct appearances of specific localised diseases, which may
make disease outbreak prediction more difficult.Machine learning techniques are provided in the
proposed system for effective prediction of various illness occurrences in disease-prone populations.
It tests the different estimation models on real-world hospital data. It uses a latent component model
to reconstruct missing data to address the challenge of incomplete data.It was tested on a cerebral
infarction that was a regional persistent condition. It employs the Machine Learning Decision Tree
technique to analyse structured and unstructured data from the hospital. It mines data sets to
anticipate likely diseases and hospitals. To our knowledge, no existing work in the field of medical
big data analytics has focused on both data types. When compared to various common estimate
techniques, our suggested algorithm has a calculation exactness of 94.8 percent and a convergence
speed that is faster than the Decision tree disease risk prediction algorithm.
80. DETECTING FORGED INFORMATION BASED ON MACHINE LEARNING

Dr. V. Geetha1,Dr C K Gomathy2

India
The Beginning of numerous years the technology is developing in many ways along with
development social networks also developed. In these social networks many news is spreading
throughout in this online world and in this news some news the real and which are fake. These fake
news for various fields that are political, commercial purposes has been arriving in huge count and
spreads through online social networks. By these fake news social networks users can get infected
easily, which has bought tremendous impact on society. An important aim is to improve the
trustworthiness of information on social networks is to detect the fake news frequently. This paper
aims to investigate the methods, principles and algorithms to identify the fake news articles, creators
and subjects from social networks and evaluating the performance. Information preciseness on
social media is an increasingly important corner, but web-scale hampers, ability to identify, evaluate
correct the information or so called "fake news", present in these platforms. In this paper, we
propose a method to detect ''fake news" and ways to apply it on social media. The outcome results
may be improved by applying various techniques and methods that are discussed in the paper.
Outcome results suggest, the fake news detection problem can be addressed with machine learning
methods.
81. UNIFIED HEALTH CARE SYSTEM USING WEB FRAMEWORKS

Karthikeyan N
Assistant Professor
Sri Krishna College of Engineering and Technology
Coimbatore, India
Shameer B*, Sukumar E, Rakul V M

UG Scholar
Sri Krishna College of Engineering and Technology
Coimbatore, India
In the advanced period, there is various measure of sickness, crises that are happening step by step.
However, innovation has grown a ton, a few arrangements are fundamental in the clinical world as
May 21st – 22nd, 2022
we can see many individuals experience the ill effects of various medical conditions. As a solution
for this, we have come up with an idea to safeguard people’s life by sitting inside their home, a
unified health care system using web frameworks where we have included technologies like react js,
bootstrap as frontend, node js, and PHP as backend, MongoDB/MySQL as backend. As we
considered associating specialists using video call which can make individuals counsel and get
analyzed and help prescriptions and other information through video call. Alongside this, we have
carried out different highlights which empower individuals to handily get to individuals in touch.
We have provided different solutions in different parts that might include pill reminder, secured
prescription storing, doctor consultant using a toy, ambulance service, video conferencing with
doctors. So that might reduce the problems in these hard times.
82. ALCOHOL DETECTION AND EMERGENCY ALERT SYSTEM

D. Prem Raja
Assistant Professor
Dept. of IT,
Kalasalingam Academy of Research and Education
Tamil Nadu, india
K. Keerthana
Dept. of IT,
Tamil Nadu, india
G. Barkavi
Dept. of IT,
Tamil Nadu, india
V. Vasudevan
Senior Professor
Dept. of IT,
Tamil Nadu, india
S. Aishwarya
Dept. of IT,
Tamil Nadu, india
It is an absolute requirement in today's modern, fast-paced, and insecure environment to be aware of

safety precautions when driving a vehicle. Furthermore, the frequency of vehicle accidents is
increasing at an alarming rate. According to many statistics, the majority of accidents are caused by
the driver's intoxication or tiredness. The main parameters to consider while building the system
include knowing where the vehicles are, what the drivers are doing, and monitoring every
occurrence in real time. Basically, any driver monitoring system is created with the use of sensors
and GSM to continuously watch the driver and issue an alarm if any suspicious behave is detected,
such as excessive speeding, intoxication, fatigue, and so on. This is also an essential step to track
down the driver's driving patterns, which might be useful in making future decisions. From the
May 21st – 22nd, 2022
perspective of truck and state transport bus drivers, this system will be implemented. As we all
know, the TNSTA is losing a lot of money due to high maintenance costs for buses caused by
reckless driving. The deployment of this system, which makes use of all modern sensors and
analysis tools, is critical for keeping track of the driver's driving pattern.
83. DEVELOPMENT OF HUMAN RESOURCE: A CRITICAL ANALYSIS OF INDIA

INNOVATION INDEX
Madhavi Damle1*, Bala Krishnamoorthy2
1* Symbiosis Institute of Digital and Telecom Management, Symbiosis International -Deemed University,
Lavale, Pune, 412115, India.
2Associate Dean (Accreditation), Professor, & Area Chairperson-Strategy, SBM, Narsee Monjee Institute of
Management Studies, Vile Parle, Mumbai, India (NMIMS),
India innovation index by NITI Aayog reports (2020) shows the Index framework. Performance
output from Macroeconomic indicators. These Indicators compared with some of the significant
innovation index studies for the difference in the outcomes and implications. The analysis also
brings out a few challenges identified. The findings of this study are that some of the areas critical as
innovation capital remain unexplored by this India innovation index framework. Knowledge
management has some lacuna, especially for following the Innovation success and outcomes within
the Indian systems. The paper identifies these challenges and problems for future work to address
them.
Furthermore, the findings are the challenges to the present index where there is no recourse to a
regulation mechanism as these are aggregates. Moreover, the inconsistencies in the measures for
yearon-year patterns are further critically observed. The implication of this study is the three
challenges identified. Moreover, human resource development has also been analyzed for five years,
with inconsistencies. The paper critically analyses the implications of the resilient Innovation Index
for India.
84. AN OPTIMAL DEEP WAVELET AUTOENCODER-BASED DEEP NEURAL

NETWORK WITH THE USE OF RIDER CUCKOO SEARCH ALGORITHM FOR
CLASSIFICATION OF LUNG CANCER ON CT IMAGES
R. Tamilarasi
Research Scholar
University of Madras, Chennai-600025
Dr.S. Gopinathan
Associate Professor Department
of Computer Science,
University of Madras, Chennai-600025.
Early detection of lung cancer is the only possible way to improve a patient’s chance for survival. A
Computed Tomography (CT) scan used to find the position of tumour and identify the level of
cancer in the body. The current study presents an innovative automated diagnosis classification
method for CT images of lungs. In this work, the CT scan of lung images was analyzed with the
assistance of Optimal Deep Wavelet Autoencoder-Based Deep Neural Network (ODWADNN).A

May 21st – 22nd, 2022
most accurate, reliable and fast automated method was implemented to segment the liver tumour
image using AcceleratedGreedy snake’s algorithm (AGSA).Next, a technique for image
compression using a deep wavelet autoencoder (DWA), which blends the basic feature reduction
property of autoencoder along with the image decomposition property of wavelet transform is
proposed. The combination of both has a tremendous effect on sinking the size of the feature set for
enduring further classification task by using DNN. Here, the proposed RCSA is employed for
training the DBN. The proposed RCSA is designed by integrating the rider optimization algorithm
(ROA) and Cuckoo Search (CS). A CT image dataset was taken and the proposed DWA-DNN
image classifier was considered to classify lung nodules as normal, benign and malignant.The
discussed method increases the lung cancer prediction rate which is examined using MATLAB-
based results such as Accuracy, Specificity, Precision, Recall and F1 score.
85. A NOVEL METHOD OF APPROACHING TRANSPORTATION PROBLEM OF

HEXAGONAL FUZZY NUMBER
V.Tharakeswari1,1*, M.Kameswari2
1Department of Mathematics, Kalasalingam Academy of Research Education, Krishnankoil,
Tamilnadu, India-626126
1*Department of Mathematics,J.N.N Institute of Engineering, Usha Garden, Kannigaipair,
Thiruvallur,Tamilnadu, India-601102
2Department of Mathematics, Kalasalingam Academy of Research and Education, Krishnankoil, Tamilnadu,
India-626126
The linear programming problem is well-known as one of the most promising mathematical
methods for efficient resource allocation. Many real-world problems can be expressed as LPPs.
Anyways in occurring each day scenarios; it is arduous to obtain appropriate accurate data for the
cost parameter, resulting in a fuzzy environment. The subject of fuzzy transportation has a lot of
consideration in present-day. It assists the decision-maker in arriving at the best answer with
appropriate data, which is frequently used in engineering and management science situations. The
prevalent goal of the transportation problem to reduce the amount of transporting a production from
multiple origins to multiple targets. It is necessary to highlight the issue of balanced and unbalanced
transportation. We used the defuzzification method for hexagonal fuzzy numbers and offered novel
approaches in this research to be helpful for determining an IBFS for balanced and unbalanced
transportation problems.
86. XGBOOST OPTIMIZED BY ADAPTIVE TREE PARZEN ESTIMATORS FOR

CREDIT RISK ANALYSIS
Sujoy Baitalik, Pradeep Sudhakaran
Department of Computing Technologies, School of Computing,
SRM Institute of Science and Technology, Kattankulathur-603203, Chengalpattu, Tamil Nadu, India
Credit Risk analysis of customers protects and ensures Banking institutions’ interests and have been
a core part of any financial system. Traditional Banks have been using decades old legacy systems
for all of their processes and these systems are no match for current Machine Learning and Deep
May 21st – 22nd, 2022
Learning algorithms. XGBoost is a tree-based distributed gradient boosting framework famous for
its efficiency and flexibility. The performance of XGBoost largely depends on its hyperparameters
which can be optimized. Methods like GridSearch and RandomSearch are largely used for
hyperparameter optimization but are inefficient in nature. This paper proposes a XGBoost based
algorithm optimized by Adaptive Tree Parzen Estimators (ATPE) for Credit Risk analysis. The
hyperparameters of the proposed algorithm, when optimized with ATPE give better results
compared to the default model and thus it can be used as a replacement for the legacy systems used
in Banks for Credit Risk analysis.
87. A LIGHTWEIGHT RUBIK’S CUBE AND CHAOTIC-BASED IMAGE

ENCRYPTION AND KEY GENERATION ALGORITHM FOR MULTIMEDIA
INTERNET OF THINGS
1st Arjun Rajasekharan
Computer Science and Engineering ,
Indian Institute of Information Technology, Design and Manufacturing, Kancheepuram
Tamil Nadu, India
2nd Panchami V
Assistant professor, Computer Science and Engineering- Cyber Security, Indian Institute of Information
Technology, Kottayam
Kerala, India
The boom of social media networks, tend to increase the magnitude of multimedia content. This
leads to the evolution of the new field called Multimedia Internet of Things. As these devices are
resourceconstrained, there is a need for lightweight image encryption techniques to safeguard
multimedia data.
The proposed lightweight encryption algorithm is based on Chaotic theory since it become more
prevalent in current multimedia cryptography. This chaotic-based multimedia encryption technique
uses 2D augmentation models for safe data transfer. The proposed technique not only preserves
the high encryption quality of chaotic maps but also have minimal residual clarity and key
sensitivity. The lightweight image encryption algorithm uses the Chaotic map model used for
achieving confusion and diffusion properties. We have also proposed a novel key generation
algorithm used on Rubik’s cube transformation and Logistic map. The initial values are generated
using the Elliptic- curve cryptography (ECC) key algorithm. The proposed system is applied on both
medical images (grayscale) as well as coloured images for analysing the computational complexity.
As per the security analysis, histograms of the encrypted images are generally flat and dispersed
over all pixel values. These pictures exhibit average correlation values of 0.0010575 (horizontal),
0.013994 (vertical), and 0.00235 (diagonal), with an entropy of 7.86046675 (encrypted image).
Therefore, the proposed lightweight image encryption exhibits a high level of security.
88. MUSIC RECOMMENDATION USING EMOTION RECOGNITION

Pranav Kizhakkevellat Nair, Anand Suraj, Pradeep Sudhakaran
Department of Computing Technologies, School of Computing, SRM Institute of Science and Technology,
Potheri, Kattankulathur-603203, Chengalpattu District, Tamilnadu, India.

May 21st – 22nd, 2022
Music has become an integral part of our lives. While emotions may be easy to describe, the
intricacies of emotions are harder to determine when one is experiencing them. Emotions are easier
to understand retrospectively. While experiencing strong emotions, one can easily get overwhelmed
and bring rise to feelings like loneliness, and other unhealthy emotions which can progress towards
mental illness. This project enables one to perceive emotions through music, thereby helping them
realize and acknowledge their emotions. This in turn serves as a medium to help one realize that they
are not alone because ultimately everyone listens to music. An emotion recognition algorithm is
used with the help of CNN, to determine the specific sentiment the individual is currently feeling
and then recommend specially made playlists of songs for them. It is not always evident that a
person going through a tough phase may open up about their feelings to people around them and
hence making use of AI/ML to identify a person’s emotion through their facial expressions captured
over a span of time can help them in getting over their emotions.
89. CLASSIFICATION OF YOGA HAND MUDRAS USING SIFT AND SURF

FEATURES
S. Abarna1, V. Rathikarani2, P. Dhanalakshmi3
1Research Scholar, 2Assistant Professor, 3Professor
1,2,3Department of Computer Science and Engineering, Annamalai University, Chidambaram,
Tamilnadu, India.
Yoga is a unique spiritual science of self-development and self-realization that teaches us how to
live our lives to their greatest potential. Yoga's integrated method creates profound harmony and
unwavering balance in the body and mind, allowing us to awaken our dormant ability for higher
consciousness, which is the true purpose of human evolution. Yoga's various acknowledged physical
and mental advantages have contributed significantly to the popularity of the practise. Due to a lack
of datasets and the requirement to detect mudra in real time, recognising yoga hand mudras is a
difficult undertaking. The yoga hand mudras are used as input in the proposed work, and the two
features Scale Invariant Feature Transform (SIFT) and Speeded-Up Robust Features (SURF) are
retrieved, followed by machine learning techniques namely Gaussian Naive Bayes (GNB) are used
for classification. In the experimental results the performance of SIFT with GNB yields better
results.
90. QUERY CACHING IN BIGDATA ANALYTICS

Dr.C.K.Gomathy 1, K.Vishnu Vardhan 2, Unnam Pavithra 3
1Assistant Professor, Dept. Of CSE, SCSVMV (Deemed to be University), Kanchipuram, TamilNadu, India
2 Student, Dept. Of CSE, SCSVMV (Deemed to be University), Kanchipuram, TamilNadu, India
3 Student, Dept. Of CSE, SCSVMV (Deemed to be University), Kanchipuram, TamilNadu, India
Queries are the heart of analytics. Without them, there is no analytics. Data analysis is helping the
enterprise to make the final decisions on customer trends and predictions and the queries would be
large and take much time to execute on the cluster. This project would help queries to execute on
different platforms based on the configuration and cache the response for future reference by using a
database. The project helps customers to analyze the data faster and they can use it further for
making the decision. In Enterprize data is important. In companies, Decisions are to be made by the
large volume of data. We cannot store the huge volume of data in Relational databases and we can't
May 21st – 22nd, 2022
update it because it is vertical Scala. But Here Hadoop in HDFS is horizontal Scala and it stores a
large volume of data in a distributed manner. Our project helps in transferring data from Relational
databases like MySQL to Hadoop and it will be accessed by the Java Database Connectivity
(JDBC). The accessed data will be converted to the JavaScript object notation (JSON) format by
using the Gson library. The converted JSON format will be stored in a database for future use.
91. DETECTION OF FACE MASK AND DISTANCE CALCULATION USING DEEP

LEARNING
Sarthak Gupta, Pradeep Sudhakaran

One of the non-pharmaceutical strategies for minimising the main origin of COVID spat by any
individual already infected by the virus, is to wear a mask. This study aims to create, real-time
system that is highly accurate as well as is capapble of detecting any non-masked faces anywhere
and force people to wear masks in order to contribute to communal health. In this suggested system,
modern deep learning algorithms were integrated with geometric techniques to develop a robust
modal that covers three aspects of detection, tracking, and validation. As a result, the proposed
approach benefits society by saving time and lowering corona virus transmission. It can be useful in
the current situation, where the lockdown is being lifted to allow people to be inspected at public
meetings, shopping malls, and other locations. By lowering the number of persons required to
inspect the public, automated inspection saves time and money. It can also be used in any situation.
92. PENETRATION TESTING TECHNIQUE WITH REVERSE SHELL ATTACK

Aswathy Mohan
1PG Scholar
Ec-council CEHv11 Certified
Francis Xavier Engineering College,
Anna University, Tirunelveli,
Tamil Nadu, India

IEEE Member: 97736154
Orchid Id : 0000-0002-0293-1656,
Dr. G Aravind Swaminathan

2Professor & Head of Department,
Computer Science and Engineering,
Francis Xavier Engineering College,
Anna University, Tirunelveli,
Tamil Nadu, India
Cybersecurity refers to the technique of using methodologies to safeguard networks from

cybercriminals which results in cyberattacks. Cyber-attacks refer to the non-legal stealing of
confidential data from organization networks by breaking network security. Network Security refers

May 21st – 22nd, 2022
to the methodologies implemented in the network of the organization both internal and external to
oppose intruder attacks. One of such topmost technology used for ensuring network security is
penetration Testing .
93. CONVERTING VOICE SIGNAL TO VISUAL INDIAN SIGN LANGUAGE

Anshul Kataria, Rajveer Mathur, Pradeep Sudhakaran
Sign language is a symbolic representation used by deaf people as their primary language. Unlike
acoustically conveyed sound patterns, sign language makes use of body posture and physical
communications to effectively express a person's words. It could be used to connect with people
who have trouble speaking, as well as non-hearing challenged people. With the development of new
technologies such as online applications, machine learning, and natural language processing, our
initiative intends to bridge the gap between these Deaf people and regular people. The main purpose
of this project is to construct a software/interface which accepts Audio/Voice as input and outputs
them to corresponding Sign Language for ill-hearing people. Hand forms, posture, and body
movements are all coordinated at the same time. The approach has 2 parts: first, it converts speech
into text using voice-to-text API; second, and then it represents the text using Parse Trees and uses
Natural Language Processing semantics (NLTK particularly) for lexical analysis of Sign Language
Grammar.. The work builds upon the rules of ISL (Indian Sign Language) and follows the ISL rules
of Grammar.
94. MODIFIED MUSIC POPULARITY PREDICTION WITH DEEP LEARNING

TECHNIQUES.
Mrs.N. Poornima
Assistant Professor
(Ordinary Grade)
Department of Computer Science
Chennai,
Ms.M. Aishwarya
Chennai,
Ms.T.Aparna
Chennai,
Ms.K.Divya
Chennai,

May 21st – 22nd, 2022
Understanding music quality is very important not just for the artists World Health Organisation
produce and perform music however conjointly for the music-related trade. it's not been studied well
however music quality are often outlined, what its characteristics square measure, and whether or
not it are often foreseen, that square measure self-addressed during this paper. AEFS will choose the
foremost vital options by excavating each linear and nonlinear info among options, that is additional
versatile than the standard self-representation methodology for unsupervised feature choice with
solely linear assumptions. Deep Neural Networks (DNNs) are gaining progressive accomplishment
compared with several ancient Machine Learning (ML) models in numerous fields. Extensive
experiment results indicate that AN improvement algorithmic program will create effects on the
DNN model sensitivity to adversarial examples. That is, once coaching models and generating
adversarial examples, Adam optimizer will generate higher quality adversarial examples for
structured datasets, and Adadelta optimizer will generate higher quality adversarial examples for
unstructured datasets. additionally, the selection of optimizers doesn't have an effect on the
fungibility of adversarial examples.
95. APPLICATION OF GIS TECHNIQUES IN CITY TRAFFIC

*M.Monica Bhavania,Dr.A.Valarmathib
Ph.D Scholar,Anna University,Trichy and 620024,Indiaa
Assistant Professor,Anna University,Trichy and 620024,Indiab
Cities are spatially, culturally, socially, economically, politically, legally, democratically and
ecologically defined as a system. A Spatial System can be obtained from the land usage and urban
networks or services. Cities have the ability to uplift even a nation, modifying a city to a smart city
to make smarter nations by planning, management, governance and development systems. Presently,
planning a system gives the master plan, zonal plan,urban land management schemes and site
planning schemes to arrive at the goals of the spatial system that needs a keen re-look for creation of
smarter city. All the Management and governing systems currently in work are the legacy colonial
with inappropriate hierarchical system that is less responsive and in real-time outmoded
cumbersome for end users nowadays.
96. AN OVERVIEW ON ADVANCED GENETIC DISEASE DIAGNOSIS AND

PREDICTION TECHNIQUES USING GENOME DATA
Anitha. M1, Mahendran Radha1
1Department of Bioinformatics, School of Life Sciences, Research scholar, Vels Institute of Science,
Technology and Advanced Studies (VISTAS), Chennai, India.
1Department of Bioinformatics, School of Life Sciences, Professor & Head, Vels Institute of Science,
A genetic disorder in individuals is caused by the inheritance of two alleles from the parents. This
review focuses on various techniques that are used to diagnose or predict the possibility of a genetic
disorder in patients. The conventional methods of prediction of genetic disorders use family histories
and lifestyle factors, this approach may decrease the prediction accuracy. Therefore, introducing

May 21st – 22nd, 2022
genetic risk score prediction based on SNP will increase the prediction accuracy and reduce the
overall screening time of medical history. These predictions are done by taking a few samples of
blood or sputum from the patient and sequencing the DNA to find the gene patterns. Genetic
disorders can be caused by both dominant and recessive alleles. The prediction is done by finding
the gene in a sequence that is increased or decreased in size; this is called Copy Number Variation
(CNV). There are many studies focused on finding the correlation between the CNV of two different
genomes. Researchers used many techniques to find the correlation between CNVs including
machine learning, signal processing techniques. We carefully analyzed more than 50 peer-review
journals and compared various methods to find the similarity in various techniques.
97. DEVELOPING A NEXT-GENERATION MULTITUDINOUS EPITOPE-BASED

VACCINE AGAINST BIOTINIDASE DEFICIENCY; AN IMMUNOINFORMATICS
BASED APPROACH
Anitha M 1, Mahendran Radha1*
*Department of Bioinformatics, School of Life Sciences, Professor & Head, Vels Institute of Science,
Biotinidase deficiency is a rare genetic disorder brought out by BTD (Biotinidase gene) mutations.
Genetic disorders are not curable but there are drugs available for suppressing Biotinidase
deficiency. To overcome this problem a vaccine that is much effective is needed. The goal of the
current research is to develop an effective multitudinous epitope-based vaccine. Biotinidase protein
was chosen as a target, and various epitopes like B-cell and T-cell were predicted. Predicted epitopes
were shown to be stable, non-allergic, antigenic, stable, and responsive. Finalist epitopes were
significantly antigenic and overlapped. Using In silico Techniques, we predicted the Primary,
secondary and tertiary, stability, MHC class I and II binding properties, and ligand-receptor
interaction for vaccine designing. This vaccine shows high affinity with the human receptors IL-2
alpha and a beta chain. Further the sequence was optimized for expression in a prokaryotic host and
cloned into the pET-28a plasmid. Docking studies further demonstrated that the forecasted peptides
interacted with HLA-B7 allele. The predicted vaccine could be a promising starting point for
vaccine development against genetic disorder. Furthermore, the suggested vaccine must be subjected
to in vitro experiment and further confirm to verify its immunogenic and safety profile.
98. CHARACTERIZATION OF PLANT DISEASE PREDICTION USING

CONVOLUTIONAL NEURAL NETWORK
Dr.C.K.Gomathy1,Kakamanu Jaya Sairam2,Illuru YadhuVamsi3 Dr V Geetha4
1,4 Assistant professor, Dept. of CSE, SCSVMV (Deemed to be university), Kanchipuram, TamilNadu
2, 3 UG Scholar, Dept. of CSE, SCSVMV (Deemed to be university)), Kanchipuram, TamilNadu
Agriculture is one of the main factors that determine the growth of a country. In India itself, about
65% of the population lives on agriculture. Due to different seasonal conditions, plants can be
infected with different kinds of diseases, which can affect the leaves of plants. First, the plant is
infected, and then the whole plant, which in turn affects the quality and quantity of plant growth.
Because there are many plants in the yard, it is difficult for human eyes to identify and classify the
diseases of each plant in the field. Since these diseases can be transmitted, it is very important to
May 21st – 22nd, 2022
diagnose each plant. Therefore, in this document, we introduce the automatic detection and
classification of plant leaf diseases based on artificial intelligence to quickly and easily identifies,
classify and perform the required tasks. Medicine to treat this disease. This is one way to achieve our
goal of increasing agricultural crop yields. In this method, we followed several steps i. Image
acquisition, preprocessing, segmentation and classification of images
99. FACEMASK DETECTION USING VIOLA–JONES ALGORITHM

1 Mr. A. Niranjan 2 Suhana Momeen 3 U. Tabusha 4 U. Vinitha
1 Assistant Professor, Department of Electronics & Communication Engineering, SCSVMV,
Kanchipuram, India.
2,3,4 UG Scholar, Department of Electronics & Communication Engineering, SCSVMV,
Kanchipuram, India.
Here, we introduce a face mask detection model which is based on computer vision. The proposed
model can be integrated with computer or laptop cameras allowing it to detect people who are
wearing masks and not wearing masks. The model has been put together using deep learning and
classical machine learning techniques with openCV, TensorFlow and keras. We have acknowledged
about a comparison between three machine learning algorithms to find the most suitable algorithm
that yields the highest accuracy. Using face mask recognition technique by using python
programming and from OpenCV library Haar cascade method. Object Detection using Haar
featurebased cascade classifiers is an effective object detection method proposed by Paul Viola and
Michael Jones. Another research in which hybrid model using deep and classical machine learning
for face mask detection is presented. A face mask detection dataset consists of with mask and
without mask images, then using OpenCV to do detection from a webcam. For this project Viola-
Jones Algorithm method is used for face mask detection.
100. AUTOMATIC DOORBELL USING ARDUINO NANO
1Mr. A. Niranjan 2V.K. Ramdas Coundinya 3R.V. Sarath Kumar 4T. Ravi Teja
1Assistant Professor, Department of Electronics & Communication Engineering, SCSVMV, Kanchipuram,
India.
2,3,4 UG Scholar, Department of Electronics & Communication Engineering, SCSVMV, Kanchipuram,
India.
In this project, we will be designing an advanced doorbell that ensures safety from COVID-19
pandemic as it is fully touch proof. What If the person who came to our house cannot find the
doorbell or if the person has fever or if they maybe covid affected, what can be done to ensure?
How will it be if we use an automatic doorbell which checks their temperature and gives us
instructions whether it is safe or not to let them inside and ring the bell only when he is good? There
are no more hassles. The person who comes to our house need not search for the doorbell and press
May 21st – 22nd, 2022
it anymore. If we install this automatic doorbell using an object detection circuit on the door itself,
the circuit will automatically sense the presence of the person and ring the doorbell.
101. UPPER LIMB ANATOMY LEARNING WITH VR
Janani S
Student, Information Science and Technology
College of Engineering Guindy, Anna University
Chennai, India
Yoheshwar S
Chennai, India
Keerthana S
Chennai, India
Dr.K.Indra Gandhi
Assistant Professor(Sl.Gr), Information Science and Technology
Chennai, India
Anatomy is the important subject for school level science students and medical students. Upper
Limb anatomy is very complex. Memorising the details about all parts of upper limb is difficult and
need much imagination. Virtual Reality (VR) is the best tool for the enhancement of learning
process of upper limb anatomy. This helps students in the thorough understanding of the upper limb
anatomical structure based on visualisation rather than memorising. Three Dimensional (3D) models
of shoulder, forearm, arm and hand are created with its complete internal structure using Computed
Tomography (CT) Digital Imaging and Communications in Medicine (DICOM). Refinement of 3D
models is done further. Models created are rendered in virtual environment in unity. Anatomical
labelling, audio description about the parts is also added. Finally well refined model is rendered in
virtual environment using which students can learn the internal structures effectively. So our
ultimate goal is to develop a VR application where students can learn upper limb anatomy through
3D models.
102. NEWS CLASSIFICATION FOR INDIAN LANGUAGES BASED ON THEIR

HEADLINES USING MACHINE LEARNING AND DEEP LEARNING MODELS
Nagani Vrudant Gopalbhai

Kattankulathur, India
Vishal Parekh
May 21st – 22nd, 2022

Anita Ramalingam
During the last few years, text mining has become increasingly significant. Users may now access
news from a number of sources, including electronic media, digital media, print media, and many
more. In this paper, we focused on text classification, classifiers, and a variety of feature extraction
strategies, but only for short texts, such as news categorization based on headlines. There are few
resources for all Indian languages, and the majority of them are not frequently utilized in text
classification. Existing classifiers are mostly focused on accurately recognizing English news
headlines, with minimal attention paid to Indian languages.
This paper presents an extensive comparison of predefined models vs a CNN model for English,
Hindi, and Gujarati news headlines. Our experimental results showed solid performance of all
models on the English dataset with a minimum accuracy of 89.33%, achieved by RFC, and top
performance of 95.04%, achieved by MNB. As for Gujarati dataset and Hindi dataset, CNN
achieved the highest overall accuracy of 89.12% and 87.58% respectively. All the models can
classify headlines into 5 categories for all datasets.
103. A/B TESTING FOR INDUSTRIAL COMAPANIES
Anjali Agrawal
B.Tech.
of SRM Institute of Science and
Technology
Chennai, India
Dr. S. Prabakeran
Department Of Networking And
Communications of SRM Institute of Science
and Technology
Chennai, India
Preeti Vijayan
Technology Consulting - BI & Analytics
Tiger Analytics
Chennai, India
User-intensive software, like mobile applications and websites, heavily depends on interactions with
many users and an unknown population. With the internet connectivity on such software, the
website allows evaluating ideas and innovations using continuous experiments like A/B tests, split
tests. We aim to study the application of A/B testing in various industrial contexts. We will also
present a brief study on different statistical tests for different assumptions and solve them using the
Bayesian algorithm. Randomized algorithms are used to address various software engineering
problems. This type of algorithm gives different results with every run for the same problem
instance. Therefore, a statistical test is important to prove the conclusions derived from the data.
May 21st – 22nd, 2022
104. IMPROVED DIGITAL TWINNING TECHNIQUES FOR SMART SIMULATIONS

Dr.SHAIK JAVED PARVEZ, AKASH VENKATESAN, VANRAJ D
HINDUSTAN INSTITUTE OF TECHNOLOGY & SCIENCE,CHENNAI
Digital twinning is a visual representation of a visual object designed to reflect exactly. The object
being investigated - that is The data is collected from sensors to generate energy and other aspects of
material activity. This data is later sent to a processing machine, where it is used in digital copy.
This paper proposes a method where digital twinning uses machine algorithm with 5G Network and
it also uses simulation program that allows replicate the orginal product or source. It creates a virtual
to physical twinning making it more effective.
105. MEDICAL CHATBOT USING MACHINE LEARNING AND NLP

B.E.S.M. VISWANADH
Department of Computer
Science and Engineering.
Hindustan Institute of Technology and Science.
Chennai, India.
PRISKILLA PERKINS A
Department of Computer Science and Engineering.
Hindustan Institute of Technology and Science. Chennai, India.
M. HANUMANTHA RAO
Dr.MAGDALENE R
Chatbots are software developed using machine learning to stimulate a conversation with a user in
real time. Chatbots have important role in news, social media, transactional and customer service.
Users engage with chatbots every day on their devices even without realizing. Chatbots can provide
news updates, sport updates and bank applications. We can implement the chatbots in medical
industry for patient care and health guidance. Chatbots will improve medical care delivery and will
also lead to significant healthcare cost savings.
106. ASTHMA PREDICTION USING MACHINE LEARNING CLASSIFIERS
Swetha.P1, Joshua Amrish.A2, G. Gnancy Subha3, Dr. Chitra Sudhakaran4, Raveena.R5, Parveen Meera.S6
1,2,5&6Department of Biomedical Engineering, SPIHER.
3Assistant Professor- Department of Biomedical Engineering, SPIHER.
4Associate Professor- Department of Biomedical Engineering, SPIHER.
Asthma is a lung illness caused by inflammation and airway obstruction that produces shortness of
breath, wheezing, and often life-threatening attacks. Patients with asthma must be able to predict
severe exacerbations caused by uncontrolled asthma. As a result, a framework that can accurately
May 21st – 22nd, 2022
forecast the likelihood of asthma in a patient is required. By mining the data comprising patients'
previous health records, an effective mechanism for asthma disease prediction has been developed in
this research. To display the results, we employ machine learning algorithms. KNN, ANN, SVM,
and Linear Regression are four machine learning classification algorithms employed in these trials to
predict asthma disease at an early stage. The efficiency of all four models is assessed using a variety
of metrics. The accuracy of a classification system is measured by the number of cases that are
classified properly and inaccurately. The results of the experiments demonstrate that SVM
techniques have the highest accuracy. MATLAB software is used to display the outcomes of the
experiments.
107. A NOVAL APPROACH FOR WSN USING SHORTEST PATH (DIJKSTRAS

ALGORITHM)
Dinesh Kumar J R
Assistant Professor Department of Electronics and Communication
Engineering Sri Krishna College of Engineering and Technology, Coimbatore,
India.
Pranesh K
Department of Electronics and Communication Engineering
Sri Krishna College of Engineering and Technology,
Coimbatore, India.
Rahul S
Coimbatore, India.
Sanjeev C P
Coimbatore, India.
To realize productivity of delaying the existence over sensor networks several plans hold been
proposed. Among this plans, a bunching convention is an wonderful approach up to expectation
drags out the existence on an organization. Notwithstanding, in applying that technique, a few hubs
burnthrough energy superfluously so a result over a local weather whereby the gathered records on
the sensor hubs effectively covers. In that paper, we endorse a Clustered Multi-hop Routing
Algorithm as decreases superfluous facts transmission amongst hubs by means of but the duplication
of information. This strategy forestalls data misfortune added respecting by means of connect
disappointment problem then consequently the records is gathered dependably. As indicated via the
consequences on the present examination, our approach lessens the electricity utilization, expands
the transmission proficiency, then delays community quick so contrasted with the cutting-edge
bunching strategies.
108. IOT BASED SMART CHARGING SYSTEM

Dr.R. Devi,
Asst. Professor Electrical and Electronics Engineering
May 21st – 22nd, 2022
Sri Krishna College of Technology

Coimbatore, India
Akash G
Electrical and Electronics Engineering
Coimbatore, India
Dinesh Kumar.M
Coimbatore, India
Hariharan D
Coimbatore, India
Coimbatore, India
This project, Smart Power Monitoring and Control Systems, is one of the few ways to use the
Internet of Things. Thanks to the Internet of Things, energy-efficient gadgets can be created,
reducing both energy and human use. As part of this project, we aim to develop a smart charging
system that improves battery life by reducing battery pressure and automatically controlling its use.
web server and battery charged. On the IoT platform, there is an ESP8266 module, and the web
server is stored in the module. Because webserver is platform-agnostic, it can be used on any device,
including smartphones. These devices send commands to the web server. This project helps with
automatic monitoring of the device and notification of the smartphone, and thus saves both the
energy and time required to maintain the equipment. The battery can be charged once the device is
connected, and it can be charged only when necessary, and the system can monitor other. Electrical
items. Battery life may be extended by minimizing the fast charge by switching over the
corresponding connection to the series.
109. TRAVALONG– A TOURISM MANAGEMENT WITH IMAGE SEARCH SYSTEM

USING PIXABAY
Dr.Arun manicka raja, CSE,
RMK College of Engineering and Technology,
Thiruvallur,Tamilnadu.
Abinayaasri M, CSE,
RMK College Of Engineering and Technology,
Thiruvallur, Tamilnadu
Anitha C, CSE,
RMK College Of Engineering and Technology
Keerthana S, CSE,
RMK College Of Engineering and Technology
The main purpose of “Online tour booking website” is to provide a convenient way for a customer to
book hotels, flight, train and bus as a single package for tour purposes. A customer finds it very
May 21st – 22nd, 2022
difficult to search for the best package as he/she has to browse multiple websites, contact many
travel agents, etc. which is a tedious process and is time consuming. There should be a system where
the user should find the best package on the Internet with a single click. To address this issue, we
created an online tour booking website which offers the best package among all the other packages
that are on the web. A customer will select a travel package for a particular place based on the
recommendations provided by the previous customers who had experience with the package. If the
customer likes to visit a particular place, they can easily search the place on the website. They can
also suggest a customized package and send it to us through the contact page. We included a map
view for the specified route. The website is very user-friendly which helps the user to book easily.
They can directly contact us through the mail facility given on the contact page.The objective of this
project is to develop a system that automates the processes and activities of a travel agency. In this
project, we will make an easier task of searching places and for booking train, flight or bus.This
application is to provide best traveling services to the customers and travel agents. We have
developed tours and travel management systems to provide a search platform where a tourist can
find their tour places according to their choices. This system also helps to promote responsible and
interesting tourism so that people can enjoy their holidays at their favorable places. This system also
helps to develop tourism with different cultures so that they enrich the tourism experience and build
pride. We develop this system to create and promote forms of tourism that provide healthy
interaction opportunities for tourists and locals and increase better understanding of different
cultures, customs, lifestyles, traditional knowledge and beliefs. This system also provides a better
way to connect with various events.
110. ML ASSISTED LOCALIZATION ALGORITHMS FOR INDOOR/OUTDOOR

SETTINGS: AN INTROSPECTION
Chirantan Ganguly1, Sagnik Nayak1, Anil Kumar Gupta2, Suresh V.2, Pradeep Kumar CH2
1 Institute of Radio Physics and Electronics, University of Calcutta, Kolkata, West Bengal, India
2Centre for Development of Advanced Computing (CDAC), Pune, Maharashtra, India
The development of novel technologies has led to an upsurge in the demand for precise localization
in outdoor as well as indoor environments. As GPS continues to provide sufficient positioning
precision in outdoor settings, its accuracy falters in indoor scenarios, which necessitates the
development of novel positioning approaches that function accurately both indoors and outdoors.
The use of various WLAN parameters for localization has been conceptualized. In this study, we
attempt to do localization using Machine Learning methods on WLAN RSSI measurements. We
compare the performance of multiple machine learning algorithms on the dataset to see which can be
used to design efficient future localization systems. The proposed study has achieved second place
for the problem statement "ITU-ML5G-PS-016: Location estimation using RSSI of wireless LAN"
in AI/ML in 5G Challenge 2021 organized by the International Telecommunication Union.
111. EVALUATION OF THE FUNCTIONAL ROLES PLAYED BY THE

DORSOLATERAL PREFRONTAL CORTEX IN HUMAN COGNITION
Sujatha Ramesh1, Sanjana Iyer2 and K. Natarajan3

May 21st – 22nd, 2022
1 Post-Graduate Student, Lovely Professional University, Phagwara,

Punjab 2 Hiranandani Foundation School, Powai, Mumbai,
Maharashtra 3 ISRO (Retd.), Tiruvananthapuram, Kerala
Investigation of human intelligence has assumed considerable importance to the cognitive scientists
during the recent past, since executive control is the single biggest differentiating quality of the
human brain, even during early childhood education. It is housed in the neo-cortex, the area that has
developed most recently in our evolution. Specifically, the large and significant complexity
developed in the prefrontal cortical area (PFC) of the brain is a unique feature of humans, as
compared to even the great apes. Within the lifetime of a human being, the PFC in human babies
has been considered to evolve later than the more primitive parts of the brain – it happens in early
childhood and into adolescence, as demonstrated by the relative lack of self-control exhibited by
babies and toddlers, but with increase of such control through the years of growth. While the PFC is
broadly responsible for executive control of all sorts, specific functional areas doing specific roles
have been delineated by studies using modern imaging techniques. Even so, there is some
variability in findings and indeed even some conflicting results. This paper has evaluated the various
roles played by the dorsolateral Pre-frontal Cortex (dlPFC) in the human brain, based on the
extensive survey of reliable published data on the subject. In young children, the role of the dlPFC
is evident in the form of temptation-resistance and aggression-regulation. Understanding of the
specifically delineated roles of the dlPFC is especially relevant now given the exposure of even very
young children to social rejection, making them vulnerable to greater aggression as a defensive
response. This investigation has clearly revealed that the dlPFC has been repeatedly implicated in
certain executive functions, while variably implicated in some others. Based on the preliminary
study, an attempt has been made to delineate the ones that have the most likely direct relationship
with the dlPFC. For adults, this has implications in the study of depression, treatment options for
Alzheimer’s, autism, Parkinson’s disease, and other prevalent neurological deficits. During early
childhood, it potentially provides the basis for helping children manage their emotional responses
and develop controlled, well-deliberated behavior in response to social challenges.
112. RECENT TRENDS IN NEURAL NETWORK-BASED APPROACHES TO

VISUAL PROCESSING VIS-A-VIS THEIR BIOLOGICAL CORRELATES IN THE
BRAIN
Sujatha Ramesh1, Sanjana Iyer2 and K. Natarajan3
1 Post Graduate Student, Lovely Professional University (LPU), Phagwara,
Punjab 2 Hiranandani Foundation School, Powai, Mumbai, Maharashtra
3 ISRO (Retd.), Tiruvananthapuram, Kerala
The rapid pace of recent advances in the realm of Artificial Neural Networks (ANN) and the
simultaneous developments in the space of brain function study have led to a dual-development of
neurosciences and neural networks in tandem. Several instances exist where the two realms have
taken inspiration from each other to make large leaps of progress. In this paper, we have analyzed
the recent convolutional neural network (CNN) algorithms that provide insights into the workings
of the biological brain in visual perception. We have also identified the challenges and possibilities
of interdisciplinary work through symbiotic collaboration between these disparate fields of research.
May 21st – 22nd, 2022
113. EXPLORATIVE STUDY ON CHILDREN’S INTERNET USAGE AND

PARENTAL CONTROL IN INDIA – USING CLASSIFICATION TREE MODEL
ANALYSIS
*K. Periyakannan and ** Dr. Arulchelvan Sriram
Department of Media Sciences, Anna University, Chennai, India
In this period of innovation, the internet has spread broadly and has been utilized for various
purposes by children. Particularly usage of the internet among school students increased in recent
years. At the same time, researchers indicate that there is an increase in internet usage and
behavioral problems. The objectives of this study are to examine the impact of internet usage and
parental mediation on school children. The participants were determined with purposive sampling
and conducted under different schools children in Chennai (N= 424) respondents and from the 14-17
age groups those were analyzed by CART used to describe decision tree algorithms that are used for
a classification tree model and regression. These results show that most children used the internet
without parental guidance. During pandemics, time children are using more than hours on the
internet and social media which impact their online activities of children. This study identified to
regular monitoring is expected to avoid psychologically harmful effects. The future lives, health, and
psychological changes of children, and online abuse.
114. A REVIEW PAPER ON DIMENSIONALITY REDUCTION TECHNIQUES

Mr. Faizan Riyaz Mulla
Bachelor of Engineering in Computer Science, AISSMS College of Engineering
Pune, India. 411007.
Dr. Anil Kumar Gupta
Centre for Development of Advanced Computing (C-DAC)
Pune, India. 411007.
Dimensionality Reduction (DR) is the process of reducing the numerous features or random
variables under consideration to a limited number of features by obtaining a set of principal
variables. These techniques cater great values in machine learning, which come in handy to simplify
a classification or a regression dataset, thereby yielding a better-performing predictive model.
Techniques used for DR include Feature Selection methods, Matrix Factorization, AutoEncoder
methods, and Manifold Learning. Merits of DR include data compression, reduced space of storage,
and removal of redundant features. This paper attempts to review various techniques used to carry
out dimensionality reduction while providing an exhaustive comparative study over the merits and
demerits of each of the techniques used under the empirical experiments performed by the authors
whose work is being reviewed.
115. SOLARPANEL DUAL MANAGEMENT SYSTEM

First A. Anandh Krushna S K, Second B. Deepananth K, Third C. Praveen Kumar S, Fourth D. Nithish
Bharathwaj K R, Fifth E. Lekshmi Kalinathan, Sixth F. Nagarajan K K
Sri Sivasubramaniya Nadar College of Engineering, Kalavakkam, Tamil Nadu, India

May 21st – 22nd, 2022
Solar Energy is a renewable source of energy and it is easy to extract electric energy from it. But
many factors such as dust deposition, size of solar panels, affect the efficiency. Mainly, the
deposition of dust on the surface of the panel reduces the output power. At present, regular manual
checking is required to monitor the accumulation of dust on the solar panel that causes a drop in
output voltage. Current theft prevention system has limitations like alerting the consumer even when
insignificant objects pass through the mechanism installed. This inefficient performance may lead to
theft and loss in investment. If above-mentioned problems are given proper attention, we can avoid
theft and prevent loss of power generated from solar panels. This paper gives a detailed review on
Solar Panel Dual Management System which can detect unauthorised activities and automates the
detection of dust deposition on solar panel.
116. SOCIAL IMPACT ASSESSMENT IN PROTECTED AREAS OF PANNA

NATIONAL PARK, MADHYA PRADESH
Ms. Sushmi Nimje
Architecture and Planning department, MNIT, Jaipur, India
According to The International Ecotourism Society, Ecotourism is defined as “responsible travel to

natural areas that conserves the environment, sustains the well-being of the local people, and
involves interpretation and education”. Some recent studies strongly suggest ecotourism as a tool for
environmental conservation and the socio-economic development of the local community. India
being a biodiverse nation, has various potential destinations for ecotourism development, such as
Panna National Park, India. This study focuses on understanding the local community of Panna
National Park by conducting a social impact assessment and the effect these tourist places have on
the development of these communities. The assessment will help develop recommendations and
strategies that will help maintain environmental integrity and sustainability with the involvement of
the local community of this ecotourism destination. The study describes the perceived benefits and
costs of ecotourism mainly depend on the level of community participation in ecotourism activities
and the availability of resources. In this context, the study aims to critically explore the local
environmental issues, dynamics of tourism planning, and its relationship with and contribution
towards sustainability in the Protected Areas of Panna National Park.
117. ATTRIBUTE BASED ACCESS CONTROL MODEL FOR EHR

Dr.K.Gurunathan1,G.Arun2,B.Karthick3,R.Srikathirvel4,S.Sugi5
1 Associate Professor, Department of Information Technology, M.Kumarasamy College of
Engineering,Karur,TamilNadu,India-639113
2,3,4,5 UG Students, Department of Information Technology, M.Kumarasamy College of
Engineering,Karur,TamilNadu,India-639113
The transmission of digital medical images for e-health services has recently piqued people's
curiosity. However, current implementations of this service do not place a high priority on patient
confidentiality and data security. In the medical field, the necessity for quick and secure
communication is critical. Nowadays, picture transmission is a daily occurrence, and it is vital to
establish an efficient method of doing so via the internet. We present a new technique for ciphering a
picture for safe and denoised transmission in this study. Image cryptography, data hiding, and
May 21st – 22nd, 2022
Steganography are all discussed in this paper. We introduce a new medical image Steganography
technique for securing patients' sensitive information by embedding it in the image itself while
keeping excellent image quality and embedding capacity, and we use LSB substitution based nine
pixels differencing in this project. In addition, a hybrid paradigm based on key there in public sphere
Steganography and Elli Curve Cryptography (ECC) have already been developed. Which
approaches are much more secured as using an unique ECC or steganography.
118. IMPLEMENTING FUZZY LOGIC TO IMPROVE THE NETWORK SELECTION

PROCESS FOR HETEROGENEOUS HANDOVER
Priyanka A*, Jayachithra N, Dr Chandrasekar C
Department of Computer Science, Periyar University, Salem, Tamil Nadu, India.
Ubiquitous user demands are increasing and they commonly opt for networks that always offer a
pleasant seamless connection. A well-known and vital requirement is an unbroken drift of voice and
data with transceivers. In a state of affairs just like the non-stop mobility of the mobile terminal, the
attached access point may relinquish its connection because of the fluctuation of signal power. A
single operator to handle the network connection for the mobile terminal is not capable to fulfill the
user demands when they are in high mobility. So the handover process is crucial for seamless
communication. The mobile terminal's vital task is to select the best network from the number of
available networks. In the network selection process right here we use a combined technique called
Fuzzy TOPSIS (Fuzzy Technique for Order Preference by Similarity to Ideal Solution) and Fuzzy
AHP (Fuzzy Analytic Hierarchy Process). Different parameters are also considered to recognize
which one is best to carry the devices.
119. MEGAPSO: MOBILITY AWARE ENERGY EFFICIENT HYBRID GA & PSO

BASED CLUSTERING FOR MANET
Priyanka A*, Jayachithra N, Dr Chandrasekar C
Department of Computer Science, Periyar University, Salem, Tamil Nadu, India.
In this work, hybrid GA & PSO is used to group the mobile nodes based on the similar behaviour in
MANET. The devices in this network have the freedom to go anywhere in any direction at different
speeds. The unpredictable movement of a node makes a non-stable network, loss of energy, and
repeated changes in topology. This migration behaviour of the devices leads the network to become
dynamic. Non-stability may cause link failure, packet loss, repeated root-finding overhead, energy
loss, topology change, routing table update, etc. To make the network turn out to be standard, cluster
formation is proposed to assemble a number of individuals into groups. Balanced clusters are
required to give a long life for the cluster members & clustering can achieve scalability. Clustering
is organizing things into meaningful groups based on the similarities among the devices. In this
grouping system, there is a cluster head selected for each & every group based on the parameters
like mobility, distance, energy, degree of connectivity, and transmission range. The cluster members
are associated with the CH based on the mobility of the individuals. The Mobility Aware Energy
Efficient Hybrid GA & PSO is proposed here to form the cluster.
120. WATER CONSUMPTION IN THE POWER SECTOR IN INDIA - PROBLEMS

AND PROSPECTS
Ms Sushmi Nimje1, Dr Niruti Gupta2 and Dr Yash Kumar Mittal3
May 21st – 22nd, 2022
1Architecture & Planning department, Malaviya National Institute of Technology, Jaipur, India
India is a water-stressed country where demand will outstrip availability by a ratio of two by 2030,
putting hundreds of millions of people at risk of acute water scarcity (Niti Aayog). The power sector
in India is heavily reliant on water. Thermal power plants generate over 80% of India's electricity
(CEA 2017), and they rely on water for cooling. Another 10% of electricity is produced by
hydropower plants, which purely run on water. Any modifications to either the energy or the water
system are likely to affect the other system directly or indirectly. Water and energy are intrinsically
linked. There are multiple cases where power plants have been forced to shut down permanently or
temporarily due to a lack of water. As the population grows, so does the demand for energy, putting
strain on our water supplies. This paper aims to identify the problems and prospects in power sector
in relation to water consumption. There are few rules and regulations concerning water usage in the
power sector. There is limited policy for water consumption in power plants and no perspective for
tackling water scarcity in the future; hence, the study on water energy nexus becomes vital for
managing water resources across all countries.
121. RFID BASED WIRELESS CHARGING STATION FOR ELECTRIC VEHICLES

USING ARDUNIO AND PAYMENT SYSTEM
Dr.Kyung Tae Kim
Kalasalingam Acadamy of Research and Education
Krishnankoil, India
Aleti Saimanohar Reddy

Krishnankoil, India
Naddiboyian Naga Surendra

Krishnankoil, India
Mallireddy Ashok Reddy

Krishnankoil, India
In this paper, we are going to propose “iot&rfid wireless vehicles charging using arduino Due to
increase in price of petrol and to decrease ecological contamination, the utilization of electric
vehicles (EV) is quickly expanding, rather than ignition motor vehicles. The primary concern issue
of the EV clients is being tended to through different charging strategies. The vital trouble in
execution of EV is course of action of charging framework. The remote charging framework (WCS)
is a most loved choice in the developing EV market.
As the new time of the auto, the business is quickly changing from an IC motor vehicle to an electric
vehicle. The interest for an electric vehicle is expanding, these lead to an expansion in charging
May 21st – 22nd, 2022
station too. In this task, a remote charging framework is utilized to charge the vehicle remotely by
means of inductive coupling. we essentially need to leave the vehicle on the charging spot. The
transmission of electrical energy from source to stack from a distance with next to no directing wire
or links is called Wireless Power Transmission. This framework doesn't need any human
cooperation. The aftereffect of this venture is we can charge our vehicles remotely by means of
inductive coupling and can transfer the vehicle numbers to the cloud which are put away in RFID
labels. Remote power transmission may be one of the advancements that are one stage towards
what's to come. This undertaking can open up additional opportunities of remote charging that can
use in our day-to-day routines.
122. ASSESSMENT OF VARIOUS MILK PARAMETER MONITORING SYSTEM

USING AN IOT
1Mr. Jenyfal Sampson , 2J.Santhosh kumar, 3P.Aswin Bharathi, 4 B.Subash Raj,
1,2,3,4Electronics and Communcation Engineering, Kalasalingam Acamedy Of Research and Education,
Tamil Nadu, India
The project presents a modern device for Assessment of various milk parameter monitoring system
using an IOT. As milk is the major food for all the infants, it has to be monitored for the safety of
the child and old ages people. The main objective of the project is to bring out the product (i.e.
compact device) which determines the quality and the safety of milk for consumption. This project
determines many parameters of milk by using five sensors. Here, we consider parameters like
temperature , pH, Color, Gas, Air sensors to determine the quality of the milk. The Temperature
sensor is used to determine the hotness or coldness of milk. The pH sensor is used to determine the
pH of the milk (i.e. whether Fat, Ammonia, Parteurised or Fresh or Rotten in its nature). The Gas
sensor is used to determine the smell of the milk. The Air sensor is used to determine the ethanol
level of the milk. The Color sensor is used to identifying the colour milk. By using the temperature ,
pH, Color, Gas, Air sensors we can get output by displaying using LCD. By using tempature sensor
Blynk Application will send report to the mobile.
123. ONLINE FOOD ORDERING SYSTEM

Sruthi R, Vellore Institute of Technology, Chennai
This project proposes the Online Food Ordering System. The case study aims to design and develop
a database maintaining the records of users. Our website provides management of users and their
orders along with the payment details and the delivery partners assigned bringing in an ease in the
entire functioning of online food ordering systems. By keeping efficient tracks of who was assigned
to deliver which order and what was the order amount along with customer details it first helps keep
a track of service and sale, and other, helps to manage complains, refunds and salaries of the deliver
agents.
124. DATA AUGMENTATION TECHNIQUES FOR DIAGNOSIS OF

KERATOCONUS USING CONVOLUTIONAL NEURAL NETWORKS
Subramanian P and Ramesh G P
St Peters Institute of Higher Education and Research, Chennai, India

May 21st – 22nd, 2022
Keratoconus is a medical condition wherein the cornea attains a conical shape caused by the thinning
of the corneal stroma. This thinning of the stroma results in irregular astigmatism, short sightedness
and corneal protrusion causing deterioration in the vision and even blindness. Symptoms depend on
the stage of the keratoconus with the early stages getting unnoticed. The advanced stages, though,
can be identified by vision loss and protrusion. Diagnosis and classification of keratoconus has
gained significance with the advent of imaging technologies and machine learning. Machine
learning when applied to the automatic diagnosis or screening of Keratoconus requires a large
dataset, for better accuracy, which is difficult to obtain. Data augmentation techniques have been
employed to increase the size of dataset. This paper presents a review of the augmentation
techniques available in the literature and compares the performances for arriving at a suitable
technique for improved accuracy of classification.
125. IMPLEMENTATION OF DEEP NEURAL NETWORK MODEL FOR DETECTION

OF HUNTINGTON’S DISEASE IN HUMAN DNA SEQUENCE
G.Tamilpavai, C.Vishnuppriya
Department of Computer Science and Engineering,
Government College of Engineering, Tirunelveli, Tamil Nadu, India
Huntington Disease (HD) is one type of neurodegenerative disorder which creates problems like
psychiatric disturbances, movement problem, weight loss and problem in sleep. It needs to be
address in earlier stage of human life. Nowadays Deep Learning (DL) based system could aid
physicians for providing second opinion in patient’s disease treating. In this work, human Deoxyribo
Nucleic Acid (DNA) sequence is analysed using Deep Neural Network (DNN) algorithm to predict
the HD disease. The main objective of this work is to identify whether the human DNA is affected
by HD or not.
Human DNA sequences are collected from National Center for Biotechnology Information (NCBI)
and synthetic human DNA data are also constructed for process. Then numerical conversion of
human DNA sequence data is done by Chaos Game Representation (CGR) method. After that,
numerical values of DNA data are used for feature extraction. Mean, median, standard deviation,
entropy, contrast, correlation, energy and homogeneity are extracted. Additionally, the following
features such as counts of adenine, thymine, guanine and cytosine are extracted from the DNA
sequence data itself. The extracted features are used as input to the DNN classifier and other
machine learning based classifiers such as NN (Neural Network), Support Vector Machine (SVM),
Random Forest (RF) and Classification Tree with Forward Pruning (CTWFP). Five performance
measures are used such as Accuracy, Sensitivity, Specificity, Precision and F1 score. DNN, NN,
SVM, RF achieve 100% accuracy and CTWFP achieves accuracy of 87%.

May 21st – 22nd, 2022
126. DELAY OF NETWORK ISSUES AND CHALLENGES

S. KAVITHA1, Dr. R. KANIEZHIL2
1Research scholar(P.T), Department of Computer Science, Loyola College of Arts and Science College,
Periyar University.
2PRINCIPAL, MIT College of Arts &Science for Women, Musiri
The demand for improved and more rapid communication is growing because of the
widespread use and advancement of wireless communication in today's world. People's
need to send data as quickly as possible creates an automatic problem with spectrum
shortage. The network is heavily utilised during this period, resulting in increased
network traffic and resource strain. The utilisation of the network is delayed as a result.
The node became busy for the reasons mentioned above, which led to traffic. To spread
the workload and ensure that each node is balanced in this scenario, the load balancing
technique is used. However, the use of wireless communication has grown in recent
years, and load balancing is no longer sufficient to prevent traffic and impede network
performance.
127. IMPLEMENTATION OF DATA INTEGRITY EVALUATION PROCESS IN

CLOUD COMPUTING
Dr.R.Poorvadevi1
Assistant Professor, Department of CSE,
SCSVMV University, Kanchipuram
Tamilnadu, India
M.Tejasri2
UG Student, Department of CSE
Tamilnadu, India
N.Ramya3
UG Student, Department of CSE
Tamilnadu, India
DML stands for Distributed Machine Learning. One of the most important AI technologies is DML
(distributed machine learning). Nonetheless, in the current distributed machine learning
arrangement, data integrity isn't taken into account. However, if network attackers alter or delete the
data, the data will be falsified. As a result, it's vital to maintain data integrity in the DM. It proposes
a distributed machine-learned ness initiates data integrity checking approach to verify that the
training data is correct. To begin, it recommends using a Provable Data Possession (PDP) selection
checking approach to guarantee data integrity. As a consequence, the DML-DIV strategy will be
able to withstand manipulation and fraud attacks. Second, as part of the TPA verification procedure,
researchers generate a unique integer known as the blinded variable and utilize the discrete
logarithm problem (DLP) to build proof and ensure data security. It uses two-step authentication and
identitybased cryptographic key generation technology to activate the data possessor's public/private
pair of keys. As a result, our DML-DIV method may be able to address the important escrow issue
May 21st – 22nd, 2022
while also cutting document maintenance costs. Lastly, Finally, a thorough theoretical examination
is carried out, with formal theoretical analysis and practical evidence confirming the DML-DIV
system's safety and efficacy.
128. A REVIEW ON IMAGE CLASSIFICATION USING DEEP LEARNING

Anand Vijay Rajsri1, Dr. Abhishek Gupta2
1,2 Shri Mata Vaishno Devi University,India
Deep learning is currently an interesting topic in machine learning and there are many benefits
associated with Convolutional Neural Networks (CNN). A wide range of computer vision tasks have
been successfully completed by CNNs trained with deep learning algorithms. A CNN has shown
good performance in many areas, such as image processing, classification with the continuous
development of deep learning technology. CNN has recently received awards from image
recognition and classification competitions, providing good results compared to other networks. We
summarize the various CNN approaches that have been used to classify images. Furthermore, we
focus on the optimal configuration of these deep learning approaches according to the quality of the
overall result. The review concludes by describing the challenges of deep learning in image
classification as well as a future direction of research. The evaluation shows that deep learning is
likely to remain the dominant approach to image classification.
129. GIS-BASED QUANTITATIVE ASSESSMENT OF PUBLIC GREEN SPACE

DISTRIBUTION IN LUCKNOW CITY, INDIA
Jyoti Yadav1, Niruti Gupta2
1&2 Malaviya National Institute of Technology Jaipur
Green and blue spaces are fundamental for urban settlements to maintain the ecosystem quality.
However, public green spaces are crucial in enhancing the quality of life of citizens. Rapid
urbanization has imposed tremendous pressure on available land resources and led to the
encroachment of existing open and green spaces within and around cities. Consequently, the per
capita green in cities is reducing across the globe and pressure on public green spaces has increased.
Most developing cities are unable to fulfil the green space norms defined by their own. The study
derives the per-capita space requirement for publicly accessible green for million-plus cities in India.
Further analyses the per capita availability of public green space to assess the service level gap in
Lucknow city. The results show that the areas with higher density have lesser per capita share,
whereas areas with lower density have comparatively higher per capita public green space.
130. DESIGN AND IMPLEMENTATION OF U-SLOT MICROSTRIP PATCH

ANTENNA USING CO-AXIAL FEEDING TECHNIQUE FOR WLAN APPLICATIONS
1st A.Dineshkumar , 2nd K.Immanuel , 3rd Mr. T. Siva, 4th Mr. D. Suresh
Dept of Electronics and Communication Engineering St.Joseph’s Institute of Technology
Chennai, India

May 21st – 22nd, 2022
U-slot patch antenna design and implementation are discussed in this paper for 5.5GHz WLAN use.
The VSWR and S11 of this suggested construction, which is based on FDTD techniques to attain the
desired specification, are computed using the Computer Simulation Technology (CST) microwave
studio. The maximum bandwidth while utilizing coaxial feed is 260 MHz, which corresponds to the
WLAN frequency range of 5.36-5.62 GHz, and at its center resonant frequency, the gain is 5.3 dB.
This antenna is designed to be used in conjunction with the coaxial feed antenna. In order to obtain a
stable VSWR and low return loss, a frequency band adjacent to the present system must be used.
131. EVALUATION OF CORE POINT DESCRIPTORS ENFORCED IN THE

PEDESTRIAN IDENTIFICATION IN LOWER QUALITY IMAGES
Aswathy RH1 ,Dinesh Arjun2, Hari Prasath3, Jaichiddharth4
Assistant Professor1, Department of computer science and engineering, KPR Institute of engineering and
Technology,Coimbatore,Tamilnadu, India
Student2,3,4,Department of computer science and engineering, KPR Institute of engineering and
Technology,Coimbatore,Tamilnadu, India
Pedestrian identification is an inexhaustible and sufficiently large task in any hyper intelligent
videotape monitoring, as it supports abecedarian information for a more in-depth understanding of
the videotape footage. Because of the possibility of perfecting safety systems, it contains an
egregious extension to automotive operations. This is available as an ADAS option from a number
of automakers. The pedestrian identification network was trained using images of pedestrians and
nonpedestrians. The trainPedNet.m coadjutor script is used to train this network in Computer Vision.
The sliding window method reduced the size of the image patches to (64*32). The patch dimensions
are derived from a heatmap that depicts the dispersion of pedestrians in the images in the data set.
The patch dimensions are derived from a heatmap that depicts the dispersion of pedestrians in the
images in the data set. It denotes the absence of pedestrians at various scales and locations in the
images. Patches of pedestrians closer to the camera are cut and processed in this illustration. Non-
Maximal Suppression (NMS) refers to the patches that have been obtained in order to combine them
and display entire pedestrians. The composition focuses on the achievement of the linear binary
pattern (LBP) abstract background fluctuations capturing and Histogram of Oriented Gradients
(HOG) alternatives with altered parameters to categorising is attained by Support vector machine
Classifier in this evaluation.
132. IN SILICO INTERACTION STUDIES OF SALACINOL WITH INHIBITORS OF

IAP PROTEINS
Anitha M 1, Mahendran Radha1*, Ram kumar2

May 21st – 22nd, 2022
Salacia reticulata is an important medicinal plant, is widely distributed in some parts of the
subcontinent of India. The plant possesses bioactive compound salacinol which is proved to have
antitumor activity and is confirmed through various pharmaceutical studies. Apoptosis plays a key
role in the regulation of tissue homeostasis. Apoptosis, triggered by exogenous and endogenous
stimuli such as ultraviolet radiation, oxidative stress, and genotoxic chemicals is a crucial
phenomenon within biological systems. DNA damage activates and stabilizes p53 in the nucleus and
cytoplasm and regulates other proteins that stimulate intrinsic and extrinsic apoptotic pathways.
Malfunctioning of the apoptotic pathway may cause diseases like cancer, neurodegenerative and
autoimmune disorders. Recently, potent apoptosis inducing compounds associated with human
health have been recorded that prevent tumor promotion, progression, and the occurrence of cellular
inflammatory responses. Certain drugs are being employed in therapy to induce apoptosis for the
treatment of cancer and non-cancerous cells. The Inhibitors of Apoptosis (IAP) are a family of
functionally and structurally related proteins, which serve as endogenous inhibitors of programmed
cell death. In the current study, salacinol was analyzed and subjected to pharmacokinetics screening.
An attempt was made to study the interaction between three IAP proteins (BIRC5-3UIH, 3UII,
3UIK) with salacinol using Auto dock software. Salacinol was found to have the best interaction
with the IAP target protein.
133. IN SILICO SCREENING OF SNP'S IN PRION PROTEIN AND

COMPARATIVE STUDIES BETWEEN PHYTOCHEMICAL COMPOUNDS THAT
INHIBIT PRION PROTEIN CAUSING NEURO-DEGENERATIVE DISEASES.
Anitha. M1, Mahendran Radha1*, Dhanaya2, Jeyabaskar Suganya3
The neurodegenerative diseases are highly fatal and they are characterized by the aggregation and
deposition of the misfolded prion protein (PrP) in the brain. The prion like prionoids leads to a
variety of neurodegenerative diseases in humans as well as in animals. Prion protein is active in the
brain and several other tissues which are made by receiving instructions from the PRNP gene. Prion
diseases are progressive, fatal, affect the structure of the brain or other neural tissue, and lack
effective treatment. Hence there is an urgent need to develop new and more effective therapeutic
strategies to combat these fatal diseases. The main objective of the current study is to identify
mutated SNP and its position which is responsible for the main cause of the disease. We have
predicted the protein stability, Amino acid function, Secondary structure of the amino acid, Mutated

May 21st – 22nd, 2022
prion protein Homology modelling, and analysis of Protein 3D structure. Through the literature
survey, we have identified phytochemical compounds from medicinal plants which possess
pharmacological activities to cure neurodegenerative disease. And we have also retrieved few
currently used synthetic compounds which are used for the treatment of neurodegenerative disease.
Both the set of compounds were further analysed for Molecular docking against the target prion
protein. The docking results and comparative docking studies reveal that the natural compounds
exhibit the best binding energy than the synthetic compounds. Natural compounds are more effective
in the treatment and cure of various Neurodegenerative diseases without any side effects.
134. MULTI-OBJECTIVE VIRTUAL MACHINE PLACEMENT USING ORDER

EXCHANGE AND MIGRATION ANT COLONY SYSTEM ALGORITHM
Saripalli krishnaveni1 , Arundhathi Lakkireddy2, and S Vasavi3*,4A.Anu Gokhale
1,2,3Department of Computer Science & Engineering, VR Siddhartha Engineering College,
Vijayawada, India
4 Illinois State University, Normal, USA
Order Exchange and Migration Ant Colony System (OEMACS) and Ant Colony System OEMACS
approach based on ACS that finds solutions by combining order exchange and migration local
search strategies, was developed after examining the ant colony system's (ACS) promising
performance for combinatorial issues. The ACO algorithm, which is regarded as best approach
solving for virtual machine placement (VMP) problem, is capable of significantly reducing the
number of active servers from a optimization standpoint. These bio-inspired algorithms help in
solving VMP problem. Using the pheromone deposition method, artificial ants are guided to the
feasible solution in ACO. It also arranges virtual machines in a way that reduces resource waste.
This strategy is used on servers with homogeneous and heterogeneous VM sizes. The VMP is
implemented using the ACO algorithm and OEMACS, and a comparison is drawn between them
that shows their efficiency in terms of execution time and Energy Conserved
135. EFFICIENT DESIGN OF SRAM CELL USING 90 NM TECHNOLOGY

MOHIT GARG
Galgotias College of Engineering &Technology
Greater Noida, INDIA
Hilal Ahmad Mir
Galgotias College of Engineering & Technology
HARSHIT CHAUDHARY
Galgotias College of Engineering &Technology
Mr.Amit Gupta (Professor)
Galgotias College,of Engineering & Technology ,Greater Noida, INDIA
NIKHIL KUMAR

May 21st – 22nd, 2022
Now a day, low power and high-speed SRAMs have become the discriminatory part of many VLSI
chips. This is especially true for microchips, where on-chip reserve sizes are constantly evolving to
expand the growing diversity in processor speeds and Cache memory. SRAM (Static Remote Access
Memory) is used as fast cache memory and is used to speed up the processor allocation and visual
memory connector. In high-speed PCs, SRAM memory is essential for improving performance
speed and DRAM is used as part of Primary Memory where Crowding is more important than
Speed.
136. THIOPHENE AS A GAS SENSOR FOR THE ADSORPTION OF CARBONYL

SULFIDE: DFT CALCULATIONS
HARIPRIYA P a , AMRUTHA R b , CHANDRAN P c
a Research Scholar, Bharathiar University, Coimbatore, PIN-641046, Tamil Nadu, India b
Department of Physics, KCG College of Technology, Rajiv Gandhi Salai, Karapakkam, Chennai, PIN-
600097, Tamil Nadu, India
c Madras Christian College, Tambaram East, Tambaram , Chennai, PIN-600059, Tamil Nadu, India
Gas sensing ability of Thiophene with Carbonyl sulphide (COS) is investigated using hybrid
functional DFT B3LYP/6-31+G(d) level of theory. The interaction energy, the counterpoise
corrected energy and the charge transfer through Mulliken and Natural bond orbital ( NBO ) analysis
between the analyte and the monomer reveal the gas sensing parameter of thiophene. When carbonyl
sulfide is exposed to thiophene, it alters the electronic charge concentration which changes the
electrical properties of thiophene. This opens the possibility of thiophene as an efficient gas sensor.
The results are supported by UV-vis spectroscopic analysis . The reduction in the band gap for
thiophene with COS than that of thiophene enhances its electrical conductivity that makes it suitable
in the design of organic semiconducting devices. By COS adsorption on thiophene, the gas sensing
ability is proved which predicts thiophene as an excellent gas sensor to replace the traditional solid
state sensor.
137. RASPBERRY PI BASED READER FOR BLIND

Pratik Dete1 , Rahul More2 , Omkar Khandekar3
Department of Electronics and Telecommunication Engineering
Rajiv Gandhi Institute of Technology, Mumbai
The Braille machines are expensive and as a result are not accessible to many. In particular, there is
a need for a portable text reader that is affordable and readily available to the blind community. The
concept of automatic document reader means blind reader is for visually impaired peoples. There are
millions of visually impaired peoples in the world. Reading is one of the major necessities of
visually impaired people. In this paper we designed system is a portable, low-cost, reading device
made for the blind people. Also overcomes the limitation of conventional Braille machine by
making it affordable for the common masses.
We proposes a smart reader for visually challenged people using raspberry pi. This system consists
of a camera interfaced with raspberry pi which accepts a page of printed text. This system uses OCR
technology to convert images into text and reads out the text by using Text-to-Speech conversion.
This system supports audio output via Speakers as well as headphone.
May 21st – 22nd, 2022
138. TEMPERATURE PREDICTION USING MACHINE LEARNING ALGORITHM

Shanmathi.E, and G.B.Hema Malini, SDNB Vaishnav College for Women.
Weather forecasting is the prediction of what the atmosphere will be like in a specific place by using
technology and scientific knowledge to observe the weather. This paper aims to predict temperature
using linear regression model, random forest regression model and support vector regression model.
This study was limited to collecting and analyzing datasets that contained missing data, as the
missing data can immensely hamper the analysis. The data collected is incomplete and the
temperature prediction model is used to refine the incorrect and missing data. The random forest
regression model is used to predict the average temperature of that particular year we get the
predicted temperature with an accuracy of 99.58%.
139. CLASSIFICATION OF PADDY AND TOMATO CROPS DISEASES DETECTION

USING CNN
1 Maamidi Yaswanth Kumar Reddy
Department of CSE
Anand Nagar, Krishnankoil, India.
2 Kedari Lakshmi Narasimha Department of
CSE
Anand Nagar, Krishnankoil, India
Paddy is one of the major cultivated crops in India which is affected by various diseases at various
stages of its cultivation. Tomato is an a major crop in India that every year production is nearly
2744.32 thousand metric tonnes. It is very difficult for the farmers to manually identify these
diseases accurately with their knowledge.
Recent developments in Deep Learning show that Automatic Image Recognition systems using
Convolutional Neural Network (CNN) with Transfer Learning (TL) and Support Vector Machine
(SVM) models can be very beneficial in such problems. we have used Transfer Learning to develop
our deep learning model. The proposed CNN architecture is based on MobileNet and is trained and
tested on the dataset collected from fields and the internet.
140. DEEP LEARNING AND SWARM INTELLIGENCE BASED SYSTEM FOR

SMART AGRICULTURE USING WIRELESS SENSOR NETWORKS
Kaniga.B
M.E-STUDENT, Dept. ECE,
Paavai Engineering College (Autonomous), Tamil Nadu, India
Dr.M.Sudha
M.E., MBA., Ph.D., HEAD OF THE DEPARTMENT, Dept. ECE,
Mr.Ravisankar Kandasamy
M.E., (Ph.D)., Asst. Professor, Dept. ECE,

May 21st – 22nd, 2022
The WSN (Wireless Sensor Network) based agricultural drainage management system is a
convergence technology to enhance agricultural production of crops from storm control, erosion
control and water table control. In addition, implementing smart agriculture, through the
convergence technology allows to determine the requirement of water resources, real-time
monitoring of the field and support farmers in decision process. In addition, sensors help in
collecting information on circumstances like level and flow of water in drainage area, soil moisture
content and rainfall condition. We presented the survey about the WSN based smart agriculture and
its techniques to future along future enhancement. The work further extends with the deep learning
approach of analysing the data collected from the level and flow sensors and camera. The CNN
based design can improve the performance of the system. The proposed ideation combines the
swarm intelligence and CNN for better performance.
141. SIGN LANGUAGE RECOGNITION USING DEEP LEARNING

Anushka Ray, Shahbaz Syed, S. Poornima, and M. Pushpalatha
Department of Computing Technologies, School of Computing,
SRM Institute of Science and Technology, Kattankulathur-603203, Chengalpattu, Tamil Nadu, India
Sign language is the primary language of the people with speech and hearing impairment. Hearing
impaired people use sign language to express themselves, participate in conversations, learn, and
live as normal a life as possible. When deaf and dumb persons try to converse in sign language with
those who aren't familiar with it, a problem occurs. This is where modern technology can step in.
Although many existing projects have proposed methods to alleviate the problem but most of these
have used external sensor or certain algorithms that do not work well under certain conditions like
variation in skin color, inclusion of facial data and similarity between few signs and gestures. In our
project we have used MediaPipe to solve the problem of variation in skin color as it is a very
accurate hand tracking framework by Google. We have also created 3 sets of customs datasets to
train 3 different deep learning models — 2 of these models are used specifically used for predicting
particular groups of letters which are similar to each other. This fixes the issue arising when similar
signs are encountered. The propose prototype can be used to help educate the elementary school
children using which they can learn to fingerspell the American sign language alphabets and string
them into basic words and learn them in a picturized manner. At the end our prototype should be
able to detect the sign alphabets from the live video, print them on screen and display the image of
the words formed using the sign alphabets.
142. DETECTION OF OBJECTS USING MACHINE LEARNING

K.Muni Vardhan
Department of Computer Science and engineering
Kalasalingam Academy of Research & Education
Krishnankovil, TamilNadu
Dr.A.Sarvanan
Associate professor
Department of Computer Science and engineering
Krishnankovil, TamilNadu
May 21st – 22nd, 2022
M.Girish Reddy
KrishsnanKovil, TamilNadu
G. Mahesh
Krishanakovil, TamilNadu
Due to object detection’s close relationship with video analysis and image understanding, it has
attracted much research attention in recent years. Traditional object detection methods are built on
handcrafted features and shallow trainable architectures. Their performance easily stagnates by
constructing complex ensembles which combine multiple low-level image features with high-level
context from object detectors and scene classifiers. With the rapid development in deep learning,
more powerful tools, which are able to learn semantic, high-level, deeper features, are introduced to
address the problems existing in traditional architectures. In this paper, we propose a two-step
solution to this problem. In the first step, our goal is to provide a model to optimize memory usage
and, in the second step, we propose a multi-thread approach that uses YOLOv3 to perform real-time
object detection on multiple, concurrent, live streams on a single GPU. In this approach, GPU
resources are optimally used. The proposed approach is evaluated on a public dataset and the result
shows improvements in performance by an average of 12% in Central Processing Unit (CPU) usage
and 13% in frames per second (FPS) compared to the YOLOv3.
143. CODE COVERAGE - STUDY ON APPLICATIONS OF CODE COVERAGE

METRICS FOR IMPROVING TEST EFFECTIVENESS
Kavitha Chellappan1, Dr MK Jayanthi Kannan2
1Research Scholar, Department of Computer Science, JAIN( Deemed to be University), Bangalore
2Professor, Department of Computer Science, Jain ( Deemed to be University, (Karnataka), INDIA
Code coverage represents the percentage of software code exercised by the test system. The
aforementioned “test system” is expected to include all tests designed to evaluate the quality of the
software code. Code coverage assessment is one of the most objective indicators that can accurately
illustrate the effectiveness of the tests in a project. Besides the primary intent to measure to test
effectiveness, coverage metrics data can be leveraged for additional purpose: improve test
effectiveness and/or to also boost test efficiency. Significant research and practical application of
repurposing the coverage metrics has been ongoing in the last few decades. The core of such
applications centers around Process and Software Quality enhancements. Such applications deploy a
variety of software tools and techniques, including static analysis tools(such as cppchecker,
Coverity) , data integration techniques (to combine data from software source control system, defect
management system and test antifactory) , business intelligence tools and techniques , and machine
learning algorithms to percolate undiscovered patterns. This paper reviews the landscape of such
applications and identifies opportunities for areas that warrant further research.

May 21st – 22nd, 2022
144. DETECTING SKIN CANCER USING DEEP LEARNING

Dr.T.C.Kalaiselvi,R.Niventhra, R.Praneshkumar, P.Palanivel
Department of Electronics and Communiction
Perundurai, ErodeTamil Nadu
Skin cancer is one of the most common cancers that affects people all over the world. The three most
dangerous types of cancer that are caused by broken DNA and result in death. Damaged DNA
causes cells to expand uncontrollably, and the rate of growth is currently increasing rapidly. The
interpretation of these images is difficult due to the difficulties of the various shapes and sizes of the
lesions. One of the most difficult aspects of developing a good automatic categorization system is
the absence of large datasets. To distinguish the seven various types of moles in skin cancer, the
suggested method uses a deep convolutional neural network. The dataset, a large collection of
dermartoscopic images, was utilised in trials to increase performance using data augmantation
techniques. Melanocytic nevi (mv), Melanoma (mel), Benign keratosis-like lesions (bkl), Basal cell
carcinoma (bcc), Actinic keratoses (akiec), Vascular lesions (vas), and Dermatofibroma are the
seven forms of moles (df). The VGG algorithm is used to train and test the dataset images. The
shown findings are ideal for achieving high classification accuracy and lower false negatives when
employed with the DenseNet network. The models perform well in large dataset classification and
can discriminate between the seven categories of moles in a dataset.
145. COVID-19 INDOOR SAFETY MONITORING SYSTEM USING MACHINE

LEARNING
Dr.T.C.Kalaiselvi,R.Niventhra ,R.Praneshkumar ,P.Palanivel
Department of Electronics and Communiction,
Perundurai, Erode
Tamil Nadu
The Covid-19 pandemic is causing a global health epidemic. The most powerful safety tool is
to wear a face mask in public places, schools, supermarkets to reduce the spread of Covid-19. There
are many Convolutional face recognition technologies to distinguish effective photographic imagery
to monitor face mask detection. Therefore, it is very important to improve the performance of
existing detection methods in the existing system. In the proposed input of the data set number
increases to improve the maximum range of accuracy. The proposed method is used to determine the
body temperature, face mask and social distance maintenance using advanced machine learning
methods. Using the EM -18 RFID scanner human data such as temperature value, face mask
identification and public distance detection are collected. It is used to indicate the state of a person's
health in a cloud platform. The wireless temperature sensor is used to detect the body temperature of
a person using MLX90614 without anyone. Raspberry Pi integrated with Raspberry pi camera is
used to monitor face mask detection and community distance. Raspberry pi captures an image and
detects with the Convolutional Neural Network algorithm if a person is not wearing a mask, not
following a social distance. So, the authorities have to monitor the human condition in the cloud

May 21st – 22nd, 2022
platform. By applying this concept, the spread of Covid-19 can be greatly reduced and it is easier to
identify people with Covid-19 symptoms.
146. SENTIMENT ANALYSIS ON USER REVIEWS AND TWITTER DATA TO

PREDICT VEHICLE SALES
Nithiavathy.R
Coimbatore,India
Blessy A V
Sri Krishna College Technology
Coimbatore,India
Kavya R
Coimbatore,India
Because of the growth of social media, writing comments or expressing thoughts on products online
is now simpler than ever. Data from user evaluations might be one of the most important sources for
anticipating vehicle sales.Sentiment Analysis is another way to understand if the user is satisfied
with the product which will help the manufacturer to improve the features in the upcoming
release.User reviews play a vital role in analysis and predict which product will have much
production and sales in future.In this paper a set of algorithms are used to predict the sentiment
analysis including the Naive Bayes,Logistic Regression and RandomForestClassifier. The
categorization and prediction will be based on user-defined reviews and data from Twitter.
147. CHILD ADOPTION BASED ON FACIAL FEATURES AND SKIN COLOR

Venkatesh V
Dept of computer science and
Engineering
Sri Krishna college of technology
Coimbatore, India
Aathi Vasanthan V
Dept of computer science and Engineering
Coimbatore, India
Daraneeswaran M
Coimbatore, India
Dharaneeish S
Coimbatore, India
May 21st – 22nd, 2022
Matching is the process of collecting a young person's down and out enactment, accompanying
skillfully anticipated persons. It is a main stage superior to landing at a maintenance resolution.
Matching includes choices that concede the possibility to endlessly change a youth’s growth in an
adoptive classification. It is a blame usually pretended by a crew of artists loyal to the care and
welfare of youth, including psychologists, friendly traders, and jurists. Matching is a weighty and
complex task that grants permission to influence negative results for two people: an adoptive
classification and an adolescent if an inappropriate competition is created. This White Paper
examines, by virtue of what creative data processing can do, how it can considerably embellish the
ratification process. Specifically, what it will show is that there is a need for a very large size of
culture in the corresponding philosophy of a new order to mirror the complex process that happens
through uniformly cruel judgement and interpretation. When created and achieved correctly, creative
equal electronics may be very valuable, more conversant selections that bring about better effects
and a sleeker and understandable maintenance process. Using countenance treat science, the
countenance in the face of the adopting father and mother is written, and the kid whose equal is
equate is recognized and assigned to the ruling class.
148. SOCIAL MEDIA ADVERTISEMENT RESULT ANALYSIS

Nishitha Velagala
Computer Science and Engineering B. Tech of SRM Institute of Science and Technology Chennai, India
M Parijat Chandra
Computer Science and Engineering B. Tech of SRM Institute of Science and Technology Chennai, India
Dr. Poovammal E
Department of Computing Technologies of SRM Institute of Science and Technology Chennai, India
Our research focuses on the issue of social media advertisement, which refers to the placement of
advertisements based on the social information and behaviours of individual users. As social
network services (such as Instagram, Facebook, and Twitter) become the primary venue for social
interactions, social network sites now account for more than 20% of all internet advertisement. The
need of advertisement is getting more and more important because of the competition that
companies are facing in their respective sector. We have done our analysis by understanding the
individual information with their behaviour on networking sites. We are presenting the analysis of
outcome for any advertisement data whereas our database consists of three parameters such as age,
salary and purchase or not. And we are predicting the outcome for other users. We have used and
compared various machine learning models and compared the accuracy of those models. For this
research we have used a dataset taken from the Kaggle website but this model can be used for any
dataset so this model can be used widely by anyone to get the result and to understand the behaviour
of users. Prior work that we have gone through is based on a click through model. Which predicts
whether the user will check or not but this model is focused on whether the user will purchase or not
which is totally based on the outcome of the advertisement. Our research will help the companies to
understand the behaviour of users and with the help of the data they can make changes to their
product and advertisement in order to accurately reach out their target audience which will help
them a lot because to reach their consumer and to expand the reach among the people is what any
business is all about.

May 21st – 22nd, 2022
149. DETECTION OF VIRULENT MESSAGES ON PROPOSED DATASET WRITTEN

IN CODE-MIXED HINDI-ENGLISH LANGUAGE USING NLP TECHNIQUES
Nirav Agarwal1, Harshvardhan Arvind Singh1, Anita Ramalingam1
1SRM Institute of Science and Technology, Kattankulathur
Social media as a medium of communication has opened countless possibilities for individuals
around the globe to express their opinions on various topics ranging from politics to entertainment,
thus reducing the gap in the accessibility of information. However, this has also led to social media
becoming a hub for spreading hateful messages. These virulent messages spread online result in the
rise of cases of cyberbullying, leaving an enduring impact upon the victims. Manually classifying
and reporting these messages is a challenging task and has therefore created an interest in the
research community to incorporate machine learning-based techniques to ease up the hand-operated
process. The study's goal is to play a small auxiliary role in the prevention and ultimate eradication
of cyberbullying. The proposed work in this research primarily focuses on the users from the Indian
subcontinent by experimenting with Code-Mixed Hindi-English language. The study achieves this
by performing the task of developing a Code-Mixed Hindi-English hate speech dataset containing
4014 tweets, out which 2000 tweet contain hate, using the publically available tweets from the
Twitter platform and, furthermore, utilising the created data to fine-tune a collection of pre-trained
deep learning models.
150. INTELLIGENT TRAFFIC FLOW USING MACHINE LEARNING

Anirudh. P
Vishnu Vijay
T Senthil Kumar
Traffic congestion is a serious problem of rising concern especially in India where the population is
already over a billion and the common choices of transport for commuting being a four wheeler and
a two wheeler. India is urbanizing at a great pace and it is competing with the fastest growing
countries in the world which also resulted in high vehicular growth. In order to tackle this some
research has been conducted on the existing methods and technologies for relieving traffic
congestion and was identified with a lot of improvement to make. Hence the main aim is to make it
now intelligent enough to take into consideration accidents/breakdowns, clearing way for
ambulances and firetrucks and better reliability of the existing system in detecting vehicles. Using
the live feed from cameras with image processing, deep learning and an algorithm with a timer logic
for the signals, this project tackles the issue.
151. DESIGN AND DEVELOPMENT OF SOLAR WATER PUMP WITH DIFFERENT

TIME SLOTS FOR POWER SAVING
Balaram Jagatha1
Electronics and Communication Engineering, Vardhaman college of Engineering Affiliated by JNTU
Shamshabad, Hyderabad, India
Sathwik Rapala2

May 21st – 22nd, 2022

Krishna Chaithanya. J
From the olden days, we have been using the non-renewable resources like coal for our needs which
is leading to the exhausting of these resources. Technology has developed to such an extent that it
has become helpful for us in all aspects of life. By utilizing this technology we can use the
renewable energy resources and reduce the usage of nonrenewable resources for achieving a
sustainable development. The photo voltaic cells convert the solar energy into electrical energy and
this energy is stored in batteries to run the water pump during the day time. If a machine is to be
switched off after operating for few minutes, it becomes difficult and many times we don’t
remember to turn off the system after the prescribed time. Our project gives a solution for not only
saving the energy but also builds a brilliant framework to reduce the wastage of water and increment
the beneficial of the product.
152. DETECTING PASSIVE ATTACKS IN FINGERPRINT SPOOFING USING

CONVOLUTION NEURAL NETWORK
Mr.R.RAJA SEKAR, G.Poojitha Sree Vandana, C.Glory Devakirubai, G.Ravindra Babu
Computer Science and Engineering, Kalasalingam Academy of Research and Education
This paper endorse the idea of “Detecting Passive Attacks in Fingerprint Spoofing Using
Convolution Neural Network”. More recently, in the field of cybercrime getting fake fingerprints is
a daunting task in any developed country. Biometric authentication grows in many areas such as
online banking, private file locks, etc. Where fingerprint detection is an important factor used to
obtain analysis of fingerprints. By using Convolutional Neural Network Algorithm to detect the
liveness of fingerprint whether fingerprint is real or fake. This paper will explain the design and
features of detection of passive attacks in fingerprint spoofing using convolutional neural networks.
Furthermore, we discuss the major research challenges and future work that need to be addressed in
this active field of fingerprint liveness detection.
153. PREDICTION OF COVID-19 & PNEUMONIA BY USING CONVOLUTIONAL

NEURAL NETWORK AND DEPLOYMENT BY USING THE WEB FRAMEWORK
A. Antonidoss1, Bishal Sarkar2
1&2Dept of CSE, Hindustan Institute of Technology & Science, Chennai, Tamil Nadu

May 21st – 22nd, 2022
The symptoms of covid-19 and Pneumonia are similar. Steps must be taken to distinguish between
covid-19 and other types of pneumonia patients. In this paper a model of Convolutional Neural
network(CNN) is presented by using Tensorflow, keras, numpy and Matplotlib to detect covid-19,
Pneumonia, both covid-19 & Pneumonia and normal Patients. It is based on classification which is a
supervised learning. There are four classes, they are normal, covid-19, Pneumonia, both covid-19 &
Pneumonia. The paper is deployed by using by using the django framework in pycharm. It is a
example of chest x-ray image classification by using tensor flow, Matplotlib, keras and numpy
library and alexnet.
154. TEMPERATURE CONTROLLED DC FAN USING SOLAR ENERGY

Niharika Ganji
Vardhaman College of Engineering Hyderabad, India
Shriya Gandreti
Mr. T. Rama Krishnaiah Electronics and Communication Engineering

Automated control is a vast area of technology, whose prime objective is to build control strategies
to increase functioning capabilities and efficiency when they are used on a system. The remarkable
feature of automation is that it reduced the user’s workload. An example of those utilities is the
Automatic fan. Fans are generally designed with control equipments, based on speed demands.
Generally, when the heat is high, the fan is switched to rotate at a higher speed and at cooler or
lower surrounding temperatures, the fan will work at a slower speed, which is mostly done with the
help of a regulator. Therefore, in this paper, an automatic control design is presented for fan speed.
A circuit consisting of a temperature sensor along with a microcontroller inside, a DC motor, and
some electronic components is used and implemented to automatically change the speed of the fan.
As an extra feature, the LCD screen is also placed which is used to display the temperature and fan
speed at a particular moment. When the temperature of the detected environment exceeds the
threshold value, the fan starts up, and the temperature drops. The fan will remain on until the
temperature drops below the threshold value. Automation of fan speed saves time and maintains an
always comfortable environment.
155. SECURED DATA TRANSMISSION FOR SMART GRID SYSTEM USING

BLOCKCHAIN TECHNOLOGY WITH A TRUSTED GAMING TOKEN HONEYPOT
ALGORITHM
Deeptha R. 1*, Rajeswari Mukesh2
1Assistant Professor,
Department of Information Technology,
SRM Institute of Science and Technology, Ramapuram.
2Professor and HOD,
May 21st – 22nd, 2022
Implementing blockchain techniques has enabled secure smart trading in many realms including
neighboring energy trading. The information stored on the blockchain will face privacy security
issues over attackers. In this paper, a Smart contract based blockchain-oriented approach to solve the
problem of privacy leakage without restricting trading functions in smart grid is presented. The
proposed approach mainly addresses energy trading users' privacy in smart grid and screens the
distribution of energy sale of sellers deriving from the fact that various energy trading volumes can
be mined to detect its relationships with other information, such as physical location and energy
usage. Then for the purpose of secure data transmission, Conviction Diffie Hellman Algorithm
(CDHA) was used for generating the keys. Finally for detecting the untrusted user, Trusted Gaming
Token Honeypot Algorithm (TGTHA) was suggested. The whole experimentation was carried out in
a MATLAB environment. Experimental evaluations have demonstrated the effectiveness of the
proposed approach.
156. FRAUD CLAIMS DETECTION IN INSURANCE USING MACHINE LEARNING

Hritik Kalra
Dept. of Computing Science Eng., School of Computing,
SRM Institute of Science and Technology, Kattankulathur
Chennai, India
Ranvir Singh
Dept. of Computer Science Eng., School of Computing,
Chennai, India
Dr. T. Senthil Kumar
Dept. of Computing Technologies, School of Computing,
Chennai, India
Insurance fraud is an illegal activity that is deliberately done for financial gain. In the present era,
this is the prime severe problem encountered by various insurance companies all over the globe. In
most cases, some shallow or deep holes in process of fraudulent claims investigation have been
identified as a major cause. Due to this reason, the need of using computer techniques to curb fraud
activities grew as a motivation, providing not only a reliable and robust environment for customers
but also reducing fraud claims to a large scale. We presented our research by automating the
assessment process of insurance claims through various data techniques, where identification of
false claims will be automatized by utilizing Data Analytics and Machine Learning methods.
Moreover, the system has the ability to potentially develop heuristics around fraud indicators.
Therefore, this model creates a positive impact on the whole insurance industry with increased
company credibility as well as customer satisfaction.
157. PHISHING SCAM IN THE MIDST OF THE UKRAINE CRISIS

Ashok Kumar R, Rakshitha P and Nisha Setty A R, [email protected]
Phishing is a cybercrime that involves the use of virtual networks. Phishing is used to attain
nefarious goals, such as stealing money and personal information from internet users. To perform
such tasks, phishers use emails or SMS as bait. The rate of such crimes is rising all the time. It's
May 21st – 22nd, 2022
essential to realize and be informed of such a crime. Donation camps have begun to support Russia
and Ukraine as a result of the ongoing hostilities. To steal this money, scammers have established
harmful websites. They're sending out emails with these kinds of website links in them. This study
outlines the recent phishing attempts, how to protect oneself from phishing and the email patterns
used by such criminals. Machine learning algorithms are used to distinguish between phishing and
authentic websites. The accuracy of test set prediction was taken after fitting model to the training
data. The Logistic Regression, SVM, and Naïve Bayes Algorithms were outperformed by the KNN
and Random Forest Algorithms. KNN and Random Forest provided 91% and 91.96% accuracy,
respectively.
158. CROP ANALYSIS AND PREDICTION USING ENSEMBLE LEARNING

Dharini Grover1, *S.Iniyan1
1Department of Computer Science and Engineering, School of Computing, SRM institute of Science and
Technology,Kattankulathur,Chengalpattu-603203,India
2Department of Computing Technologies, School of Computing, SRM institute of Science and

Technology,Kattankulathur,Chengalpattu-603203,India
Nowadays, unpredictable climate change has led to poor yields of the crops affecting their
performance and growth .Crop prediction performed in advance of the harvest, would help the
farmers adopt appropriate measures, in terms of storage, etc. It will also be advantageous for the
government, policymakers, etc. This project will help counsel the farmer in sowing reasonable crops
by deploying machine learning techniques like Ensemble learning, one of the best technologies in
crop prediction. Machine learning will help decide what crops to grow, what to do during the
growing season of the crops, after harvesting, etc. Factors like temperature, rainfall, area, etc.will be
considered. Different machine learning algorithms will be applied to support crop yield prediction
research.The predictions achieved by this system will largely help the farmers decide which crop to
grow. The accuracy of ensemble prediction is 98.9 while that of gradient boosting is 99.45.
159. PREDICTING STUDENT MARKS USING MACHINE LEARNING

GOUTHAM GK
B.Tech (CSE)
SRM Institute of Science and Technology, KTR
Chennai, India
PENDYALA REVENTH SETYA
B.Tech (CSE)
Chennai, India
Dr.T SENTHIL KUMAR
Associate Professor

May 21st – 22nd, 2022

Department of Computing Technologies
School of Computing
Chennai, India
In the realm of data mining strategies, a massive amount of computerized data is produced using a
variety of methods. An important area of education data mining development is the establishment of
students achievements predictions modelss to forecast student performance in academic institutions.
Using their prior semester grades, a prediction system has been presented. Binomial logical
regression, Decision tree, linear regression, and the KNNs classifier are used to analyze the study.
This framework would aid the student in recognizing their finals grade and improving theirs
academic conduct in orders to achieves their higher score. When a student's gradesfalls below 50%,
our app sends a message to their parents.
160. MODELING OF 2DMULTILAYER MOLYBDENUM DISULFIDE MOS2

TRANSISTOR
Dr.R.Sridevi
M.Kumarasamy College of Engineering,
Karur-Tamil Nadu,
S.Angalaparameswari
Karur-TamilNadu,
D.Indhuja
Karur-Tamil Nadu,
D.Poovitha
Karur-Tamil Nadu,
Unique Atomically slight Molybdenum disulfide (MoS2) is a semiconductor material suitable for
field-effect transistors (FETs) with channel lengths of tens of microns or less than ten nanometers.
The bulk and massive band hole of MoS2 limit the level of difficulty possible in direct source-drain
in extremely short-channel semiconductors. Its molecularly slim body improves the efficacy of door
balance. Our results show that when the thickness of the MoS2 channel layer is increased, it plays a
significant role in settling and executing debasement.
161. A STUDY TO EXPLORE HINDRANCES IN IMPLEMENTATION OF DATA

MINING IN BANKS
Vivek Bhambri

May 21st – 22nd, 2022
Assistant Prof, Department of Computer Science and Applications

A.S.College, Khanna
The growing size of data bases in all types of organizations is making it difficult for them to analyze
it and use it for making effective decisions well in time. In such a scenario, where making effective
and timely decisions is of utmost importance, the techniques and tools available in data mining has
proved to be an asset for the organizations. Data mining techniques are used in all types of
organizations for customer relationship management, detection of frauds and predicting future
trends. But still it is in a stage of infancy, where implementation of these techniques has to face
economic, administrative and behavioural problems. Readiness to adopt the new technology is a
major issue in the implementation of new technologies. This paper tries to explore the relationship
of age and gender with the readiness to adopt the data mining techniques in the banks with the help
of association rule mining. The analysis of the results will be fruitful for the decision makers in the
banking sector to design a strategy for the successful implementation of these techniques.
162. DESIGN & IMPLEMENTATION OF BUS TRACKING SYSTEM UNDER IOT

PLATFORM
Divya Sravani Bandi1 Electronics and Communication Engineering
Vardhaman College of Engineering Affiliated to JNTU Hyderabad Shamshabad, Hyderabad, India
Vishal Reddy Chewalla2 Electronics and Communication Engineering

Krishna Chaithanya. J3 Electronics and Communication Engineering

In the system, technologies have been demonstrated to make individual’s lives more secure and safe
every day. The proposal is developed on cutting-edge GPS technology and Internet of Things (IoT)
and employs an application to manage all the college buses, enabling students to track the movement
of buses. Additionally, daily issues such as traffic, abrupt delays, and faulty vehicle arrival and
departure incidents tend to happen, interrupting the students' and faculty’s schedules and requiring
them to wait for their particular bus arrival or otherwise attempting to miss their bus. The application
allows all of an institution's staff and students to track where each college bus is at any given time. A
GPS module has to be installed in every bus that is helpful in assisting users to locate the bus, and
the data is collected via an Android app that uses Google Maps to track each bus's travel. The
location coordinates of bus are updated every second in the application as per the current location of
the bus, paving the way for a real-time system
163. SENTIMENT ANALAYSIS OF PRODUCT REVIEW

N.Veerendra
Department of computer science and engineering
Kalasalingam academy of research and education
Krishnankovil, Tamilnadu
V.Arun Kumar Reddy
May 21st – 22nd, 2022
G.Gangadhar
Dr.Mohad Usama
Associate professor
Kalasalingam academy of research and education Krishnankovil, Tamilnadu
Social network, e-commerce sites, blogs are new emerging platforms for people to express their
opinion. These websites provide a large volume of text that can be utilized for a variety of purposes,
including sentiment analysis. The major purpose of sentiment anal ysis is to aid in the improvement
of a firm. Here NPL algorithm is used. In proposed system, NLP algorithm shows
outstanding performance. To create sentiment analysis a using the analysis of emotions, there is a
need to use polarity obtained through the reviews and finally we shown the graph based on
reviews and comparing two products reviews using natural language processing.
164. VOICE BASED EMAIL FOR BLIND USING ARTIFICIAL INTELLIGENCE

AND PYTHON
B. Bhavana1, T. Bala Bhavagna2, B. Lasya Priya3, M M Mastan Durga4
Assistant Professor1,2,3,4 Department of Computer Science and Engineering
Velagapudi Ramakrishna Siddhartha Engineering College, Kanuru
As technologies are enhancing, human beings are habituating to virtual life and virtual verbal
exchange. There are numerous methods to communicate with others through the net in this
contemporary technology. Most people always select and use the easiest manner of communique
commonly i.e., electronic mail. Electronic mail is the era that allows customers to contact others via
sending mails and additionally enables in the business global conversation. There are folks who can't
use this generation due to the fact that they're illiterate or do not have the visible capability on the
screen. So, to make this era towards visually challenged human beings, a Voice-based totally email
gadget is proposed by using Artificial Intelligence and Python. This device gives visually challenged
people the power of communication and makes them a great deal stronger and more unbiased. This
software will assist blind people to get the right of entry to e-mail much like different everyday
humans. Voice-primarily based e-mail structures use new technology as a way to be without
difficulty proper through visually challenged people.
165. PREDICTION OF DIABETES USING MACHINE LEARNING ALGORITHM

Heet Patel
Dept of Computer Science and Engineering SRM Institute of Technology and Science Chennai, India
Briskilal J
Assistant Professor
Dept of Computing Technologies
SRM Institute of Technology and Science
Diabetes is a disease caused by elevated blood sugar levels in the body. Diabetes shouldn't really be
overlooked. If left untreated, it can cause serious complications such as heart disease, kidney
May 21st – 22nd, 2022
disease, high blood pressure, loss of vision, and damage to other organs in the body. Early detection
of diabetes can be treated. To achieve this goal, we use a variety of machine learning techniques to
perform early predictions of diabetes in the human body or patients for greater accuracy. Machine
Learning Techniques You can improve predictions by building models from patient datasets. Given
a dataset, this study uses machine learning classification and ensemble strategies to predict diabetes.
K Nearest Neighbor (KNN), Logistic Regression (LR), Decision Tree (DT), Support Vector
Machine (SVM), Gradient Boosting (GB), and Random Forest are abbreviations for K Nearest
Neighbor (KNN), Logistic Regression (LR). Decision Tree (DT), Support Vector Machine (SVM),
Gradient Boosting (GB), or Random Forest (RF). Compared to other process, the outcome of each
model is distinguish. Project work gives an precise or more accurate model that gives that the model
can precisely predict diabetes. Our results show that Random Forest is dominating to other artificial
algorithms in terms of accuracy.
166. RSSI BASED SAND THEFT PREVENTION

1st Dhayapulle Sai Pavan
Dept name (Electronics and Communications Engineering)
(SRMIST)
Chennai, India
2nd k.Naga Pramodh
(SRMIST)
Chennai, India
3rd P.Thilak
(SRMIST)
Chennai, India
4th M.K.Srilekha
(SRMIST)
Cloud computing has a number of security issues, with the SAND theft attack being one of the most
significant. Existing security measures, like as encryption and decryption, are ineffective against
insider SAND theft. This study weaves a web of malicious fake expertise to prevent insider data
theft attacks. The approach is demonstrated by creating an example product based on a web
blueprint that demonstrates the usage of decoy technology. In addition to spreading disinformation
through decoy numbers, this importance puts doubt on illegal access and presents problematic
considerations. As a result, insider theft may be averted by using the Received Signal Strength
Indicator(RSSI) to measure the power received from the receiver’s returned signal.
167. A BRIEF REVIEW OF IMAGE CAPTIONING METHODS USING DEEP

LEARNING APPROACH
Rahul Kumar1, Manoj Kumar2
1,2 School of Computer Science & Engineering, Shri Mata Vaishno Devi University, India
Recent advances in deep learning have led to an increased interest in image captioning. A caption for
an image should adequately describe the image and provide a concise description. Moreover, the
generated captions should be syntactically and semantically correct. To achieve this goal numerous
May 21st – 22nd, 2022
architecture and techniques have been used. The baseline architecture for captioning contains an
encoder and decoder as its components. Where Convolutional Neural Network (CNN) is tested and
trained for extracting features(encoder) from the image and Long Short-Term Memory (LSTM) is
for generating the captions(decoder). In this study, we briefly discuss the methods and architectures
which are successful and efficient in recent years. MSCOCO, Flickr30k, and Flickr8k are the
prominent datasets for captioning and are widely used. To test methods quantitatively and
qualitatively various evaluation metrics have been used. Based on the resultant score of methods on
evaluation metrics we learn and compare the efficiency of those methods in this study.
168. DETECTION OF CYBER ATTACK IN NETWORK USING DIFFERENT

MACHINE LEARNING APPROACHES
B.ReddyBharath
Dept. Of Computer Science and Engineering
SRMIST
Chennai, India
G.Yaswanth
SRMIST
Chennai, India
Dr.C.Santhankrishnan
SRMIST
Chennai, India
In contrast to the past, advancements in computer and communication technologies have resulted in
broad and rapid transformations. People, organizations, and governments all benefit from new
inventions, yet some people, organizations, and governments are harmed . For example, security of
important data, the security of stored information , and the accessibility of the data, among other
things. Digital fear is one of the most critical challenges in today's world, based on these difficulties.
Digital apprehension, which has caused a slew of concerns for individuals and organizations, has
reached a point where it might jeopardize open and national security by many groups, including
criminal organizations, Intrusion Detection Systems (IDS) were developed in this vein to keep a
strategic distance from digital attacks. Currently, support vector machine computations were used to
detect port sweep initiatives based on the new CICIDS2017
In the this paper we use the ensembled based hybrid classification which can enhance the better
detection rate since here we use weak and strong classifiers
169. A CONTEMPLATION OF CLOUD SECURITY MENACES AND MITIGATION

TECHNIQUES
Satyam Kastwal1, Sanjay Sharma2
1,2School of computer Science & Engineering,Shri Mata Vaishno Devi University,India
In recent times, the influence of cloud over every single individual as well as over enterprises is
quite evident. As much as it eased the work of various IT firms in providing access to the data, an
exponential rise in the security threat of the data has also been increased simultaneously. The
chances of data loss and other such issues arise with shifting to cloud and there is a simultaneous
May 21st – 22nd, 2022
need to tackle such issues. This paper is a literature survey of different security mitigation
techniques used for clouds by various researchers. As much the world is dependent on the cloud,
there is an equal increase in potential threat to the data on cloud. Different types of attacks tend to
harm the critical data of cloud. Various researches have proposed different mitigation techniques to
the possible threats to the cloud. Latest of all, one of the research projects proposed the use of
blockchain to mitigate security threat to cloud.
170. BLOOD BANK MOBILE APPLICATION OF IOT-BASED ANDROID STUDIO

FOR COVID-19
Basetty Mallikarjuna, Associate Professor, Galgotias University, Greater Noida, Gautham Budha Nagar, Uttar
Pradesh 203201,
M. Lakshmi Prasad, Associate Professor, Department of Computer Science & Engineering(Data Science),
Institute of Aeronautical Engineering, Dundigal, 500043,Hydearabad,Telangana,India.
Munish Sabharwal, Professor and Dean, Galgotias University, Greater Noida, Gautham Budha Nagar, Uttar
Pradesh 203201
It is impossible to manufacture the blood, as it can be given by the donors. Blood bank retrieval
information can be given through the Android studio application, but there is not much work on the
integrated environment like IoT sensor connected with Android studio application development.
This paper provides the IoT health care sensors connected to the Android studio mobile application
development blood donors and blood receivers. The mobile application is most useful in an
emergency during the COVID-19 pandemic. This observational study gives the web-based
application development and also Android-studio mobile application development for blood bank
information retrieval system. The results are carried out in a real-time environment and updated
features of the blood bank mobile application.
171. VIRTUALLY CONTROLLING COMPUTERS USING HAND GESTURE AND

VOICE COMMANDS
Mayank Sharma1*, Akilesh G2, Vishwaa VS3, Sharon Femi P4, Kala A5
1,2,3UG Scholar, Department Of Information Technology, Sri Venkateswara College Of Engineering

4,5Assistant Professor, Department Of Information Technology, Sri Venkateswara College Of Engineering
In recent years, using a computer in a virtual environment has attracted the interest of researchers.
This paper presents an application that allows gadgets to be more interactive and functional with
little or no physical interaction. To interface with the system, the program provides voice gesture
recognition and hand gesture recognition functions. The interactive system begins with the user
being greeted by a voice assistant named "Proton." The assistant uses the user's input to accomplish
activities like as search, file navigation, date and time, and to start and stop the hand gesture
recognition system. A camera records a live video stream in a Hand Gesture Recognition system,
from which the input is collected via an interface. The system has been programmed to recognize a
variety of hand movements that may be used to do operations such as right-clicking, left- clicking,
opening a file, and so on. Our system detects the coordinates on the hand and recognizes the form
and motion when the user holds their hand in front of the camera. This is accomplished through the
use of the blazing palm detector model and the hand landmark model. In previous systems, data
gloves or markers were used to input data into the system. There are no such limitations in this
system. In front of the camera, the user may intuitively make hand motions. Hand gesture

May 21st – 22nd, 2022
recognition systems that are totally reliable are currently being researched and developed. The
created system acts as the basis for future work that may be expanded.
172. A UNIFIED APPROACH FOR DETECTING TRAFFIC ACCIDENT AND

ANALYSING THE ACCIDENT PRONE ZONE USING QUANVOLUTIONAL
NEURAL NETWORKS
N Suganthi1, Akarshak Bose2, Shruti Jaiswal3
1,2,3 Department of Computer Science and Engineering
Ramapuram Campus
Accidents have been a leading cause of mortality around the globe. In many situations, injuries are
claimed as a result of street consumers' irresponsibility and being stuck in remote locations.
Additionally, the great majority of persons around the accident site are concerned with snapping
pictures and films, completely unaware that their inattention might cost a life. This demonstrates
your society's insensitivity. The goal is to develop a system that can identify an accident using a live
video feed from the CCTV camera mounted on a roadway. The idea is to run each frame of a video
through a quanvolutional neural network model that has been trained to classify video frames as
either an accident or not. Furthermore, it will assist in identifying accident prone locations and
alerting vehicles to the accident prone zones ahead of them. Quanvolutional Neural Networks have
shown to be a quicker and more accurate method of picture classification. When compared to other
image classifying algorithms, QNN-based image classifiers achieve accuracy of more than 94
percent or higher for datasets and need less preprocessing. When compared to only standard CNNs,
QNN models displayed higher test set accuracy and quicker training times.
173. PREDICTION OF HEART DISEASE USING MACHINE LEARNING

ALGORITHMS
Ronak Piraka
Computer Science Engineering
Chennai, India
Karan
Chennai, India
Thenmozhi R
Chennai, India
Early detection of diseases has become a very crucial problem due to rapid growth of population in
medicinal research in the current times. With a sharp increase in the population density, the risk of
death due to heart diseases is rising at an alarming rate. The statistics shows us that at least one
person passes away every minute due to a heart disease. There are many factors that contribute to a
person having heart disease which are age, living conditions, do they smoke and much more. In this
paper, we are going to talk about multiple machine learning algorithms that are used to predict if a
person has a heart disease or not.

May 21st – 22nd, 2022
174. CODE PINK: A NEWBORN ABDUCTION DETECTION SYSTEM

Sathish G
Undergraduate: Department of Electronics & Communication Engineering
Sri Sairam Engineering College
Chennai, India
Dr.Sumathi S
Associate Professor: Department of Electronics & Communication Engineering
Chennai, India
Santhosh V
Chennai, India
Ajai Prasanna V N
Chennai, India
Child abduction incidents have long been a source of concern in our culture.In India, the number of
incidences of kidnapping or child swapping is steadily increasing. This study suggests a system for
live newborn tracking that can be recovered quickly if babies are exchanged or kidnapped.This
study suggests the creation of a prototype that can be easily worn and removed, that receives data
from satellites such as latitude and longitude, saves the baby's current location, and alerts the
parents. This wearable prototype includes a GPS module for tracking location and displaying exact
latitude and longitude, as well as a GSM cellular module for sending and receiving messages
through mobiles. If a baby is stolen or switched inside a hospital, the user can easily locate it by
sending a message to the wearable gadget, which will display the position in the browser via Google
Maps. We can defend against any crime that occurs using this technique, even if there is no guardian
present.
175. REAL TIME CRIME DETECTION USING GESTURE RECOGNITION
P. Sangeetha
Department of Information Technology
Chennai, India.
U. Shajitha Banu
Chennai, India.
K. Saranya
Department of Information
Technology SRM Valliammai
Engineering College Chennai, India.
P. Saranya
Chennai, India.

May 21st – 22nd, 2022
This paper proposes about an AI-based people security system that aims to protect people who are in
trouble. In a critical situation, this system can respond correctly both automatically and manually. To
accurately monitor the victim's status, the proposed system includes a Gesture, Speech, Violation
detection and location share with help of GPS. The collected data is processed using a python and
MediaPipe, and the protective action is done depending on the severity of the situation. The victim's
location is reported to the appropriate authority as part of the protection system's request for rescue
action. And alert will be generated.
176. PERFORMANCE ENHANCEMENT OF FULLY HOMOMORPHIC ENCRYPTION

AND DECRYPTION USING HIGH SPEED MULTIPLIER
Dr. G. SENTHILKUMAR
Associate Professor, Dept. of ECE, SCSVMV Deemed to be University, Kanchipuram, India.
A high speed fully homomorphic encryption system is intended as a key cryptographic tool for
developing a reliable and secure cloud computing environment, as it allows arbitrary evaluation of a
ciphertext without revealing the plaintext. However, existing encryption implementations stay
unrealistic due to very elevated time and resource costs. This paper presents the design of a power
and area efficient high speed 768 000-bit multiplier, based on Fast Fourier Transform multiplication
for a fully homomorphic encryption operation. By implementing a specific prime as the base for the
finite field, the radix-16 calculations were simplified to requiring only sum and shifting operations.
A two stage carry look ahead system was employed to determine carry and obtain multiplications
result. The multiplier design was implemented and verified by Verilog HDL and simulated by
Modelsim 6.4 c and MATLAB Tool. The Performance was measured by Xilinx tool synthesis
Method.
177. SMART ELECTRICITY DEMAND FORECASTING BY USING HYBRID ARIMA

WITH LSTM ALGORITHM
C.BRINDHA
Information Technology National Engineering College Kovilpatti, India
S.SIVA KUMAR
Information Technology
National Engineering College Kovilpatti, India
G.PANDIYA RAJAN
Information Technology National Engineering College Kovilpatti, India
K.RAJAKALI
National Engineering College Kovilpatti, India
For the operation and administration of power systems, demand forecasting, or the projection of
future energy demand, is required. Power supply and demand conflicts can be alleviated with
effective electricity demand forecasting. Furthermore, accurate load forecasting may help power
plants operate more efficiently while also ensuring grid safety. A drop of a few percentage points in
forecast accuracy, it is believed, would have a considerable cost impact on enterprises competing in
highly competitive power markets. We will use a Deep learning algorithm named Improved in our

May 21st – 22nd, 2022
project to estimate power consumption. We can obtain accurate future predicted output by utilising
improved hybrid ARIMA with LSTM.
178. MODIFIED M-RCNN APPROACH (AOD-NET) FOR ABANDONED OBJECT

DETECTION IN PUBLIC PLACES
Rahul Chiranjeevi V1*, Malathi D2,

1,2SRM Institute of Science and Technology, Kattankulathur, 603203, India
*Corresponding Author: Rahul Chiranjeevi
Detection of abandoned and stationary objects like luggage’s, boxes and machineries etc. in public
places is one of the challenging and critical tasks in video surveillance system. These objects may
contain weapons, bombs or other explosive materials that possess threat to public. Though various
applications have been developed to detect the stationary objects various challenges like occlusions,
change in geometrical features of objects etc. are still to be addressed. By considering the
complexity of scenarios in public places and variety of objects a context aware model is developed
based on Mask Region-Based Convolution Network (M-RCNN) for detecting the abandoned
objects. A modified convolution operation is implemented in the Backbone network to understand
features from geometric variations near objects. This modified operation layers can be adapted based
on the geometric variations in order to extract required features. Finally, a bounding box operation is
performed to locate the abandoned object and masking the particular object. Experiments have been
performed on the bench mark dataset like ABODA, which shows that a mAP of 0.6948% is
achieved. An ablation analysis has also been performed and a compared with other state of art
methods. Proposed model performed well in detecting the abandoned objects compared to existing
state of art methods.
179. AUTONOMOUS HOME GARDENING SYSTEM USING IOT

1Dr.P Suresh, 2S Maheshwaran, 3R Pavithra, 4N Monisha
1Associate Professor, 2,3,4 Students, Computer Science and Engineering, KPR Institute of
Engineering and Technology, Coimbatore
Plants have a significant role in reducing air pollution by absorbing the pollutant gases, absorbing
carbon dioxide, and releasing the oxygen that human beings can breathe. During day-to-day
activities, planting is reduced because of improper monitoring, and most people still use the physical
method for passing water to the plants. A proper irrigation system helps the plants grow well and
prevents them from the improper water supply that can cause adverse effects, leading to poor crop
development. The autonomous home gardening system uses IoT water conservation by providing
water based on the needs of the plants without any human resources. Node MCU and various
sensors are used to monitor the plants automatically. The sensors give information about the soil
moisture and temperature content connected to node MCU. It analyzes the parameter and passes the
water to the plant without the physical presence of humans. The information can be sent to mobile
devices by using the Android application.
180. THERMAL IMPACT SENSITIVITY ANALYSIS OF FLASH POWDER MIXTURE

FOR FIREWORKS APPLICATION
L.Prakash, A.Azhagurajan, R.Jegadeesan

May 21st – 22nd, 2022
Department of mechanical engineering, Mepco schlenk engineering college, Sivakasi-626005, Tamilnadu,

India
India
India
The fireworks flash powder composition containing aluminium, sulphur and potassium nitrate are
very sensitivity at room temperature. This paper suggests that the aluminium powder is replaced by
biodiesel coated dextrin powder to reduce the sensitivity of the existing flash powder. Open burning
tests were conducted to the newly synthesized flash powder and compared with the existing flash
powder mixture. The thermal impact sensitivity tests are carried out for both the existing and the
newly synthesized flash powder mixtures at temperatures 35°C, 40°C and 45°C. From the thermal
impact sensitivity results, the flash powder mixture having 20% of S, 57% of KNO3, 16.1% of Al,
and 6.9% of biodiesel coated dextrin powder is very sensitive at 40°C and 45°C.
181. EVALUATION OF GROUNDWATER CHEMISTRY AND MULTIVARIATE

STATISTICAL STUDIES IN PARTS OF TIRUPUR DISTRICT, TAMIL NADU,
INDIA
K.Arumugam*
Department of Civil Engineering,
Kongu Engineering College, Erode, India
T.Karthika
R.K.Sangeetha
J.Sujithkumar
Department of Mechanical Engineering,
Paavai Engineering College, Namakkal, India
K. Raja
S.Vikashini
Department of Pharmacy Practice,
Nandha College of Pharmacy, Erode, India
Groundwater quality at any location depends on the mutual possessions of several progressions
beside the subsurface flow path. Perceptive of natural history and degree of infectivity in particular
region involves absolute hydro-chemical statistics. In order to assess the groundwater chemistry and
its appropriateness for consumption purpose, sixty three subsurface water samples were analyzed for
evaluating to the standard values. Based on perceptive of the hydro-geological and groundwater
environment, the chemical characteristics of the groundwater were evaluated. The favored
parameters for the quality assessment are major anions and cations along with pH, turbidity, total
dissolved solids, total hardness, fluoride, and iron as trace metal. Total dissolved solids distribution,
hydrochemical facies, Piper diagram, factor analysis, correlation studies and cluster analysis were

May 21st – 22nd, 2022
carried out to evaluate the study zone. The physical parameter total dissolved solids (TDS) was
clustered independently. The assessment concludes that the groundwater of the Tirupur region is
critically deteriorated by different anthropogenic activities.
182. PREDICTION ON REAL ESTATE HOUSE PRICE USING REGRESSION

MODELS
Siddhartha Gaba1, *S. Iniyan2
1Department of Computer Science and Engineering, SRM Institute of Science and Technology,
Kattankulathur,
Chengalpattu -603203, India
2 Department of Computing Technologies, School of Computing, SRM Institute of Science and Technology,
Kattankulathur, Chengalpattu -603203, India
Machine learning has had a significant impact on image recognition, product recommendation,
clinical diagnosis, and nearly every other field of technology. Every aspect of our life is surrounded
by machine learning algorithms that help to improve security, public safety, medicine,
transportation, and so on. Because of the growth in urbanization and the volatility of property
values, there is a greater demand for a system that predicts property prices. Real estate price
prediction can assist in resolving this issue and forecasting house prices so that customers can
examine them. This paper reflects the effort made to solve the aforementioned challenge. A
comparison of the use of existing machine learning algorithms is conducted, proof of which method
is best suited for this task is supplied, and the prediction engine is developed for real-world use by
users.
183. TIME DELAY NEURAL NETWORK ANALYSIS FOR LIP READING - NEURAL
MACHINE TRANSLATION
Sangeetha R1 and Malathi D2
1,2 SRM Institute of Science and Technology Kattankulathur, Chennai
Lip-reading used to educate deaf and hard of hearing individuals to talk and communicate
successfully with other people for hundreds of years. Language content may be obtained through Lip
Reading by studying the speaker's lip shape and movement. The ability to lip-read helps deaf
individuals comprehend what others communicate. The speaker's subtitles are obtained by taking a
video of converting Hindi language to English language, importing it into the suggested model. To
get reliable results, users might employ deep learning technology to extract a vast number of various
characteristics. Our proposed approach uses sequence-to-sequence architectures built for (NMT)
Neural Machine Translation and audio speech recognition. To automatically learn nonlinear decision
surfaces, the Time-Delay Neural Network (TDNN) employs error back-propagation to construct a
three-layer hierarchy, allowing the network to discover sequences. The paper presented a TDNN
approach to lip recognition that possesses both of these essential properties. The suggested model
consists of three stages: Prepossessed, encoding, and decoding. As a result, it 1) introduces a unique
approach for inter-modality processing (before separate moralities acquire characteristics) and 2)
expands the previously suggested cross-connections. Our findings proved that a TDNN is the best
approach for predicting acoustics from sensor data. Our unique process combines the findings of
visual and audio speech to remedy the observed gaps. By lowering the word mistake rate the TDNN
achieved a detection accuracy of 98 percent after evaluating accuracy from speakers.

May 21st – 22nd, 2022
184. ON-ROAD VEHICLE BREAKDOWN ASSISTANCE

Vishesh Kumar, Mr. N Siva Kumar
Department of Master of Computer Application Hindustan Institute of Technology and Science, Chennai
603103, India
People have a strong desire to live a comfortable life free of physical labour. The Internet serves as
the backbone of all technologies, providing a very stable platform for any web- based product. The
Vehicle Service center Website and Mobile Application is a significant advancement in the field of
service centers. The goal of this project is to automate the overall process for the garage owner,
mechanic, and user by identifying the different aspects that affect the service and maintenance of
autos for everyone in order to reduce the time required for it. The project paper on 'Design Website
and Mobile Application for Vehicle Service Center' is available here. Vehicle owners can use an app
like this to communicate with the service center that does the work. Any vehicle user can utilize
these two programs, an android app and a website, at their leisure. The proposed solution is intended
to give the service center owner as much assistance as feasible. This method will aid the owner in
keeping track of the service center's general flow, including the exchange of replacement parts and
money. We're building this with user-friendliness, informativeness, and time savings in mind, but
only when the user registers with the necessary information.
185. A SURVEY ON VARIOUS MACHINE LEARNING TECHNIQUES FOR AN

EFFICIENT BRAIN TUMOR DETECTION FROM MRI IMAGES
Sanjay V1, Dr. Swarnalatha P2,*
1Research Scholar, School of Computer Science and Engineering, Vellore Institute of Technology, Vellore,
Tamil Nadu, India
2,*Associate Professor, School of Computer Science and Engineering, Vellore Institute of Technology,
Vellore, Tamil Nadu, India
On account of the uncontrolled and quick growth of cells, Brain Tumor (BT) occurs. It may bring
about death if not treated at an early phase. Brain Tumor Detection (BTD) has turned out to be a
propitious research field in the current decennia. Precise segmentation along with classification
sustains to be a difficult task in spite of several important efforts and propitious results in this field.
The main complexity of BTD emerges from the change in tumor location, shape, along with size.
Providing detailed literature on BTD via Magnetic Resonance Imaging (MRI) utilizing Machine
Learning (ML) methods to aid the researchers is the goal of this review. Diverse datasets are
mentioned which are utilized most often in the surveyed articles as a prime source of Brain Disease
(BD) data. Furthermore, a concise epitome of diverse segmentation methods that are utilized in
diagnosing BDs has been offered. Lastly, an outline of key outcomes from the surveyed articles is
exhibited, and several main problems related to ML-centered BD diagnostic methodologies are
elucidated. The most precise method to detect diverse BDs can be engaged for future advancement
via this study.

May 21st – 22nd, 2022
186. CLASSIFICATION OF DIABETIC RETINOPATHY DISEASE

Dr.Mohan Das. R
Department of CSE
Kalasalingam Academy of Research and Education.
Yetrinthala Manoj kumar
Department of CSE
Varra Goutham Reddy
Department of CSE
Indukuru Ramachartan
Department of CSE
The eye is considered as most sensitive organ in human body. The people who are suffering with
Diabetics may get affected with Diabetic Retinopathy. It is required that classification diabetic
retinopathy with the best technologies and methodologies. Artificial intelligence is having its vast
applications in various sectors. Due to its ability to perceive the problems, developing the
appropriate reasons for that, deep learning can act as a great aid in addressing the severity of the
disease. This project provide a brief overview of the application of convolutional neural network,
and highlights the various methods available for the detection of disease of diabetic retinopathy on
basis of severity.
187. COLLABORATIVE EDGE DATA CACHING - ONLINE

Nivetha N, Krishnaveni S, Pupitha P, Aswathy.R H
KPR institute of engineering and technology
Edge servers installed at base positions provide highly available figuring and storing data to
neighboring app handlers in an edge computing (EC) setting. From the standpoint of the app vendor,
reserving information on edge servers can shorten the period it takes for app users to retrieve
information. An edge server, on the other hand, usually has inadequate information due to its small
scope. By solving the collaborative caching tricky in the EC environment, the purpose of this study
is to reduce system expenses such as facts caching, information migration, and many more. The
collaborative edge data caching issue (CEDC) is an NP-complete problem with a constrained
optimization statement. OCEDC is an operational algorithm for solving the CEDC problem in all
time slots that has been suggested. OCEDC is based on Lyapunov boost and provides verifiable
nearoptimal results while operating online without the need for future data. OCEDC is put to the test
on a real-world dataset, and the findings show that it outdoes quatern sampling alternatives by a
significant margin.

May 21st – 22nd, 2022
188. MITIGATION OF FIRE AND EXPLOSION HAZARDS OF FLASH POWDER BY

ADDING MONO AMMONIUM PHOSPHATE AS A SOLID INERTANT
A.Azhagurajan, K.Kandakodeeswaran and C.Hariram
Department of mechanical engineering, Mepco Schlenk Engineering College, Sivakasi, Tamilnadu, India.
Flash powder is a chemical mixture used in the fireworks industry that is highly sensitive in nature.
According to the Petroleum and Explosives Safety Organization's (PESO) guidelines, fireworks
waste should not be stored or accumulated, thus a large pit is dug and all of the debris is burned on
the same day. A solid inertant mono-ammonium phosphate chemical is utilised as a fire
extinguishing agent and to lower the flash powder's sensitivity. The tests were conducted using a
mixture of flash powder and mono-ammonium phosphate in various proportions. It is found that as
the amount of inertant increase in the sample impact due to height increases and friction sensitivity
decreases make the sample in-sensitive and for the layer ignition temperature in the composition of
75% of monoammonium phosphate and 25% flash powder the sample forms into a lump and colour
change is observed this indicates the sample becomes in-sensitive.
189. A COMPREHENSIVE REVIEW ON TRACTION MOTORS COMPARISON,

OVERVIEW OF OPTIMISATION TECHNIQUES, RECENT AND FUTURE TRENDS
IN ELECTRIC VEHICLE
Krishnakumar C1 Sarojini L2
1Professor of Electrical and Electronics Engineering, Saranathan College of Engineering,
Anna University, Trichy-620012, India.
2Research Scholar of Electrical and Electronics Engineering Department, Saranathan College of Engineering,
Anna university, Trichy-620012,India.
The increasing inclination towards electric vehicles over the last periods is due to various factors.
The motors used in electric vehicles are substantially different from those used in other applications.
The motor in EV applications necessitate excellent efficiency, torque density, reliability, high power
density and high speed, This paper reviews different electrical motor for EV propulsion applications.
A comparative analysis of the DC motor, induction motor, switching reluctance motor, and
permanent magnet synchronous motor is performed to evaluate the best suitable electric motor
drives for electric vehicle applications based on efficiency, power density, cost, maximum speed,
torque ripple, range and fault tolerance. A detailed review is presented on recent existing motors
which include axial flux motors, Flux switching permanent magnet motor, surface mount and
interior permanent magnet synchronous motor. In addition, overviews of control, optimization
algorithms and evaluation of performance for EV electric motor drives are provided. Permanent
magnet synchronous motor has been preferred choice in electric vehicles, according to the review.
190. AI – BASED SMART IRRIGATION SYSTEM

Sardhi Hawa.J1*, Priyanka.S2*, Subhashree.S3*, Dr. Jackins.V4*
1 Under Graduate, Information Technology, National Engineering College, Kovilpatti.
2 Under Graduate, Information Technology, National Engineering College, Kovilpatti.
. 3 Under Graduate, Information Technology, National Engineering College, Kovilpatti.

May 21st – 22nd, 2022
4 Assistant Professor(SG), Information Technology, National Engineering College, Kovilpatti .
Globally, irrigation is scheduled according to crop visual inspection by farmers and as a result, about
50% of water is wasted by traditional irrigation systems. Sprinkle irrigation, drip irrigation, and
furrow irrigation are examples of controlled irrigation methods that decrease water wastage by 30 –
70%. These systems, however, fail to hold precise water in the soil owing to their open structure,
which affects the quality and number of plants by irrigating less or dramatically reducing soil
nutritive. Therefore, there’s a desire for feedback on integrated precision irrigation practices to
efficiently utilize water without harming the development of crops. Precision irrigation takes soil
moisture, climate information, and crop type to estimate water is required or not. If therefore reduces
the cost of farm workers. The proposed system uses a hybrid random forest with a linear model
(HRFLM) for prediction work. For the hybrid random forest and linear models, we employed an
ensemble voting classifier. Due to the architecture of the hybrid algorithm, we achieved accuracy for
prediction. To get more than 96% of accuracy.
191. CUSTOMER SEGMENTATION USING NATURAL LANGUAGE

PROCESSING
Sabari Balamurugan, Sahid Akil , Sathyan
Sri Krishna college of technology, Coimbatore,India
SA is commonly known as Sentimental analysis is a continuous field of research on analysis

consumer emotions on purchase of products. This survey paper tackles an inclusive survey of
mobile phone reviews. One of the most significant uses of unsupervised learning is customer
segmentation. Companies can discover different categories of clients using clustering algorithms,
allowing them to target the prospective user base. In this machine learning project, we will utilize
nltk for clustering, which is the most important approach for grouping unlabeled datasets.
The acquired findings illustrate the efficacy of the solution, which has a high level of
accuracy in both mobile classification and user segmentation.
192. ENERGY EFFICIENT CH SELECTION & OPTIMAL RELAY NODE SELECTION

ALGORITHM AND ITS PERFORMACE ANALYSIS FOR WIRELESS SENSOR
NETWORKS
S ANURADHA, Dr. D.Sreenivasa rao
Research Scholar, ECE, Professor ,ECE
JNTUH Hyderabad, India
Based on the application type, a wireless sensor network would be a heterogeneous type. The energy
consumption is one of the major issues for WSNs. The overall energy conservation of WSNs would
be reduced by the existing communication protocols. The network is partitioned into the clusters
using the low-energy adaptive clustering hierarchy (LEACH), which is a cluster-based protocol that
chooses a few sensor nodes randomly as cluster heads (CHs) according to the generation of random
numbers in each sensor node. The CHs role rotates for even distribution of energy load among
May 21st – 22nd, 2022
sensor nodes of a network. The improved version of LEACH protocol has been proposed to achieve
the energyefficient and stable CH selection. The random number is generated using a proposed work
based on the inclusion of different parameters like average residual energy (ARE) and CH coverage
(CHC). The chosen CHs would have more neighbours in the coverage circle by using the CH
coverage parameter (CHC) that reduces the energy consumption by member nodes towards reaching
the CH. The higher energy should have included in the selected CHs that ensures by the average
residual energy (ARE). The multi-objective selection parameters have been improved the Particle
Swarm Optimizer (PSO)’s fitness function to enhance the selection of a relay node and energy
efficiency during the phase of data transmission. To carry out the relay selection, the proposed
enhanced-PSO algorithm is considered the parameters like Average Energy Load (AEL), Link Error
Rate (LER), and Traffic Index (TI). Based on the analysis of simulation results, the proposed
technique achieves increased performance in terms of communication quality via dynamic and static
scenarios, network lifetime, and energy consumption.
193. ENERGY EFFICIENT STREET LIGHTING SYSTEM

Dharshini V, Saranya S, Sethuraman V K, Jawahar A, Vinoth Kumar .C
Department of Electronics and Communication Engineering Sri Sivasubramaniya Nadar College of
Engineering,
Rajiv Gandhi Salai (OMR), Kalavakkam, Tamil Nadu, India
To reduce consumption of energy, energy efficient smart street lighting system is implemented
which saves a lot of energy compared with conventional street lighting system. The main function of
energy efficient street lighting system is light will glow, if there is no light and no motion of vehicles
and human. Most of the countries using fluorescent lamps which consumes a lot of energy. In this
work, energy efficient smart street lighting system is proposed to save the energy by using light
Emitting Diode(LED), LDR sensor, Ultrasonic sensor driven by raspberry pi. By comparing the
proposed street light and conventional street light, there is huge difference in energy consumption.
In this proposed model, LDR sensor senses the intensity of light. ultrasonic sensor is used to detect
the movements of the object. Based on the LDR and Ultrasonic sensor output values, we can
determine either led should glow or not. From this proposed system, the energy usage drastically
reduced.
194. DEEP LEARNING BASED NEUROIMAGING ANALYSIS FOR PARKINSON’S

DISEASE DETECTION
Mitali Kargwal
Student
B.Tech, Computer Science Engineering
Dr. M. Suchithra
Associate Professor (0.G) Department of Computing Technologies
Mrs. Mahalakshmi.P
Assistant Professor (0.G) Department of Networking and Communications

May 21st – 22nd, 2022
Parkinson’s disease (PD) is the second most prevalent neurodegenerative disorder affecting more
than 10 million people globally. Being an incurable disease, its onset prediction and analysis is
crucial. Deep learning techniques have been progressing significantly towards neuroimaging and
other medical analyses. In this study, we leverage the use of various Convolution Neural Network
models such as 2D-CNN, VGG16, ResNet50 and LeNet-5 to classify the intricate functional
Magnetic Resonance Imaging(fMRI) scans for distinguishing the Parkinson’s Disease affected
brains from healthy ones. By achieving the highest accuracy of 99.70%, this endeavour concluded
that CNN architectures have the prospect to draw out extensive discerning features and novel
biomarkers from complex clinical data and therefore has the competence to be beneficially utilized
to a greater extent in more advanced systems.
195. RECOMMENDATION SYSTEM FOR COLLEGE ELECTIVES USING ALS

IMPLICIT COLLABORATIVE FILTERING AND LSTM NETWORK
Shreyas Raorane1, Smitha Sampathkumar1, Supraja P2
1UG Student, Department of Networking and communications, SRM Institute of Science and Technology,
Kattankulathur
2Associate Professor, Department of Networking and communications, SRM Institute of Science and
Technology, Kattankulathur
Elective recommendation systems have become an essential technology in the field of higher
education in today’s world. The spectrum of subjects and technologies included in today’s education
system has increased drastically [1] [2]. Over the span of a century, courses and subjects have
penetrated their way in and out of an educational institute’s curriculum. The aim of this system is to
help students make an informed decision about their electives through the use of machine learning
algorithms and historical data. [4]In this model, we use ALS Implicit Collaborative Filtering and
LSTM network to put in place an effective structure that will provide recommendations based on the
student's interest, rating, and historical data. The system allows the students to prioritize one
category of rating over the other based on their interests. This enhances the system’s performance by
ensuring that the subjects that have a higher average rating are being given preference.
196. IOT BASED SMART SECURITY AND SURVEILLANCE SYSTEM

Aishwarya Jagadiswaran
SIES Graduate School of Technology
University of Mumbai
Navi Mumbai, India
Ritika Karthikeyan
Navi Mumbai, India
Seema S.Redekar
Navi Mumbai, India
May 21st – 22nd, 2022
Jagthish Thevar
Navi Mumbai, India
IoT is an escalating trend in today’s world and has proved to be a game-changer in the field of
technology. There is a high need for security systems in our everyday life. For example, security in a
business space, an organization is important now. Currently, security cameras are being used in
order to develop secure places in the company. Sensors and devices are used to develop new systems
to overcome existing security issues. Lately, there are many security systems, which make use of
various motion sensors to detect any motion and notify the owner. However, most of these systems
do not provide the features of remote camera surveillance, power efficiency, advanced security
combined with ease of use. The main aim of this paper is to enhance the traditional security system
and overcome all these problems by developing a smart security and surveillance system that makes
use of multiple sensors, provides a warning to the person who has intentionally or unintentionally
entered the company. The main advantage of the system is the ability to remotely manage the
organization 24/7. We aim to design lightweight, low-cost, extensible, flexible, smart security
systems using IoT which employs the integration of various latest technologies that can be used to
control a system in an organization. All these features combined create a reliable security system.
197. MITIGATING SHATTER BEHAVIOR OF FIREWORKS BUILDING

STRUCTURE USING REINFORCED HOLLOW BLOCK
A.Azhagurajan1, L.Prakash2, T.Ganesh Kumar3
1Department of mechanical engineering, Mepco schlenk engineering college, Sivakasi -626005, Tamilnadu,
India
India
India
In this paper, the load carrying capacity of reinforced hollow block (Steel fibre) was analyzed by
conducting varies experiment like compression and impact test. The experiment was carried out by
varying the reinforced percentage (1%,2%&1% chopped) and different types of hollow block used
like hollow block, hollow mould block. The reinforced hollow block was manufactured under the
guidance of authorized hollow block manufactured. The steel fibres are randomly distributed along
with raw materials during the mixing process. After the curing process, the hollow block was tested
by using universal compression test equipment and concrete impact test setup. The results show that
the compression strength increased by 33.45 % and impact resistance improved up to 13.22% for the
2% steel fibre reinforced hollow mould block. The finite element analysis using Ansys explicit
dynamics also reflects the more or less equivalence result.
198. GESTURE RECOGNITION SYSTEM USING CONVOLUTED NEURAL

NETWORKS AND TRANSFER LEARNING
Varun R Swaminathan
100 Proceeding of International Conference on Contemporary Engineering and
Technology
May 21st – 22nd, 2022
Computing Technologies
Shubhaam Agrrawal
Dr. C Santhanakrishnan
Webserver is a web application designed to convert images to text, which may then be said out using
the anticipated text generated by the image converter. For this application, images are captured
through webcam as hand gesture inputs. In addition, Google Home and Amazon's Alexa may be
used in conjunction with this software. The collected movements may be used for specific smart
device functions, and the software can also call out the gesture inputs in various operating system
voices (OS). Browsers as well as smart devices are able to handle these multiple OS voices.
TensorFlow was used to create this HTML and JavaScript application. In order to run this JavaScript
model, we need the Node.js runtime environment. Using a Machine Learning (ML) method, there is
no need to install drivers or libraries in this application. Once a user has found the webpage, the
software is ready for use.
199. PRODUCT RECOMMENDATION ENGINE USING KNN AND K-MEANS

Arun Prashath D
UG Scholar, Department of CSE,
KPRIET, Coimbatore, India.
Dhivakar PS
Dr. Devi Priya A
Assistant Professor, Department of CSE,
Gokul G
The sale and purchase of items is beginning to shift from being done in person to being done online.
The goal of this research is to create a product recommendation system on an e-commerce platform
that is tailored to the demands of the users. This research focuses on providing recommendations
based on the 'rating' or 'preference' that a user would give to an item and provides recommendations
to existing users and new users. These users can get personalized recommendations from the
recommendation system, which can help them make better decisions in their online purchases, boost
sales, redefine their user experience, retain consumers, and improve their shopping experience. Thus
the Seller can determine which product to launch in the market based on consumer preferences to
gain more benefits.

May 21st – 22nd, 2022
200. VISION AND INFRARED SENSOR BASED DISTANCE ESTIMATION FOR

MICRO AERIAL VEHICLE NAVIGATION
K. Mohith Raj1 , Dr.B.Anbarasu2, M. Bhuvanesh3, J. Dinesh4
Hindustan Institute of Technology And Science, Padur, Chennai
1. UG student, Department of Aeronautical Engineering, Hindustan Institute of Technology and
Science, Chennai, Tamil Nadu, India.
2. Assistant Professor, Department of Aeronautical Engineering, Hindustan Institute of Technology and
To know the distance of things without using the old school methods of scales are something that’s
fascinating. Technology development has led to bringing numerous methods to find the distance of
an object using computer technology. The method most used is the one involving camera. This
method has proved to work well but one of the major drawbacks is in identifying the image. On the
other hand, distance sensors help us accomplish this task more efficiently. Therefore, this research
work is based on how a camera and a distance sensor is used to measure the distance of an object or
an obstacle.
201. INDOOR MICRO AERIAL VEHICLE NAVIGATION USING OPTICAL FLOW

Dr.B. Anbarasu1, M. Bhuvanesh2, J. Dinesh3, K. Mohith Raj 4
Hindustan Institute of Technology And Science, Padur, Chennai
1.Assistant Professor, Department of Aeronautical Engineering, Hindustan Institute of Technology and
2.UG student, Department of Aeronautical Engineering, Hindustan Institute of Technology and Science,
Chennai, Tamil Nadu, India.
Micro Aerial Vehicle (MAV) is widely employed in the global industry, and it is used in
various of sectors for a variety of applications. The MAV is used by a variety of companies to
transport packages. They're also deployed in a variety of military and civilian applications. For long
range and non-viewable locations, the MAV utilizing image processing. The MAV are unmanned
vehicle they depend upon the information sent by the sensor for the navigation. Sometimes failure
may occur and the MAV may face some damages. Until date, the work in this field has focused on
object detection or distance estimate; however, in this study, the algorithm that developed can
compute both. This combined process will be more useful in the indoor environment for the
navigating the MAV (can be used for avoiding the collision or an obstacle easily while navigating in
the indoor environment). This combined method may reduce the chances of failure. Detailed
explanation and experimental results are explained below.

May 21st – 22nd, 2022
202. BRAIN TUMOR DETECTION USING DEEP LEARNING

M.Raja,Department of Computer Science and Engineering,Kalasalingam Academy of Research and
Education,Anand Nagar, Krishnankoil
Venkata sai Ramanjaneya Reddy Pagadala,Department of Computer Science and Engineering,Kalasalingam
Academy of Research and Education,Anand Nagar, Krishnankoil
Leela nath Reddy Pereddy,Department of Computer Science and Engineering,Kalasalingam Academy of
Research and Education,Anand Nagar, Krishnankoil
Surya Prakash Narahari,Department of Computer Science and Engineering,Kalasalingam Academy of
Research and Education,Anand Nagar, Krishnankoil
Brain tumour analysis without human intervention is a critical field of research. Convolutional
neural networks, on the other hand, can be used to do this (CNNs). They have excelled at solving
computer vision problems as well as a variety of others, including visual object recognition,
detection, and segmentation. It detects brain tumours by improving brain images utilising
segmentation algorithms that are extremely resistant to noise and cluster size sensitivity issues, as
well as automatic area of Interest (ROI) recognition. One of the key arguments for using CNNs is
that they have a high level of accuracy and do not require manual feature extraction. Detecting a
brain tumour and correctly identifying its type is a difficult undertaking. The performance of CCN is
great.than others thanks to its wide usage in recognising images. neoplasm Manual classification
with It's possible to make blunders in prediction and diagnosis with human assistance. Furthermore,
it is a frustrating chore when there is an excessive amount of knowledge to be aided. Because of the
wide diversity of appearances of brain tumours and the resemblance of tumour and normal tissues,
extracting tumour regions from pictures becomes difficult.
203. REVIEW ON “SEISMIC RESPONSE CONTROL OF IRREGULAR RCC

BUILDING USING BRACING SYSTEM.”
Yogesh Dhamge1, a), Prof. Sanket Sanghai2, b),
1, a), 2, b)Department of Civil Engineering, G H Raisoni College of Engineering, Nagpur, Maharashtra, India
A large number of buildings are now being built and planned based on the needs of the buildings and
aesthetic considerations. Most buildings are built with a local structure such as X shape or V shape
with inconsistent x and y links. Earthquakes cause extensive damage to various buildings, the main
problem being thin. The main purpose of this study is to make a comparative analysis of the
dynamic behavior of buildings with different structural properties in all earthquake zones and soil
types.
RC structures with an unusual G + 10 floor are also analyzed without the cross-bracing integration
system, which is one of the best ways to concentrate, in the presented problem. The structural
response to various earth movements is determined using analytical time history. For analysis, the
ETABS 2016 program has been hired. It will investigate the effects of the models (most narrative
shifts, multi-storey shear, and multi-storey rigidity).

May 21st – 22nd, 2022
204. PERFORMANCE ANALYSIS OF SPATIO-TEMPORAL HUMAN DETECTED

KEYFRAME EXTRACTION
Victoria Priscilla C1 and Rajeshwari D2
1Associate Professor & Head , PG Department of Computer Science, Shrimathi Devkunvar Nanalal Bhatt
Vaishnav College for Women, University of Madras, Chennai (TamilNadu), India.
2Research Scholar, Research Department of Computer Science, Shrimathi Devkunvar Nanalal Bhatt Vaishnav
College for Women, University of Madras, Chennai (TamilNadu), India.
Closed circuit television (CCTV) surveillance for detecting the humans involves an expanded
research analysis especially for crime scene detection due to various restraints such as crowded
annotation, night footages, and rainy (noisy) clips. The main visualization of the crime scene is to
recognize the person in particular obtained in all frames is a challenging task. For this occurrence,
Content-Based Video Retrieval (CBVR) method refines the collection of these video frames
resulting keyframes to reduce the burden of huge storage. Here, Spatio-Temporal classifiers method
as an added advantage with frame differencing and edge detection method reports the human
detected keyframes without the termination of background regions in order to negotiate the crime
scene more efficiently. The main objective of this paper is to analyze the obtained keyframes with
Human detection pointing a distinctive between Spatio-Temporal HOG-SVM and HAAR-like
classifier to survey the optimum. Finally, the resulting keyframes mutated with the canny edge
detection method by HOG-SVM sequel with greater accuracy level of 98.21% compared to HAAR-
like classifier.
205. AN INTERACTIVE PHYSIOTHERAPIC EQUIPMENTS FOR CEREBRAL

PALSIC CHILDREN
M.Kathiravan1,Department of Biomedical Engineering Jerusalem college of Engineering Chennai,
Tamilnadu,Dharshini Venkatesan2, Department of Biomedical Engineering SSN college of
Engineering,Chennai, Tamilnadu
R.Arulselvi2,Department of Biomedical Engineering SSN college of Engineering Chennai, Tamilnadu,

Jeevitha R2,Deparment of Biomedical Engineering SSN college of EngineeringChennai, Tamilnadu, R.Merlin
Jeba Limsa2, Department of Biomedical Engineering Jerusalem college of Engineering Chennai, Tamilnadu
World Health Organisation states that 20% of children is affected by cerebral palsy which is a
neurological disorder that affects the muscular movement in prenatal, perinatal and postnatal time
period. Rehabilitation for cerebral palsy children involves physiotherapy in which the sessions are
extremely long, tedious and the children are reluctant to continue the entire process . This drawback
can be overcome by Interactive physiotherapic equipment. This helps the children to deviate from
their pain during the therapy. In this project, shoulder wheel is modified as an interactive equipment
by mounting Infra Red sensors and Light Emitting Diode. Infra Red sensors pick signal from eight
sectors in board which triggers the Light Emitting Diode to glow that act as visual feedback for
cerebral palsy children.In addition to that Liquid Crystal Display gives the score for the session done
by Cerebral Palsy children which will help for evaluation purpose and each rotation is accompanied
by a chocolate from automatic chocolate dispenser. The parallel bar is modified as an interactive
equipment by fixing pressure sensors and the audio, robotic action is given as Feedback. The entire
operation is controlled by Arduino microcontroller in both the equipments and it is found that it
produces better response and reduce reluctance during physiotherapy session.
May 21st – 22nd, 2022
206. FLY ASH AS MINERAL ADMIXTURE IN CONCRETE MATRIX – A STATE OF

THE ART
Dr.Brindha Sathiaseelan1,a), Mrs.M.HannahAngelin and Dr.P.Sathiaseelan2, 3, b)
1, 2Assistant Professor, Department of Civil Engineering, Coimbatore Institute of Technology, Coimbatore,
India. 3Professor and Head, Department of Civil Engineering, PPGIT, Coimbatore, India.
Among recent research, the use of fly ash in partial replacement of cement has steadily increased as
an inexpensive and practical method to dispose of the thermal station by-product, namely fly ash.
This extra ingredient influences the qualities of concrete based on a variety of elements such as
cement grade, % of replacement, environmental conditions, and curing times. As fly ash might affect
the compressive strength and rheological properties of concrete, a review of prior reports on fly ash
concrete mixes subjected to various experimental settings are done. When the substitution
proportion is appropriate, it is proven that a low calcium fly ash concentration increases the
compressive strength of the concrete of concrete matrix over a range of 28 to 56 days and optimizes
the concrete rheological properties. When fly ash is used as a viable substitute for cement, it has a
low-cost and long-term function on the concrete.
207. A REVIEW: BLOCKCHAIN & CRYPTOGRAPHY USED TO SECURE DATA
Kavitha C V∗, Lokamathe P∗, M Yamuna∗, Meghana M S∗, Mrs. Tanuja Kayarga (Assistant
professor) †, ∗†Department of Computer Science and Engineering, Vidyavardhaka College of
Engineering (VVCE) Mysuru,
Block chain technology is gaining more popularity over the past few years. Blockchain is a
distributed site that is shared between computer network nodes. In today's world of advanced
technology, many people are victims of cybercrime. Data protection becomes one of the most
important things. One of the key technologies of Blockchain is cryptography. Cryptography is a
technique which transform legible data to an ambiguous conformation with the help of keys and
does the same in conversely. The performance of the algorithm steady primarily based on the
complication and rotation of the CPU needed for the Crypto analysis in it. By combining
Cryptography techniques and Blockchain technologies we will see how data can be effectively
protected.
208. BIOMETRIC SYSTEM FOR FACE RECOGNITION USING TENSOR FLOW ON

RASPBERRYPI FOR EDUCATIONAL INSTITUTIONS
K. Krishna Vamsee Vardhaman College of Engineering Hyderabad, India
Kalyan Jalaneela Vardhaman College of Engineering,Hyderabad, India
Mr. T. Ramakrishnaiah Vardhaman College of Engineering Hyderabad, India
The proposal's main purpose is to develop a facial recognition-based attendance management device
for academic institutions in enhancing and replace the traditional attendance system and boost
productivity. The existing outdated method has several flaws. There was a lot of ambiguity, which
May 21st – 22nd, 2022
resulted in erroneous and inefficient attendance taking. There are numerous issues. When the
authority is unable to enforce the previous system's regulations, problems develop. Thus, this project
will use technology to fix the weaknesses in the current system while remaining cost-effective. By
automating the majority of the tasks, you can take attendance to a whole new level. The face
recognition system will be the technology at work.
The face recognition system will be the technology at work. One of the fundamental features that
could be used to define someone is their face. As a result, it is utilized to trace individual because the
chances of a face deviating or being replicated are limited. Face databases will be built in this project
to feed data into the recognizer algorithm. The likelihood of data being repeated is low. Throughout
the attendance-taking process, the features would be linked to the system to verify identity.
Whenever an user is spotted, their attendance is recorded right away, as well as the necessary data is
maintained in a relational database. The attendance data is collected after the day. A web server
operated by the Raspberry Pi can be used to obtain information about an individual.
In brief, this improved generation of the attendance tracking model not only saved a lot of money
but also gave the authority a lot of ease because many processes were automated.
209. CLIENT-SERVER ARCHITECTURE, IOT DEVICES, FACIAL BIOMETRIC

SYSTEM, LOCAL DESCRIPTORS, TENSOR FLOW, CONVOLUTIONAL NEURAL
NETWORKS (CNN), DEEP LEARNING
Dr. S. Arunvinthan Pillai,Department of Aeronautical Engineering,Kalasalingam Academy of Research and
Education ,Krishnankoil, Virudhunagar, India
V Manoj,Department of Aeronautical Engineering,Kalasalingam Academy of Research and Education
Krishnankoil, Virudhunagar, India
C Hari Babu, Department of Aeronautical Engineering, Kalasalingam Academy of Research and Education,
The project aims to use passive flow control technology on a leading edge protuberant wing and its
impact on aerodynamic performance at different Reynolds number varying surface roughness. Due
to its wide applications in areas such as aircraft wings, wind turbine blades and propellers. The
aerodynamic performance characteristics of the leading-edge protuberanced wing at different
Reynolds Number with surface roughness have been tested. Adding protuberances to the leading
edge of the wing is considered a viable passive control method for separating the flow and
increasing aerodynamic efficiency by positively affecting lift characteristics without introducing
additional drag. The aerodynamic performance of the modified airfoil with leading-edge
protuberances was examined and compared with the baseline NACA 63(4)-021 airfoil. Airfoils were
pitched from an initial angle of attack of 0 ∘ to 45 ∘ at constant pitch rates with Reynolds number
from to and surface roughness of 10 different grit sizes. Flow parameters such as velocity is
chosen to find the optimised configuration keeping the amplitude constant as 0.12c and varying
wavelength of the leading-edge protuberances as 0.25c,0.5c,0.75c.

May 21st – 22nd, 2022
210. HYBRID RECOMMENDATION SYSTEM WITH GRAPH BASED AND

COLLABORATIVE FILTERING RECOMMENDATION SYSTEMS
Shubham Khanduri, SRM Institute of Science and Technology, KTR, Chennai, S. Prabakeran, SRM Institute
of
Science and Technology, KTR, Chennai
Every online platform on the internet needs to have some kind of recommendation system. A
recommendation system will suggest products or services on a platform to a user, based on their
preference and history. This increases user engagement on the platform, as they help the user in their
choices. Hence, there is a need for a simple recommendation system solution.
Recommendation system works by creating relations between products or the users; and by
evaluation of those relations, it can suggest relevant products related to any product. Hybrid
Recommendation combines the predictions of two or more recommendation techniques. This system
combines the predictions from product - based predictions; implemented using a Graph based
recommendation system, and users - based predictions; implemented using Collaborative Filtering
recommendation system.
211. DRAIN CURRENT ANALYSIS OF Mg0.25Zn0.75O/ Mg0.6Zn0.4O/ZnO

DOUBLE-CAVITY MOSHEMT BASED BIOSENSOR
Kiran G, and Rohit Singh, Department of Electrical Engineering, Shiv Nadar University, G. B.
Nagar -201314, India,
In this work, Mg0.25Zn0.75O/Mg0.6Zn0.4O/ZnO MOS-HEMT (Metal Oxide Semiconductor -High

Electron Mobility Transistor) with a double-cavity below the gate region in both source and drain
sides is proposed for its drain current (IDS) analysis as a bio-sensing device. The analysis is done by
dielectric modulation of the cavity regions to emulate the presence of different dielectric
biomolecules by interfacing capacitance variation. The effective capacitance is obtained by using
dielectric modulation which varies the overall 2DEG density in the device channel so drain-current
will get changed. Interestingly, the changes in the drain current of the device are used as the sensing
metric for the detection of bio-molecules in the cavity region. The predicted sensing metric values of
the developed analytical model are in good agreement with the result of other similar works, thus
confirming the validity of the proposed model.
212. DECENTRALIZED PAYMENT ARCHITECTURE FOR E-COMMERCE AND

UTILITY TRANSACTIONS WITH GOVERNMENT VERIFIED IDENTITIES
Shiny Duela, Department of Computer Science Engineering, SRM Institute of Science and
Technology,Chennai, India
Dr K Raja,Department of Computer Science Engineering,SRM Institute of Science and Technology,Chennai,
India
Prashanth Umapathy,Department of Computer Science Engineering,SRM Institute of Science and
Rahul Rangnani,Department of Computer Science Engineering,SRM Institute of Science and
Ashish Patel,Department of Computer Science Engineering,SRM Institute of Science and

May 21st – 22nd, 2022
In recent years, blockchain technologies have begun to be adopted by a number of industries as a

means of improving security and efficiency. One area where blockchain technology is beginning to
be used is in the area of payment gateways. Payment gateways are used to process electronic
payments between buyers and sellers. By using blockchain, payment gateways can be made more
secure and efficient. However, there are concerns that blockchain technologies may be used to
launder money or finance terrorism. To address these concerns, some countries have proposed
making blockchain technologies only accessible to government-identified individuals. This research
paper examines the feasibility of such a proposal and demonstrates how our proposed architecture
model can perform in real-time. It will discuss the benefits of using blockchain technology in this
context and how it can help to improve security and efficiency. The paper focuses on two essential
aspects of the design: how to verify the identities of individuals in transactions and how to maximize
efficiency of this architecture while promising high security. In addition, the paper will also explore
the potential challenges associated with using blockchain technology in payment gateways.
213. A REVIEW ON SECURE TECHNIQUES FOR KEYWORD-BASED SEARCH

AND DATA SHARING
Dr. Kavitha Esther Rajakumari,Vignesh A,Arjith KS
Hindustan Institute of Technology and Science, Chennai
Server dependability and client security doubt expect data to be revealed prior to sending it to
Aiming, and we want to distinguish watchwords in the data that are stowed away from interlopers
and malignant cloud suppliers that we consider to be a method for impeding the organization. SSE
distributed storage is utilized to look for symmetric encryption schemes (SES) that can't be viewed
as genuine mists. Furthermore, cloud-based servers are intended to be secure by presenting a
cloudbased information stockpiling framework, and will perceive a cloud-based framework that
gives all around conveyed data and essential pursuit administrations. The Internet characterizes a
framework that utilizes information intrude on innovation. furthermore, fundamental pursuit
administrations. TKSE perceives the server close to the action word to forestall the making of a
reasonable server from individuals with terrible data in the information stockpiling region. What's
more, blockchain innovation and hash capacities are utilized to pay web indexes without informing
any other individual as to whether the client or the cloud is malevolent. Security investigation and
execution assessments show that TKSE is protected, effective, and appropriate for distributed
computing.
214. DETECTING MALIGN REQUESTS USING HYBRID DEEP LEARNING MODEL

WITH SEQUENCE-BASED LSTM
Dr S. Saranya,Department of Information Technology, Hindustan Institute of Technology and Science,
Chennai, India,
Wesley Livingston CA, Department of Information Technology, Hindustan Institute of Technology and
Science, Chennai, India
Nagappan CT, Department of Information Technology, Hindustan Institute of Technology and Science,
Chennai, India,
Haja Wasim Ahamed N,Department of Information Technology, Hindustan Institute of Technology and

May 21st – 22nd, 2022
As of today, a huge amount of sensitive data is being stored at various connected services and
platforms. so, the need for high-end user security mechanisms that maintains the safety and integrity
of these sensitive user's data. The Long Short-Term Memory neural network and Word Embedding
can be combined and used to identify the malicious user request. The proposed combination of Long
Short-Term Memory, Word Embeddings, Tokenization with the help of Python’s Keras and
TensorFlow achieved satisfactory results about the malign percentile of the request.
215. LOW WEIGHTED HUMAN ACTIVITY RECOGNITION USING DEEP

LEARNING
Dr. Kavitha Esther Rajakumari, Department of Information technology, Hindustan Institute of Technology and
B Monica, Department of Information Technology, Hindustan Institute of Technology and Science, Chennai,
India
Bellamkonda Vamshi, Department of Information Technology, Hindustan Institute of Technology and
R Jawahar Ganesh, Department of Information Technology, Hindustan Institute of Technology and Science,
Chennai, India
Perceiving human exercises from video groupings or still pictures might be a troublesome
assignment in view of issues, similar to foundation mess. Human action acknowledgment, or HAR,
might be a troublesome measurement grouping task. It includes anticipating the development of a
singular utilizing sensor information. Human activity Recognition assumes a major part in human-
to-human and human-PC connection. Subsequently, this strategy won't exclusively be esteem
powerful anyway also as a utility- based framework which will be consolidated in an entirely sizable
measure of utilizations that might save time and help in various exercises that need acknowledgment
technique, and save heaps of your experience with reasonable precision.
216. CARDIAC DISEASE ANALYSIS USING MACHINE LEARNING

Dr. R. Mohandas*1
Associate Professor, Department of Computer Science and Engineering, Kalasalingam Academy of
Research And Education, Anand Nagar, Krishnankoil, India.
Maniteja Akinapalli*2, Deepak Chiluveru*3, Sainath Madadi*4
Students, Department of Computer Science and Engineering, Kalasalingam Academy of Research
and Education, Anand Nagar, Krishnankoil, India.
Our project main aim is to analyse the accuracies of all the algorithm and declare which algorithm is
suitable for predicting the heart disease in human body. we took a consideration amount of data and
trained it using python for better performance and identification of the effected part of the heart for
every algorithm. We are using python because it is very good and supportive for the visualization
and understanding the complex patterns of the data. We have another alternative of using r language
but many don’t suggest to implement the project in it and suggest python to do it. Recently, there are
some of the diseases which cannot be predicted by the existing system so due to which many
medical universities are conducting challenges to solve them and predict such type of diseases with
machine learning, For example aorta separation and atrial tubes overflow are some of the challenges
May 21st – 22nd, 2022
held. Prediction of cardiovascular disease results is not accurate in each and every case because the
patterns to be followed by algorithm are limited and if a new problem is encountered by the
algorithm, then the expected output may not recurred. ML process starts from a pre-processing data
phase followed by feature selection based on data cleaning, classification of modelling performance
evaluation. XGBoost is used to improve the accuracy of the result.
217. REAL TIME CROWD ANALYSIS

Abhinaya Jaikumar
BE Information Technology
Navi Mumbai, India
Dhruv Manoj
Navi Mumbai, India
Ms. Samundiswary Srinivasan
Assistant Professor
Navi Mumbai, India
Prathamesh Salaskar
Navi Mumbai, India
In the context of computer vision , a crowd can be expressed in terms of a large number of people
demonstrating some sort of coherent movement. Crowd analysis remains an active area of research
in the field of computer vision. Crowd analysis can be used to monitor and control the safety and
security of crowds. The applications of crowd analysis, however, are not restricted to the detection
of overcrowded spaces. Crowd Count and Crowd Motion Detection are considered to be important
attributes of Crowd Analysis. Convolutional Neural Networks are widely used in deep learning
architectures for object detection. The proposed approach utilizes two CNN-based techniques called
CSRNet and YOLOv4 to determine the number of people in a given input video. CSRNet analyses
periodic frames of an input video to provide the crowd count. YOLOv4 performs the task of object
detection along with determining the crowd count by evaluating each frame of the input video. The
detections obtained from the execution of the YOLOv4 model are transformed to provide a Bird’s
Eye View or a Top-Down view. The Bird’s Eye View of the detections produced by YOLOv4 is
obtained by implementing Perspective Transform. With the use of background subtraction, a Motion
Heatmap can also be generated for the corresponding input video.
218. SMART BLIND STICK WITH IMAGE TO SPEECH CONVERSION

Sripathi Lalitha priya, Vardhaman College of Engineering, Hyderabad, India
Ganesh Reddy Gummi, Vardhaman College of Engineering, Hyderabad, India

May 21st – 22nd, 2022
Mr. T. Ramakrishnaiah, Vardhaman College of Engineering, Hyderabad, India
Blind people need supervision in all daily life activities.They need to be able to navigate in their
surroundings avoid any obstacles ahead of them. “Smart Blind stick” can be very useful to them as
they can be aware of their surroundings. Our model consists of camera connected to Raspberry pi
3B+. The live video captured by the USB Camera is processed using Tensor flow software, which is
used for object detection. These detected objects are recognized and the information is relayed in the
form of speech through earphones. GPS and GSM module are also designed is this model so that If
the person is injured, then the current location of the person is sent to his/her family and close
friends about his emergency. Blind stick also has soil sensor to alert us about the presence of any
puddles and to take any further precautions.
219. IMPLICATIONS OF THE EXTERNAL FLOW AND ACOUSTICS ON LATERAL

UPWARD SMOLDERING
Ankit Rout1, Anandita Tomar2, Kancharla Pragathi3, and Vinayak Malhotra4
1,2,3,4 Department of Aerospace Engineering,
Kattankulathur, Chennai, India, 603203.
Combustion is a scientific term for burning. Combustion is a process of fuel reacts with the air to
produce heat energy. This heat energy can produce light in the form of a flame. This is the visible
part of the reaction. Smoldering is considered a type of low-intensity fire, that meaning fire slowly
spreads that can be expected to last for several days, releasing small amounts of energy the most
persistent type of combustion. Smoldering is a low temperature, slow, flameless form of
combustion, with heat is involved with oxygen. Combustion is of two types: Flaming combustion
and smoldering combustion. This type of smoldering combustion can cause number of accidents. A
simple experimental setup was upraised to investigate the effect of external energy source(s) viz., air
flow and acoustics on upward spread smoldering. The external energy source effect was articulated
in form of different flow velocities and acoustics. The experimental predictions were systematically
validated with the conventional theoretical and numerical result. A parametric investigation was
carried out by linearly changing surface orientation, flow velocity, location of external source and
the frequency. The results obtained so far state that the external energy source(s) have significant
effect on the upward smoldering regression rate. The knowledge from the present work can be
effectively applied to various functional, practical, engineering systems ranging from aircrafts,
spacecrafts, buildings fires, wildfires and help us in better understanding of the phenomenon.
220. DISEASE PREDICTION USING MACHINE LEARNING ALGORITHMS

Ayushman Mookherjee | Abhishek Harikrishnan | Guide: Prof Sanjay Kumar Singh
Vellore Institute Of Technology
The healthcare sector is one of the most important research issues in the modern era, because of
rapid advances in technology and data. It's difficult to keep track of such a large amount of patient
data. The use of Big Data Analytics makes it easier to manage this data. Various ailments can be

May 21st – 22nd, 2022
treated in a variety of ways all throughout the world. Machine Learning is a novel way for predicting
and diagnosing diseases.
This study demonstrates how symptoms can be used to predict disease using machine learning.
Machine learning approaches such as Gaussian Naive Bayes, Decision Tree, Random Forest, and
KNN are used to predict sickness on the supplied dataset. The programming language python is used
to implement it. The research reveals which algorithm is the most accurate based on its accuracy.
The accuracy of an algorithm is determined by how well it performs on a particular dataset. We'll
then create a user interface that allows the user to submit disease symptoms and have the system
predict the ailment using various machine learning methods.
221. STUDENTS LIVE BEHAVIOUR MONITORING IN ONLINE CLASSES USING

ARTIFICIAL INTELLIGENCE
DR. A ANTONIDOSS, DENNY AGUSSY GADI , ANJANEYULU CHEREDDY, SUNEEL
KONAKANCHI
Dept of Computer Science and Engineering
Hindustan Institute of Technology and Science
Dept of Computer Science and Engineering
Hindustan Institution of Technology and Science
Many colleges have selected for remote education is part of the health crises that compelled them to
quit using their facilities as a means of teaching. Changing student' knowledge acquisition, which
has predisposed many of them to get familiar with this contemporary learning method, resulting in
the increased use of message boards. Discord, Google Meet, Microsoft Team, Skype, and Zoom are
among the digital platforms used by many educational institutions.The purpose of the study is to
document the influence of conferencing tools on learning outcomes. Parents and learners were
polled, and 66% said their educational growth was unaffected. The majority of them are now
acquainted with the networks; however, less than half of them did Although 24% of instructors
believe their scientific performance is good as a result of this new instructional approach, some
professors are still experiencing psychological challenges as a result of it. Finally, both teachers and
students believe that even these capabilities are quite beneficial in online courses. The main goal of
this project is to develop a self-contained agent that can provide knowledge to both faculty and
students. Established intellectual outcomes such as critical thinking and the grades students receive
in a topic are closely tied to the extent of active learning.
222. DETECTION OF POTHOLES USING IOT

Kalva Suneel,Department of Computer Science and engineering, Kalasalingam Academy of Research &
Education, Krishnankovil, TamilNadu.
Dr.R.Mohandas,Associate professor,Department of Computer Science and engineering,Kalasalingam

Academy of Research & Education,Krishnankovil, TamilNadu
Most of the Road accidents are caused due to potholes and bumps on the road. The driver will drive
the vehicle carefully by seeing the road. Sometimes when the driver is driving fast he can't find the
potholes or bumps due to this, accidents may occur. At night time, the driver isn't able to see the
May 21st – 22nd, 2022
potholes or bumps clearly due to low light at that time major road accidents are occurring. To avoid
this accidents we came up with an idea that can detect the potholes or bumps and alerts the driver
about that. This can reduce road accidents and save lives. And the pothole coordinates are sent to the
cloud database. The program is divided into three sections: underlying systems. For the first, the
underground system hears the holes encountered, which have no prior knowledge about it. Then
there is the communication subsystem that transfers information between a Wi-Fi hotspot and a
mobile node. The point distributes data about holes in its location. Access Point updates its website
with new holes installation. And finally a sub-local system that reads data given access points once
warns the driver about the occurrence of a hole And it also sends the location of the potholes to the
road authority so that they can modify the roads.
223. STEGANALYSIS TECHNIQUES: A SYSTEMATIC REVIEW

Victoria Priscilla C1 and HemaMalini V2
1 Associate Professor & Head, PG Department of Computer Science, SDNB Vaishnav College for Women,
University of Madras, Chennai, India
2 Research Scholar, Research Department of Computer Science, SDNB Vaishnav College for Women,
University of Madras, Chennai, India
World at present is full of data. Data plays a major role in all aspects in development in any field,
future predictions, decision making etc. Securing data is very important not only to avoid breaches
but also to shield the confidential messages. There comes the concept of Steganography.
Steganography is the art of masking data under an ordinary text or image or video. When
steganography is misused by hiding malware in safe files, there comes the art of detecting
steganography called Steganalysis. Stego-only attack, Known cover attack, Known message attack,
Chosen stego attack and Chosen message attack are some known methods in Steganalysis. This
work presents a survey on recent trends in steganography and use of machine learning algorithms in
steganalysis.
224. ONLINE SECURE VOTING SYSTEM USING SMART CONTRACTS

Dr. Shaik javed parvez , Department of Information Technology, Hindustan Institute of Technology and
Shaik Izaaz, Department of Information Technology, Hindustan Institute of Technology and Science, Chennai,
India
Divyanshi Govil A, Department of Information Technology, Hindustan Institute of Technology and Science,
Chennai, India
Derrick, Department of Information Technology, Hindustan Institute of Technology and Science, Chennai,
India
An online voting system using blockchain and ML is to develop technical solutions in the country to
spread awareness among the citizens. In this project we are using two algorithms Haar Cascade
algorithm for face detection. It will take 128 embedding features from the picture that are captured
and stored in text data. Those text dta will be trained by the SVM algorithm. MongoDB is used to
store the data in the backend. Blockchain will provide the security for the backend data. Our main
motto is to decrease the fake voting system and ensure real votes in the country.
May 21st – 22nd, 2022
225. SUICIDAL IDEATION DETECTION USING MACHINE LEARNING

TECHNIQUES
Dr B.V.Baiju1, Department of Information technology, Hindustan Institute of technology and Science,
Chennai,India
Dr S Saranya2, Department of Information technology, Hindustan Institute of technology and Science,

Chennai,India
Abishek Beno Vetrivel P3,Department of Information technology, Hindustan Institute of technology and
science, Chennai, India
Ganesh Prabhu A4, Department of Information technology, Hindustan Institute of technology and science,
Chennai, India
Rakuleeshwar SR5, Department of Information technology, Hindustan Institute of technology and science,
Chennai, India
XGBOOST aids in dealing with incomplete, uncertain, and imprecise data by providing multi-
valued subranges for many types of data. As a popular and generally available source of
information, social media data is critical for the development and growth of human knowledge.
While people were still talking about how to use and abuse social media, researchers noticed that
there was a lot of interest in social media use. This study arose from my interest in developing,
applying, and analysing more sophisticated methods for sentiment analysis.India is evolving as a
result of the adoption of digital India. Demonetization and online shopping are also critical
milestones on the path to paperless transactions, so why not look to social media sites like Facebook
and Twitter for comparable examples? It provided the researcher with the opportunity to become
acquainted with R, a robust data analysis platform. By implementing many XGBOOST and machine
learning- based principles, we were able to get fresh insights using the conceive and implement
technique. Due to their complexity, various types of organised, semi-structured, and unstructured
data entice academics to handle their difficulties with the same aim.
226. AN ADAPTIVE SEMANTIC APPROACH FOR SOCIAL RECOMMENDATION

USING CONTENT BASED ALGORITHM
NOOR ALLEMA Department of Information Technology, SRM Institute of Science and Technology,
KEERTHIVASAN B Department of Information Technology SRM Institute of Science and Technology
SHANE AS, Department of Information Technology, SRM Institute of Science and Technology,
SIVANANDHAM.M, Department of Information Technology, SRM Institute of Science and Technology
Based on the idea that the recommendation system is fundamentally used to create the user and the
system interaction. Recommendation system helps the user to suggest according to their history here
the methods which includes a social recommending for multiple users with logistic regression and
clustering algorithm as their primary method and utility matrix supporting databases for the genuine
accuracy. Additionally the system provides efficiently improve time efficiency involves the
computation time and communication time. With the help of dataset the data are extracted and
suggested to the user. here we develop the intent of the users with providing significant security

May 21st – 22nd, 2022
needs. the further enhancement is set to keep the bar high for the standard of the users to work in
content based by huge margin using by data set and support vector classifier.
227. AUTONOMOUS FIRE DETECTING AND EXTINGUISHING ROBOT

Kajol Mohanty [1]Student, Final year, B.E.,Department of Electronics and Communication, Sri sairam
engineering college, Chennai, Tamil Nadu, India
Kirthikka.S [2]Student, Final year, B.E.,Department of Electronics and Communication, Sri sairam
engineering college, Chennai,Tamil Nadu, India.
Subiksha.S[3]Student, Final year, B.E.Department of Electronics and Communication, Sri sairam engineering
college Chennai, Tamil Nadu, India
Dr.S.Sumathi Professor,Department of Electronics and Communication, Sri sairam engineering

college,Chennai, Tamil Nadu, India
Fires can start from the smallest spark. If left unchecked, wildfires can go on to devastate people and
property. An effective and reliable means of detecting and extinguishing fire is necessary. The
firefighting robots can save a lot of lives someday, the lives of those affected by the fire disaster as
well as lives of those people working as the firefighters, It can be useful in certain types of incidents
where the environment will be very dangerous for the humans such as the hazardous materials, the
radioactivity or the propane tank which can explode. In Our Proposed System we are using Arduino
with an automatic system to continuously monitor the fire in a predefined path. Proposed system
consists of Gas sensor, fire sensor, Motor driver, DC motor, water pump motor, Servo motor and
Arduino board. The Servo Motor system SG90 is used for sprinkling the water. Different specialized
sensors will be connected to the Arduino board. As for programming, Arduino IDE coding will be
implemented to give sensor inputs so that we can govern the movement of the automated system.
GSM Modules alongside GPS will be connected to the Arduino board so that it can immediately
notify the fire-station which is within the reach.
228. AN IMPROVED E-COURT SYSTEM WITH SUSPECT PREDICTION USING

HAAR CASCADE ALGORITHM
U. Nithya Dept of Computer Science and Engineering Hindustan Institute of Technology and Science
Chennai, Tamil Nadu
K. Naveen Kumar Dept of Computer Science and Engineering Hindustan Institute of Technology and Science
Chennai, Tamil Nadu
J. Dheeraj Reddy Dept of Computer Science and Engineering Hindustan Institute of Technology and Science
Chennai, Tamil Nadu
Renjith P N, Associate Professor, Dept. of Computer Science and Engineering, Hindustan Institute of
Technology and Science, Chennai
Corruption and injustice are the most serious issues in the courts. Every state's high court has a chief
justice member. The chief justice of the court is the one who decides the bench (as in the judge). As
we can see, there is no transparency here, and the controversy is that the chief justice is allotting
cases to judges in their favour. To avoid this type of corruption and injustice, we will be creating an
online portal with a system where we will create a system to randomly select judges with full
transparency and no involvement of persons. To make this system more transparent, we are adding
features such as providing the history of all previous cases and the judgments rendered by the judges
May 21st – 22nd, 2022
and lawyers, which will be accessible to the supreme court; we will also provide suspect data so that
when we input the suspect photo in the current case, it will display whether he or she has any
criminal history in the past; and the supreme court collegium can assess individual judges and
lawyers by viewing their profiles.
229. MODIFIED PARABOLIC SOLAR COLLECTOR HOT-AIR TRAY DRYER

Rohit khedkar1, Krushang Goti1, Brindesh Jodhani1, Jenish Hirpara1, Prashant Giri2.
SOE P P SAVNI UNIVERSITY SURAT GUJARAT
The present work deals with the performance evaluation of an active indirect solar dryer integrated
with a parabolic dish collector to dry Indian high quality onion. The system mainly consists of a
parabolic collector which is provided with is provided with an insulator made of plywood with air
gaps and aluminum foil wrapping, and a drier consisting of five trays where the onions are dried.
The flow of hot air through the tray is maintained using an exhaust fan positioned at the bottom of
drier. The fan exhausts the moisture extracted from the product to make the flow of air without any
localization. From the analysis it is found that the parabolic dish collector generates a maximum
temperature of 40’C and dries 3 kilograms of onions 4 to 6 hours of a day. Thus, hot air from the
parabolic dish concentrator reduces moisture content of onions.
230. CONTINUOUS SIGN LANGUAGE ESTIMATION USING LSTM

AKSHIT CHAUHAN Department Of Computer Science and Engineering, SRMISTKattankulathur, India.
RAHUL KUMAR.,Department Of Computer, Science and Engineering, SRMIST, Kattankulathur, India
C Santhanakrishnan,Department of Computer science and Engineering ,SRM Institute of Science and
Technology , Kattankulathur, India
Estimating sign language can help hearing challenged people talk with the rest of society.
Experimenters have made a number of Sign Language Estimation project, they only work for
insulated sign moments. We present a LSTM model for continuous SLR that accepts a succession
of gestures in this study. The main goal of this paper is to show how to use mediapipe opensource
model and lstm algorithm to simplify Sign Language Recognition The proposed model is efficient,
precise, and robust, with an average accuracy of 99 percent. The usage of this technology is made
more comfortable and convenient by real-time precise detection utilising the lstm algorithm without
the use of any wearable sensors.
231. LARGE SCALE PROCESSING OF SIGNALS IN NETWORK ENVIRONMENT

Asmitha Shree R,Dept of computer science and Engineering, Sri Krishna college of technology Coimbatore,
India
Asif B, Dept of computer science and Engineering, Sri Krishna college of technology Coimbatore, India
Balaji J A, Dept of computer science and Engineering,Sri Krishna college of technology Coimbatore, India
Bharath Vaj S, Dept of computer science and Engineering, Sri Krishna college of technology Coimbatore,
India
The overview of 5th generation and beyond technologies is discussed nowadays. The 5th generation
and beyond technologies is known for its speed of data transmission. At the same time, consumption
of data usage getting higher amd higher. So the implementation of 5G and beyond technologies
May 21st – 22nd, 2022
become critical. To overcome the consumption of data usage, a methodology of Cognitive Radio is
used with its Spectrum Sensing Algorithm. The classical graph embedding algorithms on Cognitive
Radio environment and their relationship with network embedding is discussed. Afterwards and
primarily, a comprehensive overview of a large number of network embedding methods in a
systematic manner, covering the structure- and property-preserving network embedding methods,
the network embedding methods with side information and the advanced information preserving
network embedding methods is provided. As a result, a simulation prototype is done showing the
data usage is consumed efficiently on the transmission of data.
232. SECURED BLOCKCHAIN TECHNOLOGY FOR AGRICULTURE FOOD

SUPPLY CHAIN
Dr.R.Parvathi, Associate Professor, Department of Information technology Hindustan Institute of technology
and science, Chennai, India.
Abhiram K, Department of Information technology Hindustan Institute of technology and science, Chennai,
India.
Girish M, Department of Information technology Hindustan Institute of technology and science, Chennai,
India.
Gnana Sandeep M, Department of Information technology Hindustan Institute of technology and science,
Chennai, India.
As we all know when we talk about marketing the thing is it needs a plan of action that relates peer
to peer relationships of farmers as well as consumers, whereas this process was followed and there
will be no loss for consumers or dealers, the problem comes to only the farmer in ways of both
financially and work issues during selling. so this paper proposes the work of getting more gain for
cultivators by including the blockchain technology in the food chain supply by using FERNET (AES
encryption algorithm) whereas this one doesn’t have any vulnerabilities we know, that did it not
secure and this has not been broken when compared to another popular hashing algorithms.
thereafter by using a software infrastructure that includes in pre-processing of data, making
transactions secure between cultivators as well as consumers/buyers, whereas blockchain is less time
taken as well as a way of security for all kinds of transactions by using smart contracts for
transactions for valuation of goods by using this algorithm we can securely protect digital signatures.
using this model or the application we develop, all farmers are good and it is designed in such a way
that every farmer can easily log in and check the details of his product and sell their crop and make
secure payments.
233. SMART HEALTH PREDICTION USING MACHINE LEARNING ALGORITHMS

Yerra Anil Kumar , Department of Computer Science and engineering Kalasalingam Academy of Research &
Education Krishnankovil, TamilNadu
Ketham Samara Simha Reddy, Department of Computer Science and Engineering Kalasalingam Academy of
Research & Education Krishanakovil, TamilNadu
Yarrabothula Praveen Reddy, Department of Computer Science and Engineering Kalasalingam Academy of
Research & Education KrishsnanKovil, TamilNadu

May 21st – 22nd, 2022
Dr. Mohad Usama, Associate professor , Department of Computer Science &engineering Kalasalingam
Academy of Research & Education Krishnankovil, TamilNadu
Based on the predicted model, the “Smart Health Prediction Using Machine Learning” system
predicts patient/user disease by depending on the user inputs-provided signals as included in the
database. This application contains two logins: one for the manager and one for the patient. The
patient can log on to the website and post the symptoms, and the patient's disease results can be
obtained using trained machine learning models. Here, the administrator oversees the provision of
information for patients' access and viewing of symptoms and conditions. Naive Bayes, Random
Forest, and XG Boost are available machine learning algorithms that can be used in brain screening
and breast cancer data sets to diagnose a patient and can tell if a patient has malignant or malignant
breast cancer, and whether they have symptoms of a stroke.
234. ANALYZING IMAGE DENOISING USING GENERATIVE ADVERSARIAL

NETWORK
Dr.S.Saranya, Department of Information Technology, Hindustan Institute of Technology and Science,
Chennai, India
Venkateshwar Rao Velichala
Pavan Kumar Vellaturi, Department of Information Technology, Hindustan Institute of Technology and
Science,Chennai, India.
Motion images they have great use in television and film making and virtual reality technology. This
paper gives the solution of reconstruction the dynamic blur images.it converts from blur images to
actual images by the mapping, it can be happened by using an image reconstruction algorithm using
the GAN convolution model. The model runs on the input sets and the output will be an actual
image that came from the blur images. The restoration effect will be good and quality is high in
terms of stability, phase manipulation, conversion efficiency, noise performance.
235. ENGINEERING A NOVEL RECRUITMENT SYSTEM USING GRADIENT

BOOSTED DECISION TREE ALGORITHM AND BUSINESS INTELLIGENCE
PRINCIPLES
Dr. Shaik Javed Parvez,Department of Information Technology, Hindustan Institute of Technology and
Abarna Arulprakasam, Department of Information Technology, Hindustan Institute of Technology and
Thowbik Ahamed,Department of Information Technology, Hindustan Institute of Technology and Science,
Chennai, India
J Jane Carolyn,Department of Information Technology, Hindustan Institute of Technology and Science,
Chennai, India
Using Business Intelligence to strategize the use of technology and reduce the rate of an Indian
economic issue of unemployment by developing a Start-up web application using By coordinating
with the Indian Sports and Cultural Institutions, and establishing agreements with NGOs/Corporates
and other sponsorship providers to raise stipends/employment for underprivileged Indian youth
May 21st – 22nd, 2022
using a novel algorithm with the GBDT algorithm for a suitable and secure recruiter
recommendation system using firebase encryption. Since the Indian youth require a vast stage of
opportunities and growth in variety of employment, the creatives and the sports industry are
primarily focused to display limelight on those nurtured with inborn talents using our recruiter
recommendation system using the 3 BI principles namely relevance, query intelligence and
personalization. This app will seek to eradicate insufficient assessments, location barriers,
inaccessibility of academies due to lack of funds, inadequate cultural categories, loopholes in the
database due to weak security, and manual screening and verification of UUP details, this paper
aims to give an overview and a studied literature review including proposed concepts and describes
the need, significance and goal of developing an algorithm using business intelligence and machine
learning to solve a socio-economic crisis.
236. A NOVEL APPROACH FOR NEXT GENERATION BUSINESS BY USING

PARALLEL AND DISTRIBUTED DEEP LEARNING FRAMEWORK
S.Varalakshmi, Assistant Professor, Department of Computer Science, Bharathidasan Govt.College for
Women, Pondicherry University, Pondicherry
The goal of this paper is to explore a way to improve performance and accuracy in future business
with large scale of datasets by using Parallel and distributed Deep Learning framework. The
processing of huge data and attain a good performance is a cumbersome task. It will take long time
to complete the task in the traditional model. And performance is the combination of speed and
accuracy. These two facts can easily be achieved through parallel and distributed infrastructure in
Deep learning framework. This paper focuses on training the huge voluminous data in parallel and
distributed platforms to attain a “high scalability” which will fuel the next generation business.
237. DECENTRALIZING ELECTRONIC MEDICAL RECORDS ON THE

BLOCKCHAIN USING SMART CONTRACTS
B.V Baiju,Department of Information Technology, Hindustan Institute of Technology and Science, Chennai,
India.
S. Saranya, Department of Information Technology, Hindustan Institute of Technology and Science, Chennai,
India.
Sriram D, Rifath Ahmed M, Adnan Mohammed, Department of Information Technology, Hindustan Institute
of
Technology and Science, Chennai, India
The blockchain technology is considered as the most secure and decentralized way to manage your
payments with cryptocurrencies. Furthermore, with the advent of NFT's which are a way to make
owning digital art a possibility we can see how blockchain has multiple purposes. Blockchain gained
traction because of its attributes such as security, immutability, expandability and decentralism. We
can see blockchain being used more and more for non-financial applications. On the topic of
electronic medical records which need to be secure and be accessed globally, it is still being stored
locally or on the cloud which is confined to an organization or a group of organizations. We can
observe that healthcare is a field that can benefit massively from blockchain technology. It can make
electronic medical records be securely accessed in case of emergencies using smart contracts making

May 21st – 22nd, 2022
the system effectively be used hand in hand with conventional database systems. Smart contracts can
be used to manage the accessibility of electronic medical data by third party organizations globally
without the data being tampered, manipulated or misused. The system can be used with the current
database as it is interoperable. With the successful implementation of this system, it will give access
to patients, health care organizations and other third parties the ability to view and manage the
electronic medical data.
238. PRESERVING DATA PRIVACY IN WIRELESS COMMUNICATION IN IOT

Vennapusa Pavan Kumar Reddy, Department of Computer Science and engineering, Kalasalingam Academy
of Research & Education, Krishnankovil, TamilNadu.
P.Velmurugadass, Assitant professor, Department of Computer Science and engineering, Kalasalingam
Academy of Research & Education, Krishnankovil, TamilNadu.
Sayd Arshad Ali, Department of Computer Science and Engineering, Kalasalingam Academy of Research &
Education, KrishsnanKovil, TamilNadu,
Batta Hemanth, Department of Computer Science and Engineering, Kalasalingam Academy of Research &
Education, Krishanakovil, TamilNadu
The technologies now-a-days are increasing gradually. In our digitalized society, with respect to
technologies, the data is also increasing rapidly. It is important to provide privacy for the sensitive
data. So, in this project, we are providing privacy for this data by using Dynamic Privacy Protection
model rather than using conventional encryption.In a present scenario, with mobile devices and
other Internet connected devices (e.g: sensors) becoming the normal in our digitalized society, the
capacity to ensure the security of data-in-transit and at-rest without involving in unrealistic
performance overheads is crucial. If adequate security and privacy-preserving measures are not in
place, a user’s privacy could be compromised due to data leakage. Here, we are discussing the
Dynamic Privacy Protection model as it is designed for ensuring user’s privacy even in large
volumes of data.
239. MUSIC GENRE CLASSIFICATION

Prithviraj Yadav
Department of Software Engineering,
Delhi Technological University
New Delhi, India
Dr. Shailender Kumar (Professor)
Delhi Technological University
New Delhi, India
The Music Genre classification have many real-world applications. The number of songs being
launched these days is growing each day on internet and cloud platforms such as Spotify, JioSaavn
etc. Having feature of instant classification of songs in playlist or library by genre makes user
accessibility to their songs easy. In this research work, we have compared the performance of
machine learning classifier and deep learning neural networks. The deep learning approach used are
a Feedforward Neural Network (multi-layered perceptron approach) and CNN model. The machine
learning approach used is KNN algorithm which is a supervised machine learning model. To predict
May 21st – 22nd, 2022
the genre label of an audio signal input, we have applied pre-processing techniques to convert the
audios which is in .wav format to feature vectors to make it understandable to machine for reading
audio files. We have trained these neurons and classifiers on data after pre-processing it and
compare their performances. We have transformed and classified these audio files using Mel-
Spectrograms and MFCC techniques. The data on which we’ve worked taken from GTZAN dataset
which contains audio file in .wav format, therefore, there is a need to change into machine
understandable language i.e., vector form. So, we’ve Extracted MFCCs vectors from music dataset
and saves them into a file along with genre labels. In this paper, we have also tried to solved the
problem of overfitting which is often occur in neural network models and we have also put more
emphasis on test data accuracy along with validation accuracy for CNN because it gives accuracy
results on unseen data precisely and CNN gives us the most promising and good results.
240. THE UNUSED POTENTIAL OF PROGRESSIVE WEB APPLICATIONS (PWA)

SUDARSAN S, Dr. P Supraja ,Dr,UG Student
Dept of Networking and communications, SRM Institute of Science and Technology, Kattankulathur
In this modern era, computational thinking, artificial intelligence, and other advanced concepts are
possible due to the evolution of technology. Automation is also a concept that was not possible
before the invention of computers but is currently possible.
Automation is the method by which tasks can be programmed to be performed automatically, with
little to no human intervention. Even though in today’s modern age, we are able to program things to
a great extent by using various applications, the involvement of humans in manual work is still high
due to the lack of required applications to suit the users specific needs. Automation is also a process,
where all companies should heavily invest in to cut costs
This aim of this paper is to survey the capabilities of Progressive web applications.
241. DEDUCTION MANAGEMENT SOFTWARE

Raveena Susan John
Department of Computer Science and Engineering with Specialization in Artificial Intelligence and Machine
Learning,SRM Institute of Science and Technology, Kattankulathur, India
Dr. R Siva, Assistant Professor,Department of Computational Intelligence,School of Computing,SRM
Institute of Science and Technology,Kattankulathur, India
Deduction Management Software works on the account receivables side of the transaction process
between the client and buyer. The traditional approach of collecting receivables from customers was
a cumbersome process for which the analysts required a lot of time and. The issue of dealing with
claims raised by customers is paramount to any company. Deduction Cloud is a system designed to
support companies in active deduction management. Configurations in the Deduction Management
Software result in a streamlined workflow to obtain a quicker, optimized and more efficient
deduction resolution process. As soon as a deduction case is created from a claim in the system, it
will get routed to the worklist of the appropriate owner/processor based on certain case attributes
like customer,company etc. The user will then have to perform role specific tasks and update the
deduction case. In this way the deduction case will ultimately reach an acceptable closure state.

May 21st – 22nd, 2022
Closure can be achieved through denial, which results in repayment and write-off, as well as credit
issuance and offset. The goal behind my project is to utilize the streamlined workflow of Deduction
Management Software to obtain a quicker, optimized and more efficient deduction resolution
process.
242. GROUPING OF CUSTOMERS USING CLASSIFICATION ALGORITHMS

Ravi teja Yamsani1
Student, Department of Computer Science and Engineering, Kalasalingam Academy of Research and
Education, Anand Nagar, Krishnankoil, India
Meeting of many competitors at once entrepreneurs have created a major conflict between
competitors businesses to find new customers and keep old ones. As a result formerly, the need for
unique customer service becomes relevant regardless of the size of the business. In addition, the
ability of any business to understand the needs of each customer will supply a larger customer
support for the provision of targeted customer services and development customized customer
service plans. This understanding possible through systematic customer service. Each part has
customers who share similar market features. Big data ideas and machine learning have encouraged
greater acceptance of automatic customer segregation methods in favor traditional market statistics
that often do not work there the customer base is very large. Customer segregation is important in
customer relationship management software. The most common way to differentiate one customer
from another is to develop a customer group according to their interests and factors like
demographic factors, Psychological factors, Behavioral factors, and geographical factors . In this
work, the company's hand-separated customer data is analyzed. As machine learning methods are
useful for solving problems with data management and data processing, the solution is searched
within machine learning methods. The different classification methods are used to differentiate the
people into different groups and help with suitable solutions. For increasing the efficiency and
precision we used all the classification algorithms as a ensemble algorithm and created a model.
243. WHEEL RIM RADIAL FATIGUE LIFE ANALYSIS CONSIDERING STEERING

GEOMETRY
Sakshi Pawar1, Saloni Thakur2*, Dr. Raju Ladhwe3, Kshitija Jiddewar4, Vinayak Marudwar5
1Department of Mechanical Engineering, College Of Engineering Pune, Pune-445001, India
Day by day, the competition is increasing with innovations and ideas in the automobile sector. With
these innovations, a new path is created in product development. There is enormous scope in
optimizing the overall time required in wheel testing processes. This paper highlights the effect of
steering geometry on fatigue life and stress distribution of the wheel rim of a standard passenger car
under radial load conditions which arises due to road unevenness and off-road field area. The test
conditions are simulated to carry out FEA and analyse stress distribution of the steel wheel rim.
Experimental analysis performed by radial fatigue testing machine for evaluating fatigue life under
May 21st – 22nd, 2022
the influence of altered steering geometry. For radial fatigue testing, SAE J328 standard is used.
This SAE recommended practice provides minimum performance requirements and uniform
procedures for fatigue testing of wheels intended for temporary use and normal highway use on
passenger cars. In particular, experimental fatigue life calculations of a wheel rim are compared to
predictions of finite element analysis.
244. A SURVEY ON KEY TECHNOLOGY AND DECISION-MAKING IN

SELFDRIVING CARS
1Dr.S.Gomathi, Asst.Professor, Department of Computer Science, SDNB Vaishnav College for Women,
Chrompet
2R.Divya Bharathi, 3G.Abinaya, 4K.Keerthana, UG Students, Department of Computer Science, SDNB

Vaishnav College for Women, Chrompet
A self-driving car is a vehicle that does not require human intervention to operate. Instead, it
employs advanced sensing technology such as lidar, sonar, GPS, radar, or odometry and inertial
measurements to more establish and adapt to them, re-establishing a safe speed or distance. In this
survey paper of research on self-driving cars that are capable of sensing their surroundings and
driving safely, the architecture involved in this self-driving car autonomy system is typically
organized into two types of systems: one is the perception system, and the other is the decision-
making system. People are running from the past to the future in search of new inventions.
Similarly, automakers and technology firms are racing toward the future of advanced self-driving
cars. Automakers have begun their efforts to achieve this goal.
245. FEATURE EXTRACTION FOR ENGLISH LANGUAGE USING NLP

1J.Juslin Sega 2Praveen Raja N, 3Srikanth S, 4Yaswanth K
1Assistant professor 2,3,4Students
1,2,3,4Department of Computer Science & Engineering
1,2,3,4SRM Institute of Science and Technology, Ramapuram, Chennai
In this modern world NLP has plays crucial role on text based process. NLP is used for text
preprocessing and from the untrusted data getting more amount of insight of information such as
reviews, feedbacks. In our proposed we are going to implement NLP and Naive Bayes. In Feature
Extraction we extract and produce feature representations that are appropriate for the type of NLP
task you are trying to accomplish and the type of model you are planning to use. After getting the
structure data we divide into two parts like Training and Testing data’s. then we feed the data to
Naive Bayes finally we got more amount either it’s an positive or negative review.
246. STOCK ANALYSIS AND AUTOMATION USING MACHINE LEARNING

Dr. S. Surya
Assistant Professor, Department of Computer Science and Engineering, Kalasalingam Academy of Research
and Education, Anand Nagar, Krishnankoil, India
Saddikuti Jeevan Reddy

May 21st – 22nd, 2022
Education,
Madakala Badrinadh Reddy
Education,
Gattupalli Maheshwar Reddy
Share market is one of the complicated ways to do business. Small ownerships, brokerages,
corporations, all depend on this very body to make revenue. The main theme of the project is to
generate a bot to automatically analyze or automatically settle the things like selling the current
stocks or buying new ones. We are planning to develop it by customizing it for our required
conditions like, at which scale to sell the stock or at what point to buy the stock. In technical analysis
can be formulated as a pattern recognition problem, where the inputs are derived from the history of
prices and the output is an estimate of the price or an estimate of the prices trend. Here, we are going
to use some of the Machine learning algorithms like K-Nearest Neighbour (K-NN), Navies Bayes,
SVM`s, and Google APIs to create a desired bot that can be automated by Python commands. Its
ability to extract options from an outsized set of data while not hoping on previous information of
predictors makes deep learning doubtless enticing for securities market prediction at high
frequencies. Deep learning algorithms vary significantly within the selection of network structure,
activation performance, and different model parameters, and their performance is thought to rely
heavily on the tactic of knowledge illustration. Our study tries to provide a comprehensive and
objective assessment of each of the benefits and downsides of deep learning algorithms for securities
market analysis and prediction. mistreatment high frequency intraday stock returns as input files, we
tend to examine the consequences of 3 unattended feature extraction methods principal part analysis,
autoencoder, and also the restricted physicist machine on the network’s overall ability to predict
future market behavior. Empirical results counsel that deep neural networks will extract further info
from the residuals of the autoregressive model and improve prediction performance; a similar cannot
be the same once the autoregressive model is applied to the residuals of the network. variance
estimation is additionally perceptibly improved once the prognosticative network is applied to
covariance-based market structure analysis. Our study offers sensible insights and doubtless helpful
directions for additional investigation into however deep learning networks will be effectively used
for securities market analysis and prediction.
247. DETECT SIGNATURE BASED MALWARE USING MACHINE LEARNING

G.Jagadeesh
Madurai, Tamilnadu
G.Hari Prasad
Madurai, Tamilnadu
C.Bhargav

May 21st – 22nd, 2022

Madurai, Tamilnadu
Malware has threatened the organizations for along time and still have not made a lot of progress in
detecting the malwareontime.Malwarecaneasilyharmthesystemby executing the unnecessary services
that will put the load on the system and hinder its smooth running. We are using signature based
malware detection technique.The signature of the malware is defined by the task the malware
performs when it gets activated in the machine, for example, running the Operating
Systemservices,downloading the infected files from the internet.The proposed algorithm detects
themalware based on its Signature. In this paper, we used Decision Trees, XGBoost and support
vector Machines for the malware detection.
248. DESIGN AND DEVELOPMENT OF ELECTRONIC VOTING MACHINE

J.Krishna Chaitanya, Professor, Department of Electronics and Communication Engineering Vardhaman
College of Engineering (Autonomous) Shamshabad, Hyderabad - 501218 , Maddela Subha Sri, Nelli Dhruthiee
The primary objective of this research is to establish an electrical mechanical device that will help to
eliminate vote fraud in both manual and computerized voting systems. Electoral fraud was decreased
in India by using Electronic Voting Machines (EVMs) to exchange paper voter support, making
elections more fair, secure, and competitive. Booth capturing, in which party adherents seize a booth
by force and load the box with false ballots, was a big worry with the use of paper ballots in India.
On the polling machine, an alcohol sensor was mounted on top. When a voter attempted to vote
while inebriated, he or she was unable to do so. A program was uploaded in preparation for the use
of EVM. It was simple to prevent those voters from voting by utilizing an alcohol detector. Index
Terms— Electronic Voting Machine (EVM), Control unit (CU), Alcohol Detector.
249. A SURVEY ON LEARNING MODELS OF REINFORCEMENT LEARNING

Subhashini.R, Ravi Ramya Sree, Anitha.S,
SDNB VAISHNAV COLLEGE FOR WOMEN
Q-learning is a frequently used and representative reinforcement learning method, as well as an

offpolicy strategy for the overall performance since its inception many articles, have described
qapplications learnings role in reinforcement learning and artificial intelligence difficulties however
understanding is scarce about how to apply and combine these complicated algorithms in a larger
artificial intelligence framework initial q-learning algorithms were ineffective in a lot of reasons and
could only be used to solve a restricted number of issues it has also been established that this rather
powerful algorithm learns inappropriately and overestimates the action values on occasion resulting
in lower overall performance several q-learning variants such as deep q-learning which incorporates
basic q learning with deep neural networks had also recently been formed as a result of the recent
breakthrough in machine learning. As a result of general improvements in machine learning, other
variants of Q-learning, such as Deep Q-learning, which incorporates basic Q learning with deep
neural networks, have been encountered and applied comprehensively in recent years. Finally, in this
work, we look at the many forms of q-learning algorithms. [16].

May 21st – 22nd, 2022
250. AN ENSEMBLE APPROACH FOR DETECTION AND CLASSIFICATION OF

FAKE NEWS
Mohd. Maaz Khan1 • Aditya Agarwal2
1,2 Student at SRM Institute of Science and Technology
*S Iniyan
Department of Computing Technologies SRM Institute of Science and Technology
Fake news is misleading information which can often damage individual’s reputation and can
change person’s opinion. Recently, the rise of social media platforms gave a boost to circulation of
fake news across the globe. Our paper presents a machine learning model using voting ensemble
with equal and different weights of each base model. We used 4 different classification algorithms
as base learners- 1- Logistic Regression, 2-Multinomial Na- ive Bayes, 3- Passive Aggressive
Classification Algorithm, 4- Random forest algorithm. All algorithms chosen works best with the
long textual in stream of data. We compared accuracy of each base model. After assigning different
weights to each base model, voting ensemble gave an accuracy of 93.5%.
251. ML-BASED OFFENSIVE TWEET ACCURACY DETECTOR ON SOCIAL

MEDIA
Dr.A.Kumaresan
Department of Information Technology and Science
Chennai, India
Thota Sai Sumanth
Chennai, India
Mekala Vasu
Chennai, India
Narayanamurthy Elayaraja
Chennai, India
Global access to the internet has surprisingly changed how we see the arena. SM is one of the
children of the internet, which can be found in many bureaucracies: online gaming platforms, dating
apps, boards, online information services, and social networks with extraordinary functions.
On Twitter and Facebook, you can share opinions make business contacts on LinkedIn, share photos
on Instagram, send videos on YouTube, and court someone on Meetic. However, they all have one
thing in common: they aim to attach people. Social networks have such great potential that by the
year 2021, it is projected that there will be 3.02 billion active social media users globally.

May 21st – 22nd, 2022
The rapid rise of social networks and microblogging sites has caused direct communication between
people of diverse cultures and mentalities, leading to an increasing number of "cyber" conflicts. As a
consequence, hate speech is used more and more, to the point where it has become a serious problem
invading these public spaces. Hate speech is defined as the use of competitive, violent, or abusive
words directed towards a specific group of people who share a common set of assets, whether that
set of assets includes their gender, ethnic group, race, or beliefs and faith. While most internet social
networks and microblogging services prohibit hate speech, the sheer size of these networks and
websites makes controlling all of their content extremely difficult. As a result, there is a need to
detect such speech automatically and remove any content that contains hateful or inciting language.
252. COMPARATIVE ANALYSIS OF DIFFERENT MACHINE LEARNING

TECHNIQUES FOR PREDICTING DIABETICS DISEASE
Venkatesh C#1, Baddili Tejesh#2, T Obulesu#3, Dr.T. Dhilipan Rajkumar*4
#Student, Department of Computer Science and Engineering, Kalasalingam Academy of Research and
*Assistant Professor, Department of Computer Science and Engineering, Kalasalingam Academy of Research
And Education, Anand Nagar, Krishnankoil, India
A major incurable disease known as diabetes mellitus. It is a global epidemic and public health
problem. Hypertension, or high blood pressure, is one of the many life frightening consequences that
occurs in diabetes that goes misdiagnosed, underestimated when the diabetes symptom changes
high. Early diagnosis of a persons at risk for diabetes complications can decrease or prevent diabetic
complications. Various ml algorithms has used to diagnose diabetes in recent years, but few
algorithm studies that has undertaken to detect the diabetic patients.
As a result, we compared various algorithms such as Decision Trees, Random Trees, and Vector
Support Machines (SVM) in our study. In previous experiment models particularly failed to provide
intelligible model sets rule. This work aims to improve the strategy of hybrid to select for rules from
SVM in order to overcome this constraint. And picking strongly the associated features from the
dataset, the feature selection process is introduced. Support vector machine model of black box was
converted into understandable XG Boost decision making tool. There were 300 diabetics in the
study, of 108 hypertensive, five general diabetes that is related to dataset which is very helping.
Experiments show that the advanced model outcome performs 10 of benchmark classification
models.
253. AN INTELLIGENT TRAFFIC LIGHT CONTROL SYSTEM USING

Swathi R
Assistant Professor
Department of Computer Science Engineering
Chirag Solanki
SRM Institute of Science and Technology Chennai, Tamil Nadu, India
Deep Vaya
May 21st – 22nd, 2022

Pulivarthi Naveen
SRM Institute of Science and Technology Chennai, Tamil Nadu, India
To ensure safe Driving at road intersections where traffic signals are required. However, because of
the queue delays at each traffic flow, they disrupt and reduce traffic flow. We present an Intelligence
Traffic Light Control System (ITLC) algorithm in this paper. While scheduling the time phases of
each traffic signal. The real-time traffic element of every flow that aims to have on that four-lane
cross junction with the use of the you only look once (YOLO) algorithm and ITLC algorithm. The
proposed algorithm intends to improve traffic flow by reducing the time moving cars spend waiting
at signalized road junctions. It also aims to increase the number of cars passing through the four-lane
cross junctions per second.
254. ONLINE STUDENT CAREER GUIDE PORTAL - A WEBSITE FOR STUDENTS

SEEKING INFO REGARDING CAREER GUIDANCE
Dr. Vigilson Prem M , Professsor, Department of CSE, RMK College of Engineering and Technology,
Bhargavi D, CSE,
RMK College of Engineering and Technology, Thiruvallur, Tamilnadu
Charitha R, CSE,
RMK College of
Engineering and Technology, Thiruvallur, Tamilnadu
Passing out of school is a big time for most students because for most it is the first time they will
have to make a life-changing choice, their undergraduate degree. Every student in our country
struggles with the questions "Which stream to choose? Which stream suits me? What the stream
actually contains? Where can I get the best of the stream I choose?” This system aims to be a helpful
and useful starting point for this confusing phase. There is a wealth of tools and data available for
students who are interested in any field but most students do not have the time to research as well as
required. This system will bring all these tools and data in one place for students.
255. AUTOMATION ON MAINTAINING OF SOIL MOISTURE LEVEL IN

EARTHING PIT
Thanikachalam J1, and Siddharth R2
1Department of Mechanical Engineering, Mepco Schlenk Engineering College, Sivakasi-626005, Tamil Nadu,
India.
2Department of Mechanical Engineering, Mepco Schlenk Engineering College, Sivakasi-626005, Tamil Nadu,
India.
In this work the main idea is to monitor and increase the soil moisture level in earth pit where it has
been placed. To monitor on daily basis this work has been made to maintain water level that is
moisture content. In this work 4 samples have been collected in which 3 samples from dry soil
region and 1 sample from wet soil region. Soil resistivity has been calculated by Wenner three point

May 21st – 22nd, 2022
method where the values of resistance and soil resistivity for dry soil is about 9 Ω and 412.64 Ωm
and for wet soil condition it is about 4 Ω and 206.32 Ωm, where it has been decreased around 50%
of the resistance value in earth pit which helps to improve good earthing and conductivity. A
prototype has been made to detect the moisture level in earth pit and GSM kit has been attached with
them for indicating the moisture level which helps to turn on and to turn off the water pump
automatically depends upon the resistance value.
256. MACHINE LEARNING AND STOCK MARKET ANALYSIS

Rishabh Dhingra, Dr.Sandeep Vanjale and Dr.Mousami Vanjale
1Dept. of Computer Engineering
Bharati Vidyapeeth (Deemed To Be University) College Of Engineering Pune, India
2Professor, Head of Dept. of Computer Engineering, Bharati Vidyapeeth (Deemed To Be University) College
of Engineering Pune, India
3Assistant Professor, Dept. of E&TC Engineering, AISSMS's Institute of Information Technology, Pune, India
This research work deals with identifying that whether fundamental or technical analysis is more
accurate for stock market price prediction by applying different Machine Learning Models like
Generalized Linear Model, Decision Tree and Support Vector Machine to calculate the feature
importance of various fundamental and technical factors like Day’s Average Price, Total traded
Quantity, Deliverable Quantity, Number of Trades, Turnover of Company and Percentage Delivery
Quantity to Traded Quantity affecting the market price of Indian Oil Corporation Limited on the
National Stock Exchange, India. Apart from the weight (Importance) of each factor, it also helps to
find out which factor favors the market price and which contradicts it. Thus, this comparative study
of different machine learning models allows us to conclude which model is more accurate based on
statistical measures like relative error, squared error, correlation, runtime comparison, standard
deviation, root mean squared error of prediction charts formed by each model.
257. PREDICTION AND DETECTION OF FOREST FIRES BASED ON DEEP

LEARNING APPROACH
Gayathri S
Department of Information Technology, Hindustan Institute of Technology and Science, Chennai, India
P V Ajay Karthi
Sourav Sunil
Forest fires are one of the crucial disrupting impact parts inside the overall forest climate, and it
causes various levels of adverse consequences on the natural climate, resources, human prosperity,
economy, etc.. Climatic changes impact are a few of the results of such pulverization. Generally,
forest fires happen due to human exercises. Inorder to control the annihilation caused by forest fires,
we try to identify forest fires at their beginning so that it does not spread.. In this paper we have
proposed a method, using image processing module and grey scaling for detecting of forest fires
using deep learning based algorithms, CNN. After the fire is being detected, an alert is sent to
control team of the forest along with location. .We have also integrated Google’s Firebase for
sending alerts through notifications of mobile or iot devices. This paper principally engaged
concerning brief prologue to the backwoods fire, related work about different strategies and
May 21st – 22nd, 2022
frameworks in timberland fires, conversation on computerized reasoning and AI calculations and

followed by forecast and location frameworks are audited.
258. IMPROVISING AIR QUALITY FORECASTING USING MACHINE

LEARNING
ANKUR RANA
B.Tech (CSE)
Chennai, India
PALAK MITTAL
B.Tech (CSE)
Chennai, India
Dr.T SENTHIL KUMAR
Associate Professor
School of Computing
Chennai, India
This paper presents a brief overview of trends in numerical weather prediction, difficulties, and the
nature of their occurrence, the existing and promising ways to overcome them. The neural network
architecture is proposed as a promising approach to increase the accuracy of the 2m temperature
forecast given by the COSMO regional model. This architecture allows predicting errors of the
atmospheric model forecasts with their further corrections. Experiments are conducted with different
histories of regional model errors. The number of epochs after which network overfitting happens is
determined. It is shown that the proposed architecture makes it possible to achieve an improvement
of a 2m temperature forecast in approximately 50% of cases.
259. INHIBITORY POTENTIAL OF NATURAL PLANT PRODUCTS AGAINST

INFLUENZA VIRUS
Abhay Kumar Jha, Monika, Smita Rastogi Verma*
Department of Biotechnology, Delhi Technological University, Delhi, India
* Correspondence
Dr. Smita Rastogi Verma
Department of Biotechnology
Delhi Technological University, Delhi-110042, India
H5N1 being highly pathogenic avian influenza virus that causes severe respiratory infection causing
avian flu spreads rapidly among the birds' population. The virus is epizootic in nature but in the last
decade several cases have been reported in humans as well. In this review, we have used five
different plant products to find their inhibitory potential against 3s12 viral influenza protein. We
have used Autodock software for performing molecular docking on 3s12 protein present in H5N1
influenza virus. The docking results were analysed and visualized using PyMol visualization tool
and then binding conformation of the protein -ligand complex is studied. We used ADME analysis
May 21st – 22nd, 2022
to find the Lycorine which is a toxic alkaloid extracted from Lycoris radiata plant has the highest
binding energy –7.72 Kcal/mol with 3s12 haemagglutinin protein making it a better inhibitor of the
protein as seen in molecular docking performed. The binding of lycorine at the active site of 3s12
gives the best drug value.
260. BIOMETRIC AUTHENTICATION ON FINGER VEIN RECOGNITION

SUBHA P, MAHALAKSHMI S, MAANSI S, DHARSHINI S, PRABAVATHI R
SRI SAIRAM INSTITUTE OF TECHNOLOGY

A finger vein authentication system is a more secure form of Authentication and Biometric
technology is playing an increasingly significant role in modern society and is a promising biometric
pattern for personal identification in terms of its security and convenience. The particular reason for
this authentication is that veins are present beneath the human skin, which makes them impossible to
replicate. Furthermore, the vein pattern remains unchanged during the span of a person's life.
Whereas fingerprints, it can be easily duplicated and it is faded due to aging. Iris authentication, the
pattern of Iris can be altered when the person undergoes cataract surgery. Here the pattern of veins is
captured under Infrared light transmission and pre-processing and extraction of vein patterns are
done using Deep Learning and CNN and the patterns are trained and verified for authentication. This
technology is currently in use or development for a wide variety of applications, including credit
card authentication, automobile security, employee time and attendance tracking, computer and
network authentication, endpoint security, and automated teller machines.
261. THERMAL WRAPPER USING T-CALL ESP32 SIM800L

Dr. K.S. Dhanalakshmi
S. Manoj Kanna
S. Sailanjali Ajitha
S. Deva Kishore Reddy
Soldiers are very important to protect our nation from other countries who are trying to occupy our
nation and to protect us from Terrorist attacks. Soldiers facing many problems who are protecting
our nation by like experiencing heavy cold conditions like soldiers who stays in Himalayas, some
soldiers got missing in some terrorist attacks and some soldiers pulse rate goes down due to loud
sounds that come from heavy bomb explosions. IoT is the field of ongoing research that measures
May 21st – 22nd, 2022
the soldier’s safety. This paper focuses on the development of an IoT device which will be inserted
in a soldier's jacket that helps to measure the pulse rate of a soldier, produce the heat all over the
jacket by sensing the temperature whenever the soldier requires and also be able to track the soldier
using GPS Sensor. Soldiers who guard in heavy cold regions like Himalayas may face excessive
cool climates due to this they can be affected by Hypothermia. This is occurred when the body
temperature falls under 35°C and may lead to death. To avoid this, we introduced a heat generation
system that supplies heat all over the jacket that can be either controlled manually or automatically.
This device measures the pulse rate and uploads the data into the website so that they are able to
check the soldier's health condition.
262. IOT BASED ENERGY MONITORING SYSTEM

1st Dr.R. Subasri,
Electronics and Instrumentation Engineering,
Kongu Engineering College,
Perundurai, Erode, Tamil Nadu.
2nd K. Prem Kumar,
3rd Y. Raghul,
4th S. S. Venkatesh,
The usage of electronic devices and gadgets has been increasing due to the present covid pandemic
situation. As one of the effects, the payment of EB bill is also increased for everyone. Since energy
consumption monitoring is normally done monthly or bimonthly, a user is not able to optimize the
energy consumption. As one of the remedial measures, the proposal should be in a position to
monitor the energy consumption daily or for a specific duration. In this paper, a system is developed
for monitoring the details of energy consumption, and those details are transferred to mobile phones
and Laptops. The consumption and power factor values are measured using PZEM -004T-V3.0
module and transmitted to the ESP-WROOM-32 module to display the cost of energy consumption.
263. A UNIFIED, SIMPLIFIED, AND INNOVATIVE TACTIC PREDICTION OF

CUSTOMER CHURN FOR BANK CUSTOMERS
1J Juslin Sega, 1Krishnan N, 1Shravan G, 1B Vignesh Ram
1,2Department of Computer Science and Engineering
1,2SRM Institute of Science and Technology, Ramapuram, Chennai
Due to the drastic increase in industry competition, product companies go through a huge loss in
revenue, thus losing a lot of valuable customers. To help the businesses understand the reason
behind the loss of consumers, customer loyalty measurement is recommended as a method to
May 21st – 22nd, 2022
suggest steps to recover lost customers, and eventually amplify their revenue. The transfer or
migration of clients from one organization to another or its competitor is called as ‘customer churn’.
By analyzing the history of prospective clients, customer churn can be brought down significantly.
On the other hand, the decision-makers in every organization have the requirement to address the
issue of inaccurate management of operations, and thus are in fervent need of an effective
mechanism that can predict customer churn and control it with frequent revision of necessary
strategies. As a part of the proposed system, a framework to predict customer churn has also been
established with three primary objectives, namely response, productivity and understanding of
customer churn. This particular framework additionally addresses the issues related to many stages
of customer yield. These mechanisms and schemes can be put to use for producing personalized and
customized product preferences and services, to augment client service productivity along with the
decision making related to it. Moreover, these frameworks would add advantage to devise promotion
strategies and campaigns targeted at consumers who are on the verge of churning out.
264. BLOCKCHAIN APPLICATION IN INCENTIVIZING STUDENTS ON

PARTICIPATION IN ONLINE CLASSES
K Ramani Navya, Yash Patel, Parminder Kaur Makode
Student of Computer Science and Engineering, Lovely Professional University, Jalandhar - Delhi, Grand
Trunk Rd, Phagwara, Punjab 144001
The world we know has changed over a short period, with the rise and spread of the Covid-19. This
affected the education sector resulting in offline classes to online classes. Technologies made it easy
with numerous websites, materials, video lectures, courses, and techniques for the students. In this
situation, the main problem arises with students not participating in the class having many reasons
such as illness, being introverted, and feeling that they may be wrong. If we are not interested in
something to talk about or are shy, we must face it so that we will make the best out of it. People
remember the things they listen to carefully, so we can probably study less if we listen. In this paper,
we proposed a design Blockchain-based application, so that students who are going to attend online
classes would be able to participate more in the class. They will be able to gain incentives that are
based on cryptocurrency and by using those cryptocurrencies they can spend it on fees or any other
resources in the university which would be beneficial for students. They will be able to gain
incentives that are based on cryptocurrency and by using those cryptocurrencies they can spend it on
fees or any other resources in the university. We are using the most popular technology which is
Blockchain technology to make sure that students who are attending online classes will be able to
pay more attention. Also, we are using the most famous functionality of Blockchain which is
incentivization. To give rewards to the students we will be using incentivization so they can pay
more attention to the classes. This design is beneficial for teachers also to look at the status of each
student and get in contact with them.
265. YIELD FORECASTING OF OILSEEDS USING MACHINE LEARNING

TECHNIQUES
Mithra C, A.Suhasini
Research Scholar,Dept. of Computer Science and Engineering,Annamalai University,Annamalai Nagar, Tamil
Nadu,India.
May 21st – 22nd, 2022
Professor,Dept.of Computer Science and Engineering,Annamalai University, Annamalai Nagar, Tamil Nadu,
India.
Agriculture is an important part of the country's economic system. The agricultural sector
contributes food and raw materials and employs a significant part of the population. Crop yield
prediction is one of the most difficult tasks. It is fundamental to decision-making at global, regional
and field levels. To overcome these shortcomings, we proposed machine learning models to predict
the future yield of oilseeds in Tamil Nadu by looking at past data from 1961 to 2017.The results of
the two proposed algorithms, namely KNN and AdaBoost, are compared and show an accuracy of
75.56% and 51%, respectively based on soil, weather and environmentalparameters. To obtain
accurate predictions, the model is evaluated using mean square error, mean absolute error, and the
R2 statistic. According to the results, KNN outperforms other proposed algorithms.
266. COMMUNITY MONITORING SYSTEM FOR COVID-19 PREVENTION

Atchaya S, Khartikesh K, Madhumithran A, Yogalakshmi S, Senthilvel A N
Coimbatore Institute of Technology
The calamitous outbreak of COVID-19 has brought a worldwide threat to human. The entire world
is putting incredible efforts to fight against the spread of this deadly disease. The second wave of
corona has led to exponential growth in the number of infectious people and death rate. While
researchers are continuously working to produce vaccines, world health organization is insisting on
following practices like wearing masks, sanitizing and social distancing as precautionary measures
to control the spread of COVID-19. In addition to this monitoring the heart rate and blood oxygen
saturation levels can enhance the detection mechanisms. The present scenario demands cost
effective methods of monitoring covid to control the expected upcoming fourth wave. The
proposed work presents a framework that evaluates the safety of a given region/community by
automated monitoring of mask/no-mask detection and social distance computation integrated with
heart rate and oxygen level monitoring using cost effective and non-intrusive methods. The
experimental results clearly reveal the ability of the proposed scheme in assessing safety of a given
community area. The system has to its credit not requiring additional hardware/software and hence
is a cost-effective method. The scheme could also be deployed in shopping malls, theatres,
restaurants, working offices etc.,
267. DECENTRALIZED SOCIAL MEDIA APPLICATION WITH NFT

MARKETPLACE
Dr.V.Ceronmani Sharmila
Hindustan Institute of Technology and Science, Chennai, India
Hrithik S Raj
Stephin T
May 21st – 22nd, 2022
Venkatapathi Raju.P
Social media is always in a threat where the data is being stored in a centralized database. Although
there are different ways of protecting the data even though there are several numbers of possibilities
for being attacked like eavesdropping, Man in the Middle attack, Account takeovers through various
attacks. In Many Countries social media and their chats and their activities are getting tracked by the
government. We Present ASFALIS, A Decentralized social media application with Realtime chat
Dapp which ensures decentralization, Rigid, censorship resistance, and data security. The data will
be stored inside of the blockchain and create a global copy of the data in each node. Only authorized
users can access those data by swapping the private key on the blockchain. There is no need for
trusted intermediaries here and the whole system is decentralized. Asfalis consists of its Own NFT
Marketplace which allows the users to commercialize their works or creativities in the form of NFT
and also, they can mint and put up their NFT as the profile pictures in ASFALIS.
268. NLP BASED QUESTION & ANSWERING SYSTEM

Umang Pandey1, Dr. P. Supraja2
1UG Student, Department of Networking and communications, SRM Institute of Science and Technology,
Kattankulathur
2Associate Professor, Department of Networking and communications, SRM Institute of Science and
Technology, Kattankulathur
As we all know the current world is progressing at a rapid rate in the field related to AI development
and machine learning. One of the most valuable technology is Chatbot which can help answering the
queries of humans rather than searching for the answers and filtering them to get the desired
outcome . This paper aims at an intelligent learning NLP system that will take a text file as an input
and gain knowledge from it to extract the desirable output. In this paper we’ve shown about chatbots
and how they are making human life easier .This paper is focused on a typical chatbot which will
answer the questions of students in college. It's created using NLTK library & flask and pycharm
libraries. .
Using the pattern matching algorithm which is a famous chatbot algorithm we’ve created this
college queries answering chatbot and deployed it to a website for better understanding and easy
going outputs.
269. A NOVEL DATA-DRIVEN OPTIMAL METHODOLOGY FOR DETECTING SHIP

FROM SAR IMAGES BASED ON ARTIFICIAL INTELLIGENCE
Dr. V. Sellam, Kalathur Hima Sai Shree, S Devashree, and Haripriya B, Department of Computer Science and
Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai,

May 21st – 22nd, 2022
A number of deep learning algorithms are currently being used in ship surveillance and detection,
but the major problems are insufficient target detection speed and accurate identification rate. Using
convolutional neural networks, an improved algorithm is presented for ship image target detection
and identification in this paper.The hyperspectral atmospheric correction was evaluated in this study
to improve accuracy in ship detection. On samples of airborne hyperspectral data, we applied two
atmospheric correction algorithms, and we then detected ship features based on the corrected
reflectance corrected by each algorithm using an unsupervised target detection algorithm.The novel
modules developed by our algorithm provide high accuracy, improve loss functions, and augment
small target data to yield fast ship identification. According to our experimental results, the detection
accuracy and detection rate of our method are dramatically higher than those of traditional machine
learning methods for target detection.
270. METHODOLOGY TO IMPLEMENT IMAGING APPLICATIONS ON FPGA

BASYS 3 BOARD
Vineeth Ajith John
Department of ECE
College of Engineering Guindy, AnnaUniversity
Chennai,India
V. Pushpalatha
Department of ECE
Chennai,India
S.Ewins pon pushpa, Department of ECE

Chennai,India
Corresponing author
Agastya Deshraju , Department of ECE
Chennai,India
R.Sittalatchoumy, Department of ECE
Chennai,India
Adithya Nair , Department of ECE, College of Engineering Guindy, Anna University
Chennai,India
This paper provides an insight to implement imaging applications on FPGA –Basys 3 board with
detailed procedures along with logics in verilog. As a technical contribution, image processing
applications such as increasing brightness, decreasing brightness, color inversion and converting to
grayscale are performed on a input image using a Basys 3 board. A Basys 3 board is an FPGA,
basedon the latest Artix-7 Field Programmable Gate Array from Xilinx. It’s a high-capacity FPGA
with a low overall cost and has a collection of USB, micro-USB, VGA and other ports which allow
it to host designs ranging from introductory combinational circuits to complex sequential circuits
like embedded processors and controller.
271. VIRTUAL CLASSROOM

Vaibhav Agarwal1, Akshat Swami1
1Department of Computer Science and Engineering ,Galgotias University, Greater Noida, India
May 21st – 22nd, 2022
Studies show that students who study at a distance may be more inclined to isolate themselves and a
lack of ‘presence’ and support, which could adversely affect their learning and performance.
Recently, the emergence of high- speed affordable , affordable and reliable broadband means a new
list of digital, support / synchronization.The interaction between students and faculty makes online
learning more effective, But it will be useful to crack those problems .By using adobe connect this
represents study of graduate students with Dance Live visual classes to represent a series of seminars
as a part of Post- Graduate Education Explores students' ideas on The importance of virtual-classes
in building the relationships and improving communication as well as interaction, and experienced
the problems that have contributed to this. The results show that while virtual the use of the
classroom is well-regarded, your value influenced by work, communication tools, multimedia
applications and technical / editing features — some of which were much more than that student
control. These problems will be examined and discussed, with suggestions presented to teachers and
trainers who are considering using virtual classes to support their online course learning programs.
272. A SURVEY OF IMAGE PROCESSING TOOLS PACKAGE

Varun Vishwakarma Department of Computer Science and
Engineering, Galgotias University, Greater Noida, India [email protected]
Mukund Moona Department of Computer Science and
Engineering, Galgotias University, Greater Noida, India
Image processing is a fascinating area since it provides better visual data for human simplicity as
well as image data processing for transmission and visualization for machine perception.For
improving the quality of digital photos,image processing is used.Image processing techniques
include grayscale transformation, segmentation of pictures, detection of edges, extraction of other
features, and classification. Advances in visual information for human exploration, as well as the
processing of image data for storage, transmission, and display for offline machine recognition have
fueled interest in digital imaging systems.The goals of this article are to describe image processing,
examine the many procedures and techniques involved in a typical image processing, and have a
check on image processing tools and processes in different areas of study.
273. INTERNET OF THINGS (IOT) AND ITS APPLICATIONS: A REVIEW

Dr.K.Chithra1, J.Jayapriya2, B.Atshaya Laxmi3
1Assistant Professor, Department of Computer Science, Shrimathi Devkunvar Nanalal Bhatt Vaishnav College
for Women, Tamil Nadu, India
2UG Student, Department of Computer Science, Shrimathi Devkunvar Nanalal Bhatt Vaishnav College For
Women, Tamil Nadu, India
3UG Student, Department of Computer Science, Shrimathi Devkunvar Nanalal Bhatt Vaishnav College For
The Internet of Things (IoT) is a rapidly growing technology in today's society (IoT). IoT is an
evolving technology that allows physical objects or things to communicate with other devices via
network gateways that can transport or receive data. Medical, agriculture, industrial, research,
wearables, gadgets, AI, smart homes, automobiles, and many other IoT applications are all part of

May 21st – 22nd, 2022
our daily lives. The goal of the article is to focus on the top five most popular areas where IoT is
applied.
274. MODELING AND CONTROL OF DC-DC CONVERTERS

Nikhil Kumar Maurya , Rishabh Vishwakarma, Rahul Jaiswal, Shreyansh Upadhyaya
Department of Electrical Engineering ,Delhi Technological University
Delhi, India
In this paper, Lag-Lead compensator design & PID tuning based on frequency domain has been
shown to control the output voltage of PWM dc-dc converters & to check the stability using
SISOTOOL. PID parameters for compensator has been adjusted & it improves the transient as well
as steady-state response of the uncompensated circuits then performance has been analyzed in order
to get feasible value of Gain margin & Phase margin at higher bandwidth & further, PID tuning has
also been done to improve robustness & it has also been verified through Simulink model.
275. AI BASED E-COMMERCE WEB APPLICATION FOR EVENT MANAGEMENT

Jyoti Kuwar Devada
Navi Mumbai, India
Devin Raina
Navi Mumbai, India
Srijani Santra
Navi Mumbai, India
Prof. Seema Redekar
Navi Mumbai, India
Even with all the e-sites available, one has to search through different applications for each special
requirement in services. For a budget friendly birthday present, buying the décor, searching for
something customized, homemade birthday cake, photographers, searching for planners, caterers.
Organizing and arranging events has often been a big task, with all the things mentioned above we
still find it difficult to rely on any of those and still the event is not up to your mark. So we
researched on this and from personal experiences we came up with a solution along with AI, that
could help many to plan an event with safety, reliability and authenticity. Bingo will boost small
businesses amidst pandemic and help introverts and people with no bargaining power to actually
have an event either planned by themselves or a professional event manager/planner. Bingo will
have a large database with a large variety of products and also the recommendation system for the
loyal users to get accurate and budget friendly packages and items.
May 21st – 22nd, 2022
276. CORRELATION BETWEEN THE DEPENDENT AND INDEPENDENT

PARAMETERS IN THE EXTENDED HODGKIN AND HUXLEY NEURON MODEL
USING PARTIAL LEAST SQUARE REGRESSION TECHNIQUE
Devi kannan1 and Akhil Ranjan Garg2 ,
Atria Institute of Technology,Bangalore1 Jai Narayan Vyas University, Jodhpur2
Parameter sensitivity analysis is the vital technique in finding the correlation between the
independent and dependent parameters in the computational modelling. Partial least square (PLS)
regression technique is used for parameter sensitivity analysis. This simple method gives a robust
model, which shows the correlation between the dependent parameters when independent
parameters are varied in certain range. The goal of the PLS regression is to generate a new
simplified and empirical model which predicts the output resulting from a new set of input
parameters. It also generates regression coefficient matrix which is reflection of parameter
sensitivities of input parameters. In this paper we are going to do parameter sensitivity analysis of
the extended Hodgkin and Huxley (HH) model with synaptic currents.
277. ONLINE ELECTRICITY BILLING SYSTEM

Ashwin Dr. Kirti Shukla
Artificial Intelligence & Machine Learning Associate Professor in
Computer Science Engineering Galgotias University
Galgotias University Noida, U.P
Bihar, India
The traditional electricity billing system is inefficient; one employee must visit each customer's
home to record metre readings and collect data. The consumed units must then be calculated, as well
as the money that must be paid. The bills that have been created will once again be delivered to
clients. Finally, individual customers must pay their dues at the electricity office. As a result, the
traditional electricity billing system is inefficient, requires a large number of employees to perform
simple tasks, and is a lengthy procedure in general. A web-based computerised system is required to
tackle this time-consuming billing process. With its features, the suggested power billing system
project eliminates all of these disadvantages. It is advantageous to both the users and the electricity
provider. The new system will result in a decrease in the working professionals employed by the
organization. The software's working speed and performance are faster with excellent performance,
saving time. Furthermore, there is virtually minimal likelihood of staff corruption and
miscalculation. The system is written in Java swings, a programming language that can be used to
create other applications.
278. MODELLING A VOTING-BASED MODEL FOR DIABETES PREDICTION

USING LEARNING MODELS
R.MURUGANANTHAM
Professor

May 21st – 22nd, 2022
TKR college of engineering and technology

Hyderabad,Telangana.
G.BHANU KIRAN
M.SOWMYA SREE
A.RUSHIK
TKR college of engineering and technology.
Diabetes mellitus is defined as a collection of metabolic problems that significantly impact human
health worldwide. Wide-ranging study into all aspects of diabetes (diagnostic, pathophysiology,
therapy, etc.) has ushered in an era of massive amounts of data. This investigation aims to provide a
prediction model using machine learning, data analysis methodologies and tools in diabetic
prediction. The primary goal of this work is to design a method that can more accurately predict
diabetes in patients. Here, a novel ensemble model is evaluated using several characteristics such as
precision, accuracy, F-measure, and recall. The machine-learning techniques are identified after
hyper-tuning and cross-validation (CV) and then employed in the Vote-based ensemble model
(vEM). According to the findings, the proposed framework can get an excellent result of
approximately 92% accuracy.
279. STUDY OF ELECTRO-MECHANICAL PROPERTIES OF FDM MATERIAL

USING DIFFERENT COATING COMPOSITIONS
1Samarth J Mangalore, 2Akhil V M
1,2The National Institute of Engineering, Mysuru-570008, Karnataka, India
Acrylonitrile Butadiene Styrene (ABS) material is engineering plastic with a uniform distribution of
butadiene on a matrix of acrylonitrile-styrene. ABS exhibits good toughness, better stability in
dimensional field, ability to process, chemical resistance and low cost. But there are inherent flaws
in terms of mechanical strength, electrical properties, upon that, the ABS is a non-conducting
material. The research is to enhance the electro-mechanical properties of the Acrylonitrile Butadiene
Styrene (ABS) material by developing a coating layer upon it. This paper aims to review the
fabrication process, pre-treatment process, techniques of coating, enhancements in the properties are
presented based on the literature published in recent years.
280. FOOTPRINTING AND RECONNAISSANCE TOOL

Hritwik Mukherjee
Dept. of Comp. Sci.
Chennai, India
Dr. Thanga Revathi S
Professor,
May 21st – 22nd, 2022
Dept. of Comp Sci

Chennai, India
Arijit Mondal
Dept. of Comp Sci
Chennai, India
Recon is an essential step in obtaining classified and confidential information by thoroughly
investigating a field or an area. Penetration testing requires it as well. Our Python-based application
aids in the collection and collation of data specific to a given company. Such sources of data will aid
in an organization's vulnerability assessment (disclosure of sensitive information).
Hackers can get access to a network by taking advantage of unpatched or out-of-date systems. They
can perform social engineering on an employee or a breach could occur even if a corporation has
good security techniques but an old update policy.
281. DESIGN AND ANALYSIS OF SOLAR AND WIND BASED CHARGING

STATION FOR ELECTRIC VEHICLES
“Dr. Balaji Madhavan¹” , “Shanmugam B ²” and “Solomon Sagayam D J ³”
1. M.E.,PHD,2. Third year ECE student,3. Third year ECE student” Agni
College of Technology
Recently, Electric Vehicles (EVs) have attracted much attention owing to their use of clean energy
and promotion in sustainable environmental growth. Large progress in lithium-ion battery has
propelled the development of EVs. However, the challenge is that the growing number of EVs leads
to huge demand in electric power, which will aggravate the power grid load. This leads to an
exploration for alternative and clean sources of energy to charge EVs. This paper presents a detailed
review of solar and wind energy based charging stations for EV application. The results of various
schemes of implementation for the charging station are discussed. Upon detailed literature review,
few inferences were made, namely, a charging station employing multi-port charging facility by
providing a constant voltage DC bus and charging controllers operated based on power balance, and
constant current/constant voltage charging.
282. REMOVAL OF CHROMIUM VI IONS FROM TANNERY EFFLUENT USING

EMULSION LIQUID MEMBRANE AND OPTIMIZATION OF PARAMETERS
USING RESPONSE SURFACE METHADOLOGY
Chandira Kumaar. P1, Harishkumar. R2, Manoj.G3, Pagidi Aruna1*
Kongu Engineering College, Perundurai, 638 060
Consistently, enormous measure of wastewater is delivered straightforwardly into the climate by the
few modern areas, for example, drug, metal covering, mining, metallurgy and so forth. In this
decade, the expulsion of poisonous metals and natural particles from the modern waste streams has
urgent significance to save human populace. The extraction of these natural or metal particles is
profoundly beneficial and helpful attributable to their extremely wide applications in a few
enterprises like food, corrective, polymer, substance, and drug and so forth. The customary division
May 21st – 22nd, 2022
methods are as precipitation, adsorption, layer partition, responsive refining, fluid extraction,
electrodialysis, and particle trade. Among different strategies, layer division is the most recent
method and exploration is going on different perspectives. In this work, among different fluid layers,
Emulsion Liquid Membrane (ELM) has been chosen dependent on their benefits. The ELM has
some appealing highlights when contrasted and other detachment process like minimal expense,
high effectiveness, short working time, basic activity, low energy utilization. In this work, we
integrated different creations of ELM. ELM can be envisioned like a 'bubble inside an air pocket'.
The inward air pocket conveys inside stage reagent and external air pocket contains
transporters/extractants. Because of the fixation slope, different solute particles are transported from
the fluid stage to the stripping stage via ELM. Changes in concentration of carrier, surfactant,
internal phase, feed, and internal phase ratio improve the membrane. The advancement of factors is
finished utilizing RSM to track down the reasonable working conditions.
283. WASTE MANAGEMENT AND WEALTH GENERATION THROUGH WASTE

ELIMINATION
Shruti Tharmia1, Riddhi Nilawar1, Nikita Aswale1, Shubhadra Kumari1, Dr. Poonam Gupta2 1Student,
Department of Computer Engineering, GH Raisoni college of engineering and management Pune.
2HOD, Department of Computer Engineering, GH Raisoni college of engineering and management Pune.
Waste management and disposal has become a matter of concern with increasing waste products due
to increase in population. There are a lot of dustbins that remain uncleaned for several days due to
which there is a spread of foul smell and various diseases. Also, there are many products which can
be reused but are not being reused properly, they are being discarded or burned which creates air
pollution and wastage of raw materials. These raw materials can be used by the industry people and
be given a new shape to it. Wealth can be generated by selling these secondary waste products.
There is lack of awareness among the people regarding proper waste management and reuse of
waste products. As a solution to all these problems the proposed system focuses on easy collection,
cleaning of the dustbins, reusability of waste products as well as spreading of knowledge and
awareness among the people.
284. IMPLEMENTATION OF ADVANCED ENCRYPTION STANDARD (AES) ON

FPGA
C Kalyana Sundaram, S Keerthana, V Krithika Laxmi Department of ECE, Mepco Schlenk Engineering
College
In the world of growing technology, security of data is playing a vital role in day-to-day life.
Nowadays hackers are equipped with advanced tools and techniques to break any security system.
Cryptographic algorithms are making use of achieving integrity, availability, identification,
confidentiality, and authentication of user. Advanced Encryption Standard (AES) can be utilized to
ensure the security of data, which overcomes the attacks of Data Encryption Standard (DES). The
proposed paper implements Image Encryption and Decryption of 128-bit AES using Verilog code
which is synthesized and simulated in FPGA.

May 21st – 22nd, 2022
285. REGULARIZATION IN NEURAL NETWORKS

Farhana Kausar1 and Dr. Aishwarya P2
1,2 Atria Institute of Technology, Visvesvaraya Technological University, India, ,
Neural networks are now important not only in the fields of computer science, but also in the but
also in countless fields of everyday life. It is one of the big technologies in artificial intelligence
development and will remain of great significance for, for instance, the growth of autonomous
driving in the future. From the data to the analysis Neural networks are obviously constrained and
thus a sound capacity for generalization is not required for the creation of neural networks for all
possible data to be treated in real-life situations. This means that a general understanding should then
be gained from the training data, in order to interpret the role associated with the data correctly and
not to merely recall the training data. Regularization is an integral part of such a generalization. A
regularization mechanism allows a neural network to (better) be generalized during data learning.
There are different, widely used and widespread regularization procedures which restrict the weight
of a neural network parameters through which the adaptation takes place or modify its structure
temporarily, and thus attempt implicitly to improve the prediction of a neural network on the unseen
data.
286. INVESTIGATION OF VARIOUS SUPERVISED MACHINE LEARNING

ALGORITHM TO CHARACTERIZE MAMMOGRAM BREAST IMAGES
N. Gayathri
PG Scholar-ME Communication Systems
Department of Electronic and Communication Engineering, Sri Sairam Engineering College, Chennai, India
Dr.J. Thamil Selvi
Associate Professor
Department of Electronic and Communication Engineering, Sri Sairam Engineering College, Chennai, India
Cancer is the prevalence global disease with statistic of 68,500 deaths coupled with 2.3 million new
cases in 2020 worldwide. Therefore, computer aided early and accurate detection of disease is the
prime concern for researchers and doctors. So, this work focuses on building a framework for
accurate detection and diagnosis of mammogram image based cancerous tissue. Different de-noising
filters, such as anisotropic diffusion-based filter (ADF), median filter, wiener filter and double
wavelet transform (DWT) techniques are attempted for noise removal. The pre-processed images are
segmented using double thresholding segmentation technique and morphological area gradient-
based segmentation (MAG). From the segmented image, grey level co- occurrence matrix (GLCM)
features are extracted and analyzed. Results shows, ADF as the most efficient filter among the
abovementioned technique. Double thresholding segmentation is observed to segment breast
efficiently. Among the 11 GLCM features, four features discriminated normal and abnormal
subjects. Supervised machine learning algorithms, viz., support vector machine, logical regression
and decision tree module were used for the purpose of classification. Quadratic support vector
machine technique proved to be the most efficient amongst the compared. Hence, the proposed
framework of ADF filter with double thresholding segmentation and GLCM feature shall be used
for characterization of mammogram breast images.

May 21st – 22nd, 2022
287. A LITERATURE REVIEW OF BLOCKCHAIN AND GAMIFICATION BASED

INCENTIVE MECHANISM IN EDUCATION SECTOR
Mahesh Undavalli, Lovely Professional University
Parminder Kaur, Lovely Professional University
Kavisha Duggal, Lovely Professional University
Technology is all about changes, enhancements, and something new. Looking into that aspect,
learning new ways of using the latest technologies is a smart move. Blockchain is an emerging
technology and its applications are wider and more helpful to move the traditional ways toward
advancements. One of its main applications is the incentive mechanism, where crypto-incentives can
be distributed as rewards. The incentive mechanism is not at all new and is also used in the
Gamification methodology for motivation. One of the major uses of incentive mechanisms is in the
education sector. Therefore, a literature review has been performed in this paper to explore the
common ways in which both blockchain and gamification can be used to incentivize students in the
education sector. The results show various means by which gamification and blockchain are
altogether similar in providing rewards for various education-related activities and enhancing the
methods of education.
288. STOCK MARKET PREDICTION USING MACHINE LEARNING

Ashish Anand, Ujjawal Pant, Dr.Nitin Mishra
Galgotias University
In finance world, Trading of stock is the most important activities. Trying to predict the future value
of another financial instrument that trades in financial markets is known as stock market guessing.
This document describes stock forecasts using Machine Learning. Basic and basic analysis or
timeline is used by most stock traders when making stock forecasts. Python is a computer language
that is used to anticipate stock market movements using machine learning. We present a Machine
Learning (ML) algorithm that will be trained using publicly available stock data in this research and
gain intelligence and use the information obtained to obtain an accurate forecast. In this respect, the
study employs a machine learning method known as Vector Support Machine (SVM) to forecast
stock values in three separate marketplaces, using prices on daily frequencies and high minutes.
289. A WEIGHTED MULTIVARIABLE ADAPTIVE REGRESSION MODEL FOR

DELIVERY TIME PREDICTION IN E-COMMERCE SUPPLY CHAIN
1 . Dr. Kavitha R 2. K. Nagur Babu
3. K. Naga Vamsi Manikanta Reddy
1. Assistant professor, Department of Information Technology, SRM Institute of Science and Technology
Ramapuram, Chennai, India
2,3,Students, Department of Information Technology, SRM Institute of Science and Technology,
Ramapuram, Chennai, India

May 21st – 22nd, 2022
Digital evolution has notably modified client buying experience and expectancies ensuing in the
firstrate increase of e-trade. The instantaneously final results of this increase become the
introduction of dynamic and turbulent surroundings with a growing density of the distribution
network. These surroundings include many transport points, a couple of transport channels, and last-
mile transport necessities. Due to its complexity and the proliferation of records challenges, e-trade
logistics is a place in which the software of Big records analytics may be demonstrated to be
extraordinarily fertile. Despite the developing interest, there are restrained research that look at the
position of Big records. By the usage of a layout technology approach, we clarified the modern e-
trade logistics practices in addition to the expected ones that can, to a huge extent, be supported with
the aid of using suitable huge records technologies. We concluded with a fixed of commercial
enterprise necessities that explicit the desires in the direction of e-trade logistics. Then, the
necessities have been translated into a fixed of use case eventualities to illustrate how they might be
supported with the aid of using huge records analytics. We finish with the aid of using featuring a
conceptual structure of a huge records analytics artifact that would cowl the e-trade logistics
necessities. Using Machine Learning computing, we advanced a version able to predict transport
instances with the aid of using schooling the algorithms on ancient cargo records and incorporating
outside reasserts of records associated with the maximum impactful elements concerning timetable
reliability.
290. MECH-HEALTH: A MACHINE LEARNING BASED FAULT DETECTION

FOR PREDICTIVE ANALYSIS OF MACHINE
Ayushman Chaturvedi1, Dr. R. Vidhya2

1Dept. of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai (India)
2 Dept. of Computing Technology, SRM Institute of Science and Technology, Chennai (India)
Predictive maintenance is a form of maintenance that looks at how well machinery works and how
efficient it is during regular operations to reduce the risk of failure. Machinery and other appliances
are subjected to less risk due to a sudden failure with the help of predictive maintenance. This form
of maintenance device alerts the owner until a malfunction or breakdown is anticipated. Predictive
maintenance schedule operations, unlike traditional preventive maintenance, are focused on data
obtained from sensors and analysis algorithms. We address the validity of using LSTM instead of
conventional approaches such as visual inspection in this paper.Two different models, the Prophet
and the Autoencoder LSTM recurrent neural network are used to train the ball-bearing dataset. The
prophet method is a forecasting technique implemented with the aid of an encoder-decoder LSTM
architecture in R and Python, while Autoencoder LSTM is an autoencoder of sequence data
implemented with the help of an encoder-decoder LSTM architecture. Using the two algorithms,
and efficient maintenance scheme can be created to limit any potential risk to the assets until they
misfire and fully collapse.
291. NON – INVASIVE METHOD TO DETECT DIABETICS

Dr. Suveetha Dhanaselvam P 1, Vijaya Selvi M 2, Dhakshitha Herculin C 2, Aruna Rajeshvari M 2
1 Associate Professor, 2 Final Year UG Students
1,2Department of Electronics and Communication Engineering, Velammal College of Engineering and
Technology

May 21st – 22nd, 2022
There is skyrocketing spread of Diabetes in the world with 463 million cases in 2019. Current
procedures for the detection of diabetes are invasive and complicated for the patients. So, there is an
urge for a rapid, non-invasive method to detect diabetes. Breath analysis is a feasible technique for
non-invasive detection with the potential to detect multiple diagnoses. Acetone in exhaled breath has
been taken as a biomarker for diabetes. Here, we present a technique for diabetic detection, that
shows the acetone level in mmol/l along with temperature and humidity levels from the exhaled
breath. Breath acetone is considered to be the accurate reflection of the body’s ketone level. The
patient is made to blow into the device and the reading from the device is sensed. Then the value of
acetone is displayed in the LCD and also the results can be stored on the websites through IoT. This
serves to be a novel method to observe diabetes in patients without pain.
292. SUPLYING ADJUSTABLE POWER SUPPLY TO THE BIOMEDICAL DEVICES

BY USING INTERNET OF THINGS (IoT).
Dr. Ganesh Perumal Dharmaraj, R. NARENDRA, V. CHENNAREDDY, B. GNANESWAR REDDY
Department of ECE
KALASALINGAM ACADEMY OF RESEARCH AND EDUCATION
Now a days most of the people are looking towards medical side for saving lives of many people.
Due to the highly requirement of health care system, these days medical facilities have been
increasing to provide better comfort to the needy peoples. For the development of these devices
Internet of Things (IoT) is playing an important role for health care services. Now a days, many of
the people are preferring operations for some kind of diseases and failure of some organs. During
this time many deadly incidents occurring due to power fluctuations. Sometimes the already existing
systems may not provide accurate output current required for the biomedical devices. To avoid the
above issues, an IoT based power supply with adjustable output for biomedical devices and safety
systems has been proposed. ESP32 DEVKIT V1 has been adapted to reduce fluctuations. Moreover,
Alternating Current sensor (AC current sensor), Alternating Current voltage sensor (AC voltage
sensor), Direct Current sensor (DC current sensor), Direct Current voltage sensor (DC voltage
sensor) also used for detecting of power (AC or DC) along with the change in the voltage.
Furthermore, pulse with modulation signal also used for adjusting the power supply in case of
fluctuations.
293. E-COMMERCE: CONCEPT AND DESIGN

Umang Singh1, Amit Singh2
1,2Student, Student, Dept. of Computer Science and Engineering, Galgotias University Greater Noida, Uttar
Pradesh, India
E-commerce can be a development within a modern business.E-commerce method trading

electronically. E_commerce entails the buying and promoting of services and products, or the
transfer of finances or records, with the aid of device, specially the internet. E-commerce may be a
paradigm shift that influences both outlets and clients. Instead e-commerce is very one of a kind to
boost present commercial enterprise processes. It results in a complete transformation of the
conventional manner of doing commercial enterprise. This vital change in the enterprise version is
proving super increase global and India is not any exception. first rate net penetration has introduced
to the boom of E-trade and especially beginners are an increasing number of using this selection as a
differentiated business version. Further E-trade has a big effect at the surroundings. even though the
May 21st – 22nd, 2022
version is broadly used inside the cutting-edge commercial enterprise surroundings however the
choice has now not been absolutely evaluated. current research and analysis has been done to outline
the fame of E-trade websites, studying E-trade patterns
294. BITCOIN PRICE PREDICTION USING MACHINE LEARNING ALGORITHMS

Tarun kumar, B-Tech, Computer Science and Engineering,
SRM Institute of Science and Technology.
Dr. T. Sam Pradeepraj, Department of Computer Science and Engineering,
SRM Institute of Science and Technology.
In this article, we have proposed to accurately predict the price of Bitcoin, taking into account
various parameters that affect the value of Bitcoin. By collecting information from various
references and applying it in real time, we found the strengths and weaknesses of Bitcoin Price
Prediction. Every paper has its own way of predicting the price of Bitcoin. Many papers have
accurate prices, while others do not, but these predictions are time-complex.
This white paper compares the accuracy of mutiple artificial intelligence-related algorithm. to reduce
time complexity and increasing the accuracy for the predicted Bitcoin price. Here we have used
algorithms such as Logistic regression, Linear regression algorithms, Random Forest Regressor,
Decision Tree Regressor, SVR and Lasso algorithms.
These are the attributes used for comparing the algorithms that is mean_absolute_error, r2_score,
median_absolute_error, mean_squared_error and explained_variance_score. Our goal is to
understand and find the day-to-day trends in the Bitcoin market, while providing insights into the
optimal features surrounding Bitcoin prices. Our dataset consists of various characteristics related to
Bitcoin prices and payment networks and is recorded daily. By pre-processing the dataset, data
mining technology is applied to reduce data noise. Then, in the second moment of area of the survey,
we use the available information to predict the signs of daily price fluctuations with the highest
possible accuracy.
295. A STUDY ON IMPROVING BUSINESS EXCELLENCE IN SUPERSTORES.

Shreya Jingade D, Sindhu M K, and Ashok Kumar R
BMS College of Engineering
Businesses, and investors prioritize sales and profits because they convey information about the
company's overall health. As huge wholesale markets have been developed, the emergence of local
superstores has become critical. The dataset contains around 9,990 rows of data on a superstore's
sales, profitability, discounts, and other features. The sales and profit analysis is carried out here
using properties like discounts, states, cities, and categories. Using business analytics visualization
approaches, city-level target areas for enhancing superstore sales and profits across the United States
are discovered. The results of four years of data collection from all superstores around the country
show that New York City and Los Angeles produce the greatest income and sales in all categories.

May 21st – 22nd, 2022
296. ANALYZING FACTORS AFFECTING YOUTUBE VIDEOS TO BECOME

TRENDING VIDEO
P M Shreenidhi, Sindhu M K, and Ashok Kumar R
BMS College of Engineering
Social media is one of the platforms, that provides content creation, ideas, information, access to
news, social interaction, and interest sharing platform to everyone. It’s not only for communication
but also used for business life. YouTube is a social media platform used to share videos by content
creators. Contents of the video may be of news, short videos, movie teaser trailers, music,
documentaries, live streams, and more. Most discussed videos on social media becomes trending and
the factors which affect them also vary, it can be based on views, likes, share, comments, and tags.
The analysis is made based on the collected data to know which factor affect a video to become
popular.
297. COVID-19 PREDICTION USING MACHINE LEARNING USING REAL WORLD

DATA
Abhishek Harikrishnan, Ayushman Mookherjee, and Dr. Sanjay Kumar Singh.
Vellore Institute of Technology
Several cases of covid-19 went undiagnosed worldwide due to poor testing infrastructure. With the
growth in big data in healthcare industry, accurate prediction using medical data is possible.
Prediction using machine learning is gaining widespread popularity. With the help of a prediction
system using machine learning algorithm, a low-cost solution for disease prediction can be provided
to the general public. The proposed web-based detection system utilizes 6 different machine learning
algorithms to provide a sophisticated detection system where the mode of all the systems will give
the outcome. Logistic regression, decision tree, Random Forest, Gaussian Naive bayes, Support
Vector classification and K-Nearest Neighbors are used to classify whether a person has covid or not
based on the symptoms. This paper contains an in-depth comparison based on the accuracy,
precision and recall of the various algorithms.
298. COST EFFECTIVE, 3D PRINTED BINIC ARM MECHANICAL DESIGN

CONTROLLED BY ELECTROMYOGRAPHIC SIGNALS
Atul John Abraham,Gouri Lakshmy Radhakrishnan,Melvyn Mathews,Dr.Varshini Karthik
SRMIST
It is recognized that the existing bionic arms are expensive ranging from $4000 to around $20000. It
is estimated by the world health organization that out of the 30 million amputees only 20% use
prosthetics to satisfy their needs. In India, it is observed that 1.8% of the population are arm
amputees, out of which 60% are from a rural background. There is also an addition of at least 23500
new amputees every year with a backlog of 400,000 amputees without any aid. Since prosthetic
arms are costly due to its sophisticated technology, it is not feasible for them to own one. Only 3 out
10 people invest in them though it is very beneficial to an amputee.
May 21st – 22nd, 2022
Understanding the fact that the existing myoelectric prosthetics are expensive and heavy, we are
developing a bionic arm which is low cost, lightweight with 5 ranges of motion. This is being
developed with the help of 3D printing where we are using PLA (Poly Lactic Acid) filament for
printing. The flexion and extension of the existing prosthetic arm are controlled using strings which
in time loosens up or causes wear and tear. So, we have come up with a unique design for the
controlling of the fingers through a series of mechanical movements. As the design of the arm gets
more complex, the cost increases. But we are creating a bionic arm which runs on a complex
mechanical design and is yet cost effective. Based on the electromyography pattern recognition of
the movement, the sensors make the development of the process much simpler. The myoelectric
signal acquisition, mapping and the application of the movement is processed by the sensors. Our
vision lies in trying to span the gap where people no longer see disabilities as a limitation, but a
super power.
299. FINGER VEIN DETECTION USING DEEP LEARNING

Dr.S.Saranya dept. of
Hindustan Institute of Technology and science
Chennai,India
Yenugu Sumanth dept. of
Chennai,India
Vellaturi Kumar Teja dept. of
Chennai,India
Alapati Sasikanth dept. of
Chennai,India
Nowadays, high security is the most essential requirement for spoofing attacks. There is more
demand for more security and more precision and more speed of authentication for buying
electronics. There is a large scope solution for security issues is human observable and physiology
characteristic in bios crypt. In any case, in phrases of both time and space the biometric systems are
highly complex, for high security this is not that much appropriate. Consequently, for high
authentication we proposed a finger vein recognition system that one give high verification
contrasted with other existing strategies.
300. MACHINE LEARNING APPROACH TO CLASSIFY AND PREDICT

OSTEOSARCOMA GRADING
Austin James1, Siddharth Baskar1, Anupama Bhan1,
1Department of Electronics and Communication Engineering, Amity University, Noida, U.P., India
Osteosarcoma is a type of bone cancer. The malignant tumors are mainly found in cells that forms
the bone. Osteosarcoma grows in long bones near metaphyseal growth plates, commonly found in
May 21st – 22nd, 2022
Humerus (10%), Tibia (19%), and femur (42%).[1] To identify a malignant bone tumor is a
strenuous task as it is difficult to find it Also, classification from histopathological images of
Osteosarcoma is a cumbersome and tedious process. Generally, Osteosarcoma is classified into
viable, Non-viable, and Non-tumor classes.
In this paper, required features were extracted from the histopathological images, which were
obtained from The Cancer Imaging Archive, using CellProfiler and the features were then fed to the
four machine learning models used in this project. Random Forest (RF), Support Vector Machine
(SVM) Decision Trees (DT) and Logistic Regression (LR) are the machine learning algorithms
used. These models were used to achieve a more accurate classification of Osteosarcoma into its
respective three classes. Finally, the outcomes of all the models were compared to select the best
performing of them. Keywords: machine learning, CellProfiler, ML algorithms, classification.
301. RAROAD SURFACE CLASSIFICATION THROUGH VEHICLE VIBRATION

SIGNAL PROCESSING
Priyam Rai1, a), Rajeev Kumar2, b), Vaibhav Mahendru3, c), Chirag Jain4, d), Tarun Kumar5,e)
1,2,3,4,5 School of Computing, DIT University, Dehradun, India
In India, road transportation is the most widely used means of transportation. Around 150,000
people die annually in India, and more than 450,000 people get injured or disabled due to
automotive accidents. Informing the organizations involved in road construction and maintenance
for the kind of road is vital and could help avoid major medical emergencies. An IoT-based method
is used in work to describe the condition and place and the kind of road condition. The method uses
the sensors available in the smartphone of the driver or the passenger mounted on the vehicle's
handle (twowheeler) using a phone mount and is kept parallel to the ground to detect the vibration of
the road condition. The vibration data that the smartphone captures are in three routes vertical,
horizontal, and depth. The data captured is recording the vibrations of the Accelerometer, which is a
three-axis X, Y, Z, and a gyroscope for the same three-axis. The Accelerometer is of both kinds,
linear and non-linear. For the accuracy of the data, this work uses the linear kind to remove the
gravitational acceleration caused by the earth and an increment of 9.8 m/s2 in the readings. The
accuracy of the work is given by KNN for classification and precision, recall, f1-score, and support.
The key intent of this research is to root out from the acquired data indispensable features and
identify the condition of the road. The work mainly focuses on two-wheeler drivers but is readily
applicable to four-wheeler vehicles. Any vehicle with a stable enough location to mount a phone
with minimum body vibration can benefit from this work.
302. PREPARATION OF HEAT RESISTANCE AND FIRE-RETARDANT COATING

USING SILICONE POLYESTER RESIN DOPED WITH Al2O3
Nagarjan J 1, Abishek P 2.
1Department of mechanical engineering, Mepco schlenk engineering college, Sivakasi -626005
2Department of mechanical engineering, Mepco schlenk engineering college, Sivakasi -626005.
In this journal discussing about the heat resistance and fire-retardant behavior of the Al2O3 is
dispersed in the silicone polyester resin and it’s obtained through the variable speed mechanical
May 21st – 22nd, 2022
agitator. The Al2O3 is crushed using the high energy ball mill of 8 hours. The coating is applied on
the wood and mild steel substrates. The formulation was analyzed using FTIR, SEM, TGA and
Contact angle measurement. The coating has good wettability due to the hydrophilic nature and the
contact angle of 82°. thermal stability of coating TMaximum weight loss = 450°C of 50 wt.%, has a
better thermal conductivity of 2.1 w/m2K, the vertical flammability test of coating is fire
extinguished at 20 seconds of its length sample and compare to bare wood is extinguished at 28
seconds of its length.
303. INTELLIGENT DIGITAL IMAGE WATERMARKING USING WAVELET

TRANSFORM
Ankur, Aman Pratap Singh
B. Tech CSE, Galgotias University, Greater Noida, India
Today, the Internet has become a suitable medium for downloading multimedia content. Digital
media can easily be copied, which can lead to security issues. Digital watermarks can protect digital
content from counterfeiting, piracy, and any unauthorized access. Digital watermarking is an
essential technology for adding copyright notices, secret messages, or verification.
304. AI BASED ALGORITHMS FOR COMPUTER AIDED DIAGNOSIS

Purushottam kumar*, Dr. kirti Shukla*
*Department of computer science and engineering, galgotiasuniversity
Many medical professionals in the field of Aided Diagnosis have started to use computer aided
design to aid in the process of diagnosis and treatment of diseases. Computer aided diagnosis is the
process of analyzing a patient's condition and then providing diagnostic information that can be used
by other medical professionals to treat the patient.Computer-Aided diagnosis (CAD) is a computer
aid used in medical image analysis to help the medical professional evaluate x-ray, MRI and blood
flow analysis.The use of CAD in medical imaging is known to be beneficial in the field of radiology.
It has allowed radiologists to better understand the physical characteristics of an individual or group
of individuals.
As factories in the modern world become increasingly complex and require more workers to
produce goods, there is a need for technologies to automate parts of the process. In order to keep up
with the growing demand for automation, technology is continuously being developed to make use
of complex algorithms for repetitive tasks.In this paper, we demonstrate that the use of AI can be
used in a CAD system to create customized 3D models called CADDS 5. Our study is focused on
how well this CAD system performs in terms of its accuracy and speed.
305. EXAMINING FAKE NEWS OBSERVATION AND DETECTION IN SOCIAL

MEDIA: USING MACHINE LEARNING AND DEEP LEARNING ALGORITHMS
1Aswiga R V, 1Sudharsan S, 1Nithishvaran T P, 1Mohan Vamsi M S
1Department of Computer Science and Engineering
Amrita School of Engineering, Amrita Vishwa Vidyapeetham,
Chennai.
May 21st – 22nd, 2022
Social media, in recent years, has become the mainstream media platform which produces a
large amount of content and news for its huge set of audience. Things like politics which have
paramount influence on the general public’s livelihood are also influenced on social media.
Unlike a century ago when only news reporters and officially recognized publications were the only
ones who could spread news to the masses, we now live in an era where literally everyone holds the
power to create and to spread news. Along with this, the problem that arises is the creation and
spreading of fake news which leads us to question the credibility of the news and the
content we consume. These fake news can change one’s perspective for the bad. In this paper,
we will have a look at the various misinformation spread across social media and the
methodologies used and being used to tackle them. This survey is mainly focused on how
traditional machine learning algorithms and deep learning techniques are being used in categorising
the news between “Fake news” and “Not fake news” from a raw text. We discussed how the
stacking model works in the field of fake news detection and about its improvements shown in the
accuracy score and also discussed how some research is done to show how feature extraction
techniques plays a vital role in improving accuracy score by reducing the complexity of the dataset.
Along with it, we also discussed fake news detection related to COVID-19 which has been the most
crucial topic in recent years.
306. ROLE OF INTELLIGENT CAD SYSTEMS FOR REAL WORD

APPLICATIONS - A REVIEW
Aswiga R V1 , Thillai Manjari T1, Sudharsan S1, Nithishvaran T P1, Mohan Vamsi M S1
1Department of Computer Science and Engineering, Amrita School of Engineering, Amrita Vishwa
Vidyapeetham, Chennai, Tamilnadu, India.
In recent years there is a rapid deployment in computer aided techniques that has been widely
adopted to solve real world problems and mostly attributed to the increasing demand in the medical
community. Recent innovations like Big data analysis, pervasive computing, and eventually
Artificial Intelligence (AI) are employed on top of the IoT to discover any new innovation to save
patients and help physicians in the medical community, which is much need of the hour. In this
context, we present a comprehensive survey of the latest innovations developed using AI techniques.
Specifically, we present an overview of applications that use AI approaches to solve the emerging
problems and show how AI plays a dominant role in wireless network functionalities. Particularly,
we provide an indepth discussion on medical applications where AI proves extremely useful for
automatic diagnosis and decision making. Finally, open issues and promising areas for future
research are also discussed.
307. AN APPROACH TO DETECT BRAIN MALIGNANCY USING BAT

ALGORITHM
Dr Suma Sira Jacob

Associate Professor - Department of Information Technology
Coimbatore , India
Sangeeth Pravin V
B.Tech Information Technology
May 21st – 22nd, 2022

Coimbatore , India
Rashmika B
Coimbatore , India
Rithikadevi P
Coimbatore , India
Brain tumor is a fatal disease and its classification is a difficult task for radiologists because of the
heterogeneous nature of the tumor cells. Recently, computer-aided diagnosis-based systems have
promised, as a rehabilitative technology, to diagnose the brain tumor, through magnetic resonance
imaging (MRI). The purpose of the paper is to model the complex interaction of brain and behavior
of the brain and to understand and diagnose brain diseases by collecting and examining large
quantities of inputs .Archiving, analyzing, and sharing the growing neuroimaging datasets posed
major difficulties. New computational techniques and technologies have emerged in the domain of
Big Data but have not been fully adapted for use in neuroimaging . This paper introduces the current
challenges of neuroimaging in a big data context. A new method is developed to overcome these
challenges . Multimodal MR images are segmented into super pixels to alleviate the sampling issue
and to improve the sample representativeness . Next , features are extracted from the super pixels
using multi-level Gabor wavelet filters . Based on the output of the GRAY LEVEL CO-
OCCURRENCE MATRIX (GLCM) and spatial affinity models, conditional random fields theories
are applied to segment the tumor in a maximum a posteriori fashion given the smoothness prior
defined by our affinity models. So the result indicates , the sample image gets preprocessed using
GLCM methodology , next the training and testing processes are performed and finally with the help
of the bat algorithm the tumor region gets segmented .
308. SINGLE IMAGE DEHAZING USING MULTI-SCALE DCP-BCP FUSION

Name: Swaroopkrishna Nair, Sandeep Menon, Manas Karlekar, MV Tejendra Prasad, Dr. Rashmi M
University/Organization: Amrita Vishwa Vidyapeetham, Amritapuri, India
Dehazing involves finding the solution to an ill-posed problem. Improvement to the traditional prior
based dehazing method has been introduced here. Considering the pivotal role of an accurate
transmission map estimate in dehazed image restoration, a fusion based dehazing algorithm is
proposed, where refined transmission maps from different scales are fused to obtain an overall
transmission map estimate, with fewer block artifacts. Transmission map obtained from a fixed
patch size is prone to block artifacts. This exemplifies the need for obtaining dehazed results with
varying patch sizes, which is computationally inefficient. Fusion provides a solution to this problem
by operating upon different scales of input image while keeping patch size fixed. This in effect
amounts to obtaining results with different patch sizes. Fusion provides a viable alternative to
obtaining highquality transmission maps without compromising on computational complexity. At
each scale, transmission is estimated using both weighted dark channel(DCP) and bright channel
priors(BCP), effectively avoiding color distortions. Additionally, Weighted Dark channel and a
Cross-Bilateral filter for refinement, helps prevent halo artifacts. The model has been compared with
other state-ofthe-art models using both ground truth and non-ground truth metrics. Experiments
show that our method can effectively remove haze, while retaining the color of original scene

May 21st – 22nd, 2022
radiance, clearly evident in sky region. MATLAB has been used for simulation and analysis
purposes.
309. PHOTOLUMINESCENCE OF ER-DOPED KNN CERAMICS

Yash Yadav, Jai Prakash, Malay Garg, Renuka Bokolia
Department of Applied Physics, Delhi Technological University
The lead-unfastened ceramics K0.5Na0.5NbO3 doped with x mol% Er3+ ions were efficiently
artificial and their photoluminescence properties, down-conversion close to infrared (200–700 nm)
were systemically examined. Because of the widespread emission levels, Er-doped K0.5Na0.5NbO3
ceramics are the hopeful applicants for optoelectronic applications.
310. ONLINE PROCTORING SYSTEM USING IMAGE PROCESSING AND

MACHINE LEARNING
Dr.K.Gopalakrishnan AP/ECE, N.Dhiyaneshwaran , P.Yugesh
Department of ECE,
Mepco Schlenk Engineering College, Sivakasi,
Tamilnadu, India.
Different forms of remote education and massive online open courses are gaining reputation. The
skill to proctor the online examination is a main essential factor for the scalability for promoting the
students for next stage education. Existing manual monitoring is the most approaching method in
education either by visually monitoring or by physically accompanying test takers to the
examination centre and monitoring them. Learner’s identity verification and proctoring of online
examinations is one of the main challenges in online learning systems. The migration and
implementation of the online exam have been accelerated during the pandemic COVID-19. So the
existing systems need a safe mechanism to authenticate and proctor online students. In this paper,
we propose a system for providing the solution for authentication and proctor by using the different
biometric technologies. The performance of the proposed system is evaluated by using real time
videos with different scenario. The experimental results provide an improved accuracy than existing
research works.
311. WEATHER FORECASTING USING MACHINE LEARNING

Ajeet Kumar Choubey School of Computing Science and
Engineering Galgotias University Greater Noida , India
Awanish Kumar
School of Computing Science and Engineering
Galgotias University Greater Noida , India
Mr Dhruv Kumar
School of Computing Science and Engineering
Galgotias University Greater Noida , India

May 21st – 22nd, 2022
Climate plays a vital role in many important areas of production, e.g. farming. Climate changes with
high charging these days, that is why ancient weather forecasts approaches are declining and
continue to be troublesome. Weather forecasting is an attempt by meteorologists to predict future
weather events as well as possible weather forecasts.. Earlier Forecasting was done on the basis of
observed patterns of events is also known as pattern recognition. For exit was seen that on a
particular day if the sunset was red it was considered to be fair weather. However not all of this
predictions prove to be authentic. Here in this Project we are making a prediction model using
Machine Learning Algorithm and the algorithm used is supervised machine learning algorithm. In
this we have collected previous data from Kaggle website and predicted the future temperature on
the basis of temperature, humidity and Pressure The application used in making this model is
Jupyter Notebook, Kaggle website and the Framework used are Pandas, sklearn, Numpy, Matplotlib.
These are the latest software and the applications used in weather Forecasting.
By using this model, we will be easily be able to forecast the weather which is very essential in
today’s day to day life for planning for any vacation, in Air traffic, Agriculture, Military application,
etc.
312. DRIVER DROWSINESS DETECTOR

Rohit Parihar
Computer Science & Engineering Galgotias University Greater Noida, India
Arham Khan
Computer Science & Engineering Galgotias University Greater Noida, India
This document is a review of the research done and the revolution made in the field of engineering
work to create a driver's system to identify drowsiness to prevent the driver from driving. Reроrt
developed results and solutions for the limited use of various techniques presented in the program.
Where project implementation provides a real-world perspective on how the system works and what
changes can be made to improve the use of the entire system.
313. AUGMENTED ANALYTICS: A REVIEW OF CURRENT TRENDS AND

FUTURE CHALLENGES
Ajai Gopal Bhartariya1, Dr. S.K. Singh2, Dr. Ajay Kumar Bharti3
1Research Scholar, Amity Institute of Information Technology, Amity University, Uttar Pradesh, Lucknow,
India
2Professor,Amity Institute of Information Technology, Amity University, Uttar Pradesh, Lucknow, India
3Professor, School of Computer Application, Babu Banarasi Das University, Lucknow, India
Analytics is a field of study that focuses on the application of scientific methods to the
transformation of raw data into actionable insights. The capacity to properly employ analytics at
many strategic and operational levels has never been more important than it is now, given the huge
growth in the volume of available data. In this article, we'll talk about how to make the leap from
analytics to AI. AI is shown as a fourth period of analytical emphasis in this discussion. According
to our observations, the analytics are now performed manually, necessitating the use of highly
skilled human resources. Augmented Analytics, on the other hand, refers to the integration of
artificial intelligence with analytics in order to automate every step in an automated workflow.
Analytics frameworks offer the door to a whole new phase of located analytics, in which users are
immersed in a physical world and presented with immersive analyses of local contextual data.
Augmented analytics is important since it allows for several perspectives on analytics to be
May 21st – 22nd, 2022
considered rather than just one. Academics, researchers and industry players may now use this work
as a springboard for a new analytics revolution.
314. AUTOGRICULTURE – AN HI-TECH AGRICULTURAL FRAMEWORK

Goutham. G1, Hrithick Krishna. KR2
Undergraduate Student, Department of Electronics and Communication Engineering, St. Joseph’s College of
Engineering, Chennai, India.
Undergraduate Student, Department of Electronics and Communication Engineering, St. Joseph’s College of
Engineering, Chennai, India.
This work deals with the mechanism in which the crops which are grown in the agricultural land is
automated by using an Arduino microcontroller and various sensors which can detect the moisture
content, level of water in the land, smoke sensing to help the farmers during adverse conditions. A
GSM (Global System for Mobile communication) module is used to send an alert message
immediately to the landlord and the concerned department of Government if any unavoidable natural
disaster affects the agricultural land. A GPS (Global Positioning System) module interfaced with the
mobile application (FARMER ASSISTANT) is also employed in order to find the availability of
seeds, fertilizers and other materials which are essential for agricultural purpose.
315. DETECTION OF PHISHING WEBSITES USING MACHINE LEARNING

Sourabh Singh, Pranav Siddhant, Amarjyot Kaur
School of Computing, DIT University, Dehradun
The Internet has become an integral part of our lives. However, it also provides opportunities for
anonymous malicious activity such as identity theft. Identity theft is a form of cybercrime that
attempts to obtain sensitive or confidential information from users by creating a fake website that
mimics an official website. Phishing scams use a variety of strategies such as link fraud, filter
escape, website fraud, encryption, redirection, and social engineering. Scammers try to manipulate
their victims through social media or to create fake websites to steal information such as account ID,
username, individual passwords, and organizations. The most common way is to set up a spoofing
web page that mimics an official website. Major resources are being squandered by organizations
monitoring and recovering from cybersecurity attacks by cybercriminals who gain access to
sensitive and sensitive user data. The purpose of this study is to develop these self-defense strategies
using a variety of website classification methods. Specifically, we will be developing a system that
uses machine learning methods to classify websites based on their URLs.
316. COMPUTER SECURITY RISK ANALYSIS IN DISTANCE LEARNING

Dr. Kirti Shukla
School Of Computer Science And Engineering
Galgotias University, Greater Noida, India
Sreesti
May 21st – 22nd, 2022
Sahil Srivastava
The Covid-19 vastly affected the education sector. There was only one way to keep the study going
and that was to take the education completely online. Technologies including different learning
managements, video calling apps, cloud computing, whose value was yet to be used in the education
field, got its importance. They are now the most important part in the education sector. Anyone can
learn anything from anywhere and that makes these technologies the best part of E-learning during
pandemic till now and will continue. But with everything being online, there is a huge risk of cyber
attacks and leaking of important data, which happened with many higher Educational Institutes.
Illegal data transfer, infection and virus from malicious applications started harming the personal
data of the user. In this paper, we are discussing the various threats of keeping everything online and
also how we can improve the technologies and protect them. It may include updating systems,
checking for malware, encryption decryption processes during file transfers over the network and so
on. This paper will be based on security issues in different networking sites used for E-learning.
317. CUSTOMER CHURN PREDICTION - A COMPARISON BETWEEN GRADIENT

DESCENT AND NEURO-EVOLUTION
Nithin Balaaji R
IV Year, B.E(CSE), Dr. Mahalingam College of Engineering and Technology
Pollachi, India
Dr. A. Noble Mary Juliet
Associate Professor, CSE Department, Dr. Mahalingam College of Engineering and Technology
Pollachi, India
Dr. G. Anupriya
Professor, CSE Department, Dr. Mahalingam College of Engineering and Technology
Pollachi, India
Customer Churn prediction is one of the most prominent problems that a business would want to
solve. In this study, we develop two models, a backpropagation model and a Genetic Algorithm
based Neuro-Evolution model. Once these models are created and built, their accuracies are tested
and compared with each other. The Backpropagation model performs better than the Neuro-
Evolution model both by accuracy and by the amount of time taken in computation. However,
further research is required in Neuro-Evolution and in the change of the structure of the Neural
Network, inspired from NEAT, Neuro-Evolution of Augmenting Topologies, may lead to Neuro-
Evolution being a more competent replacement for traditional Backpropagation models.
318: HITECH AGRI - CROP RECOMMENDATION SYSTEM WITH EFFECTIVE USE

OF FERTILIZER
Dr. VIGILSON PREM M M.E. Ph.D.,
Department of CSE,
RMK College of Engineering
and Technology,
Tiruvallur, Tamilnadu
ESAKKI MUTHU BOOPATHI P,
May 21st – 22nd, 2022
CSE,
RMK College of Engineering
and Technology,
HEMANTH KUMAR R,
CSE,
RMK College Of Engineering
and Technology,
HERAMKUMAR S,
CSE,
and Technology,
MASHKOOR D,
CSE,
and Technology,
India being a horticulture country, its economy dominatingly relies upon farming yield development
and agro-industry items. Information Mining is an growing research area in crop yield investigation.
Yield expectation is a major issue in rural. Any farmer is keen on realizing how much yield he is
going to anticipate. Various related attributes such as location, pH value are need to be analyzed
before entering into planting crops. Also, the level of supplements like Nitrogen (N), Phosphorous
(P), and Potassium (K) Location is utilized. These variety of traits of information will be examined,
and the data is trained with different suitable machine learning algorithms for making a model. The
system accompanies a model to be exact and precise in foreseeing crop yield and convey to the end
client, the appropriate proposals about required manure proportion in light of barometrical and soil
boundaries of the land which upgrade to build the harvest yield and raise the farmer’s income.
319. BIOACTIVE PHYTOCHEMICALS: A PROMISING TREATMENT FOR CANCER

1Km sakshi, Navneeta Bharadvaja*
1Plant Biotechnology Laboratory, Department of Biotechnology, Delhi Technological University, Shahbad

Daulatpur, Main Bawana Road,Delhi 110042
*Corresponding Author –Assistant Professor, Plant Biotechnology Laboratory, Department of Biotechnology,

Delhi Technological University, Shahbad Daulatpur, Main Bawana Road, Delhi 110042,
Cancer is a significant public health concern both in industrialized and developing countries. In
spite of the wide variety of treatments available, cancer is often uncomfortable due to its side effects
and ineffective due to resistance to traditional anti-cancer medications or radiation therapy. Natural
substances have been found to be effective against cancer in recent years. Chemicals derived from
plants are especially appealing because they are bioavailable, safe, usually without side effects, and,
most importantly, are inexpensive. Plant metabolites called flavonoids come from polyphenolic
chemicals found in nature and have intriguing biological properties. As well as being abundant in
May 21st – 22nd, 2022
our diets, such as green leaves, fruits, red wine, and tea vegetables due to their non-toxicity and
broad range of benefits in biological activities, flavonoids have been extensively studied for their
health benefits. In addition to having anti-cancerous properties at the beginning of the cancer,
flavonoids can also act as a preventative measure as the disease progresses. An overview of
flavonoids' potential role in cancer prevention, as well as their therapeutic application to the
prevention of cancer, is presented in this review.
320. COMPASSION: ONE MORE STEP AGRICULTURE MANAGEMENT SYSTEM
Subham Singh
Dept. – Computer Science Engineering, Organization – Galgotias University, Greater Noida, Gautam Buddha
Nagar, Uttar Pradesh
Sudhir Kumar Chaurasiya
Assistant Professor Ms. Kiran Singh
The name Agro farm stands for smart farming. Agro farm is an exemplary farm management web
application. This site helps farmers sell their produce online and provides best farming practices.
Thus, it serves a wider market and supports you to go beyond the local market. This helps
wholesalers and retailers buy products from more farmers. This allows wholesalers and retailers to
expand their business. Online shopping for fertilizers, pesticides, machinery and tools, and more.
Helps farmers track agricultural production through features such as virtual calendars and weather.
forecasting, etc. And allow them to hire workers, which in turn will help farm workers find small
jobs with a job profile on their website. In general, “agricultural farm management systems” provide
users with the concept of virtual trading of agricultural products.
321. DESIGN OF HEART RATE EQUIPMENT BASED ON IOT WHILE RIDING

VEHICLE
1Mr. S. Sridhar Raj, 2S.Jes Evans Daniel, 3B.Sanjay Kumar
1(Asst. Professor (Sr. G), 2,3UG Scholar, Department of Electronics and Communication Engineering. Mepco
Schlenk Engineering College, Sivakasi, Virudhunagar District, TN
A Beat Per Minute (BPM) is a health indicator that measures how fast a person's heart beats in a
minute or less. Athletic pursuits such as cycling and vehicle riding might benefit from a variety of
equipment. The Pulse sensors and the LM35 sensor are used to determine heart rate and temperature
of the person. The Arduino acts as the brain of this device. As a voltage source, Arduino is powered
by a 9V DC battery. The voltage source will supply voltage to the pulse sensor and the LM35
temperature sensor through Arduino. These two sensors will be powered by Arduino at 5V. Arduino
is popular because it can sense the surroundings and control lights, motors, and other actuators by
accepting input from a variety of sensors. To programme the board's microcontroller, the Arduino
programming language is used. When all of the components are working properly, the information
will be shown on the LCD on the handlebar bike. The temperature and heart rate are shown on the
screen. ThingSpeak then sends the information to the hospital and family members through email
using the ESP8266 wifi module.

May 21st – 22nd, 2022
322. SEWAGE BLOCK DETECTION SYSTEM

1Alaguvathana P, 2Subash R, 3Vineeth S, 4 Rahul Kumar R
1,2,3,4 Department of Information Technology
1,2,3,4 Sri Krishna College of Technology, Coimbatore, Tamil Nadu, India
The main objective of this project is to safeguard the front line workers from the dreadful toxic
gas emitting from the sewage by alerting them that the toxic gas present in it through SMS alert.
Additionally in our project the sewage block is also detected with the help of ultrasonic sensors and
report is automatically sent to the corporation office where the sewage block has been occurred
through the SMS alert. In the SMS alert message the unique URL is sent stating that where the
sewage block has been occurred in the Google Map. By pin pointing the sewage block location in
google map we can easily determine the location of sewage blockage and send the rescue team for
sewage block removal easily.
323. DATA SCIENCE AND AI/ML IN CONSTRAINING TAX EVASION

Eshika Bijpuria Aman Shukla Hradesh Kumar(Proff.)
Galgotias University Galgotias University Galgotias University
Greater Noida, India Greater Noida, India Greater Noida, India
Innovation in data science and artificial intelligence (Ai) could bring momentous changes in the
way how we handle and process large quantity of data. This paper assesses the necessity of data
science and AI/ML in constraining tax evaders. The key question addressed in the paper is: how data
mining technique and machine learning algorithms are used to detect tax evasion. This paper
summarizes that with further enhancement and modification of the existing technologies and
outspread usage among all the countries, there will be no tax gap or budget impaired economies.
324. A REVIEW AND PROPOSAL OF AN INNOVATIVE APPROACH FOR

ELEARNING PLATFORM
Hariom Tripathi
School of Computer Science & Engg
Greater Noida, U.P. , INDIA
Mohd Faizan Alam
School of Computer Science & Engg
Greater Noida, U.P. , INDIA
Technology is rapidly evolving, and the same is true in the world of education. People are
increasingly turning to online learning platforms and resources. Many universities, like Galgotias
University in Greater Noida, have embraced open source e-learning platforms like Moodle. E-
learning is being implemented in a number of educational departments and institutions. We hope to
encourage undergraduates, postgraduates, and even school pupils about the interactive
characteristics of ELearning through this work. E-learning is currently transforming the educational

May 21st – 22nd, 2022
system, resulting in highly sought-after subjects for researchers. This E-learning contains a variety
of research papers. It is currently one of the most popular research topics among academics.
325. CONVERSION OF SIGN LANGUAGE INTO TEXT AND SPEECH USING DEEP
LEARNING
D.Ranjani
Assistant Professor
Coimbatore, India
Thilipan S R
Coimbatore, India
Sriram Kalidoss V
Coimbatore , India
Srinivasan B
Coimbatore , India
Communication is very vital for every individual in this world. For people who have physical
difficulties it is hard to communicate with others. This makes them feel inferior as they already
possess physical disability. Therefore, there is a need to develop a system that will improve the
communication between differently abled persons and with normal individuals. This makes them
confident and makes their life better. However, the development of such systems encounters many
difficulties related to quick and proper recognition of a sign language and to convert this content into
an audio so that a blind individual can hear it. One of the technical possibilities to implement sign
language conversion is to capture 3D hand gestures and process it. This project recognizes the
differently abled individual's hand gestures and converts them into sound. This audio enables
blind individuals to seek their information from dumb individuals.
326. THERMAL DEGRADATION OF POLYVINYL CHLORIDE MODIFIED BY

HYDROXYAPATITE NANO PARTICLES
K. Maheswaran (1), Vishnuprasad Varma D(2) Dr. M.K Marichelvam(3)

1. Assistant professor (Sl Gr), Dept. of Mechanical Engineering, Mepco Schlenk Engineering College
2. M.E. Industrial Safety student, Dept. of Mechanical Engineering, Mepco Schlenk Engineering College
3. Assistant professor (Sr Gr), Dept. of Mechanical Engineering, Mepco Schlenk Engineering College
As the development of technology escalated drastically, enterprises of our country came into the
stage of all-round risk management. Various polymers that are used in our day to day applications
are creating a huge threat to the environment. In this work, a modified polymer of Poly Vinyl
Chloride (PVC) by the application of Hydroxyapatite Nano particles is developed and studied.
Hydroxyapatite is a naturally available mineral, which are abundantly present in the bones of human
beings and animals. This work investigates the extent of thermal degradation is quantified using

May 21st – 22nd, 2022
Thermogravimetry Analysis (TGA) and Differential Scanning Calorimeter. Characterisation tests for
the developed polymer is carried out using Scanning Electron Microscope imaging technique.
327. BRAIN DISEASE CLASSIFICATION & BRAIN TUMOR ESTIMATION USING

CNN
P.Alaguvathana 1 , Suma Sira Jacob 2 , GR . Aiswarya 3 , C.Dinesh 4 , A.Navin 5
1234 Department of Information Technology,Sri krishna College of Technology.
This procedure is useful to denoise, extract and detect tumours of MRI images. MRI images help
physicians investigate and diagnose brain disorders or tumours. The objective of this analysis is to
give the radiologist and the physician a second diagnostic perspective. The complexity of the
Magnetic Resonance (MR) images is easier to overcome. The computer's MRI image is analyzed in
the work. The data are used for real-time analysis. Basic preprocessing with various noise reducing
filters is achieved. The picture is segmented without noise and the characteristics removed. Features
are extracted by wavelet transformation. The transform wavelet is particularly equipped for MRI
image extraction compared to other approaches. The classifier which uses classified binary tree
labels has the features. The classification mechanism is contrasted with conventional approaches.
328. QUANTIFICATION OF MODIFIER IN POLYMER MODIFIED BITUMINOUS

BINDER USING FOURIER TRANSFORM INFRARED RADIATION (FTIR)
SPECTROMETER AN INVESTIGATIVE STUDY TO CURB ADULTERATION IN
MODIFIED BITUMEN
Navitha Govindaraju1 Dr. Godekere Visveswaraiah Ashok2 Dr. Veena Jawali3
1Inspector of Legal Metrology, Department of Legal Metrology, Government of Karnataka, India.

2Assistant Professor in Post Graduate Department of Physics, Government College (Autonomous) Mandya,
Karnataka, India
3Professor in Department of Mathematics, BMS College of Engineering, Bangalore, Karnataka, India
India has second largest road network in the world, as of August 2021 a total of 6,215,797
kilometers of road are developed, where 90 percent of the total length comprises flexible pavement
with bituminous layers. In order to enhance the performance of flexible pavement, modified
bituminous binders are preferred on surface course of flexible pavement. Since modified bituminous
binders are expensive and exhibit less shelf life, it is observed that, the adulteration of modified
bituminous binder’s increases due to above reasons, which reduces the performance of binder,
leading to decline in pavement performance during service. Department of Legal Metrology at India
is enforcing authority for any measurements and quantification. Currently quality of Polymer
Modified Bituminous (PMB) binder is measured by tests as prescribed by Bureau of Indian
Standards (BIS) such as Penetration Test, Softening Point Test, Ductility Test, Viscosity Test,
Elastic Recovery Test, Loss on Heating and Separation Test, by performing above tests one fails to
quantify the modifier quantity in PMB binder it only gives a subjective assessment of classification
of different grade of PMB’s. Current investigation study establishes a procedure using FTIR
spectrometer to quantify the modifier and address the issue arising due to adulteration of PMB
May 21st – 22nd, 2022
binder. Three (03) major polymer modifying agents viz., Crumb Rubber Modifier (CRM), Styrene
Butadiene Styrene (SBS) and Gilsonite or also known as Natural Asphalt were used to modify the
virgin bituminous binder of Viscosity Grade (VG)-10 and later subject to tests as per BIS and FTIR.
Similarly virgin VG-10 binder was blended using adulterants viz., Charcoal Powder and Print Ink
Powder and later subject to test as per BIS and FTIR. In both case the modified bituminous binder
when tested in accordance to BIS test procedures results weren’t able to distinguish between
modifying agents with that of adulterants. Further FTIR test results clearly indicated the difference
among modifying agents with that of adulterants.
329. CONTACTLESS COVID TESTING BOOTH WITH PATIENT TRACKING AND

MONITORING
Dr. K Prabhu, K Dinesh, M Ranjani, M Suhitha
Department of Electronics and Instrumentation Engineering
Perundurai-6380601
The world has turned upside down in the past two years as a result of the outbreak of COVID-19
pandemic. As the number of test centers have also significantly increased. Test taking is not a easy
job in terms of the risk factor involved in it. The laboratory professionals have faced a growing list
of challenges and one among them is the safety of test takers who are in direct contact with the
patients. The project aims to automate the covid testing booth process, track and monitor the patients
who are in home quarantine eon their temperature, oxygen and heartbeat levels using a health
monitor smart band.
330. CROP YIELD PREDICTION IN SMART FARMING USING MACHINE

LEARNING
Devansh Kumar
SRM Institute of Science and technology
Chennai, India
S. Godfrey Winster
SRM Institute of Science and technology
Chennai, India
The agriculture plays a significant role in the advancement of our country’s economy. Climate and
environmental changes has become an authentic threat inside the cultivation field. Machine learning
(ML) is a huge strategy for achieving sensible and suitable responses for this disadvantage. Reap
Yield Prediction incorporates predicting yield of the yield from Unquestionable conceivable
information like environment limit, soil limit and striking yield. We will use Random Forest, Deep
Neural Network and Support Vector Machine to get a better philosophy with high accuracy. This
paper directs the expected yields to maintain current knowledge of the Artificial Intelligence
algorithm. The actual knowledge of the country data collection was used in modelling and in
addition the models were tested with samples. Consideration can help farmers by anticipating a

May 21st – 22nd, 2022
collective harvest before creating a cultivating field. For algorithms SVM and LR accuracy is
hampered due to overfitting and clustering which is caused because the data must be handled
thoroughly and it is difficult to do so since duplicates could be present also data must be accurate,
and for Neural networks the accuracy is low because of fewer number of denses (at least 10-15 are
needed for a better accuracy) and the accuracy is lower than the other algorithms mainly because of
using 40005000 datasets (80k-100k are needed for a higher accuracy in DNN).
331. APPOINT US
Parth Pandey,Bishal ale magar
Web programs have helped in streamlining some of the obligations we carry out on a every day
basis, and feature made our lives easier. These programs are broadly used to help us in overcoming
issues with scholar studying and scheduling appointments. In the past, those appointment methods
had been accomplished manually and, due to this, there had been many times of overbooking or
forgetting to cancel an appointment that can loose up the distance to time table some other in its
place. To dispose of human mistakess because of placing appointments manually, an internet
software may be advanced to make the scheduling technique easier. Also, given the busy lives that a
lot of us lead today, a web appointments college group of workers members. The cause of this paper
is to expand and examine a web lecturer appointment gadget for students’ tasks, in which all
methods of appointments are verified. Most components of appointment control, along with
reservations, confirmations and cancellations, are managed automatically. Our on-line appointment
control gadget for students' tasks need to be capable of facilitate the assignment of reserving an
appointment with teachers. Nations defines net software as any software which could use an internet
browser as a client. The software may be simplecontrol gadget inside a college makes best
experience because it frees up treasured time, now no longer best for students, however additionally
for teachers and which include a message panel or a traveller sign-in ee-e book on a website, or
complicated like a phrase processor or a spreadsheet.
332. SENTIMENTAL CLASSIFICATION ON ONLINE CUSTOMER REVIEWS BY

USING NATURAL LANGUAGE PROCESSING
Ms. S Sajini1, Bangaru Sri Sai Varun2, Bondalapati Bhargav3, Kuppala Siva Chakravarthy4,
1 Asst. Professor, 2,3,4 Student, Department of Computer Science Engineering, SRM Institute of Science and
Technology,
Ramapuram, Chennai-89 Tamil Nadu, India
Recently the methods of Data mining and machine learning are widely used in the e- commerce
companies. These methods have given better results across the fields. In today’s world people are
showing more interest in buying products online rather than offline stores. But we are hearing many
complaints regarding the quality of these products from many customers. So, in this project we are
doing opinion mining on online customer reviews by using Natural Language Processing (NLP).
Opinion mining is also called as sentimental classification is used to determine and categorize
opinions about the product. Opinion mining can extract the amount of positivity and negativity that
the product holds. We collect dataset of about 20000 hotel reviews from Trip advisory and perform
our project with better accuracy.

May 21st – 22nd, 2022
333. AN IOT-ENABLED FLOOD INTENSITY PREDICTION VIA ENSEMBLE

MACHINE CODE MODEL
Dr Suma Sira Jacob
Associate Professor - Department of Information Technology
Coimbatore , India
Varun Gokul T
Coimbatore , India
Sujeeth K K
Coimbatore , India
Padma Sankar S R
Coimbatore , India
Stream flooding is a trademark wonder that can devastatingly influence human life likewise,
monetary incidents. There have been various systems in considering stream flooding; in any case,
lacking agreement and confined data about flooding conditions defeat the improvement of balance
and control measures for this trademark wonder. This includes one more technique for the
assumption for water level in relationship with flood earnestness using the gathering model. Our
philosophy involves the latest headways in the Internet of Things (IoT) and AI for the robotized
assessment of flood data that might be useful to prevent devastating occasions. Investigation results
show that gathering learning gives a more strong gadget to expect flood earnestness levels.
334. CONTROL OF FUEL CELL BASED SUPER-LIFT LUO CONVERTER

Kalagotla Chenchireddy, D. Navya, D Saiganesh, K.Pavan Kumar, K Sandeep
Department of Electrical And Electronic Engineering
Teegala Krishna Reddy Engineering College
This paper presents control of fuel cell based SuperliftLuo converter. Superlift converter is the
advanced version of boost converter that will produce a high output voltage. It provide higher
efficiency and nearly zero output voltage and current ripples. The voltage transfer gains of the
circuits increase in geometrical series (power series) stage by d=stage. Therefore, the super-lift
technique has attracted much worldwide attention in recent decades Simulation of the whole system
of fuel cell based superlift model with its control circuit is done in MATLAB/Simulink. Based on
the simulations, the superlift converter shows a significantly higher output voltagecompared with the
voltage of fuel cell.

May 21st – 22nd, 2022
335. PREDICTION OF STUDENT PERFORMANCE BASED ON TEXT MINING

USING NEURAL NETWORK – A REVIEW
Dr.S.Veena1, S. Lakshmanan2, B.Lokesh3, R.Ranjith4
1Associate Professor, 2,3,4Student
To see how important and influential a student's history, community service, and performance are in
forecasting student performance. Demonstrate the flow of a student performance prediction structure
that describes the categories of learners so that you can accurately articulate your learning ability.
The paper discusses the various algorithms used for predicting the students' performance.
336. A CHAOTIC SALP SWARM FEATURE SELECTION ALGORITHM FOR

ORANGE AND TOMATO PLANT LEAF DISEASE DETECTION
Mithra C, A. Suhasini
Research Scholar,Dept. of Computer Science and Engineering,Annamalai University,Annamalai Nagar, Tamil
Nadu,India.
Professor,Dept.of Computer Science and Engineering,Annamalai University, Annamalai Nagar, Tamil
Nadu,India.
Plants are considered essential because they provide energy to humans. Plant diseases can affect the
leaf at any point between planting and harvesting, leading to substantial losses in crop yield and
market value. In agricultural production, a leaf disease detection system is crucial. This process
necessitates a significant number of manpower as well as a thorough understanding of plant
diseases. Deep learning and machine learning methods that classify data based on a specified set are
required to determine the presence of illnesses in plant leaves. The Chaotic Salp Swarm algorithm
(CSSA) is used to detect disease in orange and tomato leaves, followed by the Bi-directional Long
Short-Term Memory (Bi-LSTM) technique for classification. In studies, we used the Bi-LSTM
architecture to detect disease in tomato and orange leaves. We used the Plant Village dataset of
defected and healthy plant leaves to train a deep learning network to determine the type of leaves.
The trained model is expected to achieve a test accuracy of 97 percent.
337. BIRD WATCHER USING DEEP LEARNING PLATFORM

UDIT RAJ SINGH,Anil Bameta,Dr. John A.
UG Student, Dept. of CS Engg
GALGOTIAS UNIVERSITY
Bird watching is a common interest but identifying their species requires the help of bird literature.
Provision bird watchers a useful tool to compliment the beauty of birds, we have developed an
indepth reading platform to help users identify 27 species of birds are found in Taiwan using a
mobile application called Internet of Birds (IoB). Pictures of birds read by a convolutional neural
network (CNN) to localize prominent features in images. First, we build and build boundaries an
interesting region to refine the shapes and colors of the granularity of the object and later to measure
the distribution of birds types. Then, the override connection method was used to line up the results
of previous and current layers to improve element background. Finally, we used the softmax
function to find opportunities to spread bird features. Educated Bird features restrictions were used
May 21st – 22nd, 2022
to identify images uploaded by mobile users. The proposed CNN model with a jump
Communication achieved a high accuracy of 99.00% compared to 93.98% from CNN and 89.00%
from SVM training photos. In terms of test data, average sensitivity, specificity, and accuracy were
93.79%, 96.11%, and 95.37%, respectively.
338. FACE RECOGNITION BASED ATTENDANCE MANAGEMENT SYSTEM

Piyush Kumar Jha, Sakshi Pandey
Department Of Computer Science and Engineering
Galgotias University, Greater Noida
On this virtual era, face popularity device performs an essential labour in almost every Face
recognition is a type of biometric that is widely employed in this industry. It is very useful for
safety, authentication, and identity, and is continuously giving hundreds of benefits.
However, face recognition has low percentage of being 100% accurate when compared with other
biometrics such as iris popularity and fingerprint, but then also face recognition method is being
extensively used because of Its approach is contactless, dependable, and non-invasive.
Moreover, the face popularity machines It may also be used to track attendance in schools, faculties,
workplaces, and so on. This device goals to construct a class attendance machine that makes use of
the idea of face popularity as the current guide attendance machine is time-eating and cumbersome
to preserve. There could be a probability of fake attendance and students writing attendance of their
fellow mates. As a result, such a powerful tool is required. The system grows. The four processes of
this device are database development, face detection, face popularity, and attendance updating. The
database is developed using the many images taken of the kids. The Haar-Cascade classifier and the
neighbourhood Binary sample Histogram set of rules were used to accomplish face detection and
reputation, respectively. Live videos of the classroom or the location are used to detect and identify
facial emotions. This enrollment could well be obliged to disclose to that same university only at
closing of something like the session.
339. CHALLENGES IN HYBRID APPLICATION DEVELOPMENT: A CASE STUDY

Mohd Ajaz, Ansh Gandhrav
School of Computer Science and Engineering, Galgotias University, Greater Noida, India
The smart phone is widely recognized as a watershed moment in human history, having transformed
the direction of human life in several ways. Android has surpassed iOS as the most popular software
platform for smartphones, while iOS is considered the safest operating system due to its closed
environment. Android OS is a free and open-source operating system, while iOS is a closed-source
operating system. On Android, a variety of programmes (apps) are available for entertainment and
pleasure, and users may easily install any third-party application; on iOS, this is not the case.
Consumers have turned away from the conventional manner of acquiring insurance and toward the
online market place as a result of severe competition among online policy dealers. The following
article discusses the concept and development of Policy Invest, a cross-platform programme that
enables people to shop for insurance policies with a single touch. Additionally, this article highlights
the several challenges that developers face while creating hybrid apps.
May 21st – 22nd, 2022
340. DESIGN AND DEVELOPMENT OF MULTI-PURPOSE 3D MODELLED

PORTABLE POWER BANK USING 18650 LITHIUM-ION BATTERY
Mr P Gowrishankar 1
Assistant professor,
Department of EEE,
Erode, India.
Dr P S Raghavendran2
Associate professor,
Department of EEE,
Erode, India.
P Madhumathi3
UG Scholar,
Department of EEE,
Erode, India.
V Soundhara Velan4
UG Scholar,
Department of EEE,
Erode, India.
S Vinothkumar5
UG Scholar,
Department of EEE,
Kongu Engineering College, Erode, India.
M S Naveen6
UG Scholar,
Department of EEE,
Kongu Engineering College, Erode, India.
The main objective of this project is to develop a 3D model for economical and low-priced answer
within the Hardware kind for the portable battery backup which will used to charge the electrical
appliances while travelling or we are lagging for the power for charging the appliances during the
power cut. The power bank’s main features are the ability to power at least 4 devices at once, and
having a removable battery pack that consists of 15 Lithium ion 18650 batteries which can be
replaced for extra capacity. Furthermore, 18650 batteries are easily found in things like old laptop
batteries, So it is easy to get them for free or very cheap. And if you manage to do that, the total cost
of the components goes down to Rs.16000 which is two times cheaper than anything commercially
close. For output, It has a 150W wall port that can power everything from Lights and Fans to Mini
Fridge or even TVs. All this power comes in a size slightly larger than a standard water bottle so It
can be easily carried around.

May 21st – 22nd, 2022
341. EFFECTS OF POCESS PARAMETERS THAT AFFECT THE

MICROSTRUCTURE OF ALUMINIUM ALLOY: A REVIEW
ABEY VISHNU NARAYANA
(Assistant professor department of Mechanical Engineering SNIT ADOOR )
Alen T Joy
The aim of the current study to investigate the effect of process parameters that affect the
microstructure of aluminium alloy. As the current concerns for energy efficiency and fuel economy
is increasing in the transportation industries. The industries are seeking forward to weight reduction
of the automobile components. The fusion welding of these alloys is difficult because cracks and
defects are formed in the melting and solidification phases of fusion welding. Friction stir welding is
a solidstate welding which eliminate the defect formation during the aluminium alloy welding.
Friction stir welding (FSW) is a novel green manufacturing technique due to its energy efficiency
and environmental friendliness. This solid state joining process involves a rotating tool consisting of
a shoulder and/or a probe. The shoulder applies a downward pressure to the work piece surface,
constrains the plasticised material around the probe, generates heat through the friction and causes
plastic deformation in a relatively thin layer under the bottom surface of the shoulder. The rotating
probe mainly drags along, plasticises, and mixes the adjacent material in the stir zone, creating a
joint without fusion. Aluminium alloys, such as the 2xxx, 5xxx, 6xxx and 7xxx alloy series are the
most commonly utilised materials in composite fabrication. Addition of various reinforcements such
as fly ash, TiC, SiC, Al2O3, TiO2, B4C etc. In recent years, Al based composite materials have
gained significance in aerospace, automotive and structural applications due to their enhanced
mechanical properties and good stability at high temperature. This review paper is an attempt to
describe the effect of process parameters that affect the microstructure of aluminium alloy. In this
paper discussed about the axial force, rotational speed and welding speed of FSW of aluminium
alloy.
342. IoT-BASED SYSTEM FOR DETECTING AND FORECASTING INDUSTRIAL

FIRE
Ms.S.Anitha Jebamani , Ragavi.M, Nivetha.K
Dept of Information Technology
Chennai, Tamilnadu
For the most part, fires happen in industries due to human carelessness, fault happening in machines
and changes in the environmental elements. These industry flames might cause human casualties and
significant misfortune. Thus, Industry fires should be recognized right on time to forestall more
noteworthy harm. It is our goal to foster an IOT framework that can distinguish and foresee the
industry fires and send their directions to authorities, which empower firefighting staff to smother
the fire at its earliest stages.Utilizing an assortment of IOT sensors, a systems administration
framework has been created.The Arduino UNO we used to gather this endeavor is spoken with a
temperature sensor, a smoke sensor, and a ringer. We will use GSM to give the last SMS to the
client through the given number in the diversion program.Temperature sensor which is used to
connote the temperature High and Low that will be displayed in the LCD Display. Fire sensor which
exhibits the fire range and expecting the fire range is high, the forest fire will be recognized in the
May 21st – 22nd, 2022
LCD show and if the fire range is low, the boondocks fire will not be perceived. Whenever a fire
occurs, the structure thus resources and alerts the client by sending a caution message.
343. FRICTION STIR WELDING OF ALUMINIUM ALLOY PLATES WITH

THREADED PIN TOOL: A REVIEW
Vinesh Rajendran Nair, Ajin J
Aluminium alloys have remarkably contributed to lightweight and long-lasting constructions seen in
high precision applications (marine industry, aircraft industries). This paper discusses the friction
stir welding of different aluminium plates with a threaded pin tool. FSW of aluminium alloys has the
potential to hold good mechanical and metallurgical properties. The tools used in friction stir
welding of aluminium plates have significant effects. Different types of tools are used, such as
cylindrical, square, triangular etc. The effects of threaded pin profile tools on the aluminium alloys.
Various researchers found that more mechanical and microstructural properties that given by the
thread pin tool than the other shapes. Rotational speed, welding speed, axial force, and tilt angle are
the main parameters considered for the FSW process.
344. FILTERING THEORY FOR QUANTUM STRING IN NOISE

1st Nikita Singh
Electronics and Communication Engineering Department
NSUT
Dwarka Sector-3, Delhi
2nd Harish Parthasarathy
Statistical Signal Processing Lab , ECE Department , NSUT
This article presents an application of the Extended Kalman Filter to estimate the electron wave
function in a helium atom i.e., a two-electron atom, based on noisy measurements of quantum
averages of observables like position, momentum, and spin.
The two-electron joint wave function is replaced by one electron wavefunction where dynamics is
described by the approximated Hartree-Fock equation obtained from the Variational principle. The
cost of replacing a two-electron wavefunction with one-electron wave function is that the dynamics
now becomes non-linear within the dynamics of each single electron wavefunction that takes place
in the field generated by the smeared second electron wave function. Replacing two-electron wave
function dynamics with one electron dynamics reduces the complexity and is further only an
approximate evolution equation. Non-linearity of the dynamics implies that we cannot use the KF,
rather we must use the EKF. The advantage of estimating the electronic wave function from the
measurement of quantum averages of specific observable is that knowing the wave function enables
us to calculate the averages of other observables that we have not measured.
345. NEW MOS SQUARE-LAW TRANSCONDUCTOR

Varun Sharma
May 21st – 22nd, 2022
Electronics and Communication Engineering Department

NSUT
Raj Senani
Analog Signal Processing Research Lab
ECE Department
NSUT
This article demonstrates a new circuit employing CFOAs and a pair of identical MOSFETs
operating in weak inversion region (triode region), to realise a square-law transconductor. The
proposed circuit facilitates an easy design to realise the intended function without any
approximations and constraints. The viability of the presented circuit has been shown by SPICE
simulation results.
346. DATA SECURITY AND PRIVACY PROTECTION FOR CLOUD STORAGE

Albinus Sanjeevan D
Dept. of Information Technology
Hindustan Institute of Technology and science,Chennai, India.
Dr. A. Kumaresan
Dept. of Information Technology
Hindustan Institute of Technology and science,Chennai, India.
With the quick advancement of unstructured data, the improvement of distributed computing
innovation and distributed storage innovation. Cloud administrations don't give cloud data or
distributed storage anyplace on the planet. This security strategy is generally founded on Internet
innovation. There are numerous ways of safeguarding your security to forestall data being put away
in the cloud. We give a three-level distributed storage framework. The gave framework can take full
advantage of distributed storage and safeguard the privacy of data. Here we utilize the Hash-
Solomon code to diminish data in various areas. On the off chance that the news area is missing, we
have lost news data. In such manner, we can utilize a calculation based can to safeguard data and
exhibit the security and abilities of our framework. Likewise, as indicated by the estimation
technique, this calculation can ascertain the dispersion rate put away in mists, haze, and different
nearby machines. Programming Services (SaaS): Customers discharge applications to facilitated
areas that can be gotten to online from an assortment of clients. application clients. The client
doesn't oversee or control the fundamental foundation, with the exception of the setup of the UI.
Google Apps and Microsoft Office 365 are instances of SaaS.
347. DRIVER’S DROWSINESS DETECTION USING DEEP LEARNING

KUNAL GAUTAM,HARSH GANGWAR,NIMISH YADAV,KSHITIZ CHAUDHARY,MUDIT SAXENA,
SHAHBAZ ALAM
ABES ENGINEERING COLLEGE, GHAZIABAD
As per the Automobile industry is growing every year, new technologies and innovations are also
playing a major role in its growth, whether it is about automotive performance or new features added
to it. The Driver Drowsiness Detection System using Deep Learning is a feature of Active Safety
that is installed on vehicles to prevent accidents that occur due to driver drowsiness, drowsiness or
May 21st – 22nd, 2022
inability to concentrate on the road due to fatigue. Various studies around the world have shown that
more than 40% of all road accidents are caused by these problems and in order to reduce these
accidents while driving, various new methods have been developed and introduced in modern
vehicles to deal with the effects of such accidents. Many tech bullies have worked really hard for a
long time to create a system that detects and prevents losses that occur due to driver fatigue or
drowsiness. This paper is based on an example of finding a solution to this problem. The purpose of
this paper is to help create an automated driver safety system that can detect the physical condition
of the driver and take appropriate steps to maintain safety. The Driver Drowsiness Detection System
here uses Raspberry Pi and Deep Learning and is a union of sensor, buzzer, cameras, etc. The
system is designed to detect any actions that the driver may be taking, and the system can detect
when a driver is awakened or not by the blinking of an eye. This paper explains the need of
Drowsiness Detection system, its benefits and the latest technology used how this leads to the
reduction in accidents Finally, the paper concludes on the future scope of Drowsiness Detection
system.
348. HEART DISEASE PREDICTION AND DETECTION SYSTEM

Ambalika Jaiswal a, Gunjan Purwar b, C. Pretty Diana Cyril c
a Student , Department of Computer Science and Engineering, SRM Institute of Science and Technology,
SRM Nagar, Kattankulathur, 603203, TN, India.
b Student , Department of Computer Science and Engineering, SRM Institute of Science and Technology,
SRM Nagar, Kattankulathur, 603203, TN, India.
c Assistant Professor of Department of Computer Science and Engineering, College of Engineering and
Technology, SRM Institute of Science and Technology, SRM Nagar, Kattankulathur, 603203, TN, India
Over recent years, one of the toughest issues in the medical field, has been predicting heart disease.
In the contemporary era, every minute, one person dies of heart disease. Data science is critical in
the processing of large volumes of data. There is a lot of data in the realm of healthcare. Because it
is difficult to predict cardiac illness, to avoid hazards and to notify the patient ahead of time, it is
vital to automate the operation. Machine Learning is a rapidly developing Artificial Intelligence (AI)
technology that provides a variety of Heart Disease algorithms. Many researchers contributed to this
work by employing a variety of methodologies and algorithms to foresee cardiac illness, such as
Decision Tree (DT), Logistic Regression, Support Vector Machine (SVM), KNN (K Nearest
Neighbor), Random Forest, and Neural networks. The random Forest technique outperforms other
machine learning algorithms in terms of accuracy, according to the results of the tests. The major
goal of this project was to develop and implement an effective illness prediction model and integrate
it with a user friendly website so that users could easily access it and be benefitted from it. Each
module of the project has been done efficiently.
349. BLIND IMAGE QUALITY ASSESSMENT WITH IMAGE DENOISING: A

SURVEY
Padmapriya.R and Jeyasekar.A
SRMIST
The quality of an image is most important at every stage of its processing. Images captured with
recent smart devices such as digital cameras or smart mobile phones can be affected by various types
of noise, which degrades image quality. The evaluation and elimination of noise or distortion in a
May 21st – 22nd, 2022
picture is viewed as significant as, the evaluation of the image's quality. This study presents a survey
of available techniques and algorithms for denoising images and determining image quality. This
paper studies the types of noise or distortions, techniques, parameters used by the algorithm, various
metrics used to assess the quality and the performance of the algorithm. Some of the important
filters, databases in the area of image quality assessment and denoising are discussed in this paper.
350. HOUSEHOLD MANAGEMENT WEBSITE FOR RURAL AREAS

Vaibhav Joshi,Abhishek Bagwari,
CSE Department
GALGOTIAS UNIVERSITY
Greater Noida, Uttar Pradesh, India
In this paper we are discussing about a website we made in which we will provide various household
services like plumbing and carpentry in cities and mainly in rural areas. Usually finding a carpenter
or a plumber or an electrician is a hectic task in rural areas so to solve this problem we are going to
create this platform in which we can find suitable workers for the job online. Through this platform
we will provide an option to sign with us and get a stable job profile and get employed. It is cost
effective as we will provide them with fixed hourly wages which is paid online that will lead to no
chance of fraud and customer satisfaction. Customers can also rate the workers accordingly.
The main highlights of this project are:
• It will provide services in rural areas.
• Wages are fixed, reasonable and paid online there is transparency.
• There is also premium membership option for people.
• It will provide employment -For people interested in learning these skills will be given
proper tutorials.
• It will provide contract-based services for people to be carefree.
351. SPEAKER IDENTIFICATION FROM VOICE

Yashu Kedia Nagadevi S.
Department Of Computing Technologies Department Of Computing Technologies
SRM IST Kattankulathur SRM IST Kattankulathur
Kattankulathur Kattankulathur
The speech signal conveys information about the identity of the speaker.The area of speaker
identification is concerned with extracting the identity of the person speaking the utterance.As
understanding how to recognise complex, unstructured and high-dimensional voice/speech/audio
data is one of the greatest challenges of our time.In this project, we proposed an idea for identifying
a speaker by machine on the basis of his/her voice with the help of deep learning approach.Some of
the paradigms or algorithms that will be used are MLP, CNN, RNN, LSTM and GRU.
May 21st – 22nd, 2022
352. BIOCATALYTIC POTENTIAL OF AMYLASE AND PROTEASE ENZYME ON

WASTE TANNERY EFFLUENT
Dr.Pagidi Aruna , H. Mohamed Salim, E. Rubhashree
Chemical engineering Department, Kongu Engineering College, Perundurai, Erode-638060 .
Water pollution comes from many sources including pesticides and fertilizers that get washed away
from farms, untreated industrial wastewater, and domestic wastewater. Tannery effluent contains
hazardous chemicals that are difficult to degrade. Tannery effluents are likewise found to be high in
nitrogen, particularly organic nitrogen, but low in phosphorus. Besides organic and nitrogen
molecules, tannery wastewaters contain sulphide and chromium, both of which have strong
antibacterial properties. The enzymatic approach is a biological treatment that tries to ameliorate the
polluting character of water effluent by supplying enzymes to minimize or remove water toxicity
through pollutant decomposition. We compare the efficacy of amylase and protease enzymes in the
treatment of COD elimination. The parameters (BOD, COD, TSS, TDS, pH) of the wastewater are
analyzed before and after the treatment process to determine the treatment efficiency of the
enzymes. To optimize, the process RSM is used by varying three parameters pH, Amount of the
Enzyme and Time on three levels each. The results provide the optimum conditions and the best
enzyme for the waste tannery treatment process.
353. MANAGERIAL EFFICIENCY WITH PRICE SENSITIVE DEMAND

Uttam Kumar Khedlekar
Department of Mathematics and Statistics
Dr. Harisingh Gour Vishwavidyalaya Sagar, M.P., India
(A Central University)
With effect of pricing strategy considered, an order level inventory model is proposed. The reducing
of price strategy not only maximizes the retailer’s profit but also improves the service towards the
product. In diminishing market retailer could reduce the selling price and generate the excess
demand to increase more revenue. A numerical example is demonstrated and which showed that in
every business setup, there will be an optimal number of price settings obtained. It is found that the
price reducing strategy outperform the static pricing policy.
354. EFFECTIVE TAMIL HANDWRITTEN CHARACTER RECOGNITION USING

DEEP LEARNING TECHNIQUE
V.Jayanthi
Department of Networking and Communications
Chennai, India
S.Thenmalar
Department of Networking and Communications

May 21st – 22nd, 2022
. SRM Institute of Science and Technology

Chennai, India
Computer vision applications for real-world problems include handwriting recognition. Nowadays
handwritten character recognition is commonly used for reading postal addresses, bank check
amounts, and forms. A few works have also been done for recognizing the Tamil character. The
Tamil character syllable is a segment of a Tamil word that has a single vowel sound and is spoken as
a single unit. Recently, Tamil character recognition has been implemented through deep learning
techniques. The hyperparameters used in the deep learning techniques signify the training algorithms
behavior, which has an impact on model performance in predicting the characters. One of the major
challenges in training the model is overfitting the data. This has to be overcome by choosing the
values of hyperparameters appropriately. However, the proposed system uses the bayesian
optimization algorithm for tuning the values of the hyperparameters. In addition to the
hyperparameter values, the proposed system classifies the given Tamil character using a
convolutional neural network. The proposed system performance is assessed using tuned
hyperparameters and the standard accuracy metric. The proposed system produces an accuracy of
80.3% when compared to the existing Tamil character recognition system.
355. ADAPTIVE AND INTEGRATED TEXT MINING FOR ASSESSING MASSIVE

OPEN ONLINE COURSES (MOOC) USING VECTOR SPACE MODEL
Dr. R Ramya,
SRM Institute of Science and Technology, Ramapuram Campus, Chennai, Tamil Nadu, India.
Roopak N,
V Adithya,
Anirudh Krishnan V,
There has been a surge in the popularity of MOOC (Massive Open Online Course) courses in recent
years. People independent of age group use this form of learning to augment their knowledge in their
required field and have a large collection of courses to choose from. Though the variety of courses
and the number of courses to choose from is high this availability of a large number of courses for a
particular topic also poses a problem to the learners on which course to choose from. This platform
also provides a possibility for the learners to share their reviews which helps other learners decide
whether to select the course or not based on the reviews of the previous learners. Research has been
done to evaluate the courses based on the reviews but as the number of reviews increases the
existing model and some research methods might give ambiguous results. This paper proposes a
novel approach that utilizes text-based reviews to score the courses by incorporating ANN and
archives an accuracy of 90 %.

May 21st – 22nd, 2022
356. FACE RECOGNITION BASED ATTENDANCE MONITORING SYSTEM

Dr. V. Geetha1, M.Jishnuvardhan reddy2, M.Venkata karthik3 Dr. C.K.Gomathy4
2 Student,Dept. of CSE, SCSVMV(Deemed to be University), Kancheepuram ,TamilNadu, India
3 Student,Dept. of CSE, SCSVMV(Deemed to be University), Kancheepuram ,TamilNadu, India
In this digital era, face recognition system plays a significant role in nearly each sector. Face
recognition is one in all the principally used bioscience. It will used for security, authentication,
identification, and has more blessings. Despite of getting low accuracy when put next to iris
recognition and fingerprint recognition, it's being wide used because of its contactless and
noninvasive method. what is more, face recognition system also can be used for attending marking
in colleges, colleges, offices, etc.This method aims to make a category attending system that uses the
construct of face recognition as existing manual attending system is time consuming and
cumbersome to take care of. And there could also be possibilities of proxy attending. Thus, the
requirement for this method will increase.This method consists of 4 phases- info creation, face
detection, face recognition, attending change. info is formed bythe pictures of thestudents in school.
Face detection and recognition is performed exploitation Haar-Cascade classifier and native Binary
Pattern bar chart algorithmic rule severally. Faces square measure detected and recognized from live
streaming video of the room.attending are going to be mail-clad to the various college at the top of
the session.
357. AN APPROACH TO ANALYSES AND PROTECT DATA ON UNTRUSTED

CLOUD NETWORK
Dr.V. Geetha1, G. Srinivas Reddy2. Dr.C.K.Gomathy3
1,3.Assistant professor in C.S.E Department SCSVMV.
2. Student in B.E Computer Science and Engineering SCSVMV.
The cloud server and the data privacy of users is necessary to encrypt the data before outsourcing the
cloud aiming to realize secure keyword search over encrypted data against, We find a compromised
method by into the block chain into the cloud storage accepted at provided in a private way, which
cannot be seen as a true. Moreover, the cloud server is thought to be begin by pointing out the
importance of storing the data in a public chain. We introduce a system that blockchain technology
to provide a secure distributed data storage with keyword search service The System allows the
client to upload they’re in content to cloud nodes and ensure data availability using cryptographic
encrypted form distributes the data techniques we introduce a system that leverages blockchain
technology to provide a secure distributed data storage with keyword search service. Trustworthy
Keyword Search scheme over Encrypted data without any third party Furthermore, blockchain
technologies and hash functions are used to enable payment fairness of search fees without
introducing any third party even if the user or the cloud is malicious. Our security analysis and
performance evaluation indicate that TKSE is secure and efficient and it is suitable for cloud
computing.
358. FAKE JOB FORECAST USING DATA MINING TECHNIQUES

Dr.C.K. Gomathy 1, Y.JEEVANKUMARREDDY2,T.suneel 3, Dr.V.Geetha4
1,4 Assistant Professor, Dept. Of CSE, SCSVMV (Deemed to be University), Kanchipuram, TamilNadu, India

May 21st – 22nd, 2022
2, 3 Student, Dept. Of CSE, SCSVMV (Deemed to be University), Kanchipuram, TamilNadu, India
In recent years, due to advancement in modern technology and social communication, advertising
new job posts has become very common issue in the present world. So, fake job posting prediction
task is going to be a great concern for all. Like many other classification tasks, fake job posing
prediction leaves a lot of challenges to face. The proposed framework consists of two main modules:
fake job detection module operating in real-time mode and model update module operating in batch
mode. The fake jobdetection module consists of four lightweight detectors: 1) blacklisted domain
detector to label posts containing blacklisted URLs; 2) near-duplicate detector to label posts that are
nearduplicates of confidently pre labeled posts; 3) reliable detector to label posts that are posted by
trusted users and that do not contain words; and 4) multiclassifier-based detector labels the
remaining posts.
359. CYBER ATTACK DETECTION SYSTEM

Dr.V.Geetha1, Dr C.K.Gomathy2
1,2 Assistant Professor, Dept. Of CSE, SCSVMV (Deemed to be University), Kanchipuram, TamilNadu, India
This letter develops a flexible machine learning detection method for cyberattacks in distribution
systems considering spatiotemporal patterns. The graph Laplacian recognizes spatiotemporal
patterns based on system-wide measurements. A flexible Bayes classifier (BC) trains spatiotemporal
practices, leading to a violation when cyberattacks occur. Cyberattacks can also be identified by
using flexible BCs online. The effect of the developed method is demonstrated through standard
IEEE 13- and 123node test feeders.This paper demonstrates a machine learning-based anomaly
detection (MLAD) method for load forecasting under cyber-attacks. The predicted load data is first
used to reconstruct the benchmark and scale data using k-means clustering. The naive Bayes
classification is then used to determine the specific attack template. Finally, dynamic programming
is utilized to calculate one cyberattack's occurrence and parameter on load forecasting data.
Differentiated from the widely-used SAX method, the effectiveness and robustness of the developed
MLAD method are verified by numerical simulations on the public load data. Some universal and
general lessons are shown as follows:(i) The detailed occurrence and parameter information of the
cyber attacks could be detected with considerably high accuracy using MLAD. (ii) The developed
MLAD method is strong to the load forecasting errors with a relatively high success ratio. (iii) The
improvement of load forecasts can significantly enhance the correctness of approximated attack
specifications. The expanded MLAD method is not able to find the specific derivations of the cyber
attacks, e.g., the broken topology of the forecasting model, falsified specifications of forecasting
methods, or attacked initialization techniques. In the time ahead, this research can also be further
developed by estimating the sensitivities of various load forecasting methods with the load data
tampered with cyber attacks.
360. MUSIC CLASSIFICATION MANAGEMENT SYSTEM

Dr.C.K. Gomathy 1, Dr.V.Geetha2
1, 2 Assistant Professor, Dept. Of CSE, SCSVMV (Deemed to be University), Kanchipuram, TamilNadu,
India
This study helps us to analyze the accuracy of the classification of music genre A Music genre is a
conventional category that identifies some piece of music as belonging to set of conventions. The
May 21st – 22nd, 2022
function of music labels in discovering and separating digital music resources is crucial. In the face
of a huge music database, relying on manual annotation to classify will consume a lot of cost and
time, which cannot meet the needs of the times. It can be divided into different genres in many
different ways. Categorizing music files according to their genre is a challenging task. Most of the
current music genre classification techniques use Machine Learning techniques. The same principles
are applied for Music Analysis too. ML techniques have proved to be quite successful in extracting
trends from the large pool of data. Extraction of note feature matrix, extraction of topic and segment
division based on note feature matrix, research and extraction of effective features based on segment
theme, and composition of feature sequence are all part of the process. Because of the shallow
structure of standard classification methods, it is difficult for classifiers to learn temporal and
semantic information about music. This research investigates recurrent neural networks (RNN) and
attention using the distinctive sequence of input MIDI segments. For analysis, we can extract the
features from the same sort of data. On the whole this analysis helps us to conclude that XG- Boost
Classifier is much better than other algorithms to work on. Music classification is taken into account
as an awfully challenging task thanks to selection and extraction of appropriate audio features into a
complete dataset.
361. SECURITY LEVEL DETECTION OF VARIOUS CRYPTOSYSTEMS USING

MACHINE LEARNING MODEL(ADRESS MISSING)
V Naman Khandelwal and Prof.Dr.Sanjay Kumar Singh.
Vellore Institute of Technology,Vellore
The safety of digital data has become a vital concern following recent advances in multimedia
system technology. Researchers tend to focus their efforts on modifying existing protocols to beat
the shortcomings of current security mechanisms. However, many projected encryption algorithms
have been evidenced to be insecure over the past decades, posing serious security risks to sensitive
information. using the foremost applicable encryption technique to guard against such attacks is
crucial, however, that rule is most applicable in every case and can depend upon the type of data to
be protected. However, testing potential cryptosystems one by one to seek out the most effective
option will take significant processing time. We provide a security layer detection approach for
image encryption algorithms by incorporating a support vector machine (SVM) for the quick and
accurate choice of applicable encryption algorithms.
362. DETECTION OF PARKINSON’S DISEASE USING DEEP LEARNING

ALGORITHMS
Dr.A.Christy Jeba Malar
Associate Professor Department of Information Technology
Sri Krishna College of Technology,
Coimbatore, India
Shivani Balaji Srivastava
Coimbatore, India.
Sri Kavi K

May 21st – 22nd, 2022
Technology Sri Krishna College of
Technology, Coimbatore, India.
Tinku Ram N
Coimbatore, India.
Parkinson’s is a progressive genetic neurodegenerative disease that affects people mostly in old age
but still might affect very few young people. This disease slowly eats up a part of the brain which is
responsible for body movement, resulting in a steady loss of muscle control of the entire body. For
example, frequent hand and leg tremors, body stiffness, loss of speech, bradykinesia, and dystonia.
The treatments available don’t entirely cure PD as there is no medication, but on the other side,
clinicians are trying to improve the patient’s lifetime. As the pattern recognition region of the brain
is related to PD, we are using a dataset with healthy and PD hand-drawn images from a small test
conducted. Here we have proposed a combination of deep learning algorithms of ANN and CNN
with a machine learning algorithm of Random Forest classifier to improve the accuracy rate by “74”
in finding out the person with PD. Hence, it is inferred that the expected results benefit clinicians in
identifying and treating patients with PD in an operative way.
363. TEA LEAF DISEASE DETECTION USING DEEP LEARNING

Mythili K
Assistant Professor
Sri Krishna College of Technology, Anna University.
Coimbatore, India.
Sharan R
Coimbatore, India.
Sindhiya M
Coimbatore, India.
Vasikaran R
Coimbatore, India.
In Agriculture, due to improper care, pests, diseases and undesirable weather conditions many of the
vegetation carried to the market are defected and it puts the consumers at risk of many
complications. Even though the agricultural items are good in quality, due to the country’s
unavailability of cold storage facilities and improper transportation facilities many of the goods are
getting defective and it’s reported in a recent study that the Indian agricultural market losses up to 16
percent of its products including cumulative losses. To overcome this situation, we take a dataset
containing healthy and diseased leaf images because it is difficult to cover an entire land and train
them and we have designed a prediction tool with the help of CNN which is a Deep learning

May 21st – 22nd, 2022
algorithm which predicts, detects and gives output of the quality of the agricultural plants and
vegetation.
364. REAL AND FAKE NEWS CLASSIFICATION USING NATURAL LANGUAGE

PROCESSING
Shivam Kumar
Dept. of CSE
SRMIST
Dr. C. SanthanaKrishnan
Dept. of CSE
SRMIST
The concept of Real and Fake news Classification and Detection is a domain which is still in the
initial-development stage as compared to other projects of similar kind in this domain.ML or
Machine Learning is a useful part of this project. The purpose of using these algorithms is to help
the users to understand the various difficult and unyielding problems and to build Smart Artificial
Intelligence and Machine Learning Systems to tackle problems for this concept. For the purpose of
this research, we have used the concept of NLP along with two popular Machine Learning
Algorithms for the purpose of the classification of real and fake news. They are Logistic Regression
and Decision Tree Classifier.
Other Algorithms like Random Forest, Support Vector Machine can also be used for this. The
purpose of the project that has been built here is not to simply perform classification of the news
articles as one cannot simply implement the ML algorithms and then predict whether the news is real
or not. No, what has been done here is a clear-cut implementation and a mix of Data Science Tools
as well as ML concepts for the classification as well as prediction of fake news. Various ML models
will be implemented here for the prediction of news. The process of the classification will focus on
using data science tools for pre-processing of the text and then using the results of the pre-processed
dataset to build a improved model for the project. The major obstacle which was tackled whilst
project was the lack of a properly processed dataset as well as a pre-defined model to differentiate
between the two categories of news as mentioned in the title of the paper. For simplicity’s sake,
some of the more commonly known ML algorithms and classifiers have been implemented on some
datasets that are available on the internet. The results, when the ML models were implemented on
the dataset, have been very encouraging and can prove to be very useful if any future work is done
on this project or in this particular domain.
365. A SURVEY OF CYBER CRIMES AND CYBER SECURITY

Dr K. Chithra1, M. Parijatham2
1Assistant Professor, Department of Computer Science, Shrimathi Devkunvar Nanalal Bhatt Vaishnav
College for Women, Tamil Nadu, India
2UG Student, Department of Computer Science, Shrimathi Devkunvar Nanalal Bhatt Vaishnav College for
The cyber world of the internet is a perilous area where unwitting people might fall prey to cyber
thieves. The objective of this study is to direct an exhaustive examination concerning these attacks to
May 21st – 22nd, 2022
raise information about the different kinds of attacks and how they work so that appropriate network
protection countermeasures can be taken. This helps to safeguards internet users from cybercrime
and cyberattacks, as well as to create a platform to protect themselves.
366. MODELLING A NOVEL LEARNING-BASED APPROACH FOR COVID-19

PREDICTION
R.MURUGANANTHAM, Professor
TKR College of Engineering & Technology
K.SAIRAM
TKR College of Engineering & Technology
B.SRILEKHYA
TKR college of Engineering & Technology
G.SAISUMITH
TKR College of Engineering & Technology.
This work proposes a deep learning framework for segmenting the infectious region of lung to attain
generic information regarding the severity of cell destruction in lung images. This work exploits the
efficacy of learning network for quantifying the image segmentation. Here, online available data
based on COVID-19 cases are considered for validation and to reduce the reliance of disease dataset
for further research. This work uses Convolutional Neural Network based segmentation process for
extraction of addition information regarding COVID-19 characteristics. The quantification of this
process is achieved by adopting 2DCNN model. The outcomes show that the anticipated deep
learning framework facilitates learning of image descriptors and enhances the accuracy of
segmentation process. This model shows addition image representation of cell destruction
characteristics using 2DCNN model. This research indicates the potential of deep learning model to
achieve higher prediction accuracy and to diagnosis COVID-19 at earlier stage. This assists
radiologists to make proper decision during the crucial time of disease affected person and to save
the life of human community.
367. IOT BASED REAL-TIME HEALTH CARE MONITORING SYSTEM FOR

COMPLEX EVENT PROCESSING
Angelina Geetha.G , Ravikumar.M , Antonidoss.A
Hindustan Institute Of Technology and Science,
Chennai -603103, India.
Most of the existing systems on healthcare monitoring detects the events after analysing the data
captured and stored. However, in real time situations, it is essential that we analyse the real time
stream data and assess the health status of the patients. Our approach focuses on capturing the
stream data of vital parameters of the patient, analyse and prompt the medical experts for suitable
May 21st – 22nd, 2022
assistance. Complex event processing uses an event-driven approach for the identification of
anomalies. This paper proposes IoT based Real-Time health care monitoring System for complex
event processing (RMSC). The proposed RMSC uses biosensors (Heart rate, blood oxygen,
Temperature) for collecting vital parameters. The collected parameters are sent as a live streaming to
the
Android app to identify the anomalies in the patient and generate the alert. From the experimental
results, it is proved that the proposed system considers the inter dependencies between the vital
parameters to assess the state of medical state of the patient.
368. DESIGN AND IMPLEMENTATION OF INTEGRATED e-VACCINATION
CHIP FOR COVID-19

Ms. B. Deepa Miss.S.Deepa
Assistant Professor Department of Information
Technology Sri Sairam Institute of Technology,
Sri Sairam Institute of Technology, Sai Leo Nagar, Tambaram West,
Sai Leo Nagar, Tambaram West, Chennai-600044
Chennai-600044, Tamil Nadu
Miss.S.Abinaya
Technology,
Sri Sairam Institute of Technology,
Sai Leo Nagar, Tambaram West,
Chennai-600044,
Tamil Nadu
Miss.V.Nithyasri
Technology,
Sri Sairam Institute of Technology
Sai Leo Nagar, Tambaram West,
Chennai-600044
Tamil Nadu
For Security checking of Vaccination based entry in the existing system corrupted people will show
some body's vaccination certificate and gain access to this crowdgathering places, these people
sometimes may carry covid-19 Virus and may lead to a massive spread, so in our proposed system
person who Vaccination is included Chip takes a position in front of the camera. security person he
will be having a unit which will shows the Photo Identity and the Vaccination details fetched from
the International Vaccine Page with live body temperature of the person in front of the security
check in.

May 21st – 22nd, 2022
369. CHARACTERISTIC ANALYSIS OF OFDM-OQAM SYSTEM IN BELOW DECK

SHIP ENVIRONMENT
Mrs.S.Sweetline Shamini 1 ,G.Nandhini 2 ,G.Archana 2 ,G.Kamaleshwari 2 1-
Assistant Professor Sri Sai Ram Institute Of Technology.
2-Electronics & Communication Engineering , Sri Sai Ram Institute Of Technology.
In today’s world ships plays vital role in transportation of people as well as heavy goods. So it is
necessary to ensure the better communication network in ships as it contains many decks and
compartments. Ships may face many dangerous situation such as fog, change in weather, fire on
ship. And also the communication is required for manual or automatic engine room and pilot on
boarding. During that time, it is important to have a good communication between the deck officer
and engine officer. But it is one of the most challenging thing in wireless communication. MIMO
OFDM is a wireless communication method that is being developed for 5G. When we combine
OFDM and MIMO, we get a significant boost in performance in terms of data transmission rate, bit
error rate (BER), Signal to Noise Ratio(SNR),and dependability. In this research, we compare
OFDM with OQAM using the OFDM structure to obtain good performance in terms of low BER
and SNR through the use of superior channel coding techniques and modulation schemes. To
decrease the effects of Inter Symbol Interference (ISI) generated by multiple reflections, orthogonal
frequency-division multiplexing (OFDM) has been proposed. Using numerous antennas for channel
diversity has also been found to improve communications reliability and capacity. The supplied PP-
SNR figures demonstrate the increased dependability that space-time coding can bring. The
estimation of Shannon channel capacity shows that spatial multiplexing can take use of multipath
scattering to boost performance and throughput. An enhanced strategy for wireless communication
employing intensity modulation scheme is examined in the suggested system, keeping modulation
complexity and reaching high power as significant considerations. The suggested OFDM
modulation technique incorporates the OQAM-OFDM structure and the BPSK structure, which have
been compared and developed to show the improvements on wireless communications.
370. A XTREME FEATURE-BASED BOOSTING MODEL FOR HEART DISEASE

PREDICTION USING FEATURE REPRESENTATION
R.MURUGANANTHAM
Professor
TKR College of Engineering and Technology
E.NAGARUPA
Hyderabad,Telangana
P.SAHITH
K.SHASHIKANTH
TKR College of Engineering and Technology.
Anomalies of the heart are one of the leading causes of mortality globally. Patients are typically
asymptomatic till the fatal event occurs and it is monitored by the trained staffs to detect a cardiac
abnormality. Due to the wide availability of digital Electrocardiograms (ECGs), there is an
May 21st – 22nd, 2022
increasing support of how machine learning (ML) may be used to identify such irregularities in
recent decades. New technological advancements have made it possible to use such data to create
models that can assess patterns in the incidence of heart beats and detect irregularities. We provide a
unique approach for extracting and predicting the ECG records. We use the Xtreme feature-based
boosting (XFboost) model to train models that get better accuracy, recall, precision and F1 Scores.
The proposed model establishes better trade-off compared to other approaches. The significance of
the anticipated model is extremely higher.
371. HEART RATE MONITORING, HEART ATTACK DETECTION AND

PREDICTION
Harish B
Undergraduate, Department ofElectronics andCommunication Engineering,
Sri Sairam EngineeringCollege, Chennai
Nithish P
Undergraduate, Department ofElectronics and Communication Engineering,
Sri Sairam Engineering College, Chennai
Akbar Ibrahim Musa M
Undergraduate, Department of Electronics and Communication Engineering,
Jayachandiran U
Faculty
Sri Sairam Engineering College,Chennai.
Lokkesh S
Undergraduate, Department of Electronics and Communication Engineering,
One of the most vital organs in the human body is the heart. It works as a pump to circulate oxygen
and blood throughout the body, ensuring that the body's functioning is maintained. A heartbeat is a
two-part pumping motion of the heart that lasts approximately a second. It is formed as a result of
the heart's contractions. The SA(Sino Atrial) node gives out an electrical signal when blood gathers
in the upper chambers, causing the atria to constrict. The blood is subsequently pushed through the
tricuspid and mitral valves by this contraction; this phase of the pumping mechanism is known as
diastole.When the ventricles are entirely filled with blood, the following phase begins. Electrical
impulses from the SA node reach the ventricles, causing them to contract. Heart-related health issues
are increasingly widespread in today's world.One of the leading causes of mortality in both men and
women is heart disease. Every year, it takes the lives of around 1 million people. Heart rate is an
important factor in how well the heart works. Heart rate monitoring is therefore essential in the
research of heart performance and, as a result, in preserving heart health.Using IoT and machine
learning, this research presents a heart rate monitoring detection system. The treatment of most
heartrelated disorders nowadays necessitates long-term monitoring as well as ongoing monitoring.In
this sense, IoT is highly valuable since it replaces traditional monitoring systems with a more
efficient method by giving important information about the patient's status to the doctor. Machine
learning, on the other hand, is the idea that a computer programme can learn and adapt to new data
without the need for human intervention. Machine learning is a branch of artificial intelligence
(AI) .that maintains a computer's built-in algorithms up to date despite global economic
fluctuations.Furthermore, the user or doctor present at the hospital can utilize the real-time
monitoring system to check the patient's heart rate in the serial monitor.

May 21st – 22nd, 2022
372. DEVELOPMENT OF NONINVASIVE CONTINUOUS PATIENT GLUCOSE

BOTTLE LEVEL AND HEALTH CARE MONITORING SYSTEM USING IOT
Sangeetha T1 ,Sangavi R2 , Shinega P A3 , Shobana G4
1Assistant Professor, Dept. of Information Technology, Sri Krishna College of Technology,
Coimbatore
2,3,4 B. Tech Student, Dept. of Information Technology, Sri Krishna College of Technology,
Coimbatore
Most of senior-matured people taking bed treatment are experiencing a deficiency of state of being,
an abrupt medical problem. Old individuals require constant medical services checking while in bed
therapy, which can fundamentally affect their wellbeing in the event that help isn't given promptly.
A remote correspondence framework is planned and produced for far off quiet observing in this
technique. It incorporates temperature sensors, a tension sensor, an engine unit, and a connection
point with the Node MCU for checking a patient's wellbeing. In light of sensor esteem low or high,
the glucose bottle was opened and shut by means of an associated engine unit for a patient glucose
level. The information from sensors is conveyed by means of Internet network based distributed
computing. Patient development activities, for example, low strain and high temperature, will be
saved in the information base and handled prior to being gotten back to the client organizes, web
server through Arduino UNO. The normal result is saved structure basic circumstances to make a
brief move to keep more established individuals from getting appropriate medical services. This IOT
(Internet of Thing) innovation gathers information from a patient and feeds into two separate points
of interaction that show the patient's boundaries and subtleties and dissect the basic information to
an electronic server, got to by the specialist through an application that has been used.
373. DESIGN AND DEVELOPMENT OF ANESTHESIA CONTROLLER

Sasireka.M1, Ilakkiya.S2, Santhosh Kumar.N3, Shershah Ahamed.N4
1 Assistant Professor (Selection Grade), 2,3,4 U.G Students, Department of Electronics and
Instrumentation Engineering, Kongu Engineering College, Perundurai,Erode India
The administration of high/low dose of anesthesia during surgery may cause severe effect to the
patient. To avoid such critical situation, the anesthetist administers few milliliters of anesthesia at
regular intervals of time to the patient. To overcome such problems, this project aims to design an
effective and automatically controlled anesthesia machine based on arduino. In the proposed
Automatic Anesthesia Controller System, anesthesia level is controlled by the parameters such as
temperature, heart beat rate and blood pressure. In the Automatic Anesthesia Controller, the arduino
control the anesthesia levels during surgery. Air pump blower is used to deliver anesthesia to the
patient. The anesthetist can set the value to administer the dosage of anesthesia in terms of milliliters
per hour. The anesthesia was administered based on the measured parameters. This project will play
a major role in the field of medicine and useful to the physicians during major surgery to provide the
desire amount of anesthesia to the patient.

May 21st – 22nd, 2022
374. TEXTURE BASED CLASSIFICATION OF LIVER CANCER USING

ULTRASOUND IMAGES
Dr.S.Gandhimathi @ Usha1, A.Gopikrishnan2, S.Keerthivasan3, R.Karpagamoorthy4
1AssociateProfessor, 2,3,4UG Students, Department of
Velammal College of Engineering and Technology, Madurai.
Liver cancer is one of the most malignant diseases and its diagnosis requires more computational
time. It can be minimized by applying a Machine learning algorithm for the diagnosis of cancer. The
existing machine learning technique uses only the color-based methods to classify images which is
not efficient. So, it is proposed to use texture-based classification for diagnosis. This proposed work
has three steps: pre-processing, feature extraction and classification. The input image is resized and
pre-processed by Gaussian filters. The features are extracted by applying Gray level co-occurrence
matrix (GLCM) and Local binary pattern (LBP) in the preprocessed image. The GLCM measures
the probability of the appearance of pairs of pixel values located at a distance in the image and is
used to convert a normal RGB image to a GLCM image. The Local Binary Pattern (LBP) is an
efficient texture operator which labels the pixels of an image by thresholding the neighborhood of
each pixel and considers the result as a binary number. The extracted features are classified by
multi-support vector machine (Multi SVM) and K-Nearest Neighbor (KNN) algorithms. The
Advantage of combining SVM with KNN is that SVM measures a large number of values whereas
KNN accurately measures point values. The results obtained from the proposed techniques achieved
high precision, accuracy, sensitivity and specificity than the existing method.
375. A LANGUAGE BASED SYSTEM ARCHITECTURE FOR MULTIPLANAR

CRYPTOSYSTEMS
Prathic R G
Dept. of ECE
student, final year, B.E.,
Chennai, Tamil Nadu,India.
Shanmuganathan S
Dept. of ECE
student, final year, B.E.,
Chennai,Tamil Nadu, India.
Jayachandiran U
Dept. of ECE
Assistant Professor
Sri Sairam Engineerinng College
Chennai,Tamil Nadu, India.
Dinesh Gokul S
Dept. of ECE student,
final year, B.E.,
Chennai,Tamil Nadu, India
One of the most promising branches of quantum resilient cryptography is lattice based cryptography
which offers versatility and efficiency. Discrete Gaussian samplers are a core building block in
May 21st – 22nd, 2022
latticebased cryptosystems, where optimized samplers are desirable both for high-speed and low-
area applications. The inherent structure of existing sampling methods makes lattice-based
cryptosystems vulnerable to side-channel attacks, such as timing analysis. The first comprehensive
evaluation of discrete Gaussian samplers in hardware is presented, targeting FPGA devices.
376. PROTECTION OF IOT USING BLOCKCHAIN: A SURVEY

Gulshan Nanda 1, Dr. Ajay Kaul 2
1,2 School of Computer Science & Engineering, Shri Mata Vaishno Devi University, J&K..
Internet of Things (IoT) has come up as an advancing technology in last few years. With rapid
adoption of IoT, the integration among different devices is now possible, resulting in increased
performance and accuracy. The growth of IoT has brought convenience and made human lives easy.
But with the convenience it has also brought various issues. Currently IoT systems work on
centralized cloud servers and these centralized servers have some drawbacks which results in
vulnerabilities. To mitigate the problem of centralization, various researchers have proposed
Blockchain based IoT Systems. In this paper we discussed the issues of Traditional Cloud Based IoT
systems. We reviewed various existing Blockchain based IoT systems and highlighted their
limitations and the problems focused.
377. CYBERBULLYING DETECTION IN SOCIAL MEDIA USING MACHINE

LEARNING
Ms. D Thamizhselvi
Chennai, India
Dhinesh N
Chennai, India
Mugesh G
Chennai, India
Social networking platforms have taken over our lives in today's world. The number of people who
use these platforms is steadily expanding. The person who uses these platforms becomes
increasingly exposed to the negative effects of utilizing social media as their use increases.
Cyberbullying is one of the biggest harmful repercussions of utilizing social media, among many
others. People are bullied online, which has a significant impact on their mental health. Detecting
cyberbullying on social media platforms is a difficult task, especially given the slangs that people
use on a regular basis, but this paper proposes a working implementation of an application that
detects cyberbullying across multiple social media platforms using data from Twitter, Facebook.
The goal of this paper is to apply Machine learning algorithm to detect cyberbullying in High
Accuracy.

May 21st – 22nd, 2022
378. PREPARATION AND CHARACTERIZATION OF PROSOPIS JULIFLORA STEM

ACTIVATED CARBON FOR METHYLENE BLUE DYE ADSORPTION
N.Vasiraja, R.Saravana Sathiya Prabhahar, A.Joshua
Department of Mechanical Engineering, Mepco Schlenk Engineering college, Sivakasi -626005, Tamilnadu,
India
This paper focuses on preparation, physical and adsorption characterization of activated carbon from
Precursor of Prosopis juliflora stem. In this view, carbon were prepared from the Precursor of
Prosopis juliflora stem and activated by the activating agent of ZnCl2 powder as a ratio of 1:1 at a
temperature of 800°C in muffle furnace about 1 hour. The physical adsorption characteristics of
Prosopis Juliflora Stem Activated Carbon (PJSAC) were determined by American Water Works
Association (AWWA) and Conseil EuropÌ©en des FÌ©dÌ©rations de lIndustrie Chimique (CEFIC -
European Chemical Industry Council) standards. The structure of PJSAC were characterized by
Scanning Electron Microscopy (SEM) and Fourier Transform Infrared Spectroscopy (FTIR) were
studied in detail. Batch adsorption of Methylene Blue (MB) dye on PJSAC as the function of effect
on PJSAC dosage and PJSAC contact time were carried out in details. The experimental results
shows that the PJSAC are found to be effective for the removal of Methylene Blue dye.
379. FINITE ELEMENT MODELING FOR RUTTING IN FLEXIBLE PAVEMENT

Praveen Kumar P #1, Dr. B.V. Kiran Kumar *2
#1 Assistant Professor, Government SKSJ Technological Institute, Bangalore,
*2Associate Professor, Government SKSJ Technological Institute, Bangalore
In India majority of roads are being constructed by flexible pavements. But the performance of the
pavement largely depends on quality and type of materials used, construction methodology adopted,
temperature, climatic conditions etc. Because of these variations in the mentioned parameters,
pavements undergo distress/ failure. The different types of failure in flexible pavement are Rutting,
Shoveling, Edge breaks, Cracks, Slippage etc. Rutting is a common phenomenon, which occurs in
flexible pavement surface due to overloading of vehicles and repeated application of wheel load.
Rutting is defined as channelized depression in the pavement surface along wheel path due to heavy
repetitive traffic load. Design of bituminous paving mixes greatly effects the performance of
pavements. In the present laboratory research work, Bituminous Concrete (BC) Gr-II is used in
wearing/ surface course and Dense Bituminous Macadam (DBM) Gr-II is used in binder course of
the layer, both constructed by VG-30 grade bitumen. Pavement layers are constructed/casted in an
indigenously designed, developed and fabricated equipment called Roller Compactor cum Rut
Analyzer (RCRA) and Rutting test was performed. The 3-D finite element model was created in the
ABAQUS for the pavement layers as per IRC 37-2018 for a CBR 10% value corresponding to
10msa design traffic. Lab/ experimental results obtained in RCRA equipment are successfully
validated with the results obtained in ABAQUS.
380. A TRIFOLD ELECTRONICS SURVEILLANCE SYSTEM USING IMAGE

PROCESSING
Mudit Saxena,Rishabh Singh,Shivani Singh, Dheeraj Singh
Electronics and Communication

May 21st – 22nd, 2022
ABES Engineering College (AKTU), Ghaziabad, India
In today’s time, records are greater exposed to net international; the role of security for connections
has additionally risen. In these day’s international, textual content based ordinarily passwords don't
seem to provide complete protection for the gadget that is likewise an outdated approach now.
Furthermore, users regularly implement easy password that hackers can wager easily. Hence, it
desires something for at ease and user-pleasant authentication schemes to conquer this problem. In
this challenge, we present trifold electronics surveillance system authentication scheme to triumph
over the trouble.
381. ANALYSIS OF PILED RING FOUNDATIONS IN MEDIUM DENSE SAND

USING EXPERIMENTAL APPROACH
First A. Girish Chand, Second B. Arvind Kumar, and Third C. Prince Karandeep Singh Sandhu
Department of Civil Engineering NIT Jalandhar
The piled ring foundations can be used in structures like over head storage reservoirs (OHSR), silos,
transmission towers, windmills and chimneys. They offer an economical foundation system because
of the less material consumption. The behaviour of these foundation systems in the sand has not
been checked yet. The aim of this study was to contribute toward the efficient designing of piled
rafts in the sand. So, in this study, an experimental setup was made which was used to get the load
settlement relation of the piled ring foundation system. Different models with the changes of pile
length and number of piles were tested and the effect of them on the parameters like load
improvement factor, ratio of settlement reduction, ratio of load sharing and axial stiffness was
comprehended. It was observed in this study that an increase in the length and number of piles
increases the load improvement factor, ratio of load sharing and axial stiffness while decreasing the
ratio of settlement reduction.
382. ROLLER COMPACTOR CUM RUT ANALYSER, AN EQUIPMENT FOR

PAVEMENT TESTING AND BITUMINOUS MIX DESIGN
Manjunatha S #1, Dr. B.V. Kiran Kumar *2
#1 Assistant Professor, Government SKSJ Technological Institute, Bangalore,
*2Associate Professor, Government SKSJ Technological Institute, Bangalore
One of the important factor which determines the performance of the bituminous pavement is
compaction. Several methods are being used for bituminous mix design all over the world. In India
Marshal Method of Mix design is widely adopted. However, to bring in actual field condition while
testing the bituminous mix an indigenously developed machine called Roller Compactor Cum Rut
Analyser” (RCRA) is being adopted. The advantage of this equipment is that the different layers of
the bituminous pavement can be constructed to simulate the actual field condition and can be
subjected the wheel load to find the Rutting and Fatigue performances. In this paper a brief note
about the importance of roads, roads in India, flexible pavement, failure of pavements, literature on
the rutting performance, and a brief introduction of the Roller Compactor cum Rut Analyser is
presented.

May 21st – 22nd, 2022
383. PROSTATE CANCER DETECTION USING DEEP LEARNING & TRADITIONAL

TECHNIQUES
S. Madumidha,R.B.K. Prithvi,D. Rixon Infanton
Sri Krishna College Of Technology
Prostate disease (PCa) is a serious sort of malignant growth and makes significant passing among
men due its poor demonstrative framework. The pictures got from patients with carcinoma comprise
of intricate and fundamental highlights that can't be extricated promptly by customary symptomatic
procedures. There have been only very few assessments that have portrayed the periods of partition
of prostate CT pictures. As needs be, in this article, we propose an Ensembled Transfer Learning
(ETL) design to arrange well, moderate and deficient isolated prostate CT pictures.
384. ROAD SIGNS CLASSIFICATION USING TRANSFER LEARNING

Sindhu.R
Department of Instrumentation and Control Engineering St. Joseph’s College of Engineering
Chennai,India
Mrs T. Abimala
Department of Instrumentation and Control Engineering St. Joseph’s College of Engineering
Chennai, India
Traffic signs are important in communicating information to drivers. Thus, comprehension of
traffic signs is essential for road safety and ignorance may result in road accidents.This study
presents a voice-assisted real-time traffic sign classification system which is capable of assisting
drivers. This system functions under two subsystems. Initially, the detection and recognition of the
traffic signs are carried out using Visual Geometry Group (VGG16) is a transfer learning approach
which has an advantage of classifying 1000 different categories. After recognizing the specific
traffic sign, it is narrated to the driver as a voice message using a text-to-speech engine. The
advantage of this system is that even if the driver misses a traffic sign, or does not look at the traffic
sign, or is unable to comprehend the sign, the system detects it and narrates it to the driver. A system
of this type is also important in the development of autonomous vehicles.
385. DETECTION OF DDOS ATTACK USING MACHINE LEARNING

1Ankush Sharma, 2Mr. Sudesh kumar
Shri Mata Vaishno Devi University, Katra, India
A distributed denial of service (DDoS) attempt is a sort of cyber-attack in which the perpetrator
attempts to starve a network/server of services by flooding the network/server with unnecessary
requests, rendering it unable to process valid requests. This attack is most dangerous attack in
security domain because this attack deny the availability of the service. With the false requests done
by hacker using the Bots army that makes system busy and service unavailable for genuine user.
Bots can be the devices that is connected by internet i.e Webcam, mobiles, laptops smart tv etc. As
these cyberattacks grow more common, it is becoming more difficult to identify them and protect
online businesses against them. Every day, the rate of cybercrime is increases. Many sorts of
May 21st – 22nd, 2022
cyberattacks are becoming increasingly widespread, including ransomware, phishing, and DOS /
DDOS attacks. Researchers are working hard to figure out how to defend against these attacks.
Machine Learning is contributing in the training of models to identify attacks and then prevent them
from causing maximum harm. Many machine learning algorithms for detecting DoS/DDoS threats
are covered in this research and also we study. We also discussed recent years big cyber attacks and
their peak traffic. We discussed data sets that widely used by researcher And there accuracy, And
finally we take past year research papers and analyze their approach used by detection of DDoS.
386. ADDITIVE MANUFACTURING ASPECTS FOR DESIGN AND DEVELOPMENT

OF GARBAGE PICKING MACHINE COMPONENTS FOR COMPACT DESIGN
Pallavi Upare1 , Abhay Kalatkar2, Dr. R.Shelke3
1. M.Tech CAD/CAM student, G H Raisoni College Of Engineering Nagpur, Digdoh Hills,CRPF Gate
No.3, Hingna road, Nagpur,440016,India
2. Assistan Professor, Mechanical Engineering Department, G H Raisoni College Of Engineering
Nagpur, Digdoh Hills,CRPF Gate No.3, Hingna road, Nagpur,440016,India.
3. Head Of Department, Mechanical Engineering Department, G H Raisoni College Of Engineering
Nagpur, Digdoh Hills,CRPF Gate No.3, Hingna road, Nagpur,440016,India
In India, garbage collecting is one of the major problem which covers all major cities and villages.
The government employ labor to gather all waste. But, as of now, there are of lot of safety and
health related problems. Our project mainly focuses on portable, durable machine to the common
problems of inefficient garbage collecting. This device is constructed on a steel chassis. This device
is manually operated (pushing or pulling). This equipment is made to collect Garbage from uneven
roads, rocky, hilly areas etc. Our machine includes a roller with brushes and many metal plates
connected to the axis. Which is 100% mechanically operated with excessive velocity rotates, which
pushes the strong wastes to get into conveyor belts, for the reason that conveyor belts. Some parts of
the machine have been made by additive manufacturing process in FDM machine.
387. MORPHOLOGICAL DESCRIPTORS BASED URBAN CHANGE DETECTION

USING REMOTE SENSING IMAGES
Dr.S.Gandhimathi @ Usha1, PT.Kishore2, S.VarunPrasath3, A.Yuvaraj4,
1Associate professor, 2,3,4 UG students Department of Electronics and communication Engineering,
Velammal college of Engineering and technology, Madurai
Change Detection describes the significant changes that occured between two satellite images of
same land acquired at different period of time. There is a lack of efficient automatic techniques for
discriminating between changed and unchanged pixels in the image, which affects the accuracy and
reliability. So, it is proposed an efficient algorithm for land cover classification and change detection
of multispectral images using CNN. In this work, QuickBird satellite image is preprocessed and then
segmented by binary descriptors using Hamming distance. Morphological operators are used to
eliminate the over segmented region. To improve the performance of the proposed method, the
robust feature extraction methods such as, Gabor, Zernike moment and Local Features are extracted
from the segmented images. Then, the above features are combined together to generate the overall
features set, it enhances the performance of the proposed method. Based on this feature set, the
May 21st – 22nd, 2022
image is classified using Convolutional Neural Networks (CNN) and their effectives is calculated by
Accuracy, Sensitivity, Specificity and Precision.
388. NUMERICAL INVESTIGATION OF FLOW OVER AN OSCILLATING

CIRCULAR CYLINDER AT LOW REYNOLDS NUMBER
Arun Narayanan. A1, Anand A. Kumar2, and Ajith Kumar. S*
Department of Mechanical Engineering, Amrita Vishwa Vidyapeetham, Amritapuri, India
A numerical investigation of flow over a transversely oscillating circular cylinder at a low Reynolds
number (75 ≤ Re ≤ 175) is conferred in this paper. The open source CFD software, OpenFOAM is
used to simulate the flow. The cylinder was made to oscillate transversely with frequency ratios
ranging from 0.8 to 1.2 for different Reynolds numbers. Here, the frequency ratio (fr) is the ratio of
external excitation frequency (fe) to that of Strouhal frequency (fo). For all cases, the oscillation
amplitude was constant and equal to 0.4. The Strouhal number and the aerodynamic coefficients
were calculated for different frequency ratios and Reynolds numbers.
389. INVESTIGATION OF FIBRE-BASED BAG FILTER COATED WITH METAL

OXIDES FOR DUST ADSORPTION
M.K.Marichelvam (a), S.Chella Goutham (a)
(a) Department of Mechanical Engineering, Mepco Schlenk Engineering College, Sivakasi- 625005,
Tamil Nadu, India
Now a days the production of Cement in industries gradually increasing dust / gas / particulate
matter. So, ecological contamination turning out to be more serious due to unfiltered dust. In the
control of residue contamination, the air filter and Electrostatic precipitator (ESP) assume a
significant part in controlling pollutants. Based on the comparison air filter shows the best result of
controlling small size particulates. Hence, it is essential to develop air filter material. Therefore,
developed an environment-friendly filter material that controls dust pollution and makes the filter
material an absolutely environment-friendly product was developed. In this project, the development
of new type of bag filter by using natural fibre such as sisal and hemp fibre as filtering medium was
discussed, which can be easily disposable when compared to synthetic polyester bags used in the
industry at present. The natural fibres are coated with zinc oxide and Iron oxide nanoparticles in
order to improve surface adsorbent ability other than synthetic materials. The fabrication of bag
filter was done by natural fibres to observe the performance.
390. NUMERICAL ANALYSIS OF FLUID FLOW OVER PLUNGING NACA0012

AIRFOIL AT LOW REYNOLDS NUMBER
Jaya Surya K1, Rego Hentry Shin H S2, and Ajith Kumar S3
Department of Mechanical Engineering, Amrita Vishwa Vidyapeetham, Amritapuri, India.
The primary objective of this computational research is to investigate on the effects of Reynolds
number, angle of attack, frequency, and amplitude of the plunging movement of NACA0012 airfoil
May 21st – 22nd, 2022
on aerodynamic performance over time for a constant angle of attack. This paper deals with the
laminar flow over the plunging NACA0012 airfoil at low Reynolds number 10000 with a plunging
amplitude ratio of 0.2c and with three different frequency ratios of 0.9, 1.0, and 1.1 . The influence
of the angle of attack (constant over time) was studied in the range of 0° to 6°. OpenFOAM an
opensource CFD software is used to simulate this problem computationally.
391. EFFICIENT TOPOLOGY DESIGN IN THE TIME EVOLVING AND ENERGY

HARVESTING
Dr.G.Saravanana ,V.K.Vigneshb ,M.Jayanthc ,T.Suresh
kumard aAssistant Professor ,Sri Sairam Institute of
Technology b,c,dSri Sairam Institute of Technology
The network's connectivity will fall down as soon as node energy is consumed, nodes in networks
are bound with limited power for their important operations. Due to power limits, node failures
induce system failures, reducing network end-to-end connection. Furthermore, node mobility and
congestion cause frequent link failures and packet losses, negatively impacting the protocol's QoS
performance. For efficient and routing and transmission for wireless networks, we used an effective
proposed technique called Hybrid Optimization System and denial of service in this work. The
protocol used in a wireless sensor network should be energy efficient in order to increase network
lifetime and provide reliable and efficient communication. After information theft, distributed denial
of service (DDoS) attacks are the most common cybercrime. DDoS TCP flood assaults can quickly
deplete the cloud's resources, take the majority of its bandwidth, and ruin an entire cloud project. As
a result, prompt detection and prevention of such assaults in cloud projects, particularly in eHealth
clouds, is critical. We offer a new classifier system for identifying and preventing DDoS TCP flood
attacks in public clouds (CS DDoS) in this work. By classifying incoming packets and making a
choice based on the classification findings, the proposed CS DDoS system provides a method for
safeguarding stored records. CS DDOS recognises and determines whether a packet is normal or
originates from an attacker during the detection phase.
392. CRYPTOCURRENCY PREDICTION: METHODS, TECHNIQUES AND TOOLS

Fardin Hassan
School of Computing Science & Engineering,
Galgotias University, Greater Noida, UP
Kapil Bhardwaj
School of Computing Science & Engineering,
Galgotias University, Greater Noida, UP
Bitcoin is a kind of Cryptocurrency and now is certainly one of form of investment within the stock
marketplace. inventory markets are encouraged by means of many chances of thing. And bitcoin is
one type of cryptocurrency that preserve rising in latest few years, and sometimes fall without
knowing have an effect on in the back of it, on stock market. because it’s fluctuations, there’s a need
Automated tool to prediction of bitcoin on inventory marketplace. This research study learns how to
make bitcoin stock marketplace prediction, approach to observe and prediction and also method and

May 21st – 22nd, 2022
equipment that used from preceding studies paper, many authors, web-sources and other to be had
sources. For the closing segment we discussed and make to provide true technique on future works
393. NUMERICAL STUDIES ON THE INFLUENCE OF SURFACE HEATING ON

THE FLOW CHARACTERISTICS OF A CONVENTIONAL TESLA VALVE
Pranav R. Menon1, Praveen Vijay2, Rahul V. Kaimal3 and Ajith Kumar S*.
Department of Mechanical Engineering, Amrita Vishwa Vidyapeetham, Amritapuri, India
In this paper we study in detail the effects surface heating on the fluid flow characteristics in a
conventional tesla valve. We simulate a two dimensional, laminar, incompressible flow through the
tesla valve using OpenFOAM. We use buoyantPimpleFoam solver for obtaining the necessary
resuts. The numerical study revealed that the surface heating on a vertically mounted tesla valve
resulted in an increase in the diodicity of the valve. A peak in diodicity is observed at a Reynolds
number of 30 and Richardson number of 0.8.
394. ENHANCING OBJECT DETECTION IN AUTONOMOUS DRIVING USING

YOLO-V4 AND DEHAZING ALGORITHMS
Dr.S.Gandhimathi @ Usha1, Dr.S.Vasuki2, Dr.P. Suveetha Dhanaselvam3, G P S Aakash Pawan 4, P M

Balaji5, J Jackson Solomon6, V Vanniarajan7
2Professor, 1,3Associate Professors, 4,5,6,7 UG Students,
Department of ECE, Velammal College of Engineering and Technology, Madurai.
In Autonomous vehicles, most of the decisions are taken by object detection algorithms, which
depends upon the images provided. Hence the quality of the image plays a vital role in decision
making. The quality of the image may be degraded due to various factors such as pollution, fog, mist
and smog. This leads to poor visibility, which not only degrades the perceptual image quality but it
also affects the performance of the object detection algorithms. This project proposes an algorithm
that performs dehazing using parameters such as Estimated Air-light, Transmission and boundary
constraints and provides a fog-less image that can be fed to object detection algorithms so as to
improve the efficiency of the decision made. In this project YOLO-v4 algorithm is used for object
detection because of its speed and accuracy. YOLO-v4 being a single staged object detector works
more accurate and faster than two staged detectors like R-CNN, Fast R-CNN. The main scope of this
proposed dehazed object detection is to improve the accuracy of the YOLOv4 object detection
algorithm during bad weather conditions and enhance the performance of autonomous driving.
395. SMART ATTENDANCE MANAGEMENT SYSTEM

S. Susila Sakthy 1, Nimish M S2, Vignesh B 3
1,2,3 Department of Information Technology, Sri Sairam Engineering College
Chennai, India

May 21st – 22nd, 2022
Face recognition is among the most productive image processing applications and has a pivotal role
in the technical field. Recognition of the human face is an active issue for authentication purposes
specifically in the context of attendance. An attendance system using face recognition is a procedure
of recognizing by using face biostatistics based on the mid-level camera / front camera and others.
The development of this system is aimed to accomplish the digitization of the traditional system of
taking attendance and maintaining pen-paper records. Present strategies for taking attendance are
tedious and time-consuming. Attendance records can beeasily manipulated by manual recording.
The traditional process of making attendance and presenting biometric systems is vulnerable to
proxies. This paper is therefore proposed to tackle all these problems. After face recognition
attendance reports will be generated and stored in excel format. The system is tested under various
conditions like illumination, head movements, the variation of distance between the cameras, and
matching the Longitude & Latitude. After vigorous testing overall complexity and accuracy are
calculated. The Proposed system proved to be an efficient and mobile device for taking attendance
without any time consumption and manual work. The system developed is cost- efficient and needs
less installation.
396. A NEW OTA-BASED SHADOW FILTER PROVIDING INDEPENDENT

ELECTRONIC CONTROLS OF Ω0 AND BANDWIDTH
Amisha Jha
ECE Department,NSUT
Sector -3 Dwarka,New Delhi, India
Raj Senani
ECE Department,NSUT
Sector-3 Dwarka,New Delhi, India
This paper presents a new shadow filter architecture based on OTAs and two capacitors both of
which are grounded which is feasible for the IC fabrication. The presented topology is based upon a
minimal realization of a state-variable biquad of order two (capable of realizing low-pass (LP),
highpass (HP) and band-pass (BP) filters from the same structure) using four OTAs and the two
capacitors (both of which are grounded) having basic core around which, as per the theory proposed
by Lakys and Fabre, two additional amplifiers possessing electronically-controllable gains AL and
AB that have been incorporated in the additional feedback paths. This results in the proposed OTA-
C shadow filter which enables independent electronic controls of ω0 and bandwidth both, through
the gains AL and AB respectively. The workability of the presented circuit has been confirmed with
bipolar OTAs LM3080 as well as with CMOS OTAs and some sample results of LTSPICE
simulations, using 180 nm CMOS technology parameters have been presented.
397. BLOOD VESSEL SEGMENTATION AND CLASSIFICATION IN FUNDUS

IMAGE
〖Kavitha P〗^1, 〖Gandhimathi alias Usha S〗^2, 〖Srinaath VR〗^3, 〖Sivaram Kumar M〗^4, 〖
Yoshuva P〗^5
(_ ^1)Associate Professor,(_ ^2)Associate Professor,(_ ^3,4,5)Final Year UG students Department of
Electronics and Communication Engineering,
Velammal College Of Engineering and Technology
May 21st – 22nd, 2022
Blood vessel segmentation of retinal images plays an important role in the diagnosis of eye diseases.
The segmentation of retinal blood vessels is critical for the diagnosis of diabetic retinopathy,
hypertension, and cardio-vascular disorders. This paper proposes a new approach for the automatic
segmentation of blood vessels in retinal images. The three major modules used are Segmentation,
Feature Extraction and Classification. In Segmentation process, the morphological operation plays a
major role in creating a multidimensional feature vector using the green channel intensity. Then the
segmented result undergoes Feature Extraction where the intensity of the blood vessel is enhanced
using the pattern and color feature from Convolutional Neural Network(CNN). CNN is a Deep
Learning system that can take an image as input and assign importance to various objects in the
picture, as well as differentiate between them. Using CNN the abnormality in the retinal region is
well identified.
398. DETECTION OF CYBERBULLYING IN SOCIAL MEDIA USING RECURRENT

NEURAL NETWORK ARCHITECTURES
FHA. Shibly
PhD Research Scholar,
Assam Don Bosco University, Guwahati, India
Senior Lecturer, South Eastern University of Sri Lanka
Dr. Uzzal Sharma,
Assistant Professor
Assam Don Bosco University
Guwahati, India
Dr. HMM. Naleer
Head, Dept. of Computer Sciences
Faculty of Applied Sciences
South Eastern University of Sri Lanka
Cyberbullying is a kind of bullying that takes place via social media with digital gadgets. It may
cause significant, long-lasting trauma and difficulties with mental and physical issues. This
condition may also result in difficulties in personal lives. Because of this, detecting and reporting
such offensive posts may help avoid the harmful repercussions of cyberbullying. Some studies are
available in this context, and it was found that deep learning algorithms are more efficient in
detecting abusive texts on social media. Some studies were carried out on deep learning algorithms.
The issue was not solved, and more studies are needed to find the most efficient model to control
this problem. It was observed from the literature survey that the performance of Recurrent Neural
Network (RNN) architectures is good. Therefore, the objective of this study was to find and develop
an efficient model to detect cyberbullying texts on social media. For this research, the Twitter
dataset was used, which was uploaded by Munki Albright in Kaggle. The dataset consists
suspicious, cyberbullying, hate and suicidal classifications. This research was focused on
cyberbullying texts. Sentiment analysis was implemented to classify the cyberbullying posts. For
that, sexism (Class 2), racism (Class 1) and either (Class 0) classification were analyzed.Researchers
applied the LSTM, GRU and Bidirectional LSTM architectures of RNN to perform the sentiment
analysis. All models gave somewhat similar results. When we consider f1, GRU shows a better F1
score for all the classes when compared to others (Class 0 was 95%, Class 1 – 70%, and Class 0 was
May 21st – 22nd, 2022
56%). GRU shows the best results for this data set compared to other models. Based on the
Accuracy, all three architectures were obtained at around 90%. However, GRU has outperformed the
other two models in accuracy as well. After that, researchers developed an ensemble model using all
three models. In the ensemble, each model has been given weight. Since the GRU is the best
performing model, it was given 0.4, and the other two models were assigned 0.3 each. The ensemble
model performed well in F1 measurement (57%) and accuracy (91.17%). When we compared the
previous works against the proposed model, our ensemble model obtained the highest values and
performed well in detecting cyberbullying in sentiment analysis.
399. APPLICATION OF DEEP LEARNING CONCEPT IN DERMOSCOPIC IMAGES

FOR HAIR SEGMENTATION AND REMOVAL
W.Besty
UG Student
Department of CSE
Chennai, Tamilnadu
P.Gopika
UG Student
Department of CSE
Chennai, Tamilnadu
A.Shali
Assistant Professor
Department of CSE
Chennai, Tamilnadu
A.Sangeerani Devi
Assistant Professor , Department of CSE
Chennai, Tamilnadu
E.Swathi
UG Student
Department of CSE
Chennai, Tamilnadu
D.Kavitha
Assistant Professor , Department of CSE
Chennai, Tamilnadu
In light of the rising number of cases of skin cancer, this is an important public health issue. The
proposed UNET-based model is employed which is established on convolutional neural networks
(CNNs).This loss function uses the structural similarity index, L1 distance, and total variation loss as
May 21st – 22nd, 2022
well as other new metrics introduced during the network's initial training phase. Hairless images
from publicly available datasets are used to generate hair simulations. Both qualitative and
quantitative results show that our model is effective, and that our loss function has a significant
impact on the proposed model's ability to restore itself.
400. VISUAL CUES TO VOICE CUES

1,3 J. Thamil Selvi, 2,3S. Keerthana Devi, 2,3S. Shreya, 2,3 M. Sanjana.
1 Professor, 2IV Year UG students,
3Department of Electronics and Communication Engineering,Sri Sairam Engineering College, Chennai-44,
India.
The project is aimed at developing an object identifier for visually challenged. This work proposes
creation of a system that implements sensory substitution of vision through ESP32CAM. When
visually challenged people navigate in known or unknown, indoor or outdoor environments, as they
face troubles due to inaccessible infrastructure and social challenges, this can be used to improve
quality of Visually Impaired life through visual information to the users. The idea is to capture
realtime images of objects through ESP32CAM. Using the machine learning technique, object
detection is done. The captured image is analyzed and converted into text. Further, the rule of text-
to-speech is used, the text is then converted into sound using the Python module. The output of Text-
to-speech is then amplified using an audio amplifier which is then heard through an earphone.
401. A NEW COMPUTATIONAL INTELLIGENCE APPROACH TO UNDERSTAND

AUTISM SPECTRUM DISORDERS
Rakhee Kundu Amity University
Computer Engineering Jaipur, India
Dr.Sunil Kumar
Amity University Computer Engineering Jaipur, India
Autism Spectrum Disorder is the neurodevelopmental disorder which is the fastest growing
disability in the world. The first point of contact to identify this type of disorder is the clinical
practitioners, therapist or sometimes it is observed in themselves or their family members.
Therefore, an early diagnosis and treatment strategy for ASD is highly necessary. To that end, we
seek a machine intelligence approach that takes into account function-to-class correlations instead of
function-to-fact. This research aims to classify features that are less relevant in popular ASD
assessments, which have an effect on ASD in current ASD tools. The use of intelligent approaches
focused on machine learning to minimize screening time for ASD has been explored in behavioral
science, increasing sensitivity, precision and accuracy. In order to accomplish one or more of the
aforementioned objectives, these models are to be used in a screening tool.
402. IOT BASED MONITORING AND CONSERVATION OF NATURAL

RESOURCES IN DAY-TO-DAY LIFE
〖Anand P〗^1, 〖Gandhimathi alias Usha S〗^2, 〖Suvetha Parameshwari R〗^3, 〖Sarunivetha R〗^4, 〖
May 21st – 22nd, 2022
Shanmugapriya R〗^5
(^1)Assistant Professor,(^2)Associate Professor,(^3,4,5)Final Year UG students
Department of Electronics and Communication Engineering, Velammal College Of Engineering and
Technology
This paper provides an insight into the event of an IOT based prototype to conserve energy resources
in day-to-day life. This has to be started from the kitchen as it is the most important place in every
household. Management of water, electricity and fuel in the cooking space is a major concern. The
idea behind our project is to monitor the usage pattern of water, electricity - to give alert at the end
of the month. And to automatically book cylinder when the fuel gets exhausted. Conservation is also
one of the important parameters in energy management. Hence the electricity can be conserved by
automatically switching off LED when not in use and the water can be conserved by using sensor to
automatically regulate water according to the usage. Apart from this, this project incorporate few
sensors for the convenience of the users to keep check on pest in the kitchen. The main scope of this
work is to bring a resource efficient cooking place in the hands of the user through a mobile
application.
403. PREVENTION OF SQL INJECTION ATTACKS IN WEB APPLICATIONS

1Vaibhav Srivastava, 2Abhinav Majumdar, 3Jeyasekar A
1 UG Students, Dept. of B-Tech, SRM Institute of Science and Technology, Kattankulathur, Chennai, Tamil
Nadu, India2 UG Students, Dept. of B-Tech, SRM Institute of Science and Technology, Kattankulathur,
Chennai, Tamil Nadu, India3 Associate Professor, Dept. of B-Tech, SRM Institute of Science and Technology,
Kattankulathur, Chennai, Tamil Nadu, India
SQL injection attacks are the most basic type of cyber-attacks that execute arbitrary malicious code
to retrieve confidential information from a SQL database. This paper aims to study and analyze three
types of injection attacks and find out their vulnerabilities. Based on that, we propose a new measure
to prevent the occurrence of SQL injection attacks. The proposed measure was experimented with
and tested using a local webserver and found that it accurately detects and prevents SQL injection
attacks. The proposed system could be implemented in the Web Application Firewall to detect and
prevent malicious SQL traffic.
404. LUNG CANCER DETECTION USING DEEP LEARNING TECHNIQUE

Dr.K. Kavitha1, R. Jeevaraj2, R.M. Sanjay Aravind3, M. Ram Kumar4
1Associate professor, 2,3,4UG Students Department of Electronics and Communication Engineering
Velammal College of Engineering and Technology, Madurai - 625009
Lung cancer has been one of the leading causes of mortality in humans worldwide. Early detection
can enhance survival chances amid humans. It usually causes death for both genders, so it is
necessary to take care immediately & examine nodules. Accordingly, several techniques are
implemented to detect carcinoma within the early stages. In this detection process, a comparative
analysis of different techniques based on machine learning for the detection of lung cancer has been
proposed. There are too many methods developed in recent years to diagnose carcinoma, most of
them utilizing CT scan images. In addition, multiple classifier methods are paired with threshold
May 21st – 22nd, 2022
segmentation algorithms to use image recognition to identify lung cancer nodules. From this study,
it has been found that CT scan images are more suitable to get accurate results. Therefore, mostly
CT scan images are used for the detection of cancer. In Addition, the results obtained from the
methods that supported deep learning techniques achieved higher accuracy than the methods using
classical machine learning techniques. The deep learning algorithm implemented and predicts the
result with better accuracy, precision, recall, and Fl score.
405. MRI SEGMENTATION OF LUNGS USING LEARNED SEMANTIC

KNOWLEDGE AND GRAPH CUTS
S . Boojashree 1, R . Annsheny 2, M . Aparrna 3 , K . Lohitha 4, Dr.G.ThamaraiSelvi 5.
1,2,3,4 Student -Bachelor of Engineering, Electronics & Communication Engineering Department, Sri Sai
Institute Of Technology, Chennai , Tamil Nadu - 600044, India.
5 Head Of Department , Electronics & communication Engineering Department, Sri Sai Ram Institute Of
Technology, Chennai , Tamil Nadu - 600044, India.
Detection of lung diseases and disorders is a challenging task in radiology and oncology. Computer
Aided Diagnosis (CAD) systems read the inputs from a medical imaging modality such as Computer
Tomography (CT) and Magnetic Resonance Imaging (MRI) scans and tends to highlight the
suspicious or abnormal patterns which is otherwise referred as Region of Interest (ROI). The ROI is
evaluated on the attributes like, shape, size, current pattern, and the growth pattern.
Classification is a key process in any CAD system. The aim of the classification system is to
classify the input data into different classes with high accuracy rate. Classification accuracy depends
on the classifier that is used. Moreover, two deep Multiple Instance Learning (MIL) frameworks
with dynamic max pooling and sparse representation is proposed for classification of Tuberculosis.
MIL, in general is a supervised machine learning framework that deals with uncertainty of instance
labels. The approach does not require detailed information for optimization. It is applied to
conventional CAD system to obtain better results. The proposed method was compared with other
multiple instance learning approaches like Diverse Density (DD) and Maximum pattern bag
formulation of Support Vector Machine. Numerical analysis indicates the accuracy of the proposed
work which was found to be better than the widely used techniques.
The approaches proposed by the current study have helped in achieving the overall objective of
developing a CAD system with better accuracy. The performance of the proposed algorithms was
evaluated using standard evaluation metrics and datasets. The result was validated in terms of
improved accuracy, specificity, sensitivity, and precision rate compared with existing approaches.
406. DESIGN OF D FLIP – FLOP USING KISAN GATE
¹Kirankumar Manivannan ²Dr.M. Santhi ³Jennithra.S 4Iniya.R

¹ Faculty, Department of ECE, Easwari Engineering College, Chennai

May 21st – 22nd, 2022
² Faculty, Department of ECE, Saranathan College of Engineering, Trichy ³

Student, Department of ECE, Easwari Engineering College, Chennai
4Student, Department of ECE, Easwari Engineering College, Chennai
D Flip-flops are sent anyplace in a remote sensor organization. So their correspondence can be
effortlessly checked. In these organizations, message security and hub recognizable proof are normal
issues. Consequently, security of enormous scope such organizations requires proficient key
dissemination and the board systems. Significant steganography and especially significant key
dispersion is such a procedure that dispenses secure keys just for brief distances. The significant
steganography offers enormous benefits over conventional strategies by the utilization of trash yield
decrease and significant transmission, even though it is not totally secure. These issues can be
recuperated by utilizing the reversible rationale entryways in effective way. Here another reversible
4x4 rationale entryway called KISAN door has been utilized. This has been joined with QKDP to
accomplish secure correspondence in WSN. In our examination, we embraced C-R plot as far as
these significant entryways to defeat the weakness brought about by noxious hubs. As the qubits put
away in a sensor hub can be utilized just a single time and can't be copied, subsequently hazard of
data spillage diminished regardless of whether the hub are compromised.
407. FISHING VESSELS MONITORING SYSTEM

Dr C. Gnana Kousalya, M.E., PhD, D.Baby Nivetha and Blessy Esther Selvin Jacob Department of
Electronics and Communication Engineering, St. Joseph’s Institute of Technology, Chennai, India.
The technology proliferation of Received Signal Strength Indicator (RSSI) is used to identify the
movements of the boats that cross the harbour mouth. The entry and exit of the fishing boats are
monitored. The System extant follows only manual updating of information in registers. However,
the existing system is not powerful enough for monitoring the boats as it is done manually. In the
proposed system, the boat entry and exit can be identified using Quick Response (QR) code, Radio
Frequency Identification (RFID) and the Received Signal Strength received from the RSSI. The QR
code and RFID tag are placed on the boat and it is scanned by the QR scanner and RFID reader. The
NodeMCU is placed on the boat with a power supply from the battery and power board. Using the
RSSI technology, the movement of the boat’s passageway can be identified using received signal
strength. Using this input, the time of arrival and departure of the boat are noted. The information
can be stored in the cloud. Whenever the user wants the data, the data sheet can be retrieved from
the cloud. Thus, the proposed system will help in monitoring the fishing vessel automatically by
updating the data in the cloud.
408. PERFORMANCE EVALUATION OF CNN-SVM CLASSIFIER FOR

DIAGNOSING CORONARY ARTERY BLOCKAGE
Muthiah GK1, Akash K2
Department of Instrumentation and Control Engineering St. Joseph’s college of engineering
Chennai, India
Dr. E. Ahila Devi

Department of Instrumentation and Control Engineering St. Joseph’s college of engineering
Chennai, India
May 21st – 22nd, 2022
Coronary artery disease is a form of cardiovascular illness that is the world’s largest cause of
mortality. Doctors diagnose the disease by performing many tests, scanning, and then reviewing all
of the findings, which takes time. A modern diagnostic process has been developed with the help of
artificial intelligence to diagnose the disease immediately without the assistance of a doctor. This
benefits the medical industry by allowing for faster disease diagnosis and treatment implementation
for a large number of people. A hybrid Convolutional Neural Network (CNN) - Support Vector
Machine (SVM) classifier is used to diagnose coronary artery disease. A 10-layer Convolutional
Neural Network is built using three convolutional 2D layers with relu activation functions, three
maxpooling layers, one dropout layer, one flatten layer, and two dense layers.The last layer consists
of linear SVM which classifies the coronary artery disease affected and non-affected images. The
results obtained by the hybrid CNN-SVM classifier on diagnosing the disease is 90.29 percent
accuracy.
409. INSECT CLASSIFICATION USING DEEP LEARNING

Immanuel Danie1, Dinesh R2
Department of Instrumentation & Control Engineering
St. Joseph’s College of Engineering, Chennai
Dr. M. Suresh (Associate Professor) Department of Instrumentation & Control Engineering
St. Joseph’s College of Engineering, Chennai
Agriculture has been the heart of the Indian economy. In India, rice is still the most common crop.
India is the second-largest rice grower in the world, behind China. Agriculture contrary, is a
laborintensive industry susceptible to natural disasters such as drought, poor rainfall, and pest
infestations. Agriculture's biggest threat is insects. They cause severe crop damage, resulting in
lower crop yields. As a result, it's vital to protect crops from insects, which involves accurate insect
identification and the use of specialized insecticides and control measures. The major purpose of the
study is to utilize deep learning to determine which insects are useful to rice fields and which are
harmful. The proposed solution uses the Convolution Neural Network (CNN) algorithm. If the insect
discovered is harmful to the plant, the proposed idea can assist in classifying it and providing
strategies for its eradication. The technique used in this study offers a faster execution time and
higher prediction accuracy than traditional machine learning algorithms. It also provides an excellent
balance of speed and precision, which is useful for finding and completing specified activities.
410. ADVANCED MULTI SWARM INTELLIGENT ANT COLONY OPTIMIZATION

ALGORITHM FOR VIRTUAL MACHINE PLACEMENT PROBLEM IN CLOUD
PLATFORM
Nagadevi S and Dr.S.V.Kasmir Raja
SRM Institute of Science and Technology,Chennai
Cloud computing provides users to process their data and to share resources in a shared pool
environment. Virtual machines are considered as one of the emerging approach where cloud
computing is used to separate the underlying hardware to manage the resources in shared pool. In
the virtualization technology in Cloud environment the significant issue is virtual machine
May 21st – 22nd, 2022
placement problem (VMPP). Virtual Machine Placement is placing a virtual machine on a suitable
physical machine in order to improve the resource utilizationand to avoid Non-deterministic
polynomial time hardness (NP) hard problem withless energy consumption andprocessing time
delay to minimize Service level Agreement (SLA) violation. In this research, Advanced Multi
Swarm Intelligent Ant Colony Optimization (MSI-ACO) Algorithm has been proposed to balance
the usage of resources in multiple cloud service platforms and aimed to reduce the risk of VMPP.
This article presents a stateof-the-art VM placement and consolidation techniques which focus on
improving the energy efficiency and resource utilization through hybridization of swarm intelligent
technique with ACO. A detailed comparison is presented which reveals the pitfalls and suggesting
improvisation methods along this direction.
411. PERFORMANCE ANALYSIS OF MACHINE LEARNING AND DEEP

LEARNING ALGORITHMS IN CREDIT CARD FRAUD DETECTION
SENTHOOR LEENA D
Department of computer science and engineering (Anna University)
Thiagarajar College of Engineering (Anna University)
Madurai, India
SUVETHA S
Madurai, India
DHANUSRI B
Madurai, India
Dr. C. SENTHILKUMAR
Madurai, India
Credit cards are now being used for most of the online transactions. As the credit card users are
increasing, the number of cheating cases are also increasing. It results in finding fraudulent activities
using algorithms that find the fraud cases accurately. This research work proposes different machine
learning algorithms and deep learning algorithms and helps to find out which is best for finding out
fraudulent transactions. Here, the machine learning algorithms that have been implemented are
Logistic Regression, Bernoulli Naive Bayes and Gradient Boosting Classifier. The deep learning
algorithms that have been implemented are Convolutional Neural Networks(CNN) and Deep Neural
Networks(DNN). The algorithms have been compared using comparison metrics such as accuracy,
precision, recall, F1 score and confusion matrix. The result shows that the deep learning algorithm is
better than all the three machine learning algorithms. The CNN and DNN results in 99% accuracy
which are higher than all other machine learning algorithm’s accuracy.
412. DESIGN AND IMPLEMENTATION OF AN AUTOMATED AGRICULTURE

MONITORING AND CONTROL.
M. Swara Raga Sudha
May 21st – 22nd, 2022
Sri Sairam Institute of technology Chennai
G. Deepika
Sri Sairam Institute of technology Chennai, India
S. Varshini
Dr. D. Murugaradhadevi
Professor
Machine learning techniques are highly appeared in the field of agricultural research the agricultural
factors of weather, rain, soil, pesticides, and fertilisers are the number one factors accountable for
yield production. The fundamental part of agriculture is developing fertile soil. Soil examination is
an essential aspect of soil asset management in horticulture. Soil inquiry is beneficial for cultivators
in determining the type of harvests that need to be created in a specific soil condition. The number
one objective of this is to observe soil supplements through the usage of machine learning and
categorization techniques. The department of Agriculture and Farmers Welfare furnished a huge
records set on soil nutrient repute. This work analyses the soil nutrients and helps in reducing the
usage of pesticides and fertilizers by predicting the appropriate organic and inorganic fertilizers that
are to be used in the soil for the specific crop.The pesticides can be avoided by the usage of
ultrasonic pest repeller. The soil of different sorts consists of a large array of nutrients. The nutrients
N, P, K, Ca, Zn, Al are used in this paper to investigate soil supplements using a hybrid neural
network approach. The class algorithms’ performance is evaluated using two metrics: accuracy and
execution time. Furthermore, we are avoiding the usage of pesticides by using DHT11 and an
ultrasonic pest repeller. This system consists of number of modules that are intended to solve the
problem faced by the cultivators.
413. SURVEY ON AI BASED CAR PARKING SYSTEM

Ms.S. Kalaivani1 ,Abdul Aleem2, Abdur Rahmaan3 and Hidhayathulla Khan4
1 Assistant Professor (Sr.Gr),2,3,4 B.Tech Student
University of B.S. Abdur Rahman Crescent Institute of Science and Technology
Nowadays parking of vehicles has been an important issue and the struggle of finding parking
spaces is increasing day by day. This is due to everybody using manual vehicle parking and this
causes waste of time and fuel. To overcome this, no.of works have been proposed that uses RFID
technology for authentication and to store the details of the car in DB and AI based Alongside we
detect the car parking slot in an efficient way. In this paper a survey is performed on this to compare
and identify optimized AI algorithms that best suits for such application.
414. FRAMEWORK FOR NETWORK THROUGHPUT MAXIMIZATION TO

PROLONG LIFE TIME OF WIRELESS SENSOR NETWORKS
N.Srikanth, Research Scholar & MIEEE
May 21st – 22nd, 2022
Department of ECE, Annamalai University,

Chidambaram,Tamil Nadu, India
Dr.T.Shankar, Assistant Professor, Department of ECE,
Government College ofEngineering, Srirangam,Trichy,Tamil Nadu, India
Dr.G.Yamuna, Professor, Department of ECE,
Annamalai University, Chidambaram,Tamil Nadu, India
Due to the increased throughput and high channel utilization, the concept of clustering has been
widely used in Wireless Sensor Networks (WSNs). Each sensor node to which it belongs transmits
data to its Cluster Head (CH) to gather information more efficiently. Nodes excess energy will be
consumed faster than the farther nodes when nodes are closer to the CH, which brings a negative
influence on load balance for the entire networks. In this paper, we proposed an experimental
investigations for network throughput guaranteeing using STH-MAC to prolong the lifetime of
wireless sensor networks. Simulation results show the exceptional performance of the proposed
scheme against field-proven techniques in terms of minimum possible end-to-end delay, maximum
average packet delivery ratio; throughput is improved by 51.26% and pro- longevity of network
lifetime in the proposed STH-MAC is 95.9% as compared to the existing MAC protocols.
415. SOLAR IRRADIANCE VALUE BASED OPTIMAL OPERATING ELECTRIC

ASSEMBLY DESIGN
1 Dr. P. KANNAN, 2 L. KAVITHA, 3 S. GAYATHRI, 4 S. HARINI
1 Professor & HOD, Department of ECE, Panimalar Engineering College, Chennai, Tamil Nadu,
India
2 Student, Department of ECE, Panimalar Engineering College, Chennai, Tamil Nadu, India 3
Student, Department of ECE, Panimalar Engineering College, Chennai, Tamil Nadu, India
4 Student, Department of ECE, Panimalar Engineering College, Chennai, Tamil Nadu, India
The goal of this project is to achieve uninterrupted power supply to the load and to analyze
efficiency of solar panel. Simple and low cost method based on a mathematical equation is used to
calculate the solar irradiance in terms of voltage and current. More irradiances originated from these
data are displayed on the DIGITAL system. For accurate determination these parameters are
dependent on experimental measurements. The reliability of measurements of PV volt and intensity
of solar radiation are dependent on the sampling values of the measurements. To achieve maximum
irradiance, Maximum Power Point Tracking is used. The characterization of a solar cell by using the
sun as a source of radiation and considering its temperature and received radiation intensity as
constants requires automation of the experimental setup. By this execution the PV performance
analysed and regulated with respect to successive values of the sensors. Load protection against
surge variation in sudden cloud passing by cut off region. Irradiation performance directly reflect
the PV characteristics which give smooth signal to controller by Signal Conditioner.
416. WEB BASED INTERNSHIP PORTAL

Ms. V Narmadha , B Devadharshini, B Mahalakshmi
Assistant Professor, Department of Information Technology, Sri Sairam Engineering College, Chennai.
Student, Department of Information Technology, Sri Sairam Engineering College, Chennai.

May 21st – 22nd, 2022
Internship opportunities promote authentic learning experiences by providing opportunities to

integrate course knowledge with real-world experience in a specific context. Work experience or
internships are high on the list of criteria used by recruiters for college graduates. Providing students
with opportunities to practice and learn theory is critical to their professional development. The
platform is designed to facilitate internship opportunities among students and simplify manual tasks
such as оd аррrоvаl, inter deраrtment interасtiоn etс eаsy.
417. A VANET BASED ACCIDENT DETECTION AND MANAGEMENT SYSTEM IN

A SMART CITY
S. Sivasakthi, G. Kaviya, B.Thivya and M. Pown
Department of Electronics and Communication Engineering, Sri Sairam Institute of Technology, Chennai,
Tamil Nadu, India
Infrastructure development in India is rapidly increasing and the transportation is an important factor
for a country’s progress. As a result, several highways are being built to assist India's younger
generation. Highway construction is not only an appropriate method of infrastructure development,
but it also includes monitoring, maintenance, security and other issues. Road safety has been a major
concern for governments and automakers. The global automobile population has grown
dramatically, increasing traffic density and increasing accidents. Because India's traffic is growing
four times faster than its population. Hence, the firms, researchers, and organisations are focusing on
improving road safety. The advancement of wireless technologies has allowed researchers to
construct communication systems that include cars. Thus, this paper investigates Vehicular Ad Hoc
Networks (VANETs) and their communication methods. Also, the efficient communication in
VANET is examined. The task includes investigating and designing a communication model to help
identify network traffic concerns and unintentional probability. The proposed model is aimed to
eliminate unintentional possibilities while helping in other applications. The suggested VANET
model may feature V2V, V2R and V2I communication.
418. SENTIMENT ANALYSIS OF TWITTER DATA REGARDING RUSSIA

UKRAINE WAR
Abhinav1, Atul Kumar Srivastava2
School of Computing, DIT University, Dehradun 1, 2
Social media is one of the main platforms where people express their opinions, using various forms
of data. One such platform is Twitter which is used by people to express their thoughts with the help
of text mainly. This data is in a well-structured form, and hence can be used to do analysis and
various practices on it in order to get the idea about various aspects that impact the market or the
aspects that impact people’s lives in general. In the proposed model, sentiment analysis on twitter
data has been done which was based on a major ongoing event – The Russia Ukraine War. This
model finds out the percentage of positive and negative sentiments that are being expressed by
people through their tweets. Also, the sentiments of various users on this event has also been taken
into account. This analysis will help to know how this event is impacting the people residing in
various parts of the world.

May 21st – 22nd, 2022
419. IOT BASED ACCIDENT PREVENTION AND INTIMATION SYSTEM

Mrs. Shabana Parveen
Asst. Prof. Dept. of ECE
Sri Sai Ram Engineering College
West Tambaram, Chennai
Bharath kumar KK, Deepak S, Gokul N

Dept. of ECE
West Tambaram, Chennai
Road traffic accidents (RTAs) have emerged as an important public health issue that needs to be
tackled by a multidisciplinary approach. In countries like India, the tendency for RTA injuries and
deaths is alarming. The number of fatal and disabling road accidents happening is increasing day by
day and is a real public health challenge for all the concerned agencies to prevent. The approach
includes the proper implementation of the alert system for the accident and to prevent some death
cases which happen due to delay in taking hospitals. Exclusively we are including prevention of
accidents using an alarm to intimate the driver to be precautious to maintain the distance between
vehicles. Our Project mainly considers these scenarios where no man situation appears and tries to
reduce the death rate by a proper information system and prevention. This project implemented an
IoT system that triggered the incident and attempted to contact the person responsible to take
immediate and necessary action. This project has an implementation of an auto accident detection
system through IoT using piezoelectric sensors and a Raspberry PI. With the help of piezoelectric
sensors, we can detect whether an accident happened and further proceed with the intimation system.
The Intimation system works on the Raspberry PI's Wi-Fi module through the firebase database.
With the help of a webpage, we can notify the required personnel and hospital. In prevention we
implement a ML algorithm which detects the blink rate of the driver and if the driver feels sleepy or
in sleep the system intimates the driver by ringing a buzzer.
420. ANALYZING THE EFFECTS OF A LOTTERY SCHEDULER IN THE XV6

KERNEL
Piyush Sachdeva, and Vikas Singh Mehta
Dehradun Institute of Technology
Lottery scheduling is a revolutionary randomized resource allocation mechanism presented in this

paper. Lottery scheduling allows for efficient and responsive control of computation execution
speeds. Such control is beyond the capability of traditional schedulers, and it is desirable in systems
that fulfill a variety of requests, such as databases, media-based applications, and networks. Lottery
scheduling also helps with modular resource management by allowing concurrent modules to
separate their resource allocation policies. A currency abstraction is used to manage resources
flexibly and efficiently. For the purpose of this paper, we have implemented a lottery scheduler in
the xv6 kernel developed by MIT for studying purposes.

May 21st – 22nd, 2022
421. PREDICTION OF CROP YIELD PRODUCTION USING HYBRID RANDOM

FOREST MODEL IN MACHINE LEARNING
Aparna S1*, Valarmathie P2, Bala Gayathri D3
1Research Scholar, Department of CSE, Anna University, Chennai, Tamilnadu-600025, India
2 Centre for Artificial Intelligence, Chennai Institute of Technology, Chennai, Tamilnadu-600069, India
3Research Scholar, Department of IT, Anna University, Chennai, Tamilnadu-600044,India
Most agricultural crops are not well-affected in terms of climate change. Predicting the harvest
before its harvest will help policymakers and farmers to take the necessary marketing and storage
measures. The current challenges of water scarcity, uncontrollable costs due to existing needs, and
climate uncertainty are some of the factors that enable farmers to equip farmers with smart farming.
This activity can help farmers to understand the yield of their crop before sowing in the agricultural
field and help them to make the right decisions. In our research work, a prototype of a collaborative
guessing system was developed and the results were compared between different models. The
proposed work creates a novel with the use of parameters such as Region, region, season, location
and the user can predict the crop yield during the year he needs to do so. Machine learning strategies
such as Random Forest and K-Nearest Neighbours are being developed and integrating them leads to
an integration model that predicts crop yields in a better way.
422. FOOD ALLERGY PREDICTION SYSTEM

S. Deivanayagi1, P.Shalini2, P.Shrivani3, GR.Kaviyasri4
1 - Associate Professor, Electronics & communication Engineering Department, Sri Sai Ram Institute Of
Technology, Chennai , Tamil Nadu, India.
2,3,4 - Student –B. E (Electronics & Communication Engineering), Sri Sai Ram Institute Of Technology,
Chennai , Tamil Nadu, India.
We emphasise health-related applications as an example of how advances in ubiquitous computing

are allowing new opportunities to arise in a variety of fields and industries. In this field, there is a
growing trend toward the development of numerous applications that enable care to be provided
anywhere and wherever it is needed, a trend known as ubiquitous healthcare. A thorough
examination of existing and proposed models has revealed that none of these applications adequately
addresses the needs of persons who are allergic to certain foods. As a result, this article presents an
allergy detection model that is ubiquitous and is based on context awareness (situation awareness).
Known as Allergy Detector, this approach is focused on food allergies, specifically the eight major
allergens (peanut, milk protein, egg whites, wheat, soy, fish, shellfish, and tree nut) and their
derivatives, which are responsible for around 90 percent of all food allergies. The Allergy Detector's
most significant scientific contribution is the application of situation awareness for the purpose of
identifying allergies at any time and from any location in a transparent manner. A number of case
studies were developed in order to evaluate the model. We also performed a profiling exercise,
assessing the performance of the Allergy Detector and finding the bottlenecks that were
encountered. The findings demonstrated that the model may be used to detect allergies across a wide
range of environments.

May 21st – 22nd, 2022
423. STOCK MARKET PREDICTION USING LSTM

Dr. Suma Sira Jacob,Dr. Alaguvathana, Monesh Kannan R, Jeva Vignesh T, Navin Ganesh Pandian R K.
Department of Information Technology, Sri Krishna College of Technology, Coimbatore, Tamilnadu, India
Predicting in what way or manner the stock exchange will act is an individual's ultimate troublesome
agenda. There are so many determinant factors that make the declaration -made in advance
complicated and concerning the body determinant angle. Concerning the mind, of sound mind, and a
careless manner of conduct, etc. All these elements to consider connect to form share prices that are
explosive and very difficult to express in advance with a large degree of precision or correctness.
Using facial characteristics like new proclamations about a group bound by interest/work/goal, their
income results, etc., machine intelligence methods bear the potential to dig up patterns and
awareness we didn’t visualize before, and these may be used to create accurate and correct
declarations made in advance. Here in this method, we have tried random forest, KNN, and
ensemble means to express an outcome in advance, the correct results of stock.
424. FRICTION STIR WELDING OF ALUMINIUM ALLOY PLATES USING

CYLINDRICAL PIN TOOL: A REVIEW
Mithun P Nath, Jerin Johnson
Aluminium alloys have significantly contributed to the lightweight and long-lasting constructions
seen in high-precision applications (marine industry, aircraft industries). The most often used
aluminium alloys in these industries are from 2000, 7000, and 6000 series. In aerospace, automotive,
and other industrial sites Friction stir welding is used. The mechanical and metallurgical features of
FSW aluminium alloys are promising. In the FSW process, rotational speed, welding speed, and tilt
angle are most important parameters. FSW is a solid-state technique in which metal is not melted,
and it is welded together using a cylindrical shouldered tool with a profiled pin that is rotated and
progressively plunged into the weld joint between two metal plates or sheet sections. Frictional heat
is created between the tool and the material, causing the work pieces to soften below the melting
point and then intermix both metals at the joint. Due to the high temperature, further softened metal
is joined by mechanical pressure applied by the tool. A high-quality weld is produced since the
temperature is below the melting point and the joining is below the material's melting temperature.
425. DEEP LEARNING BASED DEFORESATION CLASSIFICATION USING CNN

Sabthesh JK Sriniwas and Vineet Rajpal and Dr. Sanjay Kumar Singh.
Vellore Institute of Technology
Human-induced deforestation has a major impact on forest ecosystems and therefore its detection
and analysis methods should be improved. Where, this type of detection or classification helps us to
degrade the deforestations in future. In this project we are mainly focusing on the problem of
deforestation, performing the classification of deforestation and the healthy forests with the help of
deep learning model. CNN is the algorithm that which are been using here for the classification

May 21st – 22nd, 2022
process. We are preparing a dataset which is trained using the algorithm and classification will be
performed in testing.
426. METHODOLOGY TO IMPLEMENT IMAGING APPLICATIONS ON FPGA

BASYS 3 BOARD
Vineeth Ajith John

Department of ECE
Chennai,India
V. Pushpalatha
Department of ECE
Chennai,India
S.Ewins pon pushpa, Department of ECE

Chennai,India
[email protected]
corresponing author
Agastya Deshraju, Department of ECE

Chennai,India
R.Sittalatchoumy, Department of ECE

Chennai,India
Adithya Nair, Department of ECE, College of Engineering Guindy, Anna University
Chennai,India
This paper provides an insight to implement imaging applications on FPGA –Basys 3 board with
detailed procedures along with logics in verilog. As a technical contribution, image processing
applications such as increasing brightness, decreasing brightness, color inversion and converting to
grayscale are performed on a input image using a Basys 3 board. A Basys 3 board is an FPGA,
basedon the latest Artix-7 Field Programmable Gate Array from Xilinx. It’s a high-capacity FPGA
with a low overall cost and has a collection of USB, micro-USB, VGA and other ports which allow
it to host designs ranging from introductory combinational circuits to complex sequential circuits
like embedded processors and controller.
427. CHARACTERIZATION OF SOIL BY SCANNING ELECTRON MICROSCOPY

WITH ENERGY DISPERSIVE X-RAY SPECTROSCOPY
Bodh Ram Chaohana , Pooja Shrivastavab

May 21st – 22nd, 2022
aHead Dept. of Physics, Govt. Shyama prasad Mukherjee College Sitapur, Surguja, Chhattisgarh, India,
bResearch Scholar , Civil Engineering department, National Institute of Technology, Raipur, Chhattisgarh,
India
We need to enhance our knowledge of the variables that regulate soil characterisation by
incorporating physical, chemical and biological measurements at microscopic scale. The selection of
soil attributes and its interpretation are not straightforward, because of the intricacy and soil
specificity. This study was done to test the potential of SEM-EDX investigation. Soils have three
different phases i.e. dry, saturated, and partially saturated and need to be characterising properties at
different phases. In the present work we applied SEM-EDX technique on the uphill water conduit’s
soil having a composition of elements i.e. Fe, Mg, Si, Al, Ti, Mn, Ca, K, C, O, that were quantified
in the dispersive analysis, especially involving the transport of matter and energy. Element has
unique atomic configuration and weight, responding a distinctive series of peaks of spectrum as per
electromagnetic emission. We collected soil samples during a monsoon from water conduit of
village Visrapani, block Mainpat, and district Surguja in Chhattisgarh, India. In viewing soils of
various dispersive classifications with SEM-EDX, surface soil sample and one feet depth soil
sample with the same dispersive classifications can be a compare and we can record common
features of soil morphology and soil concentration.
428. SMART CARD BASED E-HEALTH RECORD

1,3 J.Thamil Selvi, 2,3S.N.Sunandha, 2,3V.Gayathri, 2,3 P.Kiruthiga ,2,3V.Sujitha.
1 Professor, 2IV Year UG students,
3Department of Electronics and Communication Engineering, Sri Sairam Engineering College,
Chennai-44
Health care system is one of the critical issues for developing countries and thus the information
technology is becoming progressively more important nowadays. The deployment of smart health
cards simplifies the prescription process, improves the standard of care given, and facilitates
electronic healthcare records management through a coordinated health service process. Using the
smart health card patient’s data, doctor’s prescription, patients present and former health history is
accessible. When an infant is born a record is formed with details like blood group, vaccination
dates, complications involved, allergies and other important details. This card plays a crucial role
during unexpected incidents like accidents and reduces child traffic problems. The cloud computed
system enrolled within helps to secure data of patients and prevent data intrusion. For authentication
and security purposes the hospital's unique code, Aadhar card number and Doctor's access code is
provided which could be accessed only within the Hospitals. The e-health card facilitates privacy
protection by filtering access to sensitive data which is accessible only to authorized people like
Doctors and Cardholders.
429. SMART MEDIC ASSIST IN TRAIN USING IOT

1.SRI KAMALI S,2.SIVASRI B,3.MANASHADEVI V

May 21st – 22nd, 2022
1,2,3 Student-Bachelor of Engineering,Electronics & Communication Engineering

Department,Sri Sairam Institute of Technology,Chennai,Tamilnadu,India.
GUIDE 1:Mrs.D.PUSHGARA RANI

Associate Professor,Electronics &Communication Engineering Department,
Sri Sairam Institute of Technology,Chennai,Tamilnadu,India
Mrs.V.K.G.Kalaiselvi
Assistant Professor,Department of Information Technology,
Sri Sairam Engineering College,Chennai,Tamilnadu,India
Internet of things (IoT) field has emerged due to the rapid growth of artificial intelligence and
communication technologies. Railways is one of the old and developing mode of transportation.
Railway transportation offers reduced time travel, energy conservation etc., This mode of
transportation plays a major role not only in transportation but also safety of humans traveling in it.
In this modern era of technology, we develop many technologies to protect human life but we
haven’t seen one installed in train. Our paper is mainly based on providing a emergency medical
module in trains, so that it can help people in emergency situation in train. We propose controller-
based technique to give first aid support to the emergency patient and the details will be transferred
to the loco pilot by implementing this technique to save the life of emergency patient in travel. We
use Arduino Mega as controller and heartbeat and temperature sensor to sense the patient’s heartrate
and temperature and IOT to transfer the information.
430. EAR DETECTION IN UNCONSTRAINED ENVIRONMENT USING FASTER R-

CNN
Vinny Varkey Sam1, Natarajan. K2 and Resmi K R3 [0000-0002-1307-8080]
1,2 School of Engineering & Technology, CHRIST (Deemed to be University), Karnataka,India 3 Department
of CS, Santhigiri College, Thodupuzha, Kerala,India
Biometric based human identification systems are in high demand due to growing security and
privacy concerns. The human ear is one of the important physiological biometric used extremely for
personal recognition nowadays. Ear detection is a crucial stage in the computational process of ear
recognition. How- ever, there is a significant absence of ear detection techniques in the literature.
This creates a problem in terms of growing the use and applications of ear recognition systems in
commercial systems. Uncontrolled conditions such as pose change, occlusion, and change in
illumination offers a significant barrier to clas- sic ear detection algorithms. This study provides an
efficient method for automatically detecting ears from 2D profile images using Faster R-CNN. The
system goes through two stages of training: First, pretrained CNN model ResNet50 is trained to
classify ear and non-ear segments. Second, the unified Region Proposal Network (RPN) with
ResNet50, is trained to detect ear from profile face images.
431. IMPROVEMENT IN ENGINEERING BEHAVIOUR OF BLACK COTTON SOIL

USING GGBS
First A. Sonali Pawar and Second B. Mahesh Patel
Department of Civil Engineering NIT Jalandhar
May 21st – 22nd, 2022
Rapid population increase and urbanization have resulted in a decrease in the high quality of
accessible land. Black cotton soils (BCS) are often problematic in nature due to their swelling and
shrinkage characteristics and are not suitable for constructing civil engineering projects. For using
them for construction they need to be stabilized. The traditional methods of stabilization usually
lead to substantial emissions of CO2 to the environment. This can be avoided by using industrial
byproducts which can give similar results. Hence utilization of this waste can reduce the carbon
footprint to some extent. GGBS is one of the by-products of the steel industry whose stabilizing
potential is assessed in this study. The impact of its addition to the black cotton soil was studied in
terms of compaction, Atterberg’s Limits, specific gravity, STS (split tensile strength) and UCS
(unconfined compressive strength). The optimum content of GGBS (Ground Granulated Blast
furnace Slag) for the purpose of stabilization was found to be 30 % which led to a 180 % increase in
28 days UCS and a 195 % increase in 28 days STS.
432. AGRICULTURAL OPTIMISATION AND LIFE ANALYSIS OF A PLANT

USING IOT
RANJANI.D, VIJAY G DEEPTHI, CHARU GUPTA.R, SAAI KARTHIK V.M.

Department of Information Technology, Sri Krishna College of Technology.
Traditional life of agriculture is back-breaking work in land manually and has a lot of challenges to
be faced by agriculturists. The natural challenges faced by agriculturists is soil condition will destroy
due to climatic change,water-supply to the soil and monitoring 24/7 in real-practice is hard to check
whether the plant is growing healthy or not. This project is engrossed in optimisation of agriculture.
Smart plant monitoring was launched many years ago but most of the products will always check the
soil condition,temperature ,humidity and water- level condition.The efficient manner of smart
agriculture is to level up the optimisation for plants by detecting the notification in BLYNK APP
with WI-FI and the plant will report to us for life time in excel sheet. Lifetime values will be
calculated in each plant and takes maximum value from the reported plants. The maximum value of
the plant checks the lagged life plants and tells the valid reason for healthy growth. Owner can check
the optimization and atlast he/she will resolve the challenges faced in future.
433. REAL TIME IMPLEMENTATION OF FACIAL RECOGNITION

Dharani. J
B. Tech Information Technology
Dr. N. G. P. Institute of Technology
Coimbatore, India
Anusurya. P
Coimbatore, India
Keerthivasan. S
Coimbatore, India

May 21st – 22nd, 2022
Navin Prasanth. R
Coimbatore, India
Sivanesan. J
Coimbatore, India
Face is an individual identity of everyone and recognition of face helps in authenticating a person's
identity. Facial recognition is a category of biometric recognition system that maps an individual’s
entity to the training features. In this paper, we trained Elman Neural Network to recognize faces
within intended limits. The Elman brain network with its great powerful properties and solid
worldwide soundness is most generally used to take care of issues containing nonlinear, dynamic,
and complex information.The proposed software uses deep learning techniques to compare live
features of captured images from videos to the face print of training set to verify one’s identity. Face
recognition systems can successfully match the training features in photos, videos or in real-time.
Geometric, photometric stereo and localization are the different approaches of facial recognition.
This system detects and recognizes multiple facial features for batch input in dynamic formats like
videos. It is recorded with 94.6% of accuracy towards facial recognition thereby enhancing the
pixel's inception rate.
434. A STUDY ON IOT - BOTNET DETECTION TECHNIQUES

Rahul Sharma1, Khyati2, Harshit Singh2, Chirag Joshi3 School of Computing, DIT University, Dehradun1
School of Computing, DIT University, Dehradun2
School of Computing, DIT University, Dehradun3
With a rapid growth in the area of communication, it is easy for people to connect with each other.
The accumulation of data is also increased exponentially and it also facilitate many people around
the world. Access to information is easier than ever before, which increases the amount of cyber
attacks. Many attacks, such as viruses, worms, and trojan horses, have occurred in recent years.
Attackers conceal themselves using many strategies, one of which is the use of a botnet. They are
preparing and deploying Botnet-based attacks. Botnet assaults are not just aimed at computer
systems, but also at IoT devices. One current attack that must be addressed is the IoT-Botnet. In this
paper, we have studied various methods and techniques for the detection of IoT Botnet. We have
also compared different methods on the basis of their methodology, dataset and performances. The
study tells us the important of IoT Botnet and its detection methods.
435. MOVABLE PLATFORM FOR SOLAR POWER GENERATION

Rajini A R
Sri Sairam Engineering College Kanchipuram, India
Kaviyasri C A
May 21st – 22nd, 2022

Shanmuga Priya R
Halima Sahnas H
In this paper we introduced a movable platform for generation of solar energy by using Arduino, IR
sensor, Temperature sensor and dust sensor. The platform is meant to move to a location where solar
power intensity is relatively high, and is capable of self-assembling in the form of a solar panel,
allowing for optimal solar power consumption. The robots are designed to move to the place where
the intensity of the solar power is relatively high and hence there could be optimum consumption of
solar power. This project helps to overcome manual effort for solar energy conservation, inefficient
usage of solar energy due to immovable solar panels and insufficient electricity to rural areas.
Selfcleaning processes are also implemented in this project.
436. A SYSTEMATIC LITERATURE REVIEW ON DEEP LEARNING BASED

MEDICAL IMAGE SEGMENTATION
Vetriselvi D1 and Dr.R.Thenmozhi2*
Kattankulathur.
Medical imaging becoming an essential life supporting aspect in the current world. It having various
types of modalities and each one serves for specific applications. Lot of applications and necessities
are there to figure out the various life-threatening diseases. But the identification of abnormalities is
not so easy doing manually. It is error prone and time consuming. And also requires lot of
proficiency and experience. Deep learning is a state of art methodology which having a huge span of
applications especially in medical field. Particularly, in medical imaging the deep learning methods
can be applied and can make huge differences in the accuracy of findings. They can be used for
synthesis, segmentation, and classification. This study is aimed to focus on the different types of
medical imaging modalities and the various deep learning algorithms on medical imaging. The
performances of different methods were compared by means of various evaluation metrics.
437. IOT BASED ASSIST DEVICE FOR MONITORING STROKE PATIENTS

Mohammed Adhil S1, Mahesh Veezhinathan2, PG Student1, Associate Professor2, Department of
Biomedical Engineering, SSNCE
Stroke is the leading cause of long-term disability worldwide. In India stroke is found to be fourth
biggest cause of death and disability, wreaking havoc on stroke survivors' quality of life. Mirror
therapy, electrical TENS, and physical therapies are all examples of stroke rehabilitation techniques
that help patients regain control of their activities and avoid paralysis. So, daily physical activity like
exercising can help to reduce muscle spasticity, joint stiffness, and improve or regain motion, and
continuous monitoring can enable them and their caregivers provide exercise. Our Ideation proposal
May 21st – 22nd, 2022
focuses on the development of a monitoring device for stroke patients with damaged arms and
hands. The patients wear a glove with a flex sensor embedded in their hand and parallel EMG
electrodes on the inside of their fore-arm (flexor carpi radialis, flexor carpi ulnaris and flexor
digitorium superficials). When the angle of rotation of the sensor changes, this flex sensor can
measure angular displacement, resistivity, and voltage. When a patient tries to do an activity, the
angular displacement is detected, and EMG is used to capture muscle activity. These real-time data
are collected and stored in the cloud, where they may be downloaded at any moment using an API or
software (IOT integration).
438. INSTANT VEHICLE INSURANCE RENEWAL SYSTEM

SWAROOP SHELAR, SHANTANU NAGTHANE, KUNAL PAWAR, ANIKET ATAKARE
DEPARTMENT OF COMPUTER ENGINEERING. SAVITRIBAI PHULE PUNE UNIVERISITY, PUNE.
For continuous coverage, every vehicle owner must renew their car's insurance before the due date.
If a person fails to renew their vehicle insurance by the due date, the coverage will immediately
expire. As a result, the policy would terminate the insurance coverage provided under the insurance
plan. The insurance renewal procedure can be arduous at times. It may entail making phone calls to
the insurance agent and personally meeting with them in the insurance office to provide policy
details and renew the insurance. We want to create and develop a system where users can get a fast
quote for their policy renewal by simply uploading a photo of their existing/old policy paper.
439. DATA SCIENCE APROACH TO STOCK PRICE FORECASTING IN INDIA

Himanshu Aggarwal
Electronics & Communication Engineering
Galgotia’s College of Engineering & Technology
Nitin Dubey
Divyam Kumar
Shivam Gupta
Kapil Saini
One of the most challengingKjobs in the field of computers is predicting the stock market. Physical
vs. physiological elements, rational vs. illogicalKconduct, investor emotion, market rumours, and
otherK factors, all play a role in the forecast. All of these factors combineK to make stock values
May 21st – 22nd, 2022
very volatile and difficult to anticipateKaccurately. In this arena, we look into data analysisK as a
game changer. When all information about a firmKand stock market events is promptly available to
all stakeholders/market participants,K the impacts of those events are already embeddedKin the
stock price, according to efficientK market theory. As a result, it is claimed that only the historicalK
spot price reflects all other market occurrences andK may be used. to forecast its future behaviour
As a result, we use Machine Learning (ML) techniques on historical stock priceKdata to infer future
trend, treating the previous stock price as the final expression of all affecting aspects. ML
approachesK have the ability to uncover previously unseen patterns and insights, whichKmay then
be utilised to generate imprecise predictions. We provideK a framework for analysing and
predictingK a company's future growth utilising the LSTM (Long Short Term Memory) modelKand
the net growth calculationKtechnique.
440. IOT CONTROLLED LAWNMOWER

Shivani S, Yaswanthini R,Kanini P,Mrs B Rajalakshmi
The IoT Controlled lawnmower that performs grass trimming and water sprinkling uses modern
technologies that consist of sensors, robotics, and microprocessor for shifting from tedious
operations to advanced automation processes. It consists of two modules, i.e., One for controlling
and another for executing with camera monitoring. An IoT Controlled lawn mower is controlled
through an IoT Web Page which has control features like moving the lawn in all directions,
trimming the grass along with a water sprinkler and the GPS tracking system which shows the
latitude and longitude values. The DC motor is used for movement and navigation which gives
precise control. The camera feature which is present in the executing module gives a clear visual of
the mower’s surroundings. By this, we can see and control the lawnmower from anywhere. The
Ultrasonic Sensor is used for detecting the obstacles in a short range and the mower stops operating
if it detects any object or person. For this, the main requirement is Good Network Connectivity. The
Webpage is then interfaced to the Raspberry pi microcontroller which allows to run and control the
mower while cutting the grass and also sprinkle the water which keeps the grass hydrated.
441. SMART APPLICATION BASED PARKING SYSTEM USING INTERNET OF

THINGS
Dr.Y.Chalapathi Rao1, P.Lavanya2, B.Koteswara Rao3, Sohila.Kanaparthi4
1 Associate Professor, Department of ECE, VNR VJIET, Hyderabad, Telangana, India
2,3Assistant Professor, Department of ECE, MLRITM, Hyderabad, Telangana, India
4Student, Department of ECE, VNR VJIET, Hyderabad, Telangana, India
Parking is a key issue in today's congested cities. With an ever-increasing population and a scarcity
of parking spots, there are just too many vehicles on the road. We require parking management
systems that are effective. We use Internet of Things (IOT) technology to demonstrate the
deployment of a parking management system that allows for efficient parking space utilization. We
use infrared sensors to demonstrate the notion. To detect the presence of a parking spot for internet
access, a Wi-Fi modem is employed, and the system is controlled by a microcontroller. For verifying
the status of parking spaces, we used an open application programme. With the use of infrared
sensors, the system recognizes occupied parking spaces. With the open application software
May 21st – 22nd, 2022
installed, the device updates data stored on the smart phone by reading the number of open and
occupied parking spaces. This allows customers to conveniently search for available parking spaces
online and find unrestricted parking. As a result, the technology aids in the resolution of the parking
problem and provides users with an effective IOT-based parking management system.
442. DESIGN AND DEVELOPMENT OF A NOVEL ACCIDENT DETECTION AND

EMERGENCY REPORTING SYSTEM
Sohila Kanaparthi1, Y Chalapathi Rao2, Divya Kadam3 1,3Student,
Department of ECE, VNR VJIET, Hyderabad, Telangana, India.
2Associate Professor, Department of ECE, VNR VJIET, Hyderabad, Telangana, India.
A novel Accident Detection and Emergency Reporting System (ADERS) is designed to automate
speed detection system which can detect and analyse the speed of vehicles using infrared and
ultrasonic sensors. If the over-speeded vehicles occur, the inbuilt components infrared and ultrasonic
sensors will calculate the corresponding speed, distance and compare with permitted fixed
measurements. In the case of beyond limit, the sensors give a buzzer to the vehicle, driver and
generate a virtual map from Google Maps which automatically sends the information to available
help line centres with the help of GSM (Global System for Mobile communication) or GPRS
(General Packet Radio Service). The complete criteria of vehicle detection are only suitable for
scenario i.e., when the speed of the vehicle is greater than usual or exceeding the speed of an
adjacent vehicle and approaching at less distance in no time gives rise to a crash or emergency. In
such advanced cases, this ADERS built-in system is used. The result hereby shows that the method
successfully detects over-speeded vehicles, differentiates by distance from ultrasonic sensor, and
extracts the map. It can automatically decrease the vehicle speed with the corresponding
specifications. Also, it has more efficient performance which can track and also be deployed on
roads to range for over speeding.
443. BIO-CERAMIC COATING FOR KNEE IMPLANTS FOR BETTER INTERFACE

BETWEEN BONE TISSUES AND TITANIUM METALS
Praveen kumar M1, Geetha balasubramani1* , Rajasekar V2 , Ragavi.M.S3 , Logambika P3
1,2,3IV year student, B.Tech, Biomedical Engineering, Department of Biomedical
Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and
Technology, Avadi
1*Co-Founder & Lead-Operations, Medcuore Medical Solution Private Limited.
3Assistant Professor, Department of Biomedical Engineering, Vel Tech Rangarajan Dr.
Sagunthala R&D Institute of Science and Technology, Avadi
Total Knee arthroplasty or partial knee arthroplasty is conducted when the patients knee region starts
to pain due to osteoarthritis and other bone related diseases, in this surgery a new knee implant
made of metal on metal (titanium, cobalt-chromium) or polymer on metal (polyethylene on
titanium) is used. A huge disadvantage of this kind of knee implants it causes inflammation,
infections due to the metal or polymer debris generated on the implant. Infections / inflammation
caused by bacterium adherence to an implant surface a biofilm formation occurs at the implantation
site, as well as infections caused by metal debris generated from friction and movement of the knee
joint, are referred as implant-associated infections. So, in this project we are developed a bio-
May 21st – 22nd, 2022
ceramic coating on titanium comprising of betaTricalcium phosphate, pectin, gelatine and

polyvinylpyrrolidone on the titanium screw to increase antibacterial effects and biocompatibility of
the implant materials. Bone implant coating will hence a cell growth around the implant and it give
a viable environment for the implanted site. The primary characterization of the coating materials is
done by SEM with EDX, FTIR analysis, in-vitro antibacterial testing, anti-inflammatory testing and
the in-vitro degradation study is done for the determination of stability of the coating. In the above
testing’s it is concluded that our coating materials have an increased antibacterial effects and
biocompatibility in nature. However further In-Vivo testing process is needed to confirm the use of
bio-ceramic coating for knee implant.
444. SMART SURVEILLANCE USING MOBILENETV2

Dhanvanth P
Sri Sairam Engineering college
Chennai, India
Rajkumar S
Chennai, India
Mrs. K. Subhashini
Sri Sairam Engieering College
Chennai, Imdia
Aravind Siddharth L
Chennai, India
Surveillance systems have existed for long and have been executed with humans playing an integral
part in the system. Hence, making the system obsolescent at the absence or heedlessness of the
guard. According to studies, an average adult’s maximum attention span is 20 minutes. With the
growing population and an alarming increase of the physical intrusion gathering the limelight, the
human surveillance systems are in dire need of upgrade. Technology has evolved exponentially
since the invention of the first surveillance system in 1969. Invention of IOT has made it possible to
centralize a surveillance system, making the processing ability of the person assigned as the only
limitation. With the recent advances in Convolutional Neural Networks and formulation of the CNN
models, even a basic classification model capable of detecting a few threats can serve as a decent
live surveillance system.
445. ANOMALY BASED INTRUSION DETECTION SYSTEM USING HYBRID

MACHINE LEARNING ALGORITHMS
S. Revathy,S. Sathya Priya , M.Shakeel
Department of Computer Science & Engineering,
Hindustan Institute of Technology and Science,
Chennai.

May 21st – 22nd, 2022
Intrusion detection technologies have grown in popularity in recent years using machine learning.
The variety of new security attacks are increasing, necessitating the development of effective and
intelligent countermeasures. The existing intrusion detection system (IDS) uses Signature or
Anomaly based detection systems with machine learning algorithms to detect malicious activities.
The Signature-based detection rely only on signatures that have been pre-programmed into the
systems, detect known attacks and cannot detect any new or unusual activity. On the other hand,
Anomaly based detection using supervised machine learning algorithm detects only known threats.
To overcome this challenge, the proposed model is constructed using an unsupervised machine
learning anomaly detection method which uses a combination of Sub Space Clustering and One
Class Support Vector Machine algorithms with feature selection method like Chi-square and Feature
Discretization Methods like Equal Width Discretization to identify both known and undiscovered
assaults. The results of the experiments using proposed model outperforms several of the existing
system in terms of accuracy and detection rate.
446. IOT-BASED INDUSTRIAL ADAPTATION CONTROL FOR WIND ENERGY

CONVERSION SYSTEM
Rathnavel.P¹, Vignesh.M², Logesh.S³, Kirubanandhanan.P⁴
¹Department Of Electrical and Electronic Engineering, Sri Sai Ram Institute Of Technology, Chennai-44.
²Department Of Electrical and Electronic Engineering, Sri Sai Ram Institute Of Technology, Chennai-44.
³Department Of Electrical and Electronic Engineering, Sri Sai Ram Institute Of Technology, Chennai-44.
⁴Department Of Electrical and Electronic Engineering, Sri Sai Ram Institute Of Technology, Chennai-44.
Wind energy is the world's second largest source of renewable energy generation, after only hydro
power, and is also the most extensively employed. According to the International Energy Agency,
wind energy conversion and distribution have experienced a technical revolution as a result of the
development of advanced wind energy conversion systems (WECS), which include multilevel
inverters (MLIs). Ripples are visible in the output waveforms of the traditional rectifier, and the MLI
exhibits voltage balance concerns across the DC-link capacitor, which is likewise a problem with the
conventional rectifier. These designs are based on the "Vienna rectifier" architecture for power
electronics evaluation and design, which can be used for a range of applications, including
variablespeed and continuous sinusoidal current and voltage control, as well as continuous
sinusoidal current and voltage control. This article proposes a simplified proportional-integral (PI)
algorithm for removing output waveform ripples, fixing voltage difficulties, and producing higher-
quality waveforms at the output. The work on this project has resulted in the development of a
significant alternative in the field of high-power medium-voltage energy regulation. Through the use
of a new design, low-harmonic multilayer voltage source inverters with low harmonics are able to
provide enormous voltages without the use of synchronised switching devices or series-connected
transformers.
447. IOT BASED ENVIRONMENTAL MONITORING SYSTEM

Mohanraj V, Soorya R, Saivignesh C R, Sangeeth Gopinath
Sri Venkateswara College of Engineering , Sriperumbudur

May 21st – 22nd, 2022
We propose an environment monitoring system that allows us to monitor and live check the
environment in particular areas through IoT. The System uses air sensors to sense the presence of
harmful gases/compounds in the air and constantly transmit this data to the microcontroller and
reports it to the online server over IoT. The sensors interact with the microcontroller which
processes this data and transmits it over the internet.
448. PROTECTING CROPS FROM WILD AND DOMESTIC ANIMALS INTRUSION

USING DEEP LEARNING
1Murugesan G, 2Abhinivesh K V, and 3Anandh B
St. Joseph’s College of Engineering, Chennai – 600 119, Tamil Nadu, India
Agricultural automation is on the rise, with deep neural networks (DNNs) and the Internet of Things
(IoT) being used to design and implement precise control, monitoring, and tracking systems.
Managing relationships with factors outside of agro ecosystems, such as wildlife, is a relevant open
topic in this rapidly changing landscape. Protecting crops from wild animal attack is one of the key
worries of today's farmers. Traditional approaches to this problem can be fatal (e.g., shot, trap) or
non-lethal (e.g., scarecrow, chemical repellent, organic materials). Some old methods, on the other
hand, pollute both people and ungulates, while others are prohibitively expensive due to high
maintenance costs, lack of reliability, and limited effectiveness. We are working on a system that
leverages AI Computer Vision and DCNN to detect and recognise animals while generating
customized (i.e. different for each species) sonar waves to repel them. If an animal is detected, the
advanced computing device activates the camera and then runs its DCNN software to identify the
target, sending a message to the animal repellent module with the type of ultrasonic wave to be
detect, which is created based on animal species.
449. AUTOMATIC NUMBER PLATE RECOGNITION SYSTEM USING DEEP

LEARNING ALGORITHMS AND IMAGE PROCESSING FOR SURVEILLANCE
Dr. A. Christy Jeba Malar

Associate Professor
Coimbatore, India
Jayasree S
Coimbatore, India
Keerthana G
Coimbatore, India
Deepika S

May 21st – 22nd, 2022

Coimbatore, India
The increase in amount of vehicles in the past few years have made traffic management a difficult
job. Technologies play an important role in these systems to regulate the traffic. Number plates are
distinguished by their shape, bulk, and color in different countries. In India we have norms for
number plates and usually has light color as background and black as the foreground color. In this
paper we have proposed a method to locate the number plate on a vehicle by using histogram
equalization and then identifying the alphabets and digits on the plates using character segmentation
using Sobel and Canny edge detection image processing technique.
450. SYSTEMATIC INVESTIGATION ON CLOUD ASSISTED IOT DEVICES,

CHALLENGES AND APPLICATIONS
Salil Bharany 1 [0000-0002-2282-0419] and Sandeep Sharma2
1,2 CET Department, Guru Nanak Dev University, Amritsar, India
Over the past few years, technology has progressed and developed rapidly, where ad hoc networks
have contributed a significant part to innovative development. There are four varieties of ad hoc
networks such as mobile (MANET), vehicular (VANET), flying (FANET), and sea (SANET). Due
to the variation of specifications, these four ad hoc networks have to turn into an alternative for
providing connectivity in areas where infrastructure-based networks cannot be deployed. Therefore,
this paper sets out a review of these four ad hoc networks, particularly focusing on the
characteristics such as routing, density, and mobility. In addition, the variances between the four ad
hoc networks are discussed with related works. This paper also outlines the challenges to be
addressed in the deployments of these ad hoc networks.
451. CARBON EMISSION CONSIDERATION IN A FUZZY INVENTORY MODEL

UNDER HYBRID PAYMENT POLICY AND TRADE CREDIT FINANCING
A.Thangam*
Department of Mathematics, Pondicherry University – Community College, Lawspet ,
Pondicherry - 605 008, India.
Global manufacturers are sensitive on the emissions of carbon dioxide and other green house
gases while material handling in logistics, transport, production of inventory items and waste
disposal in their industries. Today’s business environment is mostly associated with permissible
delay payments under the situations like Covid-19 Pandemic. In business, the prices of commodities
are imprecise due to fluctuations in production and marketing. The purpose of this study is to model
an inventory system with hybrid payment strategy under fuzzy inventory costs and fuzzy carbon
emission cost. The supplier is offering their retailers permissible delay for the payment of
imperishable inventory items. The retailer’s profit function is defuzzified using signed distance
method. A solution method is proposed to obtain optimal replenishment policies for maximizing the
retailer’s profit. Several managerial insights are also obtained with several numerical examples.
Keywords; Inventory; Fuzzy ; Carbon Emission; Trade credit.
May 21st – 22nd, 2022
452. BUS TICKET BOOKING SYSTEM USING CLOUD COMPUTING

Dr.Su.Suganthi
Associate Professor,
Dept. of CCE,
Sri Sairam Institute of technology,Chennai
Karthik.B
Student (B.E)
Dept of CCE,
Sri Sairam Institute of
Technology,Chennai
Aakaash.M
Student (B.E)
Dept. of CCE,
Sri Sairam Institute of technology,Chennai.
Uthirasree.M
Student (B.E)
Dept of CCE,
Sri Sairam Institute of
Technology,Chennai
Aravindh.T.S,
Student (B.E)
Dept. of CCE
Sri Sairam Institute Of technology, Chennai.
Kishore Kumar.S
Student (B.E)
Dept of CCE, Sri
Sairam Institute of
Technology,Chennai
Sathish.S
Student (B.E)
Dept. of IT
Sri Sairam Institute of technology,Chennai
Presently a day's public local transportation system is as yet involving the customary ways for ticket
booking. Individuals need to stand in queues for extended periods of time. Some peoples travel via
local transport even without buying the ticket. There online bus ticket booking system plays a major
role in easing the process of buying bus ticket for the customer by booking the ticket from where
ever they are. This project is a cloud based online bus ticket booking system. Traditionally, online
bus ticket booking exists by which the customer can book their ticket, but we are adding some other
features such as authenticating the passengers by QR-code while entering the bus, and then we will
know the number of boarded passengers by utilising this software.
453. PERPETUAL PRESERVATION OF PESTICIDE FREE PRODUCTS USING IOT

Ms.G.Manimala, Dhanashanthine M, T.Keerthana, T.Subhashree

May 21st – 22nd, 2022
For the majority of the population, grains constitute the most essential source of daily nutrition. One
of the most important aspects of our economy and our way of life is grain hoarding. A large number
of people in India are responsible for ensuring the quality and security of grain stockpiles. For
example, a large connection scope, increased impedance and lighting strike, and a high care cost
were all disadvantages of traditional grain storage. Sensors and other devices may help to reduce
waste to a manageable level. Because they affect hundreds of millions of people, the storage of
harvested grains is essential to the economies of both developed and developing countries. As a
result, it is critical that food storage facilities, such as FCI, keep an eye on and regulate
environmental conditions. As a grain storage facility, the primary goal is the safety and quality of
the grain that is being kept. Real-time monitoring of faraway sites is now possible thanks to
advances in technology, making it simpler to keep an eye on things wherever you are. In order to
increase grain storage levels and decrease grain losses while reducing manpower and labor intensity
in the process, NODEMCU has been used to build a real-time monitoring system for grain storage.
By using Embedded and WI-FI technologies together, a granary monitoring system may be created.
454.YDR’S BROWSER
Guide Name : Kannagi L
Yogeshwaran M, Dushyant Prabu D, Ramkumaar E T, Manthra K S, Jiffy Renissa A, Rishikumar V
Department Of Computer and Communication
Engineering Sri Sairam Institute ofTechnology West
Tambaram, Chennai-44.
Web browser plays an important role because it provides many valuable applications, such as data
searching, online shopping, news browsing, e-mailing and so on. It is so essential that almost every
smartphone/computer/laptop is equipped with a web browser. One of the reasons that decides users’
purchase decisions is the smoothness of human-browser interaction. This project is aimed to develop
a user friendly browser. The entire project is developed keeping in view of a new browsing
technique. Here when the browser is clicked it starts with the default webpage and then constructs
structures for webpage. Here the results are shown as per the variations are applied. As the internet
is growing very fast, everything that is in the internet is accessible to everyone. So, we have
developed a browser where the user will be able to browse in full screen and can switch over tabs for
browsing two different pages, and this is personalized based upon the search that we intend to give.
This browser is enabled with every functions as the normal browser, and so multiple sites can be
reached without any delay in a single click from its homepage, without any Ad disturbances.
455. MODERN DAY SMART HOME WITH E-BILLING

Mrs. S. Tephillah, M.E., (Ph.D.) Mr. MonishKumar N. J. Mr. NaveenKumar M.
St. Joseph’s Institute of Technology, Chennai, India
The power management system consists of digital power meters installed in every consumer unit and
an electricity e-Billing system at the energy provider side. Wireless sensor network sends its power
usage reading using information back to the energy provider wirelessly. At the power provider side,
they have the control to change priority of the devices when power distributed in low range. Human

May 21st – 22nd, 2022
operator billing is prone to reading error as sometime the houses electric power meter is place in a
location where it is not easily accessible. The concept of dynamic assignment of priorities to
interrupts is discussed which reduces the time delay for a lower priority task which under some
circumstances becomes a higher priority task. In the proposed work, all devices are controlled on
priority and time in order to manage low power automatically.
456. ETHEREUM BASED BLOCKCHAIN FOR SECURED TRANSACTION IN WSN

Ms.C.Ambika Bhuvaneswari1,
Assistant Professor1, Electronics and communication engineering dept. Vel Tech Rangarajan Dr.Sagunthala
R&D
Institute of Science and Technology
Chennai, India
S.Md.Shareef2,S.N.Javeed3,S.Md.Akram4 Student2,3,4, Electronics and communication engineering dept.
Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai, India
Wireless Sensor Networks(WSN) are extremely vulnerable to network assaults, necessitating the
implementation of adequate security measures. The block chain Technology is a reliable method of
safeguarding data processed and held by Wireless sensors networks. Many existing approaches are
based on traditional security principles. Traditional techniques, on the other hand, are more prone to
node failures. In this case, article, a unique and efficient authentication method for Wireless sensor
Networks that uses for security, block chain Technology is used. Wireless sensor network nodes are
IoT devices network They're created by the Basestation Cluster heads , and regular sensors Nodes .A
hierarchical block chain model Will emerge from the construction of a block chain network among
various network nodes, including tiny chain and global chain This hybrid's node identity safe
connection is realised in many communication scenarios, nor-User authentication for mal nodes is in
the Block chain Network . The identification of selected Cluster nodes is verified using technology.
The investigation into protection and outcomes show that the protection is strong and the results are
better. System. The results of the experiments reveal that the computational capabilities increases
the security.
457. DETECTION AND CLASSIFICATION OF CERVICAL CANCER BASED ON

ELEPHANT HERDING OPTIMIZATION ON CONVOLUTIONAL NEURAL
NETWORK
C. Meenu Kumari,"Research Scholar,Annamalai University, Chidambaram,India
Bhavani ,Professor,Annamalai University, Chidambaram,India
S. Padmashree ,Professor,GSSS Institute of Engineering and Technology for Women, Mysuru, India
R. Priya, Professor,Annamalai University, Chidambaram,India
Cervical cancer is one of the leading causes of cancer death in women. Early diagnosis and treatment
of this tumor can reduce its complications. The purpose of this article is to diagnose and classify
cervical cancer using an Elephant Herd Optimization based on Convolutional Neural Network
(EHCNN). First, the image database of the open-source system is collected. The images are sent to a
central filter to remove unwanted noise. This proposed study helps in prediction and classification of
cervical cancer at early stages to avoid complications. As part of this classification, we improved the
weight of the EHCNN algorithm to classify the normal and abnormal cells along with the type of
abnormal cells. This proposed classification method is used to improve the application of CNN
May 21st – 22nd, 2022
design parameters and is selected with the help of EHO. The project shows that the images can be
identified using deep learning techniques and they may have some relevance in the diagnosis of
early cervical cancer. The impact of the proposed technical results is evaluated and used for
performance and image quality analysis in terms of accuracy metrics.
458. STUDY ON EFFECT OF FOUNDATION SOIL ON MASS IRREGULAR

BUILDINGS UNDER SEISMIC LOADS
Ojaswi K S1 and B R Jayalekshmi2
1 National Institute of Technology Karnataka, Surathkal, Karnataka 575025, India
2 National Institute of Technology Karnataka, Surathkal, Karnataka 575025, India
All the commercial buildings and residential buildings are conceptualized as regular buildings while
designing but in reality, no building is ideally regular. All the existing buildings have one or more
types of irregularities. In this study, effect of foundation soil in the response of multistorey building
with mass irregularity is considered by incorporating two types of soil under the raft foundation .
Shear wave velocity of 150m/s and 600m/s are considered representing soft and dense soil.
Transient analysis of three dimensional building frames resting on soil stratum is carried out with
time history of acceleration of a ground motion using finite element software. The seismic response
variation in these buildings for various mass ratios are evaluated and found significant increase in
roof acceleration and base shear if mass irregularity is present at the bottom floors when the
building rests on soft soil.
459. CUSTOMER CHURN PREDICTION USING MACHINE LEARNING

Ujjwal Raj SRM UNIVERSITY
City: Chennai,Country: India
Dr. R.Srinivasan
Assistant Professor SRM,Chennai
Churn is a word that indicates no. of customers an organisation/company loses over a specific time
period .Customer churn is a major issue and one of the most necessary concerns for big organisation
as well as companies. Due to the immediate effect on the revenue of the companies, specially in the
telecom sector, companies are seeking to develop means to predict potential customer to churn. So,
finding factors that increases customer churn is important to take necessary actions to reduce the
churn. The main contribution of work is to develop a churn prediction model which assists telecom
operators or other companies operators to predict customers who are most likely subject to churn.
460. FAKE REVIEW DETECTION SYSTEM

MAYUR SHINDE, ROHIT SHINGATE, TEJAS BADGUJAR, GAUTAM ZINGADE
DEPARTMENT OF COMPUTER ENGINEERING.
SAVITRIBAI PHULE PUNE UNIVERISITY, PUNE.
Both the buyer and the seller benefit from consumer reviews on ecommerce websites, online
resources, ratings, and experience tales. A reviewer can help other customers understand their views
about a product while also increasing its trustworthiness. Similarly, when customers offer good
May 21st – 22nd, 2022
comments on product reviews, it helps marketers increase their revenues by increasing product sales.
Unfortunately, marketers can take advantage of these review approaches. A person may, for
example, fabricate reviews in order to boost a product's reputation or to disparage competitors'
products by leaving poor evaluations on their own. Different machine learning algorithms and
technologies, such as Waka, are used in existing monitored solutions. Emoji and text are used to
create emotional analysis.
461. SMART PORTABLE DUSTBIN WITH METHANE DETECTOR USING NODE

MCU
D. Harish, Y. Anas Tarik, P. Mariselvam, S. Lakshmanan, Dr.L. Kalaivani
National Engineering College, Kovilpatti
The Smart dustbin with a methane detector is node MCU based. Our primary objective is to hold the
surroundings neat and clean. The Internet of Things (iot) can be used to manipulate this smart trash
can. This project is a low-fee demonstration for small-scale projects. Smart waste bin displays the
bin’s methane ranges. This device makes use of a node MCU board and a mq4 sensor. The smart
dustbin is managed by a cell phone and is attached to the internet of things. When methane gas
builds within the trash, the mq4 sensor detects it and sends details to the connected consumer.
462. A SMART CONTRACT BASED ACCESS CONTROL FRAMEWORK

FORCLOUD BASED SMART HEALTHCARE SYSTEM
Renuka Devi B
Assistant Professor,
Mohamed Bilal N
Student,
Kannan S
Student
In current care systems,electronic medical records (EMRs)square measure invariably placed in

several hospitals and controlled by a centralized cloud supplier. However, it ends up in single
purpose of failure as patients being the $64000 owner lose track of their personal and sensitive
EMRs. Hence, this text aims to create AN access management framework supported good contract,
that is constructed on the highest of distributed ledger (blockchain), to secure the sharing of EMRs
among completely different entities concerned within the good care system. For this, we tend to
propose four styles of good contracts for user verification, access authorization, actus reus detection,
and access revocation, severally. during this framework, considering the block size of ledger and
large quantity of patient knowledge, the EMRs square measure hold on in cloud when being
encrypted through the science functions of elliptic curve cryptography (ECC) and Edwardscurve
May 21st – 22nd, 2022
digital signature formula (EdDSA), whereas their corresponding hashes square measure packed into
blockchain. The performance analysis supported a non-public Ethereum system is employed to
verify the potency of planned access management framework within the period good care system.
463. A BRIEF REVIEW OF E-NOSE FOR QUALITY ANALYSIS OF MATERIALS IN

DIFFERENT CATEGORIES
Kumar Vikas1 and Sakshi Arora2
1,2School of Computer Science and Engineering, Shri Mata Vaishno Devi University, Katra, India
In recent days, everyone is sincere about the quality of the daily use items. This review paper
focuses on the quality inspection of materials and components through e-nose based on their aromas
or odors. Quality evaluation through aroma is a challenging process because it is associated with
distinct and unique characterization of volatile organic compounds. E-nose plays an important role
as an instrument to measure and analyze the quality of different materials, so this paper has
categorized them into five parts including nature products or synthesized food, beverages, air
quality, soil and human presence. The study has also emphasized the conventional methods with
compare to the electronic nose. This research has categorized the related work of quality analysis
application of enose and results are analyzed. For pattern identification, numerous machine learning
algorithms such as PCA, SVM, ANN and others are used to determine the quality of substances.
464. SIMULATING SUBJECTIVE OPINIONS ON WOMEN’S SAFETY IN HIGHER

EDUCATION CAMPUSES USING AN ARTIFICIAL INTELLIGENCE-DRIVEN
DESIGN FRAMEWORK.
Samriddho Ghosh, Undergraduate student, Department of Architecture, Jadavpur University
Rupaz Bhadra, Undergraduate student, Department of Architecture, Jadavpur University.
This article aims to establish a theoretical design framework to design urban spaces within higher
educational campuses in India with higher levels of women safety. This is an attempt to include real
time user opinions into a design workflow than just vesting the responsible of calibrating subjective
opinions like perceived safety to the designer. Artificial intelligence and machine vision are the two
bedrock components behind this iterative workflow - imitation of the process of taking subjective
decisions by human beings is carried out with the help of these technologies. This not only removes
biases from the designers’ part but also creates a mathematical definition to subjective perception of
aspects like women safety from a spatial perspective. The framework outlined in the article tries to
explore the spatial factors that account to events related to violation of women safety in higher
educational campus in India. The rising rates of such crimes at one of the most premier locations of
this nation is not just a mere call for action but an issue requiring immediate attention and one of the
possible interventions is described in this research.
465. SYSTEMATIZED VLSI ARCHITECTURE FOR AES ALGORITHM
Dharshinipriya R
Department of Electronics and Communication
Sri Sairam Engineering College Chennai, India

May 21st – 22nd, 2022
Mr.K.Srinivasan. M.Tech (PhD)

Assisant professor
Bhuvanika G
Avula Sri Kaveri
In cryptographic applications, the advanced encryption standard (AES) algorithm is commonly used.
It has been regarded as an efficient VLSI implementation approach when compared to symmetric
cryptography. Using key expansion techniques, this work presents a low-power, small-area,
highefficiency AES implementation. Using the proposed high-performance design, we minimize the
area and critical path latency. Because increasing frequency can result in excessive power
consumption, it is preferable to reduce frequency to improve efficiency. In comparison to past
developments, the number of flip flops, adders, counters, and registers is being reduced. The
bandwidth has been increased to 254.94 Mbps. It has been lowered to 1.028ns. The proposed AES
architectures outperform the competition, according to test findings.
466. AUTO ATTENDANCE SYSTEM FROM SINGLE IMAGE OF THE CLASSROOM

USING FACE RECOGNITION
Mr. Iniyavan1 , Gamini Saikiran2, Haja Mohideen3 and Idris Mohamed4
1 Assistant Professor (Sr.Gr), 2,3,4 B.Tech Student
University of B.S. Abdur Rahman Crescent Institute of Science and Technology
Manual Attendance marking is a typical activity to monitor the presence of students day to day in
all institution organizations at all grades. These approaches are accurate without a chance of marking
false attendance marking but these are time-consuming and laborsome for a large number of
students. To conquer the drawbacks of manual attendance taking, computerized systems are created
utilizing radio frequency identification-based scanning, fingerprint scanning, Face-recognition, and
Iris scanning-based biometric systems. Every framework has its advantages and disadvantages.
Furthermore, these systems experience the ill effects of the restriction of human intercession to
stamp the attendance individually at a time. To beat the impediments of existing manual and
computerized attendance systems, in this work, we propose a strong and productive attendance
checking framework from a solitary gathering picture utilizing face identification and recognition
algorithms. In this framework, a gathering picture is caught from a high-resolution camera mounted
at a fixed location to catch a single group picture for every one of the students sitting in a classroom.
Then, the face images are removed from the gathering picture utilizing a famous Dlib algorithm
followed by recognition utilizing a Microsoft Cognitive Face API prepared on the face information
base of students. We tried our framework for various kinds of group images and sorts of databases.
Our experimental results show that the proposed structure outflanks other attendance marking
systems concerning productivity and usability and execution. The proposed framework is an
independent attendance marking framework that requires less human-machine connection, making it
conceivable to consolidate in a smart classroom easily.

May 21st – 22nd, 2022
467. A USER EXPERIENCE STUDY ON RESPONSIVE MOBILE TOURISM

WEBSITES' PERCEIVED USABILITY, ATTRACTIVENESS, AND INTUITIVENESS
Narayan Mishra and Arivolee VN under the humble guidance of Mrs. Richa Sharma
School of Computer Science and Engineering
Lovely Professional University, Punjab, India
Users increаsingly аccess the internet through their smаrtphones creаting the need for а different
web design аpproаch in order to meet the requirements of new user behаviour. Responsive Web
design (RWD) is аn аpproаch to developing websites dynаmicаlly аnd аdjusting their lаyout аnd
content to the screen size of а user’s device. This study investigаtes the impаct of RWD on
perceived usаbility, perceived аttrаctiveness аnd intuitiveness through а user experience experiment
of two touristic informаtion websites (desktop аnd mobile versions) with vаrying levels of
compliаnce with those guidelines. Results indicаte thаt there аre significаnt differences in perceived
usаbility аnd user experience between desktop computers аnd smаrtphones in generаl. Additionаlly,
mobile informаtion services following а stricter аpproаch to RWD аre perceived аs eаsier аnd more
intuitive to use, yet they fаil to creаte emotion аnd consequently аre less recommended by users.
468. REAL TIME CONVEYOR GUARD MONITORING SYSTEM USING IOT

Dr.V.Ezhilmaran ,Mohamed Naveen Shamsudeen.N
Department of mechanical enginnering, Mepco Schelenk Engineering college Virdhungar Dt,India
In industries it is reported that hundreds of accidents occur each year is machinery related. And the
incidents involved is mainly due to stationary machines such as conveyors which continue to be
among the most frequent. It is known that accidents in conveyors have occurred frequently where
the severity of the accident may be mild or it may be fatal. It is observed that the guard used in
conveyors have been bypassed or removed for maintenance purposes have not been replaced in its
existing place. Where due to absence of conveyor guard’s accidents have occurred. To prevent these
types machine guard misplacement and bypassing of guards a system has been designed and
proposed to develop a continuous monitoring system for conveyor guards consisting of limit switch
and magnetic switch sensors connected with the compact controller also wirelessly transmitting
information about the conveyor’s machine guard’s status to the control room. Also, an Infrared red
sensor has been placed to monitor the operation of the conveyor. The data transfer for alert
notification is wireless with IoT communication system and displayed where the real time guard
monitor is done using the MATLAB GUI in case of bypass of guard a mail notification is sent to the
concerned department. The prototype model developed is a wireless IoT solution for monitoring
conveyor guards during operation and maintenance time of work. The primary goals are to provide a
real-time monitoring of machine guards to facilitate the planning and execution of safety procedures
during maintenance and operation of conveyors.
469. AGRO DIRECT SELLING APP WITH VOICE BASED CHATBOT

1Priyankadevi.A.M, 2Dr.Suresh.P, 3Dr.Aswathy.RH
1PG Student, 2Associate Professor, 3Assistant Professor(SI.G), Computer Science and Engineering, KPR
Institute of Engineering and Technology, Coimbatore

May 21st – 22nd, 2022
The project entitled “Agro direct selling app with voice based chatbot” has been made for giving
offer aid to agriculturists to scatter incredible taken a toll between farmers and retailers. This app is
utilized for keeping up centralized store of all the information related to agriculturist and them
facilitate advertising things like characteristic items and vegetables. This Extend gives a stage for
wholesalers and retailers a simple communication. The framework points to create this Chabot as an
intuitively framework so that ranchers can utilize it to offer their deliver specifically to customers
who may require it. The foremost neutral of this application is to construct an app which will help
the farmer to get the best from his inputs, and customers to offer and buy the develop things that are
required without the intrigued of go betweens at a sensible and beneficial taken a toll. The system
provides us live watching of crops besides the live following of transports and harvests from grower
to consumer. The app has been made utilizing Android and SQLite. App will detect disease and
NLP is used for fruit and vegetable display through voice.
470. SIGN LANGUAGE INTO TEXT CONVERTER

#1 Dr.T.P.Rani, #2M.Priyadarshini,#3S.Priyadharshini
#1Assistant professor, Department Of Information Technology
#2#3Student, Department Of Information Technology
#1#2#3Sri Sairam Engineering College, West Tambaram, Chennai-600044
Communication is very important and basic need for human beings .It is a form of receiving or
sending information or messages to express emotions or to give information. But there are some
people in our society who was dumb and deaf . These people are facing many challenges while
communicating with others in their sign language which is difficult to understand . Sign language is
one of the commonly used method for communication. A translator is required for communication
between the person who knows the sign language and to whom they want convey their message.
This will help deaf and dumb people to communicate with others effortlessly without an interpreter.
471. E-WASTE CLASSIFICATION AND SORTING USING IOT

Dr Himanshu Shekhar
Department of ECE
Hindustan Institute of Technology and Science Chennai, India
K Indrasena Reddy
Department of ECE
Zahoor Almas
Department of ECE
B Naveen
Department of ECE
The main theme of this project is to reduce the environmental and health hazards caused by the
Ewaste. As we know small change can bring the large difference in the outcome same, we are
applying in this project. At present daily labours or workers are working in classifying the e-waste
i.e., manual sorting even though they take precautions by wearing safety dress, gloves, shoes and
May 21st – 22nd, 2022
masks some of them are attacked with different diseases as we know that there are so many health
hazards from ewaste. We get diseases easily if large amounts of normal wastes are dumped nearby
and I guess you know the differences of normal waste and e-waste. Now coming to environment if
we are dumping the e-waste even though some of its components can be reused and this is
happening due lack of information about the reusability not all but maximum people some of the
people who work in electronic shops like mobile shops, refrigerator, washing machine etc. they are
able to reuse some of its components my point here in reusing the small components will make a
small difference in environment. We have projects that classify and sort based on the type of waste
like plastic, glass, metal, dry, wet, biodegradable, non-biodegradable, etc. so we are focusing on e-
waste and in e-waste especially on pcb’s that too taking pcb image as input as we wanted to make a
real time project so we are using raspberry pi, conveyor belt for moment of input and then camera
for taking input image then raspberry pi process the image and classifies and then sorting using
servo motor and at last update this on web page. In web page we will not only able to see the
classified pcb image but the no of pcb’s collected at the blocks. The web page is to get the record of
the pcb’s and then sell them to the local electronic shops to reuse the components and use the money
to collect the waste from different locations where e-waste is dumped and no one cares about it.
472. SOLAR POWERED POND CLEANING MACHINE

V.Kumarasan 1 R. Raajeshkumaran 2 S.Sathyanayaki3 K.Vignesh4 S.Vijay Somasundaram 5
1 Assistant Professor (Senior Grade), 2,3,4,5 U.G Students, Department of Electrical and Electronics
Engineering, Kongu Engineering College, Perundurai,Erode India
Planet Earth is probably the only known planet with water in it. Almost 70% of the planet is covered
by water. This makes water an invaluable asset for life. But nowadays water bodies are vastly
polluted, especially by human activities. This affects the aquatic ecosystem and the lives that are
indirectly dependent on the water bodies. A major part of the pollution is due to solid wastes and
debris that are thrown in the water. In most cases, these wastes are left as such or sometimes cleaned
manually. Former is harmful to environment whereas the latter may risk human lives. Also it may
involve a huge effort to clean the water system if done manually. The government of India has taken
charge to clean ponds and invest huge capital in many pond cleaning projects like “Namami Gange”,
“Narmada Bachao’’ and many major and medium projects in various cities like Ahmadabad,
Varanasi etc. By taking this into consideration, this system has been designed to clear those floating
debris from the water bodies in an efficient way without risking human lives. Solar panels are used
as the power source for batteries which help to operate the motors, conveyer belt etc to collect the
waste. ESP Module and Motor driver are used to control the movement of the system inside the
water. Ultimately the p roject envisions making a way to clean the water body with the help of solar
power in an efficient way.
473. A STUDY ON METHODS AND DEVELOPMENT OF SELF LEARNING

PORTAL FOR MATHEMATICS ENTHUSIASTS
Dr S. Alamelu1, S. Gayathri2, A. Jayashree3, S. Muthushalini4, B. Vaishnavi5
1,2,3,4,5SHRIMATHI DEVKUNVAR NANALAL BHATT VAISHNAV COLLEGE FOR WOMEN
Self Learning has gained unprecedented momentum during the pandemic. Our Indian Government is
making continuous efforts to provide facilities for teaching – learning and materials suitable for Self
May 21st – 22nd, 2022
Learning.The digital initiatives like SWAYAM, SWAYAM Prabha, E pg pathshala have proved
beneficial to a suitable number of students in far flung areas of the country. The main objective of
this work is to create a portal named ‘QUARTILE’ which is aimed at developing a ‘Self Learning
Portal For Mathematics Enthusiasts’ that hosts a wide range of information on various Mathematical
topics in different forms of media.The proposed Self Learning Portal is targeted for the attainment of
additional knowledge through
Listen and Learn : Audio Watch and Learn :
Video
Read and Learn : E-Materials
This portal is not only limited to Mathematics students but also to students of other disciplines who
wish to have knowledge in Mathematics.
474. HUMAN COMPUTER INTERACTION: CONTROLLING MOUSE

FUNCTIONALITIES USING HAND GESTURES
Rajarshi Banerjee1, Tiyasa Khan2, and Anil Kumar Gupta3
1Department of Applied Optics and Photonics, University of Calcutta, Kolkata, West Bengal, India 2Indian
Institute of Engineering Science and Technology (IIEST), Shibpur, West Bengal, India 3Center for
Development of Advance Computing (CDAC), Pune, Maharashtra, India
A person is made up of and interacts with their immediate surrounding through gestures. Most
physical impairments leave a minimum of one physical gesture, which can be further tuned to make
interaction with the computer for those with disabilities easier. This paper exploits what hand
gestures can do to make their lives easier. We have combined our knowledge of Euclidian distance
and available machine learning modules to build a package which reads hand gestures and performs
mouse functions. The said Google’s handmediapipe library detects Euclidian distance and relative
positioning of the hand gestures. The resulting positions, called handlandmarks, have been
mathematically manipulated to give the desired output.
475. RECOGNITION OF WHEAT LEAF DISEASE USING TENSORFLOW AND

KERAS
*Mrs.P.Sheela Rani, **V.Archanaa, **K.Laila, **M.Swathi Sree

*AsstProfessor, Department of Information Technology, Panimalar Institute of Technology, Chennai, India
**Student, Department of Information Technology, Panimalar Institute of Technology, Chennai, India
Wheat is one of the major crops harvested in India. Wheat is considered to be important as they are
the source of energy supply to mankind. There are several diseases that affect the wheat leaves,
which result in a significant loss in crop production. In the early days, farmers were able to identify
plant diseases manually since they had knowledge about agriculture. Nowadays young people who
indulge in smart farming do not have great knowledge about agriculture. In this paper, we are using
deep learning techniques to detect wheat leaf disease. Using CNN (Convolutional Neural Network)

May 21st – 22nd, 2022
algorithm we are able to recognize the images of wheat leaves. Here there are three classification
namely Healthy, Septoria, Stripe-Rust. To find the accuracy of the wheat leaf disease, two modules
i.e.AlexNet & LeNet are used in this system. Django framework is used for the deployment process
which provides a better user interface.
476. SMART ONLINE BOOKING SYSTEM THROUGH HAND GESTURE

RECOGNITION
1Murugesan G, 2Hari Krishnan, 3Hari Prasad3
Department of Computer Science and Engineering, St. Joseph’s College of Engineering, Chennai – 600 119,
Tamil Nadu, India
Online railway reservation system inquiries about the trains available on the basis of source and
destination, booking and cancellation of tickets. It is the computerized system of reserving seats on
trains in advance. There are various ways of booking online railway tickets but the disadvantage is
we need to physically interact with the system to reserve the tickets. Gesture - controlled laptops and
computers have recently gained a lot of attraction. Hand gestures are the most natural and effortless
manner of communicating. The virtual mouse provides an infrastructure between the user and the
system using a webcam or an inbuilt camera for capturing of hand gestures and hand tip detection
using computer vision. So the hand gestures are used to control the mouse moves and perform
clicking without any electronic equipment. The Mediapipe algorithm is used in the system for
detecting the hand gestures and movements of the hand. The algorithm is based on computer vision
for detecting hands. Hence, the proposed system will avoid COVID-19 spread by eliminating the
human intervention and dependency of devices to control the computer for booking the railway
tickets
477. ANALYSIS ON ELECTRONIC SKIN

Dr . Balaji Madhavan ¹, Sri Rakshaga S R ²
1. Professor, Agni College of Technology
2. Second year ECE, Agni College of Technology
The latest advancements in electronic skin (e-skin) research are discussed in detail, with an emphasis
on the technology required for three key applications: skin-attachable electronics, robotics, and
prosthetics. Electronics are crucial in the development of simple devices for a variety of
applications. Electronic equipment is necessary for all fields. Artificial Skin is the finest
achievement as well as a future example of integrated electronics in the medical industry. It's a
super-thin electronic gadget that adheres to the skin like a tattoo and measures the electrical activity
of the heart, brain waves, and other critical information. Artificial skin is a skin created in a
laboratory. It can be utilised in robotic applications or as a skin substitute for persons who have had
serious burns or skin illnesses. This study focuses on the Artificial Skin (E-Skin), which is used to
create a skin that works similarly to human skin and is also implanted with many feelings or senses
of touch that operate on the skin. This skin is being put together right now. It is made up of millions
of implanted electronic measuring devices, including thermostats, pressure gauges, pollution
detectors, cameras, microphones, glucose sensors, EKGs, and electronic holography. This gadget
would boost developing technologies and considerably expand the use of robotic probes in locations
where humans unable to go. The sensor might open the way for a slew of new apps that can
May 21st – 22nd, 2022
remotely monitor a patient's vitals and bodily movements, sending data directly to a computer where
it can be logged and stored to help with future choices. These issues are discussed, with an emphasis
on progress, present obstacles, and future potential
478. LUNG NODULE MALIGNANCY CLASSIFICATION USING MACHINE

LEARNING TECHNIQUE CNN ALGORITHM
Mr.A.Siva Kumar
Chennai, India
Surya Prakash K
Chennai, India
Pasala Dinesh kumar Reddy
Chennai, India
Lung cancer corresponds to 29% of all deaths due to cancer in 2021,estimated more than 1.8million
of deaths globally.Considering this challenging situation, several computer aided diagnosis systems
have been developed to detect lung cancer at early stages, which increases the patients’ survival
rate.In the history of medical imaging, a number of computer-aided diagnostic (CAD) systems have
lately been introduced to assist radiologists with their patients.. For Numerous computer-aided
diagnosis (CAD) before many of them used so many algorithms. But it wont give accurate
results.For full assistance of radiologists and better analysis of magnetic resonance imaging (MRI),
multi-grade classification of lung nodule malignancy is an essential procedure. We propose a novel
convolutional neural network (CNN) based multi-grade lung cancer classification
system.Convolutional neural network (CNN) is an latest algorithm which gives accurate results in
image processing.
479. DEVELOPMENT OF AUTOMATIC POSTAL PARCEL COLLECTER WITH

PINCODE BASED SEGREGATION
Mahesh N1 , Riswan Ahamed M I2 , Sowndharyan S3, Vignesh A4
Associate Professor, UG Scholar Electronics Instrumentation Engineering,
Kongu Engineering college, Perundurai, Erode.
Now a days the amount of parcel delivered in a day has increased arithmetically. So, it is difficult to
sort the parcel according to the places to which it has to be delivered. An individual has to weigh the
parcel and calculate the amount to be collected for the parcel and making fee collection, followed by
collection of parcels. In order to eliminate this complicated process, the project makes the process
automated by initially weighing the parcel and ask user to enter the pin code. Based on pin code the
parcels are sorted into different boxes and the respective price for the parcel based on the distance
and weight is displayed in the screen and customer can easily hand over the amount to the office.

May 21st – 22nd, 2022
The project will make the process ease by sorting it into different slots and payment method will be
simplified.
480. SEGMENTATION AND CLASSIFICATION APPROACH TO IMPROVE BREAST

CANCER SCREENING USING HISTOPATHOLOGICAL IMAGES
Anupama Bhan1, Simone Singh2, Sudaksh Puri3
Electronics and Communication Department,
Amity School of Engineering and Technology Noida,
Amity University Uttar Pradesh
The most common type of cancer among women is breast cancer. The early diagnosis is crucial in a
treatment process. The radiology support system in the diagnostic process allows faster and more
accurate radiographic contouring. The aim of the paper is to enhance the result of the detection and
more accurate results could be generated. Histopathological images were used for training and
testing deep learning models. Primarily two deep learning models have been successfully
implemented, namely RESNET 50 and Keras Tensorflow. These models have helped in attaining
desirable results with accuracy of 98.3% and 99.9% respectively. Alongwith that improvements in
precision 92.7% and 93.4% for respective models was obtained. These values better help in
understanding the working and the scope of improvement in the models that were deployed.
Satisfactory numerical results as well as GUI images have been attached in this project report.
481. SMART SOLAR PANEL CLEANING SYSTEM USING IOT

(Dr Manoj Kumar P*, Kalaiarasan A [1] , Dinesh B S [2] , Gunapal C [3],
Shandeesh R S [4], Rithikanth B C [5] )
1, 2,3,4,5 -B. E Final Year Students
*-Assistant Professor Level III
Department of Electrical and Electronics Engineering
Bannari Amman Institute of Technology, Sathyamangalam-638 401, Erode District
In recent days demand for electricity has increased, gradually along with the development of
technology especially in the field of E-vehicles. Due to gradually increasing the electricity demand
people started moving to the field of renewable energy sources. Among the renewable energy
sources solar PV modules are mostly preferred, nearly 70% consumption of renewable energy
sources. The accumulation of dust particles on the solar PV modules can reduce the power supply
production from the module. In order to clean the panel effectively, here employing a cleaning
system using Node MCU ESP8266 which has an inbuilt wi-fi module to monitor the system using
IOT.
482. MUSIC RECOMMENDATION SYSTEM USING SENTIMENT ANALYSIS

THROUGH HAAR CASCADE IN MACHINE LEARNING
1Murugesan G, 2Gowtham K, and 3Balaji N
May 21st – 22nd, 2022
St. Joseph’s College of Engineering, Chennai, Tamil Nadu, India
Music streaming services now make it simple to listen to a wide variety of music. Consumers are
increasingly relying on recommendation systems to help them choose appropriate music at all times.
However, there is certainly room for improvement in terms of customization and emotion-based
suggestions. Furthermore, music tastes will change dependent on the users' current mood. If these
issues are not solved, these online services will fail to meet user expectations. In this project, we
show how to create a personalized music recommendation system based on listener thoughts,
emotions, and facial expressions. A recommendation system is being created using a combination of
artificial intelligence technology and generalized music therapy approaches to help people choose
music for different life situations while maintaining their mental and physical health.
483. FINGER VEIN RECOGNITION BASED ON DEEP NEURAL NETWORKS

Dr. Sudhakar R1,Thilagavathi S2 ,Abirami P3,Vigneshwaran M3,Kavin Sundari R3
1Head of Department, Department of ECE, Dr. Mahalingam college of Engineering and Technology, Tamil
Nadu, India
2Assistant Professor, Department of ECE, Dr. Mahalingam college of Engineering and Technology, Tamil
Nadu, India
3Students, Department of ECE, Dr. Mahalingam college of Engineering and Technology, Tamil Nadu, India
In this paper finger vein recognition is achieved with more accuracy in comparison with other
biometrics. The accuracy and working of other biometric systems like finger print, iris may be
affected by the skin condition of people. The other potential risk in biometrics includes tracking the
data from the public with or without their knowledge may also occur. To solve this kind of issue a
method of finger vein recognition based on deep neural networks is proposed. This technique is used
to analyze the patterns of blood vessels which are visible from the surface of the skin of fingers. The
system can also provide the enhanced security from fraudulent activities by identifying the blood
flow in the veins to ensure the person is alive or not. The main purpose of this project is to propose a
method of finger-vein identification system based on Convolutional neural networks(CNN) which
deals with finger-vein images of different quality. This system is used to produce stable and highly
accurate performance and also improved authentication speed. The input image is compared with the
images in the database. If it is matched with the existing image, the system will give the positive
response for authentication otherwise vice versa . This system is used for Biometric Authentication
at Hotels, Security purpose in Schools, Colleges and also secured unlocking in Mobile phones.
484. EFFECTIVE UTILIZATION OF FOOD WASTE AS CONSUMABLE

RESOURCES
Deepamahi Boddapu1, Madda Susmitha1, Biswal Pavithra Kumar1, Ch. Sekhar1
1Vignan’s Institute of Information Technology, Duvvada, Visakhapatnam.
In recent days, food waste is discharged daily from factories, restaurants, supermarkets, etc. This
waste has potential to be reused as an organic resource providing it can be transformed with the
correct technology. Collection of the food waste from such areas and categorizing them to what they
can be used into is a major task .It is found that there is discharge of food waste daily from various
May 21st – 22nd, 2022
types of places i.e. restaurants, factories, supermarkets etc. Most of this food waste are either
dumped or destroyed by some means. We want this food waste to be reused in such a manner that it
can be helpful for the society. Food waste reusability can highly boost the economical sector of the
country. We wanted to reuse food waste as a consumable organic resource or ideas in Digital
management or Effective Use of Waste Material for the Food and Beverage Industry (e.g. Factory,
Restaurant, Supermarket, and Warehouse, Building and so on) and its supply chain. Some issues
which could be tackled under this category include Biogas power generation, Compost,
Transforming food waste into animal feed.
485. AN ALTERNATIVE METHOD FOR NON-INVASIVE GLUCOSE DETECTION

USING FTIR SPECTROSCOPY
Sameera Fathimal M1, Dr. Angeline Kirubha SP2*, Jeya Prabha A3, Jothiraj S4
1,2,3 Department of Biomedical Engineering, SRM Institute of Science and Technology, Chennai, Tamilnadu,
India
4 Department of Biomedical Engineering, Rajalakshmi Engineering College, Chennai, Tamilnadu, India
Measurement of glycated hemoglobin in the blood samples collected by puncturing the vein is
considered as gold standard in the diagnosis and prognosis of diabetes mellitus (DM). Uninterrupted
and non-intrusive monitoring of glycemia is essential while the frequency of drawing blood for
glucose quantification is utmost in a brief time period. Hair, nail and urine possess the potential to
reveal the metabolic information in patients with hyperglycaemia that facilitates efficient
observation of DM. Spectroscopic analysis of the collected samples is feasible for early diagnosis as
it is possible to discriminate the diabetic from healthy subjects. The optical properties of micro-
samples can be examined with Fourier Transform Infrared Spectroscopy (FTIR) which is capable to
minimize the sample utilization. Reflection, absorption, or transmission phenomena alter the
incident light as it interacts with the sample. A spectrometer measures this change over a range of
incident wavelengths of electromagnetic radiation that has interacted with a sample. Based on the
energy of the excitation source, the molecular structure will either vibrate or absorb and emit
photons that produces a spectrum. This spectrum is capable to distinguish the diabetic and normal
subjects. In the proposed work, hair, nail, saliva and urine samples will be collected from the
hyperglycaemic patients as well as healthy controls. The samples were pre-processed prior to further
examination. Then the specimen was analysed by FTIR so as to differentiate diabetic and non-
diabetic subjects.
486. USABILITY STUDY ON VIRTUAL FITNESS APPLICATION USING

ARTIFICIAL INTELLIGENCE AND DEEP LEARNING
K J Mohamed Dhanish1, S Mohammed Ijaz2, Dr. Parnasree Chakraborty3, Ms. A. Ambika4,
1,2 -B. Tech Student | 3,4, Assistant Professor (Sr. Gr),
B.S. Abdur Rahman Crescent Institute of Science and Technology
When the Pandemic struck the world, the normal-flow gets halt. Public spots have been closed to
stay away from enormous social events. One such spot is Gym, which has made it hard for countless
individuals to stay aware of their wellness. Because of the awful place of disengagement, individuals
May 21st – 22nd, 2022
are battling with different wellbeing related issues, for example, weight, unpredictable rest designs,
eye strain, mental pressure, diminished invulnerability, and subsequently, are at a higher gamble of
getting tainted with the Coronavirus. As the gym centres are shut and individuals don't have the
necessary direction, an option is required. We learnt a lesson that we need to create a new-normal
lifestyle which later became the everyday routine, and now the ‘Future of Fitness’. In this quest for
adjusting to present day standards, an answer with gamified ideas is expected to assist individuals
with their wellness routine and keep them ready for day-to-day workout. So, we came up with the
idea of Creating a Virtual Fitness and Lifestyle Studio which gives life to our dream and makes it a
reality, but virtually.
487. ANALYTICAL DEVELOPMENT OF MICRO COMBUSTION CHAMBER
Bhavik S Barot
Department of Mechanical Engineering
C. K. Pithawala College of Engineering & Tech.
Gujarat, India
Dr. Digvijay B Kulshreshtha
Associate Professor, Dean Research & Development
C. K. Pithawala College of Engineering & Tech.
Gujarat, India
The gas turbine technology is facing challenges in terms of increased power density of the engine,
which is related to high compression ratio engines and high turbine inlet temperature. With the high
temperatures at turbine inlet, issues with emissions have increased, especially NOx. This has
compelled the designers and engineers to think of alternative methods to reduce emissions without
deteriorating performance from combustion devices (emission in terms of NOx increases as
Temperature levels increases). This has led to development of many technologies, namely DLN
(Direct Low NOx), LPP (Lean Premixed Pre-vaporized), RQL (Rich burn quick mix lean burn) and
Micro mix combustion technologies. The main aim of all these technologies is to reduce the flame
temperature and thereby reduce NOx emissions.
488. CUSTOMER SALES PREDICTION USING MACHINE LEARNING

Aniket Shukla And P. Selvaraj*
Department of Computing Technologies, College of Engineering and Technology, Faculty of Engineering and
Technology, SRM Institute of Science and Technology, Kanankulathur, 603203, Tamilnadu, India
Customer sales prediction is the process of estimating a future value based on previous values and a
variety of other criteria. I used the Keras functional API to develop a sales forecasting model in this
model. It involves calculating current or prospective sales utilising data such as previous sales,
seasonality, holidays, economic situations, and so on. As a result, after being given a set of inputs,
this model will forecast future sales of a specific product/item.Forecasting sales is a crucial part of
modern market intelligence. A solid revenue projection can help a business save money on
superfluous products, better plan for the future, and enhance profits. Grocery sales forecasting is
linked to anticipating the potential sales of establishments including supermarkets, retail outlets, and

May 21st – 22nd, 2022
bakeries. It allows firms to effectively divide cash, estimate realistic sales revenue, and plan a
stronger strategy for the store's future growth.
489. A SYSTEMATIC ASSISTANCE FOR FOOD WASTAGE REDUCTION USING

MOBILE APPLICATION
Suganya M1, Gayathri L2,Thanushri T3,Divya Dharshini N4,Divya E5,Janani U6,Monalisa B7

1Assistant Professor, 2,3,4,5,6,7Student,Department of Computer and Communication Engineering,
Sri Sairam Institute of Technology
Food is one of the basic necessities of humans, and it stands first among all basic needs- food,
shelter, and clothing. It is important as it nourishes the human body- sustaining the very existences
of humans. However, with the rising population and development of this country,food wastage has
risen to a new high.Wasting food is a common problem in our society. Food waste management is
crucial since it can improve our environmental and economic sustainability. There are many people
who wish to donate food to the needy but are unaware of how exactly they can execute that. We
have identified the use of mobile technology to reduce food waste management and built an android
mobile application that allows restaurants to donate and share their left over food with people
indeed. The application for food donation acts as an interface between the users who are looking for
a channel to give the excess food without wasting it.It is targeted in two ways, the user who is
donating the food and the person/organization who is claiming the food. The visualization impact of
the donation can create a positive impact on the users. Minimizing food wastage and feeding the
hunger is the main goal of the project.
490. LOCALIZATION APPROACH USING BLE AND GSM

Bala Sakthivel G, Akashraj D, Sai Prashanth G, Dr Raja J
Sri Sairam Engineering College affiliated to Anna University
Chennai ,India Chennai ,India Chennai ,India Chennai ,India
Wireless Localization has been a key field of research ever since the advent of Wireless Sensor
Networks (WSNs) in mobile handheld devices for almost a decade. It provides a wide range of
applications ranging from day-to-day activities like person/object tracking to remote area
localization. In this project, a Bluetooth Low Energy (BLE) and Global Systems for Mobile (GSM)
based location sensing system is proposed. The system uses Received Signal Strength Indicator
(RSSI) function of Bluetooth 5.0 in case of indoor localization to determine the signal strength
between connected node and the BLE device to determine the distance between them by using
appropriate mathematical calculations. The outdoor sensing is done by using the Wireless Access
Points (WAPs) to locate at nearest accuracy of the required object. This Novel approach brings out
the tracking system with ease of users by working in any weather condition and all around the year
without any environmental dependencies. The proposed method of remote localizing is expected to
outplay the existing contemporary methods by using suitable modern algorithms.

May 21st – 22nd, 2022
491. A SURVEY ON CHANNEL ESTIMATION IN mmWAVE MASSIVE MIMO

SYSTEMS
1 Dr. M.Vijaya Lakshmi, 2 Udataneni Divya,
1Assistant Professor, 2 Student,
Dept of ETE, G. Narayanamma institute of science and technology (for women) Hyderabad
Massive Multi-Input Multi-Output (MIMO) systems plays an important role in 5G communications.

In massive MIMO, a base station with large antennas simultaneously uses the same frequency and
time slots by multiple users, increasing the spectrum efficiency. The combination of mmWave with
Massive MIMO systems is usually used to reduce complexity and cost, but channel estimation is a
crucial task. In this paper, the channel estimation of mmWave Massive MIMO systems is studied
using different techniques and the performance is compared with deep learning-based channel
estimation.
492. PERFORMANCE ANALYSIS OF IMAGE RESTORATION FILTERS FOR

TEXTUAL IMAGES
1Grace Kuriakose, 2Remya Ajai A S
Department of Electronics and Communications Engineering,
Amrita Vishwa Vidyapeetham, Amritapuri, India
In the modern era, images are one of the most important media used to store, transmit and analyze
the information within it. True images often get degraded during the acquisition process due to
environmental disturbances. Image Restoration techniques can be applied to restore such images to
extract the information contained within the original data. Over the past years, several restoration
techniques have been developed. In this paper, analysis of some of those techniques which are
Wiener Filter, Richardson-Lucy Algorithm, Inverse Filter and Median Filter are performed for blur
and noisy blur images. These methods were used to restore the degraded images and further, their
performances were compared using factors such as Power Signal to Noise Ratio (PSNR), Structural
Similarity Index (SSI), and Mean Squared Error (MSE). The algorithms were implemented on
MATLAB(2020a) and the results show that Wiener Filter gives the best results for even blurred and
noisy images.
493. PERKSUITY – GET CONTROL AND BENEFITS FOR YOUR DATA

Name: Aman Sarda (member)
Department: Computer Engineering
College: G H Raisoni College of Engineering and Management, Pune, India
Name: Shubham Kachore (member)
Name: Manasi Garad (member)
Name: Manjushri Mahajan (Mentor)
May 21st – 22nd, 2022

Name: Amey Ghorpade (member)
The open secret within the digital advertising industry is that it's built on exploiting our data like
when you search for something on the internet or mention something to a friend and the next thing
you know you're being bombarded with advertisements for it. Everything that we do online is being
tracked to follow our movements and understand our behavior. Then they collect this information
and sell it for a huge amount making a tremendous profit.But, we believe that people should have a
choice to stop this from happening or even better to earn from it themselves.With this belief,
“PerkSuity”, is a platform/browser for users to enjoy internet browsing safely and securely with the
benefits of privacy and rewards. The main aim of this system is to give power to users for
monitoring and control personal data over the internet. Users’ privacy is maintained as users are
willing to share their preferences for the advertisement data and hence no need to track additional
personal data/activity.
494. TRAFFIC SIGN RECOGNITION AND ACCIDENT PRONE ZONE INDICATION

IN AUTONOMOUS CAR
A.Subbhapriya
PG Scholar,
ECE(Industry Integrated),
NITTTR,Chennai,India
The overarching goal of the automotive industry is to improve the safety of driving machinery and to
prevent accidents. Traffic laws and regulations written by law are not followed by many citizens.
Many citizens do not follow the rules and regulations of the road written by the law. This is another
reason for the accident. Sometimes accidents are unintentional. Other serious actions such as
drunkenness and driving, ignoring, or misinterpreting signposts, and speeding can cause serious
accidents. To avoid such situations, we need an Advanced Driver Assistance System (ADAS) in
private cars. An Autonomous vehicle (AV) also known as a self driving vehicle, non-motor vehicle
or robot vehicle is a ground vehicle that can sense its location and drive safely without user
interference. ADAS has the potential to increase safety and provide comfortable driving. Driving
conditions are electrically controlled and decisions are made easier for the driver. Older people can
also get many benefits from this technology. ADAS is built with a personal communication device
that often improves road safety slightly. Accidents caused by human error can also be minimized.
ADAS helps drivers to automate, adapt and improve the car's safe driving system. Fixed safety
technologies such as installing seat belts and airbags cannot prevent road deaths. ADAS also informs
the driver of potential problems and helps maintain the stability of the vehicle under critical
conditions. Safety features are designed to control the car during a crash. ADAS relies on input from
road sign identification, etc. Additional features can be customized based on the needs of the driver.
In this project measures to prevent speeding, collisions between vehicles,and driver warning
systems. Mathematical models such as image processing techniques and neural networks are used to
sense the speed limit on signal boards. Vehicle speed is controlled based on readings obtained from
Signs. Drivers will receive a warning message as they approach an increasingly dangerous area, and

May 21st – 22nd, 2022
the vehicle will maintain a steady speed. The proposed system is designed to achieve maximum
accuracy and real-time limits.
495. INVESTIGATION OF FAULTLESS STRESS LEVEL MONITORING SYSTEM

USING ARTIFICIAL ALGORITHMS.
Deepika B*
, Supraja G^
, Dr. Jeyalakshmi V$
*PG Student, Department of ECE, Anna University, Chennai, India
^Research Scholar, Department of ECE, Anna University, Chennai, India
$Professor, Department of ECE, Anna University, Chennai, India
In modern life, stress has emerged as frequent trouble for everyone, even from an earlier age, it has a
massive poor effect on people's health and daily activities. Stress is an inability or excessive feeling
to cope with intellectual or emotional stress. Stress motives chemical changes in the human body,
which can increase blood pressure, coronary heart rate, and blood sugar levels. Stress can be at a
severe acute or chronic stage. Often these prerequisites had been identified using emotions of
frustration, anxiety, anger, or depression. Therefore, it has to be controlled, otherwise, it will reason
a distinct variety of risky suffering. The major goal of this paper is to prevent stress-related risky
health problems by designing an IOT-based stress detector with the use of a Machine algorithm.
Mental stress can be detected based totally on the skin conductance sensor, heartbeat sensor, and
temperature sensor which are integrated into the wearable system for measurement of physiological
features. This detector can display the user’s mental stress continuously and collects accurate
information through the use of machine learning algorithms (SVM and Decision Tree algorithm).
Getting the accurate data by the comparison of SVM and Decision Tree algorithms and then it
shows the entire data in excel sheet via IoT web page.
496. EYE MOTION ANALYSIS USING COMPUTATIONAL FLUID DYNAMICS

Jeya Prabha.A1 , Dr.Angeline Kirubha.S.P*2,Sameera Fathimal.M3 , Dr.Nithiyanantham.K.K4
1,2,3Department of Biomedical Engineering, SRM Institute of Science and Technology, Chennai, India.
4Department of Aeronautical Engineering, Rajalakshmi Engineering College, Chennai, India.
Macular hole degeneration, which is treated with intravitreal injections of inhibitor medications, is
one of the most common causes of blindness in patients. Most patients require a loading course of
three injections at regular intervals of four weeks. After the first three courses of injections, the
majority of patients will require further injections depending on the leakiness of the blood vessels.
According to the need to reduce the number of injections, the development of new drug delivery
devices able to extend the therapeutical outcomes is mandatory and magnesium can be considered as
a promising candidate. A magnesium-based device is used to administer medication in the treatment
of AMD, and it is quite effective at minimising the number of injections. Magnesium is a good
candidate for developing innovative medication delivery systems that can improve treatment
outcomes. Magnesium is required for proper eye function. Because of its low corrosion resistance,
biodegradability, and biocompatibility, magnesium is particularly helpful. The purpose of this
May 21st – 22nd, 2022
research is to see how ocular fluid dynamics affect the performance of a magnesium-based
implanted device in the vitreous chamber. Between distinct eye movement times, the fluid-induced
total pressure response on the device's external surfaces was investigated.
497. SMART HOME CONTROL SYSTEM USING MACHINE LEARNING

ALGORITHM.
Kirthivasan S*, Supraja G^, Dr. Jeyalakshmi V$
*PG Student, Department of ECE, Anna University, Chennai
^Research Scholar, Department of ECE, Anna University, Chennai, India $Professor, Department of ECE,
Anna University, Chennai, India
In General, the home automation system controls lighting and other home appliances in a smart way.
Day by day the need for automation in home appliances is high and people also wish to expect more
smartness in the things surrounding them. In this work effective SVM-based home automation is
proposed with two modes of smart operations. Mode 1 is designed to be suitable for the existing
environment and Mode 2 is for a smart environment. In this system, multiple appliances are linked
into the centralized controller for a user-friendly operation. Home appliances such as motor, fan, and
bulb are used in this work for analysis. In mode 1 the system was designed to do the control
operations manually but the necessary condition for manual control was identified by an automatic
process using sensors in a smart way. In mode 2 both the condition identification and control were
done automatically. For the effective automation SVM based Face recognition was used. The
proposed work is more effective and novel in such a way that the recognition of datasets for
performing a particular task was identified with the most accurate coefficient points. The accuracy
of the automation is improved by reducing the number of classes in the dataset for performing
particular tasks.
498. BED FALL DETECTION AND PREDICTION FOR ELDERLY PEOPLE

Ms Usha S
Asst Prof. Dept of ECE
Sri Sairam EngineeringCollege West Tambaram, Chennai-44
Amritha Chandra S,Kaviya V,Prathiba P
Department of ECE
Towards the clinical and living medical care for the older and patients, tumbling from a bed is a
basic mishap that might prompt genuine wounds. To ease this, a fundamental issue is to recognize
this occasion in time for procuring the savage time. Albeit endeavors that hotel to wearable gadgets
and shrewd medical care rooms own beforehand have been paid to resolve this issue, the
presentation is as yet not palatable enough for viable applications. The two handle the fall issues
with a few detecting procedures and investigation models. There is a deficit of surveys on bed-fall-
related advancements. Hence, a conventional characterization of bed-fall-related advances including
fall location and expectation frameworks in view of their sensor contraption is advanced in this
paper. Specifically, the profundity camera is utilized as the visual sensor because of its heartlessness
toward brightening variety and limit with regard to security protection.According to the movement
May 21st – 22nd, 2022
data of the human chest area, they tumble from bed-fall discovery frameworks and all the way to
Sustainable Development Goal is accomplished.
499. DEVELOPMENT OF AN AUTOMATIC PATHWAY DIVERTER FOR THE

COUNT BASED CONVEYOR BELTS
Achiammal B, Deepadevi A, Kalaivani G and Sasi Reka P
Electronics and Instrumentation Engineering
Government College of Technology
The manual control of pathway diverter in the cement packing section is the major problem in
industries. The diverter in the common conveyor path is operated manually by using a pressurized
motor. In common conveyor belt, the pathway diverter is placed to divert the cement bags into the
branches of conveyor belts for loading it on the truck. When the packed cement bags come from the
rotary electronic packer to the conveyor path, the diverter in the midway of the conveyor path helps
to divert the cement bag into the branches of the conveyor belt. And also, the rotary electronic
packer operator cannot know whether the trucks in the loading area are loaded with an exact count.
In that situation, the operator receives information from the external person on the truck. To
overcome this problem, the proposed system of automatic pathway diverter has been developed as a
hardware prototype with the help of NodeMCU ESP8266 controller and necessary components. In
hardware prototype, the diverter’s position gets changed based on the count of cement bags passing
across the IR sensor. Whenever the IR sensor detects the cement bags, it will send the signal to the
LCD to decrease the count. By that, the rotary electronic packer operator knows whether the truck is
loaded or not and also knows which branch of the conveyor belt is loading the truck.
500. 4D PRINTING USING FUSED DEPOSITION MODELLING MECHANISM FOR

SHAPE MEMORY POLYMER: REVIEW
Ayushi Thakura , Umesh Kumar Vatesb, Sanjay Mishrac
aPhD Research Scholar, Department of Mechanical Engineering, Amity University, Noida, Uttar Pradesh-
201310, India
bAssociate Professor, Department of Mechanical Engineering, Amity University, Noida, Uttar Pradesh-
201313,
India
cAssociate Professor, Department of Mechanical Engineering, MMMUT, Gorakhpur, Uttar Pradesh-273016,
India
3D printing has emerged as an enabling technology for a wide range of applications almost in every
field. The review provides a deep understanding of 4D printing process, majorly used 4D printing
technology Fused deposition modelling (FDM) and its critical parameters such as dimensional
accuracy, surface roughness, and aspects related to mechanical behaviour. In the second half of the
paper well-known SMP is polylactic acid is discussed, which belongs to the most often used smart
material in FDM 4D printing. However, there are other SMPs which can be 3D printed by adding
fillers to PLA or blending it with other polymers. A summary of previous research on 4D printing of
PLA and PLA composite on FDM has been represented in this review article.

May 21st – 22nd, 2022
501. USING 3D PRINTING TECHNOLOGY FOR PRE-OPERATIVE SURGICAL

PLANNING AND TRAINING FOR OPTICAL NERVE DISEASES
VijayalakshmiVa, Geetha Balasubramani#a Nithyasree Bb , KiruthikaVb
aAssistant Professor, Department of Biotechnology, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of
Science and Technology, Avadi.
#aCo-Founder & Lead operations, MedCuore Medical Solutions Private Limited, Chennai. bIV
Year Student, B.Tech Biotechnology, Department of Biotechnology, Vel Tech Rangarajan Dr.
Sagunthala R&D Institute of Science and Technology, Avadi.
Preoperative surgical planning has become an important tool for any kind of surgical treatment.
Current procedures for surgical treatment planning are limited; clinicians frequently depend on
related knowledge and clinical imaging to direct the decision-making process. Moreover, two-
dimensional (2D) scanning image of anatomical structures may not precisely depict their three-
layered (3D) complexity, frequently leaving the healthcare practitioner unprepared for the
procedure. Threedimensional (3D) printing has become an economical tool for aiding the
neurosurgeons for the clear understanding of optic nerve in the human brain region, both in surgical
planning, education and training of students & residents. This technique allows the fabrication of
physical prototypes from standard medical images by converting into the anatomical information
into stereolithograhy (STL) files for the printing purpose. Here, we modeled and segmented the
brain optic nerve related to severe diseases by using 3D technology.We fabricated a patient specific
optical nerve model can determine the feasibility of the treatment and provide the relevant
information about the size and position of the nerve deformities. Our aim is to model and segment
the patient specific optic nerve for pre-operative surgical model and demonstrate a potential benefit
in cases of abnormal condition. We collected the real time patient data from the Friedrich-
Alexander-University for the disease of Diabetic Retinopathy. For the purpose of fabrication of 3D
printed prototype, we segment the optic nerve using MIMICS software. This fabricated prototype
offers a 3D view from the two-dimensionalimages of Magnetic Resonance Imaging (MRI)
permitting the better surgical planning for the surgeons in specific cases.
502. HUMAN ACTIVITY RECOGNITION USING MACHINE LEARNING

Swapnil Agrawal1, Himanshu Gupta1, Sarthak Bahuguna1, Chandra Prakash1*
1School of Computing, DIT University, Dehradun
Human Activity Recognition (HAR) has been a really well known study from the past twenty years.
Understanding human actions using machine learning has a vast number of applications using
Machine learning is an extremely challenging technique in the field of computer science. By our
paper we are proposing the features along with the Machine Learning model to recognize the various
human activities. Besides the motion features we also utilize context information from multiple
sources to improve the recognition performance. We have designed a machine learning model to
improve the activity recognition performance by comparing with baseline approaches. We have also
shown this is our approach which outperforms state- of- the- art methods on 4 activities of the
collective activities dataset.

May 21st – 22nd, 2022
503. ADVANCED HYBRID SOLID PROPELLANT COMPOSITION(S) FOR

FUTURISTIC MISSILE TECHNOLOGY
Sarah Manuel Dias
Department of Aerospace Engineering
SRMIST, Kattankulathur,
India-603203
Vinayak Malhotra
Department of Aerospace Engineering
SRMIST, Kattankulathur,
India-603203
Missile form an essential and integral component of National Defense System. That alleviates the
monotonous necessity of keeping up with the state-of-the-art technological advancements. Present
study is aimed out on developing hybrid propellant compositions for advanced and futuristic missile
systems. The work is motivated by the need to find advanced propellant composition(s) yielding
highest and lowest ‘Isp’ values. The study addresses to find an attractive alternative solution to the
excessive dependence of Cryogenics. Simulations were carried out on the NASA-CEA with
different groups (Hydrides, Hydrocarbons, Metals, NonMetals) as catalyst(s) keeping the
conventional base composition viz., Aluminium (cr) as fuel, Hydrogen Peroxide(L) as binder and
Ammonia perchlorate as the Oxidizer. From the results obtained one can infer that, the right
composition of propellent can result in an enhanced missile performance with a high Isp helping in
an enhanced missile performance while the low Isp helping in thrust termination. However, there are
hazardous implications to the environment due to the use of NH4CLO4 as an Oxidizer. The results
from the study are expected to provide useful insight about the strengthening of the missile systems.
504. AUTOMATIC PATHWAY DIVERTER FOR CONVEYOR IN PACKING

SECTION OF CEMENT FACTORY
Madhan Mohan M, Baluprithiviraj K N, Kalavathi Devi T, Renugadevi M
The diverter available in the common conveyor path of the cement packing section at Tamilnadu
Cements Corporation Limited, Alangulam manually operated using a pressurized motor. Packed
cement bags from the rotary electronic packer lead to the two branches of conveyor belts to load the
cement bags on the truck with various vehicle capacities or customer demand. A diverter in the
midway of the conveyor path directs the cement bag to the appropriate conveyor branch for filling
the trucks in the loading area. The operator needs information from the external person to track the
truck space, customer demand or count. Automation to this unit at the preliminary prototype level
for diverting the cement bags from the common conveyor path to branches of the conveyor belt
attempted to load the exact count of cement bags according to the appropriate truck capacity using
an infrared sensor placed in the branches of the conveyor belt loading the truck.
505. TRANSFER LEARNING WITH VGG16 AND RESNET50 FOR LUNG

MALIGNANCY CELLS CLASSIFICATION AND COMPARING ITS PERFORMANCE
VIA EVALUATION METRICS.
May 21st – 22nd, 2022
R. Sudha1 Research Scholar

Chengalpattu, India
Dr. Umamaheswari K M2*
Chengalpattu, India.
Pulmonary cancer caused 26% percent in all respects of cancer-related fatalities in 2017, or more
than 1.5 million deaths worldwide. Several computer-assisted detection techniques are created to dig
out lung cancer in its initial stages, hence increasing the percentage of patients who survive.
Motivated bythe development of Deep Learning in environmental and involving classification
technique, the presented scheme seeks to investigate the efficacy of the application of Transfer
Learning to classify pulmonary nodulesusing a deep learning neural network pretrained model.
Convolutional neural networks (CNN) such as “VGG16” and “ResNet50” are utilized to handle the
lung nodule datasets containing diverse types of pulmonary nodule cells. The resulting deep features
were then categorized with a novel DCNN model. In addition, the model assessment metrics
Accuracy (ACC), Loss, Area Under the Curve (AUC), Precision, and test accuracy F-Score were
computed to Compare the effectiveness of the proposed classifier to that of current systems.
ResNet50+transfer learning model was the most robust CADx(detection with classification) system.
with an ACC of 80.41 %, a loss of
0.6, and an AUC of 90.19 %. Even with a Convolutional NN's non-medical images, their findings
are equal to those of related research. Deep transfer learning has proven to be a helpful technique for
obtaining relevant imaging biomarkers for the classification of lung nodules in chest CT images.
506. A REVIEW ON ROBOTICTECHNOLOGYAND ITS APPLICATIONS

Dr.K. Chithra 1, K. Sangeetha 2
1 Assistant Professor, Department of Computer Science, ShrimathiDevkunvarNanalal Bhatt Vaishnav
College for Women, Tamil Nadu, India
2 UG Student, Department of Computer Science, ShrimathiDevkunvarNanalal Bhatt Vaishnav
Collegefor Women, Tamil Nadu, India
Abstract-The research in the subject of robotics has been undergoing a major change from
manufacturing applications to entertainment, home, rehabilitation, search and rescue, and service
applications.Because of its speed, capacity to work in hazardous situations, and ability to do
repetitive tasks with accuracy, robotics is essential. The goal of this article is to explore the detail
study on Robotics and its Applications in various fields.
507. ASSESSMENT OF POST-FIRE CHANGES IN FOREST COVER USING

SATELLITE IMAGE PROCESSING
1Murugesan G, 2Amrutha varshini V, 3Blessy Gloria D
Department of Computer Science and Engineering, St. Joseph’s College of Engineering, Chennai – 600 119,
Tamil Nadu,India

May 21st – 22nd, 2022
Wildfires have been a major problem in Brazil, since 2001. It can be a natural or man-made
disaster.Millionsoflivingspecieshavebeendestructedbythewildfireasmostoftheworld’slivingbeings
have been contributed by Amazon Forest in Brazil. Therefore, it is essential to plan and mitigate
destruction caused by such fires for future use. The primary aim of the project is to generate a burn
severity map for the assessment of areas affected by wildfire. The idea was to use Google Earth
Engine to classify burnt and non-burnt areas. Google Earth Engine (GEE) portal provides access to
satellite and other ancillary data, cloud computing, and algorithms for processing large amounts of
data with relative ease. The Normalized Difference of Vegetation Index (NDVI) is used, as it was
designed to highlight burned areas and estimate burn severity. It uses near-infrared (NIR)
wavelength. Healthy vegetation before the fire has very high NIR reflectance. In contrast, recently
burned areas have a low reflectance in the NIR. The NDVI is calculated for images before the fire
(pre-fire NDVI) and for images after the fire (post-fireNDVI) andthepost-
fireimageissubtractedfromthepre-fire image to create the differenced (or delta) NDVI (dNDVI)
image. dNDVI can be used for burned area classification, as areas with higher dNDVI values denote
more severe damage whereas areas with negative dNDVI values could depict increased vegetation
productivity. dNDVI can be classified according to ranges of burn severity proposed
bytheUnitedStatesGeologicalSurvey(USGS). Finally,thepost-fireimageisimportedintoQGIS,aopen
source software and geo referenced bys etting fourgroundcontrolpoints.Theraster
layeristhenclassified to show unique colors of the area affected by fire. Lastly the area is calculated
by counting the number of pixels based on similarcolors.
508. AN ALGORITHM FOR CRIME DETECTION IN DIGITAL FORENSICS

Arpita Singh1, S. K. Singh2, Nilu Singh3 and Sandeep K. Nayak4
1,2Amity Institute of Information Technology, Amity University, Lucknow, India
3Department of Computer Science and Engineering,
KoneruLakshmaiah Education Foundation
4Department of Computer Science & Application, Integral University, Lucknow, India
Digital forensics is a bunch of pre-defined processes or tasks during crime investigation with some
technical implementation details with traditional forensics for handling and capturing technical
details of the evidence. Although many researchers and practitioners proposed lots of digital forensic
investigation frameworks.Due to a lot of technical and legal details, the investigation process gets
itself complicated.To break the technical barrier between investigators, information technologists,
and legal practitioners, the researcher needs to propose a technical-independent framework that can
combine these all corresponding tasks together. The key principle of digital forensics investigations
was highlighted in this study (Obtaining authorization, documentation, information flow,
preservation, collection of evidence, and evidence analysis). The author defines five questions for
digital forensic investigation based on this approach. A digital forensics investigation algorithm is
created by incorporating these five sets of queries. We'll go over how this new algorithm can work
with legal counsel as part of a larger digital forensics investigation framework.
509. PERSON FALL DETECTION AND HEALTH MONITORING SYSTEM

S. Arumugam
UG Scholar, Department of EEE
May 21st – 22nd, 2022
National Engineering College, K.R.Nagar, Kovilpatti

M. Lathish Kumar
S. Arun
B. Benjamin Antony Rishi
Dr. N.B. Prakash
Associate Professor , Department of EEE
The Health care sector is one of the major sectors which need improvements day by day in terms of
efficiency of both performance and cost. Elderly people who require wheelchairs and are bedridden
are in need of constant medical attention will need some electronics assistance to monitor them
whether they fell down or need their health monitored. This system will use Node MCU as the
Microcontroller to interface sensors for health monitoring and trigger an SMS in emergency period
to the caretaker.
510. ATTEMPT TO CLASSIFY TREE SPECIES USING MACHINE LEARNING

ALGORITHM WITH SPATIAL DATA
Jayasubha S, Dr. C. SENTHILKUMAR
Department of computer science (Anna University)
Madurai, India
The identification of tree species is an important issue in forest management. Spatially explicit
information on tree species composition is important for both the forest management and
conservation sectors. In combination with machine learning algorithms, very high-resolution
satellite imagery may provide an effective solution to reduce the need for labor-intensive and time-
consuming field-based surveys. In recent years, many studies have explored this topic using
hyperspectral, multispectral, and Landsat 8 OLI/TRIS data. In this study, we evaluated the
possibility of using Landsat 8 OLI/TRIS data satellite imagery for the classification of eight main
tree species (Tectona grandis, Santalum album, Mangifera indica, Pterocarpus sp., Ficus religiosa,
Eucalyptus sp., Azadirachta indica) in Trees Outside Forest as well as agroforestry. The pixel-based
supervised classification was performed using machine learning algorithms support vector machine
(SVM). Additionally, the contribution of individual tree crown (ITC) delineation, for identifying the
location and size of individual trees; (2) alignment between field surveyed trees and ITCs delineated
on remote sensing data for tree species classification was evaluated. Overall of the 43 visually
interpreted reference polygons, 30 were used as training polygons and 13 were used as validation
polygons. The validation results show relatively high overall (accuracy of 88.1%, kappa accuracy of
75.7%, and mean class accuracy of 61.5%), although the accuracy was biased toward the most
represented species.The major issues are the selection of algorithms with training data, selection of
parameters (user-defined) with optimization, feature space impacts and reduction, and majorly
computational costs. In this paper, the issues are addressed via machine learning classification for
open-source satellite datasets.

May 21st – 22nd, 2022
511. A NOVEL METHOD FOR FAULT DETECTION AND PROTECTION IN SOLAR

PHOTO VOLTAIC ARRAYS
Mahesh N1
Associate Professor
Electronics and Instrumentation
Engineering Kongu Engineering College
Perundurai, Erode.
Deepak Prasath G S2
UG scholar,
Perundurai, Erode.
Divyadharshini E3
UG scholar,
Perundurai, Erode.
Gokul V4
UG scholar,
Perundurai, Erode.
With the increased use of photovoltaic (PV) modules, there is a greater demand for quick response
and effective PV fault detection methodologies, not just to assure system dependability, but also to
cut down on maintenance costs, computational load, and human labour. To improve the efficiency
and reliability of solar modules, fault detection and condition monitoring are critical. Manufacturing
errors are thought to be the primary cause of performance inconsistencies in some modules.
Undetected defects in photovoltaic (PV) arrays have the ability to cause permanent power loss and
severe catastrophic failures. Some of the defects that cause damage to the PV arrays are over
voltage, under voltage, over current, short circuit, open circuit, blindspot error and hotspot error.
512. BIG DATA ANALYSIS FOR HEART FAILURE: DETECTION, PREVENTION

AND PREDICTION
Ganesh kumar K, Rajalakshmi S
Professor
Saranya T
Assistant professor
Sachin V, Samya Devi V, Suganya R, Surabi D
Department of Information Technology, Velalar College of Engineering and Technology
In this era of realignment, the healthcare is upfronting with advanced technologies such as artificial
intelligence, 3D-printing, robotics and nanotechnology. However in cardiology, an estimation of
64.3 million people are living with heart failure across the world. In developed countries, the general
estimated heart failure is from 1% to 2% of the adult population. Heart failure expenditures are
considerable and will increase dramatically with an ageing population. The cardiology department in
big data becomes a demanding task. Early decision making can help to reduce the risk and cost of
May 21st – 22nd, 2022
the patients. The future depends on both the technologies and the healthcare. The present survey
analysis helps to detect, prevent and predict from a heart failure disease. Thus, the analysis will be
useful for doctors, physicians and researchers to make more actual decisions, which ultimately will
help to use the survey as indication for treating patients with heart failure.
513. REVIEW: COMPARISON OF PARAMETER ANALYSIS OF RING

OSCILLATORS FOR MEASUREMENT OF DELAY TIME
G.Elaiyaraja1
1Professor, Department of ECE, VEMU Institute of Technology, P. Kothakota, Chittoor-517112, Andhra

Pradesh, India,
A ring oscillator (RO) is used to measure the delay time and also clock frequency generation for
embedded processors. In this paper, three types of ring oscillator such as current controlled RO,
Bootstrap inverter based RO and Tunnel Field Effect Transistor (TFET) based RO are analyzed.
Bootstrap RO which operates with low voltage and low power consumption using bootstrapped
inverters and circuit was designed using 65nm Silicon on Thin Buried Oxide (SOTB) process. The
circuit area was reduced by using a MOS capacitor instead of the capacitor used in the conventional
bootstrapped inverter. The oscillation frequency range of this RO produces from 860 to 968MHz
with a supply voltage of 0.5V. A current-controlled CMOS ring oscillator topology, which exploits
the bulk voltages of the inverter stages as control terminals to tune the oscillation frequency.
Oscillators made up of 3, 5, and 7 stages designed in a standard 28-nm technology and supplied by
0.5 V, were simulated. By exploiting a programmable capacitor array, it allows a very large range of
oscillation frequencies to be set, from 1 MHz to about 1 GHz, with a limited current consumption. In
TFET based Ring Oscillator RO, Two different ring oscillator topologies are analyzed with two
different structures for TFET device. Better reliability for oscillation frequency is obtained using
Negative Skewed Delay ring oscillator (NSD RO) topology. NSD RO is resulting in lesser jitter,
more reliable frequency as compared to other ring oscillator topology.
514. DIVERSE OBSERVER’S STANDPOINT ON THE EXTRATERRESTRIAL

INTELLIGENCE -A HYPOTHETICAL NOTION
ChitralekaA, PonnuswamyB
A) Centre for Yoga, SRM Institute of Science and Technology, Chengalpattu, Tamilnadu, India.
B)Yoga for Human Excellence, WCSC-Vision, Aliyar, Tamilnadu, India.
Corresponding author: V Ponnuswamy
For the past many decades, the foremost common debates about the existence of extraterrestrial
intelligence have been prevailing in the scientific community, moreover, it is a mysterious search
that interested people. Because humans have doubted whether we are unaccompanied in the universe
or whether intelligent life is possible somewhere else. Very few people are aware of the exoplanets
on account of the existence of other stars and galaxies. But still, we are struggling to find the life
forms from the exoplanets in different galaxies. Apparently, more development is required rather
than the contemporary scientific advances to know the existence of Extraterrestrial Intelligence
(ETI) and to communicate with them. Explicated, four types of cosmological systems comprise the
whole cosmos, which are transforming relatively and cyclically. And also explained the observer’s

May 21st – 22nd, 2022
states of consciousness. The proposed concept made an attempt to conclude whether the existence of
extraterrestrial intelligence is possible or not. Elucidated, that humans have diverse standpoints with
respect to their dimensions of consciousness, in which 64 standpoints are identified. Most of the
observer’s standpoints are in uncertainty and fuzziness about the extraterrestrial intelligence, which
is relatively compared with Erwin Schrodinger’s cat experiment analogy. Discussed the feasibility of
the presence of ETI in the multiple universes which compose the cosmos cohesively.
515. AUTOMATIC LUNG SEGMENTATION USING U-NET

Jishna N Madhu1 , Angeline Kirubha S. P.2*, Shwet Thaker3
12Department of Biomedical Engineering, SRMIST ,kattankulathur ,Chennai ,Tamil Nadu ,India
3Department of Biomedical Engineering, VIT, Vellore, Tamil Nadu, India
COVID-19 is an extremely contagious disease caused by a virus called SARS COV 2 which is
otherwise called coronavirus. Sore throat, dry cough, and fever are the most typical signs of
COVID19. It can lead to severe pneumonia, severe sepsis condition which can result in own tissue
or organ damage, acute respiratory distress syndrome, and pulmonary edema. In clinical practice,
CT gives detailed diagnosis as well as COVID-19 severity information. As it is very hectic for a
radiologist to look through all patient’s CT scans, automatic segmentation and prediction of Covid
positive/negative along with severity prediction are needed. In order to detect lung lesions in 3D CT
volumes, the lung needs to be segmented from the images. In this paper, we are performing
segmentation of the air lung region of 3D CT images using U-net. Kaggle 2D and 3D images
database are used to train the model for U-net based segmentation method.
516. BATTERY MANAGEMENT SYSTEM IN ELECTRIC CYCLE

Dr. Arunarasi J1, Jyothilakshmi R2 ,Maragadhavalli L3, Kirthika A4
Assistant Professor Student
Department of ECE ,Sri Sairam Engineering
Department of ECE
Chennai, India
The goal of the project is to analyze the existing drawbacks of the E-cycles (such as Frequent battery
failures, High price, prolonged charging time) and attempt to fix them all. Rectifying these issues
would help in multiplying the amount of trust people have towards E-cycles which would in turn
boost up the usage of Electric cycles. By promoting the usage of EVs, access to affordable, reliable,
sustainable and modern energy for all can ensured. By optimizing the BMS’s (Battery Management
System) circuit, the life span of the battery can be improved and time taken by the battery pack to
get charged can be reduced. Initially Lead-acid batteries were being used for electric vehicle
batteries until they were replaced with lithium-ion cells. And in this project, we have replaced
Lithium-ion batteries with LifePo4 batteries, because of the short battery life of Lithium-ion
batteries. Requirement of costly maintenance and frequent repair of electric cycles significantly
affects the efficiency of the electric cycle, which is solved in this model.

May 21st – 22nd, 2022
517. A SURVEY ON EMOTIONAL DETECTION IN DEEP LEARNING

Anand M1, Dr.S.Babu2
1Research Scholar, Department of Computing Technologies, SRM Institute of Science & Technology,
Kattankulathur, Tamil Nadu, India
2Associate Professor, Department of Computing Technologies, SRM Institute of Science & Technology,
Kattankulathur, Tamil Nadu, India
Emotion recognition is a significant issue in the field of Big Data. Reliable categorization, analysis,
and interpretation of emotional content are highly desired in a wide range of applications.In this
paper, we look at how Deep Learning models perform on an emotion perception test. The first
section includes an introduction to emotional detection in deep learning. In the next section, related
works bring sentiment classification with a comparison of deep learning-based sentiment
classification model for E-commerce comments, convolutional neural networks, and database
available for emotional detection. In section 3, we are having adiscussion and comparison of various
deep learning techniques. In the final section, we are concluding the things and the future areas to be
concentrated in emotional detection in deep learning.
518. DETECTION OF MAN IN THE MIDDLE ATTACK IN 5G IOT USING MACHINE

LEARNING
C.Arul Stephen *1, #2,A.Vijayalakshmi.#3 .J.Broody #4..J.Sri sathish kumar #5B.Ebenezer Abishek
*1 Assistant professor Dept of ECE,VISTAS*2 Associate professor,Dept of ECE,VISTAS, *3,4UG Scholar
VISTAS *5 Associate professor,Dept of ECE,Vel Tech Multitech ,Avadi
Detection plays a vital role in the security system of the wireless network. Hence an effective
wireless intrusion detection system is proposed to detect a Man-in-the-Middle attack in a wireless
network. We show through empirical observation that the presence of a Man-in-the- Middle attack
incurs a considerably longer delay and bigger variance in measured RTT compared to other attacks
We evaluated different machine learning algorithms on the IoT device log dataset to detect Man-in-
theMiddle attack.Experimental results show that the proposed technique effectively detects and
predicts the attack with a minimal prediction time and high accuracy.
519. DIFFERENT LEVEL MULTI MEDIA CONTENT DELIVERY SYSTEM BASED

ON ADVANCE WATERMARK ALGORITHMS
Logeswaran A, Dheektchana S, Gopinath A, Kamalesh K, Kavishna G
Velalar College of Engineering and Technology
Erode, India
Wireless multimedia dispatches have developed fleetly with adding wireless access bandwidth and
popular intelligent bias. In the decade, a number of studies have been conducted to design robust and
May 21st – 22nd, 2022
effective schemes for delivering multimedia content over error-prone wireless networks. In
discrepancy, veritably many of those studies have concentrated on the security aspect of similar
transmission. There have been adding demands for the security of wireless multimedia Operations in
recent times. Wireless networks when compared to traditional wired networks, are more likely prone
to vicious attacks. Current security styles include physical layer and operation layer security
technologies singly and independently. Generally, physical layer information is dynamic in wireless
networks, and operation layer information is related to wireless multimedia content delivery; both
have significant impact on security performance. This study proposes a common frame combining
both the physical and operation layer security technologies. Specifically, by exercising the security
capacity and signal processing technologies at the physical layer and the authentication and
watermarking strategies at the operation layer, the available network coffers can be employed
efficiently. Moreover, scalable multimedia security services could be maximized within given
multimedia delivery deadlines. So, in particular, this common scheme could be enforced fluently
with low communication above, which facilitates the deployment in large-scale wireless multimedia
systems.
520. CREDIT CARD FRAUD DETECTION USING DYNAMIC BAYESIAN

NETWORK
Dr.M.P.Thiruvenkatasuresh,Ph.D,Professor
Erode Sengunthar Engineering College
Erode,TamilNadu
Ms.E.Indira,Final year
Erode,TamilNadu
Mr.S.Shajahan, Final year
Erode,TamilNadu
Mr.M.Vignesh,Final year
Erode,TamilNadu
Creditcard fraud begins either with the theft of the physical card or with the compromise of data
associated with the account, including the card account number or other information that would
routinely and necessarily be available to a merchant during a legitimate transaction. Stolen cards can
be reported quickly by cardholders, but a compromised account can be hoarded by a thief for weeks
or months before any fraudulent use, making it difficult to identify the source of the compromise.
The cardholder may not discover fraudulent use until receiving a billing statement, which may be
May 21st – 22nd, 2022
delivered infrequently.Even though fraud detection has a long history, not that much study has
appeared in this area.Nowadays digitalization gaining popularity because of seamless, easy and
convenient use of eCommerce. It became very rampant and easy mode of payment. People choose
online payment and e-shopping; because of time convenience, transport convenience, etc. As a result
of, huge amount of e-commerce use, there is a vast increment in credit card fraud also. Fraudsters try
to misuse the card and transparency of online payments. Thus, to overcome with the fraudster’s
activity becomes very essential. The main aim is to secure credit card transactions; so people can use
e-banking safely and easily. To detect the credit card fraud there are various techniques which are
based on Logistic Regression, Support Vector Machine (SVM), K Nearest Neighbour, Decision
Tree, etc.The project proposes a new fraud detection mechanism using an Artificial Neural Network
based on Deep Learning Technique. The project is designed using C# and ASP.NET.
521. SIGNATURE AND FORGERY VERIFICATION USING CONVOLUTIONAL

NEURAL NETWORK
Mrs.N.SenthamilArasi 1. S.Dhivya2. A.Dharmila chowdary3. S.Naveena 4
Assistant Professor, Department of Information Technology, Panimalar Institute of Technology, Chennai,
India1.
Student, Department of Information Technology, Panimalar Institute of Technology, Chennai,India2,3,4
Signatures are veritably important in our social and legal life for verification and authentication. A
hand can be accepted only if it's from the intended person. The probability of two autographs made
by the same person being the same is veritably less. Numerous parcels of the hand may vary indeed
when two autographs are made by the same person. So, detecting a phony becomes a grueling task.
In this paper, a result based on Convolutional Neural Network (CNN) is presented where the model
is trained with a dataset of autographs, and prognostications are made as to whether a handed hand is
genuine or forged.and also we are using harris technique for the corner detection step to give better
input features to the algorithm of convolutional neural network. Using step signature forgery
detection is simple.
522. 3D DENTAL PROSTHETIC TOOTH CROWN: DESIGNING AND

REMODELLING OF CUSTOMIZED TOOTH CROWN USING ADDITIVE
MANUFACTURING
Nisitha S1 , Geetha Balasubramani1*, Paul Pradeep J 2*
1Project Intern, MedCuore Medical Solutions Private Limited, Chennai.
1*Co-Founder & Lead operations, MedCuore Medical Solutions Private Limited, Chennai.
2*Founder & CEO, MedCuore Medical solutions private limited
This paper aims to show that what extent 3D printing can be used in dental laboratories and dental
practices at present. It attempts to present a rational evaluation of today’s application of 3d printing
technology in the context of dental restoration. In addition, the article discusses future perspectives
and examines the ongoing viability of traditional dental laboratory services and manufacturing
processes. It also shows which expertise is needed for the digital additive manufacturing of dental
restoration and a case report of prosthetic treatment builtupon 3d technologies for manufacturing of
dental restorations. We designed a customized 3D tooth cap using 3D printing technology.
May 21st – 22nd, 2022
Rhinoceros and Solid work are the software is used to design and simulated a customized 3d tooth
cap. Ultimaker Cura software and Ansys is used to slice and to test the mechanical simulation of the
tooth cap in viable environment such that there is no miscible after the implantation. The 3D printing
methodology used to print the prototype model is Fused Deposition Modelling (FDM). Bio-
compactible tooth filling / toothcement was prepared using beta tri-calcium phosphate, co-polymers
like Pectin and poly vinyl pyrrolidone (PVP) as a promising biomaterial to enhance a cell growth in
a suitable environment. The tooth filling/tooth cement slurry was characterized by FTIR and the
antibacterial testing was carried out against E. coli and S. aureus it exhibits an excellent inhibition
zone. Thus, our primary study was concluded that the bio-ceramic tooth filling could be an ideal
solution but further in-vitro cell and in-vivo studies are needed to confirm the use of tooth filling in
infection site.
523. MONITORING THETEXTILE WASTE WATER TREATMENT BY USING

NATURAL RESOURCES
Madhan Mohan M,Boobesh.S ,Gowtham K, Kishorrkumar KM,Prithivaka P
The waste water in the textile industry is one of the major threats of Erode and other nearby districts
to cause environmental and health related problemsin human beings. The main reason to resolve the
problem throughwater purification processbychemical dyeing and dissolved solvents absorption with
softening properties. Most of the waste water is released in underground or in river causes major
form of water pollution. Here, to reduce the harmfulness of the waste water out from the industry, an
attempt is made through natural resources like drum stick seeds and Canna Indica plants are used to
improve the TDS and pH property of releasing waste water. The result shows the value of TDS and
PH.
524. DEEP LEARNING APPROACH FOR ECG –.BASEDHEARTBEAT

CLASSIFICATION FOR ARRHYTHMIA DETECTION
Madhulika Dhar
Department of ECE, ASET
Amity University.
Uttar Pradesh, India
Vaishnavi Kumari
Amity University.
Dr. Shubhra Dixit
Amity .University
Millions of people from entire the world deals from cardiovascular diseases (CVDs). The
classification of heartbeats is a crucial step in determining cardiac function. An ECG is a graphical
portrayal of cardiac signals that is used to assess the heart's electrical pulses and is commonly used
to identify any abnormalities. It can diagnose various cardiac issues by analyzing and examining the
May 21st – 22nd, 2022
electrical signals generated by ECG with the use of electrodes. Heartbeat classification may be done
with a variety of classifiers. In this we applied two DL techniques i.e., CNN & CNN-LSTM, applied
for ECG-based heartbeat categorization by discussing preprocessing, Electrocardiogram dataset,
feature extraction and types of classifiers available for automatic heartbeat classification. We have
used one dataset for the classification of heartbeat that is MITBIH arrhythmia dataset having 109446
pulses of ECG recorded in which data is divided into training with the 80% and testing with the
20%. “N, S, V, F and Q” are the 5 classes that are included in the dataset. The result attained by both
the model shows the 99.37 % trainingoaccuracy and 98.46% validationoaccuracy for CNN model,
99.47% training accuracy and 98.75% validation accuracy for CNNLSTM model.
525. DOCTOR PATIENT APPOINTMENT SYSTEM

AKSHATA CHAUDHARI, GAURAV CHOUKSEY, VAIDEHI KAWAL, SUBHAM PATIL.
DEPARTMENTOFCOMPUTERENGINEERING.
SAVITRIBAI PHULE PUNE UNIVERISITY, PUNE.
Online Doctor appointments are a smart web system, this provides registration and access for both
doctors and patients. Doctors can register by providing his / her required information such as times,
fees, category, etc. After successful registration, the doctor can login by providing a username and
password. The doctor may view the patient's reservation request and upon receipt of the patient's
requests the condition will be indicated as a guaranteed reservation for the patient. He can also view
the response given to the patient. Patients must be registered and logged in to book a doctor based on
the category and type of problem they are facing and the location. The search results will show a list
of doctors who match the required patients and he can select one and send the application will be
forwarded to admin and admin to the doctor and if available he will return the confirmation request
to the admin review the booking request and say confirmed to the patient. the patient can view the
status on the status tab.
526. A PREDICTIVE APPROACH ON STOCKPILE SYSTEM USING ABC ANALYSIS

Dr.Mynavathi R Mr.RajendrenP Anand S.D, Arun P.R, Balaji J, Devi priya S
Department of Information Technology, Velalar College of Engineering and Technology
Stockpile system is a systematic approach to sourcing, storing and selling inventory both raw
materials and finished goods. In business terms inventory management means the right stock, at the
right levels, at the right place, at the right time and at the right cost.The purpose of the stockpile
system is to manage the different variety of stocks, brands, materials and also can track the goods in
time inorder to place the order for the decreasing stocks. The proposed system can be manipulated
for managing the total items in or out from the inventory which focuses in the area of stock controls
and generates the various required report. The predictive analysis helps to figure out the highest
selling product, location -based selling of products with the historical data and can also gather the
customer feedback. The stockpile system can forecast sales demand to help optimize inventory and
save significant cost due to the high or short inventory caused by inaccurate demands. Systematic
management of stock will ensure that no duplicate orders are placed that will prevent the capital loss
for the company. This system also will help the companies to make future predictions on stocks that

May 21st – 22nd, 2022
makes profit for the company and will also provide time to make business plan and strategies that
will develop the company also can easily do the job without any time lag.
527. DEVELOPMENT OF EMBEDDED CUM AI BASED SMART HELMET FOR

UNDERGROUND MINERS
V Ramkumar*1, R Balakumar*2, S P Annamali*3,V Gajemdraraj*4, B Mahendra Raj*5
*1Assistant Professor, Department of ECE
K.Ramakrishnan College of Technology
*2,*3,*4,*5IV Year, ECE Department
K.Ramakrishnan College of Technology
The main objective of this project is to identify the injured miner and to rescue them, when an
accident occurs in the mining industry. Mining is a one of the highly profitable resource for the
government but lacks in its safety, to resolve issue this we have developed a smart helmet through
which a worker’s health and safety can be tracked constantly. A CNN(Convolution Neural Network)
algorithm based on Artificial intelligence’s deep learning technique is used for helmet detection,
with this technology a firm can be ensured whether a worker wearing helmet or not.
528. CATARACT DETECTION USING DEEP LEARNING

N.Prem Kumar , K.Jagadeesh , A.Siva Kumar
Department of Information Technology Sri Sairam engineering College, Chennai, Tamil Nadu
This paper explains detection of cataract using machine learning algorithms. The combined feature
vectors from Scale-Invariant feature transform (SIFT) and Gray Level Co-Occurrence Matrix
(GLCM) algorithm applied to classifier models of Support Vector Machine (SVM), Random Forest
and Logistic Regression for detecting the image as cataract eye or normal eye. For the type
classification on the basis of severity: mild, normal and severe, again the combined feature vector of
SIFT and GLCM algorithm are used for training using pre-trained deep convolutional models like
SqueezeNet, MobileNet and VGG16. The web application is created using flask framework. When
image is inputted, it will detect it as cataract or normal eye. If cataract has been detected then it will
check intensity of cataract and displays the result as mild or severe.
529. COMPUTER BASED DIAGNOSIS AND CLASSIFICATION OF SKIN DISEASES

IN DERMOSCOPIC IMAGES
Ayush Somani
Department of ECE
ASET
Amity University
Khushi Gupta
Department of ECE
ASET

May 21st – 22nd, 2022
Amity University
Ms. Anupama Bhan
Department of ECE
ASET
Amity University
A skin lesion refers to a skin area that represents change from the surrounding skin in terms of
colour, shape, size, and texture. There are many types of skin lesions, some are cancerous that can be
severe, and some are non-cancerous that can be harmless.A physician faces many difficulties for
accurate diagnose of lesion through its characteristics. Computerized identification of malignancy in
data have been the concern of analysis. However, analysing these photos is highly challenging due
to a variety of issues such as reflections over the body surface, differences in colour and variable
shapes. As a result, evidence-based automated skin cancer detection can aid pathologists in
improving their accuracy and proficiency in the prelimnary stages of the disease.Our purposed
method uses a deep learning methodology to identify skin lesions in early stage so to distinguish
between benign (noncancerous) or severe (cancerous) lesions. Melanoma (MEL), Melanocytic nevus
(NV), Basal cell carcinoma (BCC), Actinic keratosis (AK), Benign keratosis (BKL),
Dermatofibroma (DF), Vascular lesion (VASC), Squamous cell carcinoma (SCC), None of the
above (UNK) are included in the dataset. Various high end deep learning models are constructed in
our suggested methodology to categorise malignant and non-cancerous skin lesions. We have
employed techniques like as filtering and feature extraction to improve classification and hence
expand the estimate of the final analysis.
530. IMPLEMENTATION OF WOMAN SAFETY SYSTEM USING GEO-FENCING

Pavithra M, Gowri Bharathi B, Swetha S
Guided by
Ms. S. Surya
Sri sairam engineering college
To develop a cost efficient security system for school going children to track their location and send
an alert message to their parents or guardian when a child is in distress.To calculate the distance
between the school and home which is considered as a safe zone is inputted to arduino interfaced
with u-blox NEO 6M GPS module .If the child goes beyond the safe zone an immediate alert
message is sent to their parents or guardian via gsm module.The hardware includes Arduino UNO
which is interfaced with the u-blox NEO 6M GPS module and GSM module to track the location of
the children when reached beyond the safer zone.The software includes arduino IDE which is used
to easily compile and load the arduino program (embedded C) to the board connected to the
computer.Geofencing combines awareness of the user's current location with awareness of the user's
proximity to locations that may be of interest. To mark a location of interest, you specify its latitude
and longitude. To adjust the proximity for the location, you add a radius. The latitude, longitude, and
radius define a geofence. You can have multiple active geofences at one time.When the GPS is
turned on the location of the child is pinned and through the GSM module, the message is sent from
the microcontroller to the victims authorized group stating ―”Child in distress”, if the location area
exceeds the safer zone.
To calculate the distance between the school and home which is considered as a safe zone is inputted
to an arduino interfaced with u-blox NEO 6M GPS module. If the child goes beyond the safe zone
May 21st – 22nd, 2022
an immediate alert message is sent to their parents or guardian via GSM Module.The hardware
includes Arduino mega 2560 which is interfaced with the u-blox NEO 6M GPS module and GSM
module to track the location of the children when reached beyond the safer zone. The software
includes an arduino IDE which is user to easily compile and load the arduino program (embedded C)
to the board connected to the computer.
531. CONTROLLING AND MONITORING OF BOILER PARAMETRS USING

SCADA
Vidhyalakshmi.P1, Dhanishka.K.V2, Gokulapriya.P3, Vasant.S4
1Associate Professor, 2,3,4U.G Students, Department of Electronics and Instrumentation Engineering, Kongu
Engineering College, Perundurai,Erode India
This paper discusses about the controlling and monitoring of the boiler operation. Boiler is a large
part in an industry which boils the water and generates the steam. This steam can be used to run the
engines or turbines. In this work Boiler operation is controlled by PIC microcontroller, which
controls the parameters such as Temperature and Pressure of the boiler. This is because PICs are
very flexible, cost effective, space efficient and reduce complexity. SCADA (Supervisory Control
and Data Acquisition) gets the data from the microcontroller and monitors on the screen. The
proposed boiler automation ensures safe, reliable and continuous operation of boiler system and
hence of the power generation plant.Many commercial and industrial facilities use boilers to produce
steam or hot water for space heating or for process heating. Boiler setup is made using LPG cylinder
and heater is placed inside the cylinder so that it acts as boiler. Pressure sensor and temperature
sensor is used to measure the pressure and temperature inside the boiler respectively. To control
temperature, pump is placed and when the temperature goes beyond the certain level pump
automatically turns ON and the feedwater is supplied. When the pressure goes beyond the limit
solenoid valve turns ON and steam released through the solenoid valve. LCD is placed to monitor
the values from outside. This article outlines the types of boilers used to heat facilities, while
providing an overview of basic boiler controls and the parameters that affect energy efficiency.
532. LUNG CANCER DETECTION USING MACHINELEARNING TECHNIQUES

DrJ.ARUNARASI[1]
Department of Electronics and
Communication Engineering
WestTambaram,Chennai
MONISHA S[3]
RAVULA SONALI[2]
May 21st – 22nd, 2022
DIVYA M[4]
An uncontrollable growth of body cells in lungs is called lung cancer. Lung cancer is ranked fourth
among all the cancers over the world. Diagnosing of lung cancer is a vital step at early stages for
increasing the survival rate. To detect lung cancer, CT (computed tomography) is an effective
method to track lung cancer. In this proposed system, the software model is trained with different
CT images. This lung cancer model includes methods for segmenting the lung regions out of which
GLCM parameters are collected and used for categorization of tumour stages using CNN.This
proposed work has anefficiency of 98 percentage.
533. PRIVACY PRESERVING CONFIDENTIAL HEALTH CARE ADMINISTRATION

USING BLOCKCHAIN TECHNOLOGY
Dr.K.Ganesh kumar,
Professor, Department of Information Technology,
Erode, India
Arun balaji T,Gopinath R,Kavinesh M,Krishnamoorthy S,
Erode, India
Blockchain technology has the ability to create new business models and trust issues handling in a
more efficient way. It leads to wide research opportunities and business innovations. Blockchain
based software solutions can be applied in a wide range of domains. This project presents a design in
which blockchain technology is used in the healthcare system, in which the information regarding
medical analyses are shared between hospitals and research institutes based on access policies
defined by the patients. In order to protect confidential data, the project involves the use of two types
of chains: a private one, the sidechain, which keeps information about the real ID of the patients, and
a public one, the mainchain, which stores information about patients’ health data marked with a
temporary ID. The project deals with the hospital branch and patients’ management. Since cloud
computing delivers convenient, on-demand access to shared pools of data, applications and hardware
over the internet, it provides unlimited infrastructure to store and execute patient data and programs.
The project applies blockchain technology which is a decentralized network, where the entire
database of healthcare management is being handled by many users. All the records are hashed as
entries as blocks and introduced into the blockchain so that the proof of contract is available for all
the past and current transactions. The entire transactions are said to be valid using the above
blockchain concept.
534. DESIGN AND ANALYSIS OF COMPACT MICROSTRIP PATCH ANTENNA

FOR 5G APPLICATIONS

May 21st – 22nd, 2022
Velvizhi V A1, Panchagnula Lalitha Sravya2, Konka Sreya3, Panchagnula Sai DivyaSree4
1Associate Professor, Department of ECE, Sri Sairam Engineering College, Chennai, India,
2,3,4 Student, Eletronics and Communication Engineering, Sri Sairam Engineering College, Chennai, India,
In this project, an attempt is made to design a compact microstrip patch antenna using CST software
tool. Low profile of the antenna eases the simulation. The resonating frequency 3.5GHz of the
designed antenna can serve 5G operating systems. Substrate used for the design is FR-4 Epoxy with
the dielectric constant of 4.4. Return loss determines the loss of power in the signal returned or
reflected by a discontinuity in a transmission line. Hence, it is used as a metric to evaluate the
performance of the newly designed antenna with the conventional antenna.
535. AUTOMATIC TEXT SUMMARIZER

Dr.Rani.T.P#1, Lavanya.M#2, Karthiga.G#3
#1Associate Professor, Department of Information Technology
#2#3Student, Department of Information Technology #1#2#3Sri Sairam
Engineering College, West Tambaram, Chennai-600044 [Autonomous],
Tamil Nadu, India.
“The first forty years of life give us the text; the next thirty supply the commentary on it” - Arthur
SchopenhauerText plays a vital role in each and everyone's life. In this booming world, we have an
enormous amount of text that we need to go through which is sometimes a tedious
process.Therefore, we are in a need to summarize all those texts. Ease access and availability of
smartphones and devices made text, image and hyperlink processing more reliable and convenient,
adding power to obtain all the information in a single-handled application.To accomplish this, we
created a text summarizer application that gets the most precise and necessary information from a
large document and annihilates the extraneous or less noteworthy ones. Summarization is a
technique where a computer summarizes a text. Here we are building a mobile application that will
summarize the text you give in a way that is easy to read and understand in a short period. Our
ultimate goal is to reduce the given text to a fraction of its size.
536. SURVEY OF COIN RECOGNITION AND DISCRIMINATORS

Sweety Kumari
UG Student, Dept. of Electronics and Communication, AMC Engineering College.
Sheetal Arjun Kamble
\UG student, Dept. of Electronics and Communication, AMC Engineering College.
Rachana D N
UG student, Dept. of Electronics and Communication, AMC Engineering College.
Dhruva Kumar Ck
UG student, Dept. of Electronics and Communication, AMC Engineering College.
Dr T.Kavitha
Professor, Dept. of Electronics and Communication, AMC Engineering College.
Dr G.Senbagavalli
Professor, Dept. of Electronics and Communication, AMC Engineering College.

May 21st – 22nd, 2022
A coin detector or currency validator is a device that determines whether coins or notes are genuine
or counterfeit. The automatic sorting system has been reported to be complex and a global problem.
This is because of the inability of sorting machines to incorporate flexibility in their design concept.
This paper discusses the coins classification and counting where the manual method of counting the
coins does not have any recording device for future usage. It will give a glance at several papers
related to Coin Classification and Counting. This paper will give the idea to researchers that
technology is used in coin recognition and classification and also open issues which need to be
addressed.
537. A STATE OF ART TECHNOLOGIES IN SMART MASK

Revannaradhya R
Communications AMC Engineering College
(Affiliated to VTU) Bangalore, India
Praveen B
(Affiliated to VTU)
Bangalore, India
Sudeep V
Department of Electronics and Communications
AMC Engineering College
(Affiliated to VTU)
Bangalore, India
Dr.Senbagavalli G
(Affiliated to VTU)
Bangalore, India
Shashidhar Satish Hegde
(Affiliated to VTU) Bangalore, India
Dr.Kavitha T
(Affiliated to VTU)
Bangalore, India
Face masks have been very useful in preventing the spread of infectious diseases such as the flu and
have also been used by doctors during surgeries. More recently, when the world was facing a
pandemic due to the outbreak of coronavirus, masks were an essential tool to slow its rapid spread,
and masks have also become an integral part of our lives due to increasing pollution. Today, we can
find a variety of masks on the market, such as single-use masks, reusable masks, and smart masks
that not only filter air while breathing, prevent contagious viruses such as SARS COVID-19, dust
particles, and other allergens, but also incorporate various additional characteristics such as health
monitoring, environmental sensing, and so on. This paper examines the many smart mask ideas and
devices that are already available. These masks are capable of sensing, processing, and wireless data
May 21st – 22nd, 2022
transfer for monitoring and analysis, which aids in the formulation of procedures and safeguards to
avoid future spread.
538. REAL-TIME TOMATO LEAF DISEASE CLASSIFICATION USING OPTIMIZED

PRETRAINED CNN MODEL BASED ON TRANSFER LEARNING IN DEEP
LEARNING AND PREVENTION STRATEGIES
S.Anitha,G.Nivetha, N.Vaishnavi, M.V.Priyadharshini,K.Lalitha
V.Chandrasekaran, K.Vanitha
In the era of Deep Learning, Computer vision plays an important role. Computer vision is an
interdisciplinary scientific topic that studies how computers can perceive digital images or movies at
a high level. Color, texture, and form features are used to segment RGB images of crop diseases in
traditional machine vision algorithms. However, because the symptoms of many diseases are
similar, it is difficult to distinguish between them, and disease recognition accuracy in a complex
natural environment is poor. The Convolutional Neural Network (CNN) is a high-performance deep
learning network that eliminates the need for sophisticated image preprocessing and feature
extraction. The widespread presence of illnesses in the tomato crop has an impact on the quality and
quantity of produce. Farmers will profit from early illness diagnosis utilizing a rapid, reliable,
nondestructive way to combat the problem. In the proposed work, images of tomato leaves (9
diseases and a healthy class) from the PlantVillage dataset are fed into Resnet34 model. After
completing relevant pre-processing procedures such as image augmentation, the input dataset was
fed into the model. With Adam as the optimizer, an accuracy of 92 percent was attained with
ResNet-34.
539. SMART BAG FOR WOMEN’S SAFETY USING EMBEDDED SYSTEM

S. Sathya Priya1,
St.Joseph’s Institute of Technology
Chennai,India
Bianca S.I.2
Chennai,India
Godlin Sneha E.3

Chennai,India
Women's safety is a major worry in a country like INDIA, where women excel in virtually every
sector. India is a peaceful country that is a popular tourist destination across the world. However, a
few recent cases have brought to light the necessity for women's protection. Because of the high
number of incidences of violence against women, many women in industrialized nations are still
afraid to walk outside alone. Many attempts have been made to make women's safety safer, but a
May 21st – 22nd, 2022
better and more secure system that can assure safety on public transportation and in general is still
required. This study describes a method that can provide increased security and safety.
540. PERFORMANCE ANALYSIS OF ENHANCED LIGHTWEIGHT ALGORITHM

FOR IOT DEVICES
Sheeba DM
Dept of Computer Applications, Vels Institute of
Science,Technology & Advanced Studies
Chennai, INDIA
Dr. S. Jayalakshmi
Dept of Computer Applications
Vels Institute of Science,Technology & Advanced Studies
Chennai, INDIA
The Internet of Things (IoT) is the revolutionary word in today’s world and it connects billions of
electronic devices and gadgets with data security. Internet of Things enables many industries to
connect to end customers and provide effective product services. The IoTelectronic devices are
remarkably tiny and require less power consumption. The easy network accessibility and existence
of more IoT devices, the growth is rapidly increasing. Due to this, the data security to network is still
vulnerable. In this paper, we use EDIA (Enhanced Algorithm for Data Integrity and Authentication)
Algorithm for data integrity and raspberry pi, as nodal manager to manage the integrity of nodes in
low-powered devices. The enhanced lightweight algorithm and authentication for the devices are
examined with encryption and decryption process and are demonstrated with Arduino Uno and
Raspberry Pi boardsand finally the results and discussion with the future work are discussed.
541. ECG–BASED HEARTBEAT CLASSIFICATION FOR ARRHYTHMIA

DETECTION USING TRANSFORMER MODEL
Vaishnavi Kumari
Department of ECE
ASET
Amity University
Madhulika Dhar
Department of ECE
ASET
Amity University
Dr. Shubhra Dixit
Department of ECE
ASET
Amity University
It's necessary to extract meaningful information from the massive data of electrocardiogram (ECG)
recordings of patients to generate a judgement on heart function. ECG classification with high
May 21st – 22nd, 2022
precision can save doctors and cardiologists time while also lowering the risk of misdiagnosis. Some
deep learning-based styles may now conduct point selection and classification successfully while
consuming less force. An end-to-end deep learning frame supported convolutional neural network
(CNN) for ECG data processing and arrhythmia classification is suggested in this paper. A
Transformer model is embedded in CNN to capture the temporal information of ECG signals in the
frame, and a relief link restriction is incorporated to the loss function to improve the classification
performance of the model. The testing results showed that the transformer model pays more
attention to the signals' temporal durability and captures the data's deep properties better. The link
constraint enhances the constraint on the fundamental characteristics and efficiently reduces the
outcomes' influence of imbalance. The result achieved by the model shows the 95.56% training
accuracy and 96% validation accuracy for transformer model.
542. ON-SITE WASTE MINIMIZATION PRACTICES AND EFFECTIVE QUALITY

MANAGEMENT AS A WAY OF REDUCING ENVIRONMENTAL FOOTPRINT OF
THE CONSTRUCTION INDUSTRY: A REVIEW
Dzuthono Tunyi1, Monika Verma 2
PG Student1, Assistant Professor 2
School of Civil Engineering, Lovely Professional University, Jalandhar, India.
Construction waste is one of the front-runners for ecological corruption in light of its over-utilization
of valuable natural materials. However, the waste minimization practices are left unexplored and
there is an absence of an efficient research that waste minimization practices and effective quality
management goes hand in hand to cut down the environmental footprint of the construction industry
which is needed for future undertaking. This paper intends to provide an understanding of the
behaviour of waste generation on-site and seek solutions from a sustainability perspective that is
workable for waste reduction. A systematic review of 54 quality papers were done and compared to
find feasible solutions like the various approaches of waste minimization practices, advocating
environment protection through 7R’s, leveraging smart technologies as a way of waste minimization
practice and adopting Quality Management of waste as “The New Normal” in Construction Industry.
543. OBSTACLE AVOIDANCE ROBOT USING ARDUINO

Sruthi Raghavan
Vellore Institute of Technology, Chennai
In this paper, we propose the building of an Obstacle avoidance robot using Arduino. The robot is
meant to move in a particular direction and ignore obstacles in its way. If it has an obstacle in front,
left, and right of it, then it moves in the backward direction. The robot uses an ultrasonic sensor to
detect an object and send a signal to the Arduino. The Arduino helps the robot move in a defined
direction. The microcontroller controls the motors left, right, back, front, based on ultrasonic
signals[1]. The Mechanism and Working of this robot are summarized in this work. This robot also
has a wide range of applications and can be extremely useful to mankind[2].

May 21st – 22nd, 2022
544. ARTIFICIAL INTELLIGENCE BASED VOICE ASSISTANT

SruthiRaghavan
Vellore Institute of Technology, Chennai
Voice bot is one of the most popular AI applications that became popular in the 21st Century.[1] We
must have heard at least a few voicebot names in our day-to-day life such as Alexa from Amazon,
Siri from Apple and Bixby from Samsung etc. But these types of voicebots can be only used by
people who are all rich and well versed in technology. Therefore, we have developed a project based
on voicebot. Our project is going to represent the development of cubic (specialized voice bot)
which is specifically made for providing virtual assistants. We have made the cubic in such a way
that it can be helpful through voice chat. We have developed the website in such a way that the user
can talk with the cubic (voice bot) to get the answer through the voice of cubic. The cubic uses
Google to find the answer for the questions that are asked by the users and helps in finding the date
and time (IST) and many more. In the current world where everyone depends on technology, our
cubic is a cheap and efficient way to find the things around the world for the people who are not that
much familiar with technology. We would like to conclude this project as a very great and enriching
experience to interact with the interesting field of voicebot and their application.
545. PRAGMATIC ANALYSIS OF E-COMMERCE SUCCESS FACTORS AND

DESIGN ATTRIBUTES
Debiprasad Sahoo
School of computer Science and Engineering
Lovely Professional University
Phagwara, Punjab (India) Singh Purushottam Manoj
Phagwara, Punjab (India) Tanima Thakur
Phagwara, Punjab (India)
Web designing is the creation of websites and web pages to portray a company's brand value and
information and to provide a user-friendly experience. It mainly focuses on UI/UX aspects of
website development rather than software development. Till the mid-2010s, web designers
mainlyfocus on designing websites for desktops; however, design for small screen devices such as
smartphones and tablets has became more critical after that [1].The website has become a primary
middleware used by companies, institutions, individuals to establish and elevatetheir brands. It also
has been recognized as a crucial factor for the acceptance, trust-worthiness, success of any company,
e-commerce, or individual. Web design is based on specific attributes such as; typography, color,
navigation, quality of the content, interactivity, and webpage layout. It is directly proportional to
trust, satisfaction, and loyalty. This review paper conveys the various aspects of web design
attributes addressed to get users' positive outcomes. The outcome of this review paper extends vital
support to various design attributes and other success factors of an e-commerce website.

May 21st – 22nd, 2022
546. PNEUMONIA DETECTION FROM CHEST X-RAY IMAGES BASED ON

Medhun Chandrasekaran,Hariesh Vasu, Abimanyu Ganesan,
IV-B.E CSE, Dr. Mahalingam College of Engineering and Technology
Pollachi, India.
Dr. G. Anupriya Professor, CSE Department,
Dr. Mahalingam College of Engineering and Technology Pollachi, India
Pneumonia is a life threatening illness that affects the lungs of humans. This is a kind of disease
which makes breathing very difficult for a person. Pneumonia is a menacing ailment, therefore it is
necessary to detect pneumonia as early as possible. Chest X-ray exam is one of the most frequent
and cost-effective medical imaging examinations. However clinical diagnosis of chest X-ray can be
challenging. This work focuses on pneumonia detection using three distinct algorithms such as
VGG16, DenseNet201, XceptionNet based on method of transfer learning.
547. GRID AND RENEWABLE ENERGY SOURCE DRIVEN INDUSTRIAL MOTOR

WITH POWER FLOW MANAGEMENT
Sasikala.A1, Shlok Prakash2, Rahul.V3, Divagar.R4
This paper proposes a scalable and dependable unidirectional power flow control system for a
solarassisted system with grid support. Solar photovoltaic (PV) arrays serve as a backup power
source, ensuring that the system remains operational at all times in the event of a primary power
failure. This system has a shorter payback period and is a more efficient solution in the current
energy climate. As part of this research, a brushless DC motor (BLDC) drive is used to demonstrate
the motor drive's essential switching action. Additionally, as noted earlier, this controller offers
power optimization for photovoltaic arrays as well as voltage balancing between split dc-link
capacitors under all operating conditions. The approach has been tested with a variety of industrial
motors. Additionally, this work investigates a novel method for simulating a landsman converter's
power factor adjustment. To determine the responsiveness of the proposed unidirectional grid
control system, both MATLAB simulation and actual experimental data are used. Due to the fact
that the photovoltaic array is just a supporting component of the overall design, it is a cost-effective
and energy-efficient system that requires little to no ongoing maintenance.
548. CLOUD COMPUTING

K.Sivapaquiavady Assistant professor,Computer science dept.BGCW
PUDUCHERRY
Resource sharing in a pure plug and play model that dramatically simplifies infrastructure planning
is the promise of „cloud computing‟. The two key advantages of this model are ease-of-use and

May 21st – 22nd, 2022
costeffectiveness. Though there remain questions on aspects such as security and vendor lock-in, the
benefits this model offers are many. This paper explores some of the basics of cloud computing with
the aim of introducing aspects such as: Components in the model Characteristics and Usage of the
model and exploring options available for complementing your technology and infrastructure needs.
549. ANALYSIS OF A PROJECT COST OVERRUNS AND DELAYS USING RII

METHOD
Hardeep Singh Jassal1, Monika Verma2
PG Student1, Assistant Professor2
School of Civil Engineering, Lovely Professional University, Jalandhar, India.
The construction industry covers agreat extent of the industry world and it contributes largely to the
economy section globally. Time and Cost are the main parameters including Quality which
contribute to the success of any vision of construction but simultaneously, cost overruns and delays
are the dreaded factors that affect itdetrimentally. Considering the present scenario, it is very
important to analyse the causes for which the runover of cost and delay are occurring. In this paper,
the literature review is carried out to figure out the various reasons and aspects of holdup and
runover of cost inconstructionprojectsbyemploying the RelativeImportantIndex(RIImethod). The
study is grouped into 8 factors to signify the role of these factors in Time Delay and runover of
Costs in construction undertaking. The study of literature review enlightens how these factors
impact the construction project and, on that basis, a research gap is found. Moreover, suggestions
and proposals are made for future research.
550. FINITE ELEMENT MODELLING AND SIMULATION OF GLASS FIBRE

REINFORCED INDUSTRIAL SAFETY SHOES USING ANSYS
S.Vishnu1, A.Vasanthanathan2
1PG Scholar, Department of Mechanical Engineering, MEPCO SCHLENK Engineering College, Sivakasi,
India
2Associate Professor (Sr.Grade), Department of Mechanical Engineering, MEPCO SCHLENK Engineering

College, Sivakasi, India
Safety shoes are one of the most use personnel protective equipment which are worn at all times
inside an industry unlike other PPE which can be removed when not in use. Hence safety shoes stays
with the worker at all times, when the steel toe shoes are worn it provides additional weight for the
legs which can cause musculo-skeletal disorders with the prolonged use. The project focus on
replacing the steel toe caps with composite toe caps. In this paper, composite toe caps are modelled
in Solidworks software. The 3D model of composite toe cap is imported into ANSYS software for
finite element analysis and simulate for compression loading test and impact load test. The results of
the analysis are then tabulated and evaluated.
551. IDENTIFYING ELECTRICAL FAULT DETECT USING MACHINE LEARNING

TECHNIQUE
Mrs.R.Jeena1, K.Prathana2, S.Ramya3 R.Subashini4
May 21st – 22nd, 2022
Associate Professor, Department of Information Technology, Panimalar Institute Of Technology,Chennai,

India.1
Student, Department of Information Technology, Panimalar Institute of
Technology Chennai,India2,3,4
Electrical fault detection is frequently utilized on each labeled and unlabeled data which is
acknowledged as unsupervised Electrical fault detection for unlabeled data. In this project, the
Electrical fault detection is done according to the dataset and the machine learning algorithms are
used on the dataset and the comparison of the model is accomplished for better prediction. Machine
learning is progressively being used to automate Electrical fault detection. In most cases, the
possible short current of a predictable defect may be estimated. Before the analysis, it simply
identifies the data for cleaning. However, in a real-world system, it is very difficult and expensive to
obtain a massive amount of images to find the fault. The proposed technique uses lots much less
information than the Deep learning technique to obtain fault analysis under variable working
conditions and does not reason an over fitting problem. Finally, experimental on a real-world dataset
using the Machine Learning technique has a fault diagnosis accuracy of 94.7 %.
552. SKIN CANCER DIAGNOSIS USING AI COUPLED WITH IMAGE PROCESSING

Dr. Nitin Mishra, Dr. Saumya Chaturvedi, and Sweta.
Galgotias University, Sharda University, Galgotias University, Greater Noida, Uttar Pradesh, India
Skin diseases consists of a wide range of ailments that affect the skin, including microbial infections,
viral, fungal, allergies, epidermis malignancies, and parasitic diseases. In South-Asian countries like
India people don’t care much about the skin conditions. In our country, people prefer home remedies
to cure skin conditions instead of visiting a dermatologist which can lead to serious skin conditions.
Early diagnosis of skin disease is very important as it can reduce the severity of the condition.
Melanoma is the deadliest type of skin cancer, and it is the most prominent form of cancer.
Melanoma could be diagnosed early, which would reduce overall illness and death. The odds of
dying from the ailment is proportional to the extent of the malignancy, which is proportional to the
length of time it has been growing. The keys to early detection are patient self-examination of the
skin, full-body skin screenings by a dermatologist, and patient engagement. This work aims to
categorize skin cancer into two types: malignant and benign. Two different approaches were used.
Starting with a simple Convolutional Neural Network and then moving on to transfer learning. The
Dataset we have used in this work consists of train data with 2637 images and test data with 660
images belonging to 2 classes, malignant and benign. This work aims to categorize skin cancer into
two types: malignant and benign. Two different approaches are used. Starting with a simple
Convolutional Neural Network and then moving on to transfer learning. The major goal is to
evaluate the performance of the two models and then see how transfer learning can improve
accuracy, particularly when there is not enough data.
553. ANALYSIS OF ACCURACY RATE IN CREDIT CARD FORGERY

Mr.N. SENTHAMILARASI¹, AYEESHA.S², SANGEETHA.D³, SNEGHAA.R.K⁴

May 21st – 22nd, 2022

India1.
Student, Department of Information Technology, Panimalar Institute of Technology, Chennai, India2,3,4
A credit card that remains a very widespread compensation technique is accepted on-line & offline
that has cashless transactions. It’s a simple, suitable then quite common to make payments and
alternative transactions. With the rise of developments credit card frauds also are growing. Monetary
deception is addictive worldwide statement improvement. These actions are accomplished so
gracefully that it is similar to real transactions. Therefore, easy style practices and alternative less
composite way are non-operation. In directive to minimalize disorder and convey order in place
having a well-organized technique of fraud detection has become a necessity for all banks. In this
paper we used Machine learning, to notice Credit Card fake transactions. To notice and prevent the
fraudulent, slightly of these approaches are often applied on bank credit card scam detection system,
to notice and prevent the scam. The proposed system uses four type of ML algorithm. The average
accuracy of the system in classifying using Random Forest machine learning algorithm is found to
be highest among others.
554. AUTOMATIC CLASSIFICATION OF EVENT RELATED POTENTIAL USING

NEURAL NETWORK
Shwet Thaker, Sivakumar Rajgopal
Department of Sensors & Biomedical Technology, School of Electronics Engineering
Vellore Institute of Technology,Vellore, Tamil Nadu, India
The P300 is an endogenous component of the event-related potential (ERP) that exhibits a positive
deflection in the scalp-recorded electroencephalogram (EEG) and is induced around 300
milliseconds after the exposure to an infrequent stimulus, generally auditory or visual. The presence
of P300 signals in an EEG signal can be used to predict whether or not the data contains any
response to a certain auditory or visual stimulus. The recurrent neural network architecture is most
suited for constructing neural networks that will receive time series signals as input, such as EEG in
our case. EEG signals could be sent directly into this neural network, and it would be possible to
anticipate whether or not the signal contains any response to auditory or visual stimuli. These
predictions will make it easier to understand the condition of individuals in comas, or those who are
incapable of responding to the surrounding stimuli physically.
555. STUDY AND ANALYSIS OF VARIOUS CONTROL ALGORITHMS FOR

COUPLED TANK SYSTEM
Ms. J. Nivetha
Assistant Professor
Department of Instrumentation and Control Engineering
Peri Institute of Technology
R. KOTTEESWARAN., M.E., Ph.D.,
Associate Professor
Department of Instrumentation and Control Engineering
St. Joseph’s College of Engineering

May 21st – 22nd, 2022
The four-tank system is reviewed in this paper and this is a common mechatronic laboratory
configuration in control theory. This research aims to identify the best controller for a four-tank
system (4TS) with dual input forces. For the level control of 4TS, the optimal control technique is
described, and it is one of the finestmethods in terms of presentations. Among the different
controller schemes created are the H2, H∞ controller, linear quadratic regulator (LQR) and linear
quadratic Gaussian regulator (LQGR) systems. The system's tank level is then controlled using a PI
controller, a PID controller, and a FOPID. To explore the influence of various controller systems on
the 4TS controlled state, these controllers were provided to this significant mechatronic system
(4TS) independently and their outputs for disturbance rejection was compared. Various
computational approaches for the control process of a connected tanks system are discussed and
analyzed in this research. The determination of the appropriate water level in the tanks might be
stated as an optimal control issue for a meaningful operating decision since the dynamics of the
connected tanks system are nonlinear. System optimization and parameter estimates are incorporated
on this foundation. For example, the numerical parameters of a connected tank system are
investigated, as well as the applicability of the methodologies.
556. DESIGN OF CRANE BEAMS & GANTRY GIRDERS

1Faique Ahmed, 2Shalaka Sharma
G. H Raisoni Institute of Engineering and Technology
Now a days in industries, Workshops, ware houses, on Construction sites, etc. Crane are generally
used. The work of this crane is to Carrying and transferring the materials form one place to another.
This crane is moving longitudinal as well as in lateral direction. These cranes are moving on the
Gantry girders. This is made up of I-Section With inverted C-Section on it and also with I Section
with channels and many more as shown in figure no.1.In India, This Gantry girder is design by LSM
(Limit State Method) method and before 2007 this is design by WSM (Working StateMethod) and
AISC (American Institute Steel Construction) is calculated at America. In this project We are going
to compare the LSM method with WSM method and with AISC. In this study we take different
capacity of crane for different span and different length of crane bridge. For same data we design the
Gantry Girder by LSM as well as WSM and AISC.At last, a Reference table is prepared which gives
the value and difference of all the three methods.
557. DESIGN OF CONNECTIONS AND DEVELOPING DESIGN GUIDELINES

Aradhana S. Bavistale, Shalaka Sharma, Dr. Ramesh V. Meghrajani
Now a days steel structure is growing fastest in all over the world because of its easy installation,
light in weight than concrete structure and its eco-friendly nature and economical at that time when
effect of global warming increasing rapidly. Connection is the important part of every steel structure.
In the steel structure long span, column free structure is the most important part of every industrial
structure and pre- engineered building take all the responsibility and fulfilling the requirements in
low cost and making structure economical. In this study based on IS method and AISC method
design of tension capacity, shear capacity and moment capacity has been done also design types of

May 21st – 22nd, 2022
connection. Then comparing IS methos with AISC method and in last prepared ready reckoner’s
table which gives the appropriate results.
558. ENHANCE DETECTION AND SECURITY IN ATMUSING HARR CASCADE

ALGORITHM
Mr.M.KrishnaRaj1U.Rogith2N.Susil3S.Venkataramanan4
Associate Professor, Department of Information Technology, Panimalar Institute of Technology, Chennai,
India1.
Student, Department of Information Technology, Panimalar Institute of Technology, Chennai,India2,3,4
Automated teller machines (ATMs) are standard devices generally used by Individuals to hold
out a spread of personal and business monetary transactions and/or banking functions. ATMs
have became very fashionable with the general public for his or her accessibility and general user
friendliness. ATMs are now found in several locations having a regular or high volume of shopper
traffic as an example, ATMs are generally found in restaurants, supermarkets, bankingcentre,
airports, entertainment establishments, transportation facilities and a myriad of other locations.
ATMs are typically available to consumers on a continual basis such consumers have the flexibility
to carryout their ATM monetary transactions and/or banking functions at any time of the day and on
any day of the week .For this purpose we are using face recognition step with haar cascade classifier
to find out the features of face .when face is detected then it will give access to do transactions
otherwise not allow.
559. A SURVEY ON FLIPPING AND PICK-AND-PLACE MECHANISMS USED IN

THE FIELD OF ROBOTICS
Anusima Das
ECE DEPARTMENT
(Affiliated of AMCEC)
AMCEC
(Affiliated of VTU)
Bangalore, India
Dr. G. Senbagavalli
ECE DEPARTMENT
AMCEC
(Affiliated of VTU)
Bangalore, India
Chithra Y
ECE DEPARTMENT
AMCEC
(Affiliated of VTU)
Bangalore, India
Dr. T. Kavitha
ECE DEPARTMENT
AMCEC

May 21st – 22nd, 2022
(Affiliated of VTU)
Bangalore, India
Inchara D
ECE DEPARTMENT
AMCEC
(Affiliated of VTU)
Bangalore, India
Machinery has been an integral part of human civilization since industrialization. Machines have
aided human beings in various works of life while retaining the accuracy of work and increasing the
speed. Down the time lane, human beings have advanced on to merge machines with an intelligence
of their own and built robots, namely merging artificial intelligence with machines to make them
into robots that can do a wide array of work. This paper is based on various academic papers and
journals to do case study on how machines and robots vary from one another, yet similar. We have
study papers based on operations like automatic parking systems, paper counting machines, aids for
people with disabilities and much more to better understand how machines and robots have aided
man kind in various ways. We take the help of microcontrollers, various kinds of motors, actuators,
sensors and various kinds of robotic manufacturing materials to build up a study automated machine.
560. EVALUATING THE THERAPEUTIC POTENTIAL AND EFFECTIVENESS OF

EXTENDED REALITY EXPOSURE THERAPY IN THE TREATMENT OF PHOBIAS
AND EMOTION REGULATION
Gautam Bhatia
School of computing,
DIT University,
Dehradun, India
Adityaraj Malhotra
DIT University,
Dehradun, India
Jahanvi Negi
School of Computing,
DIT University,
Dehradun, India
Sanjeev Sharma
DIT University,
Dehradun, India
Virtual reality, mixed reality, and augmented reality are included within the phrase Extended
Reality. Extended Reality (XR) is revolutionizing how people visualize and interact with their
physical and real or the exact opposite as artificial environments, shifting away from observation
and more towards immersion.Microsoft HoloLens 2 and Google Glass Enterprise 2 are examples of
augmented reality Head-mounted Displays (HMDs) that overlay digitally generated content on the
user's glasses combined with the ability to view real objects, the Oculus Rift is a virtual reality
Head-mounted Display (HMDs) that allows users to perceive a computer simulated 3D simulated

May 21st – 22nd, 2022
environment and experience a first-person view with the ability to interact with its
elements.Applications of Extended Reality have been acknowledged in numerous disciplines and
XR was studied and considered under and there needs to be an Eagle eye study for macro-level
understanding of the subject matter and a need for prior research work collected, analyzed and
compiled in a single work outline its achievements, factors leading to possible XR adoption,
concerns about ethics, gap highlighted in current studies and future scope of research.
561. INVESTIGATIONS ON THE COUPLED EFFECT OF SPRAY ANGLE AND

FLUIDIC PARAMETERS ON THE ONSET OF COMBUSTION INSTABILITY IN
SPACE PROPULSION
Shivangi Dhawan, Eshaan Raj, Vinayak Malhotra
SRM Institute of Science and Technology, Kattankulathur, TN
One of the major concerns in propulsion is the instability of combustion which arises due to slight
fluctuations in a flow field and can lead to tremendous loss of systems. As all combustion-operated
systems undergo this issue, it is imperative and one of the aspects of space propulsion that is yet to
be comprehensively resolved. In context to this, systematic numerical simulations were carried out
utilizing Fire Dynamic Simulator and Smoke View to understand the effects of parameters,
including spray angle, flow velocity, and flow rate, that can cause changes in the flow. The
parameters were coupled to keep the study as close to a real-life scenario. Throughout the analysis, a
spray burner was considered as an example of a system, as it resembles many other real-life
situations. The driving force for carrying out this experiment was to minimize the losses that are
caused by combustion instability and to maximize the working efficiency. Results were ought to
provide good physical insight into thorough testing, validation, and designing of space propulsion
devices.
562. DESIGN OF COMPOSITE SLAB IN MULTISTORIED BUILDING

Deepak Harinkhede
The aim of the project is to develop a spread sheet which give the value of depth of slab and decking
profile by the method of euro code. In this study different types of deck panel designed for different
shapes and different embossment. Also, the comparison of experimental and analytical results of the
load-carrying capacity of composite slabs revealed that agreements between these values are
sufficiently good.
563. PHOTO VOLTAIC POWER FORECASTING USING REGRESSION ANALYSIS

FOR POWER SYSTEMS
AyushShrivastav

May 21st – 22nd, 2022
Electrical Department Visvesvaraya National Institute ofTechnology

Nagpur, India
Mohan Khedkar
Electrical Department Visvesvaraya National Institute ofTechnology
Nagpur, India
B V Surya Vardhan
Electrical Department Visvesvaraya National Institute of Technology
Nagpur, India
With the emergence of renewable energy sources in power system tend to increase its
complexity.The demand and supply fluctuations in power grid makes it unstable and electrical
quality suffers .Thus the development of Machine learning approaches are frequently employed to
transform the way renewable energy intermittency is dealt and the issue of power forecasting is
addressed.Regression analysis can be an efficient technique to predict the PV power output
power .This manuscript analyses three regression algorithms for predicting Micro Grid power,
namely TREE, SVM (Support Vector Machine) , and ENSEMBLE. All three techniques are
examined using K fold analyses.5 fold technique used
inthismanuscripttopredictdata.ThebestapproachtopredictPV power output for given data set is Tree
Fine .The value of RMSE (Root Mean Square) is 0.0157 , R2 is 0.94 ,MSE (Mean SquareError) is
0.00024 MAE (Mean Average Error) is 0.0087 and Training Time is 0.68 seconds.
564. ADVANCED HYBRID SOLID PROPELLANT FOR FUTURISTIC MISSILE

TECHNOLOGY
A. Gangulya1 and A. Kumara2
a Department of Aerospace, SRM Institute of Science and Technology
In Today’s time the rocket engines for the missiles generally depend on the cryogenic and liquid
propellants because of their high Specific Impulses (ISP ~ 450s) and control. The cryogenics have a
major issue of storage and handling. Over the past few years, the dependence on the cryogenic
propellants {i.e., fuel-> Liquid Hydrogen (LH) and oxidizer- >Liquid Oxygen (LO2)} is increasing
at a very high pace. Hybrid Solid propellants provide a good alternative to these. In this project, we
aim to enhance the performance of rocket engine by increasing the ISP of the propellants by using
various energetic materials {that are readily available}, and varying proportions of fuel & oxidizer to
obtain a new composition that can be prepared in a room temperature to create a Hybrid Solid
Propellant. Our work is driven by the need to reduce the over-dependency on liquid and cryogenic
propellants, and find an alternative to these propellants. The performance is evaluated in terms of
Specific impulses (Isp) and Characteristic velocity (C*) both of which are significant parameter.
565. DATA MINING AND NEW AGE ELECTIONS

Arjun M.V
SJC bangalore
Elections, universal adult franchise, free thinking.These are the pillars for what democracy stands
for, to uphold and protect these values should be the ultimate debitum for any democracy. The title
May 21st – 22nd, 2022
stands for the imperilment technological interception i.e., data mining poses to free thinking which is
one of the pillars of democratic ideals. “Thought defines what man chooses” so datamining is
euphemized manipulation. Election rigging has evolved over the years, earlier it was a voluntary
choice i.e., whether you wanted to sell your vote but now it has become a very involuntary action
i.e., you are manipulated without you ever knowing you were. The negation of choice being the
main point. Data Mining companies collect your data from various sources which may include social
media applications, internet activity, income levels etc. The principal on how to influence your
decisions I.e., choices is pretty easy. The human mind is intangible but certain factors determine
what we think which is the demographical factors. The essence of datamining is to exploit these
factors and influence elections. One such incident what’s the Cambridge Analytica scandal in which
the data of Facebook users were mined and the data which is used to split you into a psycho-graph
which is then used to formulate personalized advertisements which appealed to the demographic
dividend. The trump administration allegedly used search data mining techniques to win the
elections. The term paper will try to study how elections were swung in Donald trump’s favor. The
IT cell of the BJP too has resorted to similar tactics to garner sentiments. We have reached the
pinnacle off election rigging. To correlate data mining with elections the paper will also look into a
little bit of guerilla marketing psychology and political science. So, what are elections today have
they become predictable have we lost democratic relevance or values because of the technological
interface? In the era of fake news and data mining, our choices have become obsolete to some
extent. With new advancements like these “Political Science” is going to explore new spheres.
566. INTELLIGENT CRUTCH TOOL FOR BODY POSTURE MANAGEMENT

SYSTEM
Mrs.R.JEENA¹, B.ESHITHA², VISHNU PRIYA.M³, YAZHINI KUMAR.S⁴
Associate Professor, Department of Information Technology, Panimalar Institute of Technology,
Chennai, India1
Student, Department of Information Technology, Panimalar Institute of Technology,
Chennai, India2,3,4
In the last decade, the clinical reasoning in physical therapy has been to develop systems for
physiotherapists to make clinical decisions effectively and efficiently. Some studies show the
importance of walking aids during rehabilitation from some diseases and elderly patients with
balance disorders. In this project we design a solution for the patient using crutch tool, this device
helps them to maintain the proper balance. This smart crutch tool is equipped with sensors. The
sensor embedded in the walker were chosen to extract the relevant information related to the
walker’s use during the physiotherapists sessions like applied forces on the walker. This device will
automatically warn the walker’s when they apply more pressure on a single side using haptic
feedback. Along with this we included the automatic fall detection system using MEMS Sensor, if
the sensor reaches its threshold value, it’ll automatically alert the surrounding people using buzzer
sound. This walker has an automatic internal lighting system based on the surrounding environment
and obstacle detection using ultrasound for blind people.
567. PREDICTION OF HOUSE PRICE USING SML TECHNIQUE

Mr.KRISHNARAJ¹, HARINI PRIYA.R², HARINI.A³, HARINI.P⁴

May 21st – 22nd, 2022

India1.
Student, Department of Information Technology, Panimalar Institute of Technology, Chennai, India2,3,4
Predicting the growth of the real estate market is one of the most challenging problems. It can be
described as one of the most important processes to anticipate. This is a difficult endeavour with
many unknowns. To avoid this issue in the future, Using machine learning algorithms, one of the
most fascinating (or possibly most profitable) time series data. As a result, house price forecasting
has become a popular research topic. The goal is to forecast house price prediction outcomes using
machine learning techniques in an error-based approach. Additionally, the proposed machine
learning algorithm technique can be compared with best accuracy MAE,MSE,R¬2.
568. MULTI OUTPUT FLY BACK CONVERTER USING SOLAR ENERGY

Anbazhagan.A¹,Sanjana.S.M2, Rakshana.P3, Srimahalakshmi.S.R⁴
Flyback converter is known by everyone because it is simple in design ,count of parts is lesser and
it has isolation. So, it occupies less volume and cost can also be saved. DC-DC converter is used to
increase or decrease the voltage level and divided into non –isolated and isolated categories .
Flyback converter is an isolation transformer, when mosfet switch is ON that stores energy from
the MPPT controller and it delivers energy to the load side if mosfet switch is OFF. Due to this the
need of additional inductor for the storage of energy can be eliminated unlike it can be used in the
case of forward half and full bridge topologies. Solar energy is used as an input to the Flyback
converter and multiple output is obtained for both low and high power application. MPPT controller
is used along with solar panel to extract maximum power and constant output voltage is obtained in
output side to produce higher efficiency for wide range of input solar irradiation level.
569. EOG SIGNALS – A BIOMARKER FOR NEURODEGENERATIVEDISEASES

KirubaVashiniB,Logeswari V, SangaviV K Subhashini
Department of Electronics And Communication Engineering
Neurode generative diseases(NDD)are one of the most serious health problemswith increasing age.
In the fieldof NDD it is essential to diagnose as early as possible, patients affected by NDD can be
post observed at regular intervals andtheirprogresscanbemonitored.Toimprovediagnosticaccuracy,
bio medical signal analysis is essential. Abnormalitiesof eye movements are commonly noted in
multiple NDD. The eye movement scan be detected using Electrooculogram(EOG). EOG is the
electrical activity caused by cornea-retinalstanding potential due to the movements of the eyeballs.
Thiswork proposes the design and development of a low cost EOGacquisition system that detect the
different eye ball movements.The system consists of Electromyography sensor and ArduinoUNO.
The system detects horizontal and vertical movements ofthe eyeball. In this system the Thingspeak
May 21st – 22nd, 2022
platform is used fordigital filtration, analysis and displaying the EOG signal. Thestored data is
exported in excel format where different types ofsignal analyses have to be performed. The main
objective ofEOG signal acquisition is to examine the NDD patients andmonitortheirimprovements.
570. ON THE STRUCTURE AND DYNAMICS OF LINEAR TIME VARIANT AND

NON LINEAR TIME VARIANT FLAMES
Saurabh Randive, Jayesh Sharma, Mir Raza Ali Naqviand Vinayak Malhotra
Department of Aerospace Engineering, SRM Institute of Science and Technology
We loose precious lives, natural resources and a huge amount of money every year due to large scale
fires. The proposed project aims to investigate the non-linearity and time-sensitiveness of external
energy sources in uncontrolled self igniting and propagating fires and the effect of airflow on such
fires. The experiments will be performed in a setup with matchsticks arranged in non linear
configurations. The experiments will be recorded by optical shadowgraphy, these recordings will be
used to collect useful data that would give an insight about the behaviour of fire propagation, flame
spread rate and energy transfer. The results obtained from this investigation would lead to a better
understanding of the phenomenon and the prediction of occurrence of such devastating fires might
become possible. It may help in devising solutions like designing systems that help in putting off
large scale fires even before they start or in their initial stages without much destruction.
571. FACE RECOGNITION BASED ATTENDANCE MANAGEMENT SYSTEM

IMAGE PROCESSING
Rahul Jadhav, Rupesh Padature, Sanchit Kulkarni, Padmanabh Kale, Mrs. Mansi Bhonsle
G. H. Raisoni College of Enggineering and Management- Wagholi, Pune
In this age of modern technology, which is developing at a rapid pace, there is no reason why such
an important event in education as attendance should be done in the old, boring and traditional way.
Face Recognition Attendance Management System saves a lot of time and energy for both the
teaching staff and the students. Attendance is monitored by the face recognition algorithm by
recognizing only the face of the students from the other objects and then marking the students as
present. The system is fed in advance with the images of all the students enrolled in the class. Using
this data, the algorithm recognizes the students present and matches the features with the already
stored images of the students in the database.
It uses the image of a student and compares it with a dataset of images provided during registration.
Then it identifies the students and marks their attendance. Also, the attendance report of the students
is automatically generated and forwarded to the respective recipient.
572. COMPARATIVE STUDY OF ENSEMBLE METHODS FOR CLASSIFICATION

IN HEALTHCARE DATA
S. Sathurthi, Research Scholar, Dept. of CSE, Puducherry Technological University, Puducherry

May 21st – 22nd, 2022
Electronic health record systems are adapted during a good deal of health care facility to enhance the
quality of patient care which is maintained electronically. Classification methods are the foremost
widely used algorithms in Healthcare sector because it helps predicting the status of patient by
classifying patient records and find the class that matches the new patient record. Ensemble learning
is one of the supervised learning approaches required for constructing the classification model. In
this paper, our main goal is to compare the performance of various ensemble methods used in
healthcare data classification. Experiments are conducted using Python with UCI Machine learning
repository diabetes and breast cancer datasets. Results of the conducted experiments showed that all
ensemble methods are predictive and can give relatively better result. Henceforth we conclude that
the ensemble learning method is more suitable in handling the classification problem in the
healthcare domain.
573. OPTIMIZATION OF VISUAL AND THERMAL COMFORT IN GREEN

BUILDING: A REVIEW
Shubham mishra
Mechanical Engineering
Manmohan Tiwari
Shubhank Khare
Md.Ozair Arsad
Subodh Kant Ojha
Thermal and visual comfort are very important for office culture. By optimizing thermal and visual
comfort we can increase the productivity of the occupants within the building. In this paper we have
analyzed different research papers based on thermal and visual comfort. This paper also describes
various difficulties which one is facing in optimization of a green building. There are various
research papers which shows the trends on analysis of green building which uses the various
software-based simulations. Through which we can simulate different objective functions and
achieve the desired results, which takes us more towards the green building.
574. ASSISTIVE SYSTEM FOR BLIND PERSON

Syed Amaanullah1, Rajesh Kumar DV2, Prajwal AB3,Farhana Kausar4
May 21st – 22nd, 2022
1,2,3 Atria Institute of Technology, Visvesvaraya Technological University, India ,
Visual insufficiency is the inability to recognise objects due to physiological or neurological factors.
The absence of development in an optic nerve or visual focus of the eye is referred to as incomplete
visual impairment. Human vision helps them to see their surroundings. Blind people use a device to
overcome the limitations of existing technology, allowing them to read printed and electronic textual
resources such as books, newspapers, bills, labels, and e-mails, as well as detect objects. Scanning,
image pre-processing, text extraction from the image, and narration through the speakers are the four
serially integrated procedures used by the gadget. High-quality photos in the focus window could be
taken, and then a succession of image processing techniques could be used to de-skew, Hough line
transform, and binarize the image while removing unnecessary noise to provide the best output for
OCR. The device's primary strengths include its portability, ease of use, near real-time operation,
large scanning field, rechargeable battery with long runtime, low development cost, and lack of
reliance on external assistance. Additionally, the device can be used while commuting.
575. DENSE RECURRENT NEURAL NETWORK ARCHITECTURE TOWARDS

PARTICIPANT PREDICTION FOR EVOLVING EVENTS IN EVENT BASED
SOCIAL NETWORK
VADIVAMBIGAI.S, DR.S.GEETHARANI
PSG COLLEGE OF ARTS & SCIENCE

The Event Based Social Network is a distributed platform for organizing the online social events, in
which events are submitted to the user groups on associated topic, location and time period.
Usually, event based social network systems contain abundant events and large scale of user
participating on event, there exist an overhead on clustering the user for the event. In order to model
a participant prediction, deep learning architecture for the user recommendation has become
necessary. However, many existing work using machine learning model leads to scalability and
sparsity issues. Further events and user profiles continually arrive and evolve in terms of behaviour
and knowledge, which leads to the issues of task cold-start. In order to overcome the above
mentioned challenge, a novel dense recurrent neural network architecture using deep neural network
for participant recommendation on basis of evolution of user knowledge and preference has been
formulated in this research. Initially proposed idea deals with the dynamics of the user and event in
the event based social network by extracting the features with multiple latent factors using feature
extraction technique. Latent factor extracted on basis of the event and user features according to
their experience and historical behaviors. Further deep neural network has been incorporated to
model a participant recommendation system to event on basis of latent feature extracted towards
prediction on timely basis to achieve high reliability, accuracy and latency. Deep learning
architecture schedules the participants to event on embedding latent features on generation of the
objective function to produce recommendation with minimum error. It results in significant increase
on the prediction performance for discriminative information’s of the event and user. Extensive
experiments have been conducted on real datasets to compare proposed model with conventional.
The experimental results show that deep learning architecture can achieve both effectiveness and
good scalability on large scale data.

May 21st – 22nd, 2022
576. ASSESSMENT OF SOIL MOISTURE PROFILE USING SMART REMOTE

SENSING AND IOT TECHNIQUES
Devesh Mishra1*, Krishna Kant Agrawal2, and Ram Suchit Yadav1
Department of Electronics and Communication, University of Allahabad, Prayagraj – 211002 U.P, India
School of Computing Science and Engineering, Galgotias University, Greater Noida – 203201 U.P
The world is facing an acute surge in the population, which enforces the high water and agricultural
yield leading to the need for smart technology that safeguards the scarce water re-source and at the
same time enhances agricultural yield by maintaining soil health. Specifically for agricultural-based
countries such a smart water management system is the prime need for sus-tainable development. In
such scenarios either the excess or deficiency in soil moisture content leads to poor crop health and
hence the acute need for an appropriate water balance sensor model is pivotal. Motivated by this
concern this paper focuses on a smart soil moisture monitoring system by remote sensing the
moisture level of the target patch. The advanced soil profile monitoring-based remote sensing unit is
installed which explores the Internet of things (IoT) network. The proposed experimental setup uses
electrochemical Energy Harvesting (EH) to solve the problem of constrained battery issues for
WSNs. The setup measures the moisture level of the soil and thus can be utilized for several
applications including a water conservative farming system, detecting forest fire, and also landslide
detection. The data processing unit uses NodeMCU 12E IoT Module and storage is performed by
cloud-based Blynk service. The result reveals that the deployment of the proposed smart sensing unit
can detect and monitor soil profiles using IoT networks even at remote and other inaccessible
locations.
577. DESIGN OF VEDIC MULTIPLIER USING QUANTUM DOT CELLULAR

AUTOMATA FOR CRYPTOGRAPHIC APPLICATIONS
B.Shanmugi,Dr.V.VinothThyagarajan,Dr.V.R.VenkataSubramani,Dr.S.Rajaram,bshanmugi968@gm
a il.com, [email protected], [email protected], [email protected]
Abstract—Cryptography has achieved widespread recognition as a secure method of protecting

sensitive information. Multiplication is one of the most demanding process in cryptographic
algorithms as well as most time consuming. A more efficient and faster multiplier can be utilized to
boost the cryptosystems speed. The vedic mathematics makes arithmetic operations easier. There are
sixteen algorithms available in vedic mathematics. Vedic polynomial multiplication is performed
using Urdhva-Tiryagbyham sutra from vedic mathematics. The speed of multiplication is improved
due to parallelism in the nature of the vedic method. The vedic multiplication algorithm enhances
the calculation speed by doing concurrent addition and partial product generation. Quantum-dot
Cellular Automata (QCA) is a new nano technology trend that is well suited to the construction of
high performance and low power integrated circuits. This paper employs a new architecture for a
vedic multiplier based on Urdhva-Tiryagbyham sutra. This paper discusses the usage of a vedic
multiplier to accomplish finite field multiplication. With the help of QCA Designer tool, the
functional simulation of the circuit was verified. In addition, the QCA Designer-E tool is used to
calculate the energy dissipation of QCA circuits. The area, latency, and energy dissipation of the
vedic multiplier built with QCA technology have all been thoroughly researched. The suggested
vedic multiplier provides much higher device density, lower cell count, area and clock delay than
existing multipliers cited in the literature.

May 21st – 22nd, 2022
578. ANALYSIS OF FORENSIC IMAGES DETECTION FOR SUPERIMPOSED OF AN

IMAGE AND OTHER IMAGES.
Dr. S. Suma Christal Mary
Department of Information of Technology 1. G. Monisha, 2. M. Bhuvaneshwari, 3. R. Kavya shri
Panimalar Institute of Technology
Detection of fake image remains a difficult problem. Detecting fake images can be a difficult task to
find the authenticity of an image with the naked eye. In recent years, with the improvement of digital
resources, falsification of images has become a serious social problem. Normally, overlayed digital
images have two types of counterfeit detection. Copy move forgery is an important type of digital
operation that involves duplicating a portion of an image to hide certain information. The other is the
actual fake detection on multiple overlay images. For the most common copy motion forgery
detection and multi-frame forgery, the robustness and accuracy of the existing method can be further
improved. The reliability and reliability of digital images is becoming more and more important
because digital images can be easily modified. Therefore, the ability to identify image tampering is
the focus of current research, and an important area of digital image authentication is copy transfer
counterfeit detection (CMFD). Copy move forgery is to copy and paste the same image from one
area to another. DBSCAN is a clustering technique that finds a cluster key from points in a CMF
image. Error Level Analysis (ELA) is a method of detecting counterfeit products from multiple
images. This document outlines the recent development of forensic FD and describes the entire
related FD process. In particular, it describes the normal FD (false positive) workflow of the
functional matching and mining process using a key point-based approach. Instead of listing the
records and certifications used in the document, completed process by identifying a copy of and
verifying the forgery found in the image.
579. ADVERSE DRUG REACTION ANALYSIS USING SUPPORT VECTOR

MACHINE MODEL
Mrs.D.Sathya, M.E Assistant Professor,
Ms.K.Deepika,Ms.S.Gowreeshwari,Ms.K.Divya
In medical domain, Adverse Drug Reaction (ADR) analysis is a crucial process for doctors and
medical scientists. Adverse drug reaction measures the injury occurred due to usage of a drug. The
growing concern to the ADRs has stimulated the progress of statistical, data mining methods to find
the Adverse Drug Reactions. This project proposed a hybrid model of data mining and machine
learning to classify different Adverse Reactions and foretell the outcome intensity. It used the
Proportionality Reporting Ratio (PRR) along with Chi-Square test equations to find out the
differentrelationships between drug and symptoms called the drug-ADR association. In addition,
support vector machine
method is applied to classify the data set records into either normal or adverse drug. Moreover, our
project aims infinding the percent of adversity of the drug reaction. Based on the number of
occurrences, where a specific drug P,causes a specific Adverse Reaction, R, the various terms are
May 21st – 22nd, 2022
measured for PRR and Chi Square. In addition KNN and SVM classification is made on drug
records to classify them based on data set columns. Neural network basedclassification is the
proposed system to classify drug reaction based on data set columns.
580. ARTIFICIAL INTELLIGENCE AND TELEGRAMCHAT-BOT BASED

INTELLIGENT AUTHENTICATION SYSTEM
Mrs. R.Dharani1. Nivitha Bharathi K2. Viveka M3. Yuvashree K4.
Associate Professor, Department of Information Technology, Panimalar Institute of Technology, Chennai,
India1
Student, Department of Information Technology, Panimalar Institute of Technology, Chennai,
India2,3,4
In the modern world, the IoT is at its peak. The world is getting smarter. Home automation is
emerging. Intruder attacks and burglary zones have been a serious problem in recent days. Currently,
many systems use motion sensors and devices, but these systems cannot be trusted as they can be
activated with little weather changes such as heat. The Smart Door Control System is the latest
technology in home automation. The main objective of Smart Door technology is to provide users
with complete, simple, and comfortable door security. The system can detect and identify a person
from previously provided data. The proposed system can detect a face and reorder a person to
determine if the person in the frame is a known person or an unknown person. The aim is to
strengthen the door automation technique using embedded and Android phones. The embedded
system is attached to an RGB camera and has implemented three software modules: image
capturing, face detection, and face recognition. The PC captures an image of a person as soon as the
bell is pressed and sends the still image to the Android device via the telegram app, which notifies
the owner about the person standing at the door. All programming of the system is set up in Python
and C Embedded for the Controller.
581. CLOCK BASED TRANSMITTER IDENTIFICATION AND ATTACK

RECOGNITION USING INTRUSION DETECTION
Ms.E. Angel Anna Prathiba.M.E., P.Kokilaveni, D.Hena Ranjani ,
V.Poornimadevi
Foamy Disturbance Intrusion Detection Systems (FIDSs) can assume a significant part in identifying
and forestalling security assaults. Combination of the web into the elements of the various areas of
human culture (like brilliant homes, medical services, shrewd networks, fabricating processes, item
supply chains, and ecological observing) is arising An interruption location system is viewed as a
central wellspring of security for data and interchanges innovation. Notwithstanding, regular
interruption location techniques should be adjusted and improved for application to the Internet of
Things attributable to specific limits, similar to asset obliged gadgets, the restricted memory and
battery limit of hubs, and explicit convention stacks. In this work, we foster a lightweight assault
location methodology using a regulated AI based FIDS to distinguish a foe endeavoring to infuse
superfluous information into the organization. Reproduction results show that the proposed FIDS -

May 21st – 22nd, 2022
based classifier, supported by a blend of a few in complex elements, can perform sufficiently as far
as arrangement precision and recognition time.
582. DETECTION OF PROSTATE CANCER USING J48 ALGORITHM

Mr.M.Senthil Kumar ME., A.Kalaiyarasi, Assistant professor/CSE, Final year CSE, Erode Sengunthar
Engineering College, Erode Sengunthar Engineering College, Perundurai,Erode Perundurai,Erode
C.Aboorva, C.Sudha, Final year CSE, Final year CSE, Erode Sengunthar Engineering College, Erode
Sengunthar Engineering College, Perundurai,Erode Perundurai,Erode
Prostate disease (PCa) is a serious sort of malignant growth and makes significant passings among
men due its poor analytic framework. The pictures got from patients with carcinoma comprise of
perplexing and important elements that can't be removed promptly by customary symptomatic
strategies. Free of hand-made includes, and is calibrated. Anyway, there have been relatively few
assessments that have portrayed the periods of partition of prostrate CT pictures. As needs be, in this
article, we propose an Ensembled Transfer Learning (ETL) design to arrange well, moderate and
insufficiently isolated prostrate CT pictures. We use AI calculation to portion prostate disease.
583. CUSTOMER PERCEPTION TOWARDS ONLINE FOOD BUSINESS APPS

1Dr.P. Vanitha2Dr.R.Kamaraj 3Gokul P4 Naveen C5Arulmurugan S
1Associate Professor, Department of Management Studies, M. Kumarasamy College of Engineering,
Karur, Tamil Nadu, India
2Professor ,MBA K.Ramakrishnan College of Engineering . Trichy.
3, 4, 5 Student Second Year MBA, M. Kumarasamy College of Engineering, Karur, Tamil Nadu,
India
The main aim of this research is to examine the consumer perception towards online food business
apps and to analyze the perception towards quality of apps and websites of electronic food ordering
companies among the respondents in Karur district .Both primary and secondary data has been used
to collect the information. Technology Acceptance Model questionnaire model has been used to
collect the data. Sample Size is 120.Kruskal Wallis test has been used to find the relationship
between variables.
584. SPEECH EMOTION RECOGNITION USING MACHINE LEARNING

Mrs. B.S. DEEPA PRIYA B.E, MTech, Ph.D ,
Asst. Professor /CSE ,
Erode Sengunthar Engineering College, Erode
Nandene S, Parkavi S, Surya S

Erode Sengunthar Engineering College, Erode

May 21st – 22nd, 2022
Communication is the key to express one’s thoughts and ideas clearly. Among stall forms of
communication, speech is the most preferred and powerful form of communications in human. The
era of the Internet of Things (IoT) is rapidly advancing in bringing more intelligent systems
available for everyday use. These applications range from simple wear ables and widgets to complex
seslf driving vehicles and automated systems employed in various fields. Intelligent applications are
interactive and require minimum user effort to function, and mostly function on voice-based
input.This creates the necessity for these computer applications to completely comprehend human
speech. A speech percept can reveal information about the speaker including gender, age, language,
and emotion. Several existing speech recognition systems used in IoT applications are integrated
with an emotion detection system in order to analyze the emotional state of the speaker. The
performance of the emotion detection system can greatly influence the overall performance of the
IoT application in many ways and can provide many advantages over the functionalities of these
applications. This research presents a speech emotion detection system with improvements over an
existing system in terms of data, feature selection and methodology that aims at classifying speech
percepts based on emotions, more accurately.
585. FACE MASK DETECTION SYSTEM

ISHAN CHANDEL
Electronics and communication engineering, Galgotias University, Greater Noida
The COVID-19 the virus, caused by coronavirus, is spreading rapidly around the world. COVID has
had a profound impact on almost every aspect of development. The health care system is different.
Wearing a mask is one of the steps that can be taken to prevent the spread of infection. In this study,
we suggest how to find people who do not wear a mask in public places where there are CCTV
cameras. MobileNetV1 is a machine learning tool used to create a face mask application. Collecting
photos of masked and uncovered people, or taking ordinary photos of people and using a custom CV
text to hide them, are the first steps in producing a model, followed by previous data analysis. and
data filtering, testing, and finally modeling. The algorithm is 98.71- It is 99.90 percent accurate in
determining whether a person wears a mask.
586. QoS ENABLED LOAD BALANCING IN CLOUD USING GREY WOLF

OPTIMIZATION ALGORITHM
Dr. Santhi G
Puducherry Technological University, Puducherry, India
Yaswanth B, Sesha Sai Prasanna Krishna Niladhuri, Priyadharshini E

Puducherry Technological University, Puducherry, India
Cloud computing is an emerging computer technology, that provides distributed, scalable, elastic
computer resources to the end-user over the Internet. One of the most challenging tasks in the cloud
May 21st – 22nd, 2022
computing environment is task scheduling. The main objectives of task scheduling are to identify the
appropriate resources for scheduling a specific task on time, utilize the resources more efficiently,
and reduce the total completion time of all input tasks to be executed. Many load-balancing methods
try to solve this problem by metaheuristic algorithms, and each attempts to enhance the operation
and efficiency of systems. In this paper, we propose a task scheduling algorithm using the
metaheuristics approach. The proposed scheduler is based on the grey wolf optimizer nature-inspired
algorithm. In this method, first, the GWO algorithm tries to find the unemployed or busy nodes and,
after discovering this node, tries to calculate each node’s threshold and fitness function. The
experimental results prove the quality and efficiency of the proposed method.

TENTH INTERNATIONAL CONFERENCE ON
CONTEMPORARY ENGINEERING AND
TECHNOLOGY (ICCET 2022)
LIST OF PARTICIPATING ORGANIZATIONS
S.NO INSTITUTION/ORGANISATION
1 ILLINOIS STATE UNIVERSITY, USA
2 SOUTH EASTERN UNIVERSITY OF SRI LANKA
3 A.S.COLLEGE, KHANNA
4 ABES ENGINEERING COLLEGE,GHAZIABAD
5 ACADEMY OF MARITIME EDUCATION AND TRAINING, CHENNAI
6 AGNI COLLEGE OF TECHNOLOGY, CHENNAI
7 AISSMS COLLEGE OF ENGINEERING, PUNE
8 AISSMS'S INSTITUTE OF INFORMATION TECHNOLOGY, PUNE
9 ALAGAPPA ARTS COLLEGE, KARAIKUDI
10 AMITY UNIVERSITY, JAIPUR
11 AMITY UNIVERSITY, LUCKNOW CAMPUS
12 AMITY UNIVERSITY, NOIDA

13 AMRITA SCHOOL OF ARTS AND SCIENCE, MYSURU
14 AMRITA SCHOOL OF ENGINEERING, AMRITA VISHWA VIDYAPEETHAM, CHENNAI
15 AMRITA VISHWA VIDYAPEETHAM, AMRITAPURI, KERALA
16 ANNA UNIVERSITY, CHENNAI
17 ANNAMALAI UNIVERSITY (DEPUTED TO GCE, SRIRANGAM, TRICHY)
18 ANNAMALAI UNIVERSITY, CHIDAMBARAM
19 ASSAM DON BOSCO UNIVERSITY, GUWAHATI, INDIA
20 ATRIA COLLEGE OF ENGINEERING, BANGALORE
21 ATRIA INSTITUTE OF TECHNOLOGY, BANGALORE
AVINASHILINGAM INSTITUTE FOR HOME SCIENCE AND HIGHER EDUCATION FOR WOMEN,
22 COIMBATORE
23 B.S. ABDUR RAHMAN CRESCENT INSTITUTE OF SCIENCE AND TECHNOLOGY, CHENNAI
24 BABU BANARASI DAS UNIVERSITY, LUCKNOW
25 BANNARI AMMAN INSTITUTE OF TECHNOLOGY, SATHYAMANGALAM
26 BHARATHIAR UNIVERSITY, COIMBATORE
27 BHARATHIDASAN GOVT.COLLEGE FOR WOMEN, PUDUCHERRY
28 BHARATI VIDYAPEETH (DEEMED TO BE UNIVERSITY) COLLEGE OF ENGINEERING, PUNE

29 BIT MESRA, RANCHI
30 BMS COLLEGE OF ENGINEERING, BANGALORE
31 C K COLLEGE OF ENGINEERING & TECHNOLOGY, CUDDALORE
32 C. K. PITHAWALA COLLEGE OF ENGINEERING & TECH, GUJARAT
33 CDAC, PUNE
34 CENTRE FOR DEVELOPMENT OF ADVANCED COMPUTING (C-DAC), PUNE
35 CET DEPT
36 CHENNAI INSTITUTE OF TECHNOLOGY, CHENNAI
37 CHRIST (DEEMED TO BE UNIVERSITY), BANGALORE
38 COIMBATORE INSTITUTE OF TECHNOLOGY, COIMBATORE
39 COLLEGE OF ENGINEERING GUINDY, ANNA UNIVERSITY
40 COLLEGE OF ENGINEERING PUNE, PUNE
41 DELHI TECHNOLOGICAL UNIVERSITY, NEW DELHI
42 DEPARTMENT OF LEGAL METROLOGY, GOVERNMENT OF KARNATAKA
43 DIT UNIVERSITY, DEHRADUN

44 DMI ENGINEERING COLLEGE, CHENNAI
45 DR. HARISINGH GOUR VISHWAVIDYALAYA SAGAR, M.P
46 DR. MAHALINGAM COLLEGE OF ENGINEERING AND TECHNOLOGY, POLLACHI
47 DR. N. G. P. INSTITUTE OF TECHNOLOGY, COIMBATORE
48 FAMT, RATNAGIRI, MAHARASHTRA
49 FRANCIS XAVIER ENGINEERING COLLEGE, TIRUNELVELI
50 G H RAISONI COLLEGE OF ENGINEERING AND MANAGEMENT, PUNE
51 G. NARAYANAMMA INSTITUTE OF SCIENCE AND TECHNOLOGY (FOR WOMEN), HYDERABAD
52 G.H. RAISONI COLLEGE OF ENGINNERING, NAGPUR
53 GALGOTIAS COLLEGE OF ENGINEERING AND TECHNOLOGY, GREATER NOIDA
54 GALGOTIAS UNIVERSITY, GREATER NOIDA
55 GIET UNIVERSITY, GUNUPUR
56 GITAM UNIVERSITY, BENGALURU
57 GMR INSTITUTE OF TECHNOLOGY, ANDHRA PRADESH
58 GOVERNMENT COLLEGE OF ENGINEERING TIRUNELVELI, TIRUNELVELI
59 GOVERNMENT COLLEGE OF TECHNOLOGY, COIMBATORE
60 GOVERNMENT COLLEGE(AUTONOMOUS), MANDYA

61 GOVT SKSJT INSTITUTE, KR CIRCLE, BANGALORE
62 GOVT. SHYAMA PRASAD MUKHERJEE COLLEGE SITAPUR, CHHATTISGARH
63 GSSS INSTITUTE OF ENGINEERING AND TECHNOLOGY FOR WOMEN, MYSURU
64 HINDUSTAN INSTITUTE OF TECHNOLOGY AND SCIENCE, CHENNAI
65 HIRANANDANI FOUNDATION SCHOOL, MUMBAI
66 INDIAN INSTITUTE OF ENGINEERING SCIENCE AND TECHNOLOGY (IIEST), SHIBPUR, WEST BENGAL
67 INDIAN INSTITUTE OF INFORMATION TECHNOLOGY, KANCHEEPURAM
68 INDIRA GANDHI DELHI TECHNICAL UNIVERSITY FOR WOMEN. NEW DELHI
69 INSTITUTE OF AERONAUTICAL ENGINEERING, HYDEARABAD
70 ISRO (RETD.)
71 JADAVPUR UNIVERSITY, KOLKATA
72 JAI NARAYAN VYAS UNIVERSITY, JODHPUR
73 JAIN( DEEMED TO BE UNIVERSITY), BANGALORE
74 JERUSALEM COLLEGE OF ENGINEERING, CHENNAI
75 JNTUH, HYDERABAD
76 KALASALINGAM ACADEMY OF RESEARCH AND EDUCATION, SRIVILLIPUTHUR
77 KONGU ENGINEERING COLLEGE, ERODE
78 AMC ENGINEERING COLLEGE, BANGALORE
79 KPR INSTITUTE OF ENGINEERING AND TECHNOLOGY, COIMBATORE
80 LOVELY PROFESSIONAL UNIVERSITY, PUNJAB
81 LOYOLA COLLEGE OF ARTS AND SCIENCE, NAMAKKAL
82 M S RAMAIAH INSTITUTE OF TECHNOLOGY, BANGALORE
83 M.KUMARSAMY COLLEGE OF ENGINEERING, KARUR
84 MADAN MOHAN MALAVIYA UNIVERSITY OF TECHNOLOGY, GORAKHPUR
85 MADRAS UNIVERSITY, CHENNAI
86 MAHENDRA COLLEGE OF ENGINEERING, SALEM
87 MALAVIYA NATIONAL INSTITUTE TECHNOLOGY, JAIPUR
88 MARRI LAXMAN REDDY INSTITUTE OF TECHNOLOGY, HYDERABAD
89 MEPCO SCHLENK ENGINEERING COLLEGE, SIVAKASI
90 MLRITM, HYDERABAD
91 MNIT, JAIPUR
92 MOTHER TERESA WOMENS UNIVERSITY, KODAIKANAL

93 NAGARJUNA COLLEGE OF ENGINEERING & TECHNOLOGY, BENGALURU
94 NANDHA COLLEGE OF PHARMACY, ERODE
95 NATIONAL ENGINEERING COLLEGE, KOVILPATTI
96 NATIONAL INSTITUTE OF TECHNICAL TEACHERS TRAINING & RESEARCH, CHENNAI
97 NATIONAL INSTITUTE OF TECHNOLOGY KARNATAKA, SURATHKAL
98 NATIONAL INSTITUTE OF TECHNOLOGY, RAIPUR, CHHATTISGARH
99 NETAJI SUBHAS UNIVERSITY OF TECHNOLOGY, DELHI
100 NIT JALANDHAR
101 NMIMS, MUMBAI
102 PAAVAI ENGINEERING COLLEGE (AUTONOMOUS), NAMAKKAL
103 PANIMALAR ENGINEERING COLLEGE, CHENNAI
104 PANIMALAR INSTITUTE OF TECHNOLOGY, CHENNAI
105 PERIYAR UNIVERSITY, SALEM
106 PLINTRON MOBILITY SOLUTIONS, CHENNAI
107 PONDICHERRY UNIVERSITY COMMUNITY COLLEGE, PONDICHERRY

108 PPGIT, COIMBATORE
109 PSG COLLEGE OF ARTS AND SCIENCE, COIMBATORE
110 PUNJABI UNIVERSITY, PATIALA
111 RAJALAKSHMI ENGINEERING COLLEGE, CHENNAI
112 RAJIV GANDHI INSTITURE OF TECHNOLOGY, MUMBAI
113 RAMAIAH INSTITUTE OF TECHNOLOGY, BENGALURU
114 RMK COLLEGE OF ENGINEERING AND TECHNOLOGY, THIRUVALLUR
115 SANTHIGIRI COLLEGE, THODUPUZHA, KERALA
116 SARANATHAN COLLEGE OF ENGINEERING, TRICHY
117 SATHYABAMA INSTITUTE OF SCIENCE AND TECHNOLOGY, CHENNAI
118 SAVEETHA SCHOOL OF ENGINEERING (SIMATS), CHENNAI
119 SAVEETHA SCHOOL OF ENGINEERING, CHENNAI
120 SCSVMV UNIVERSITY, KANCHEEPURAM
121 SCSVMV(DEEMED TO BE UNIVERSITY), KANCHEEPURAM
122 SDNB VAISHNAV COLLEGE FOR WOMEN, CHENNAI
123 SHIV NADAR UNIVERSITY, GREATER NOIDA
124 AMITY UNIVERSITY, UP

125 SHRI MATA VAISHNO DEVI UNIVERSITY, KATRA, J&K
126 SHRIMATHI DEVKUNVAR NANALAL BHATT VAISHNAV COLLEGE FOR WOMEN, CHENNAI
127 SIES GRADUATE SCHOOL OF TECHNOLOGY, NAVI MUMBAI
128 SIES GST, NAVI MUMBAI
129 SMVDU, KATRA
130 SNIT ADOOR, KERALA
131 SOE P P SAVNI UNIVERSITY, SURAT, GUJARAT
132 SOUTH EASTERN UNIVERSITY OF SRI LANKA
133 SREE VIDYANIKETHAN ENGINEERING COLLEGE, TIRUPATI
134 SRI KRISHNA COLLEGE OF ENGINEERING AND TECHNOLOGY, COIMBATORE
135 SRI KRISHNA COLLEGE OF TECHNOLOGY, COIMBATORE
136 SRI RAMAKRISHNA ENGINEERING COLLEGE, COIMBATORE
137 SRI SAIRAM ENGINEERING COLLEGE, CHENNAI
138 SRI SAIRAM INSTITUTE OF TECHNOLOGY, CHENNAI
139 SRI SIVASUBRAMANIYA NADAR COLLEGE OF ENGINEERING, KALVAKKAM

140 SRI VENKATESWARA COLLEGE OF ENGINEERING & TECHNOLOGY, CHITTOOR
141 SRI VENKATESWARA COLLEGE OF ENGINEERING, SRIPERUMBUDUR
142 SRM INSTITUTE OF SCIENCE AND TECHNOLOGY, KATTANKULATHUR
143 SRM INSTITUTE OF SCIENCE AND TECHNOLOGY, RAMAPURAM
144 SRM VALLIAMMAI ENGINEERING COLLEGE, CHENNAI
145 SSN COLLEGE OF ENGINEERING, KALAVAKKAM,
146 ST PETER'S INSTITUTE OF HIGHER EDUCATION AND RESEARCH, CHENNAI
147 ST.JOSEPH'S COLLEGE OF ENGINEERING, CHENNAI
148 ST.JOSEPH'S INSTITUTE OF TECHNOLOGY, CHENNAI
149 ST.PETER'S INSTITUTE OF HIGHER EDUCATION AND RESEARCH, CHENNAI
150 SURYA GROUP OF INSTITUTIONS, VIKRAVANDI
151 SYMBIOSIS INSTITUTE OF DIGITAL AND TELECOM MANAGEMENT, PUNE
152 T.K.R COLLEGE OF ENGINEERING & TECHNOLOGY, TELANGANA
153 TEEGALA KRISHNA REDDY ENGINEERING COLLEGE, MEERPET
154 THE NATIONAL INSTITUTE OF ENGINEERING, MYSURU
155 THIAGARAJAR COLLEGE OF ENGINEERING, MADURAI
156 TIGER ANALYTICS, CHENNAI

157 UNIVERSITY COLLEGE OF ENGINEERING, KANCHEEPURAM
UNIVERSITY COLLEGE OF SCIENCE TECHNOLOGY AND AGRICULTURE, CALCUTTA UNIVERSITY,

158 KOLKATA
159 UNIVERSITY OF CALCUTTA, KOLKATA
160 VARDHAMAN COLLEGE OF ENGINEERING, HYDERABAD
161 VEL TECH RANGARAJAN DR.SAGUNTHALA R&D INSTITUTE OF SCIENCE AND TECHNOLOGY, CHENNAI
162 VELAMMAL COLLEGE OF ENGINEERING AND TECHNOLOGY, MADURAI
163 VELLORE INSTITUTE OF TECHNOLOGY, CHENNAI
164 VIDYAVARDHAKA COLLEGE OF ENGINEERING (VVCE), MYSURU
165 VIGNAN'S INSTITUTE OF INFORMATION TECHNOLOGY, AP
166 VISTAS, CHENNAI
167 VIT, VELLORE
168 VNR VJIET, HYDERABAD
169 VR SIDDHARTHA ENGINEERING COLLEGE, VIJAYAWADA
170 ANNAMALAI UNIVERSITY (DEPUTED TO GCE, SRIRANGAM, TRICHY)
171 EASWARI ENGINEERING COLLEGE, CHENNAI

172 ERODE SENGUNTHAR ENGINEERING COLLEGE, ERODE
173 K.RAMAKRISHNAN COLLEGE OF TECHNOLOGY, TIRUCHIRAPALLI
174 MEDCUORE MEDICAL SOLUTION PRIVATE LIMITED, CHENNAI
175 PERI INSTITUTE OF TECHNOLOGY, CHENNAI
176 PUDUCHERRY TECHNOLOGICAL UNIVERSITY, PUDUCHERRY
177 SELIMA CORP, CHENNAI
178 SHARDA UNIVERSITY, GREATER NOIDA
179 SJC, BANGALORE
180 UNIVERSITY OF ALLAHABAD, ALLAHABAD
181 VELALAR COLLEGE OF ENGINEERING AND TECHNOLOGY, ERODE
182 VEMU IT, CHITOOR
183 VISION FOR WISDOM, TEMPLE OF CONSCIOUSNESS, ALIYAR
184 VISVESVARAYA NATIONAL INSTITUTE OF TECHNOLOGY, NAGPUR

ICCET 2022 - Proceedings Release v1

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

ICCET 2022 - Proceedings Release v1

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ICCET 2022 - Proceedings Release v1

Uploaded by

Copyright:

Available Formats

10th INTERNATIONAL CONFERENCE ON

“CONTEMPORARY ENGINEERING AND TECHNOLOGY –

May 21st – 22nd, 2022

ORGANIZATION OF SCIENCE AND INNOVATIVE

PRINCE SHRI VENKATESHWARA PADMAVATHY

PRINCE DR. K. VASUDEVAN COLLEGE OF ENGINEERING

Medavakkam - Mambakkam Road, Ponmar, Chennai – 600 127

Proceeding of International Conference on Contemporary Engineering and

21st & 22nd May, 2022

Organization of Science and Innovation Engineering & Technology

4/3, Vallalar Street, Choolaimedu, Chennai – 600094

Sabanayagam Publications, Chennai.

VICE CHAIRMAN’S MESSAGE

Dr. V.Vishnu Karthick,

Prince Shri Venkateshwara Padmavathy Engineering College.

ADMINISTRATIVE OFFICER’S MESSAGE

Prince Group of Educational Institutions

Dr. T.Sunder Selwyn

OUR GROUP OF INSTITUTIONS

Prince Shri Venkateshwara Padmavathy Engineering College

Prince Dr. K. Vasudevan College of Engineering & Technology

Prince Shri Venkateshwara Arts and Science College

Venkateshwara Higher Secondary School, 9,Kannagi Street, Madipakkam to Puzhuthivakkam, Chennai 91

Thiru. Dr. K. Vasudevan, M.A., B.Ed., Ph.D,

Thiru. Dr.V.VishnuKarthik, MD, Vice-Chairman, Prince Group

Administrative Officer, Prince Group of Educational Institutions.

Dr.Sunder Selwyn, Principal,

Dr. K. Aldrin Karunaharan, Assistant Professor, International Maritime College

Dr. Vaithyasubramanian.S, Assistant Professor, PG and Research Department of

Dr. S. Vijayakumar, IEEE Vice Chairman, PSES – Madras Chapter

Dr. M.Jeyakarthic, Assistant Director, Tamil Virtual Academy, Anna

Dr. Vaithyasubramanian.S, Assistant Professor, PG and Research Department of

Dr. G. N. R. PRASAD, Associate Professor, MCA department, Chaitanya

Dr. T Jayakumar, Assistant Professor, Department of Mechanical, AMET

Dr. H.S.Niranjana Murthy, Associate Professor, Electronics & Instrumentation,

Dr. Harshavardhan Awari, Senior Assistant Professor, Computer Science,

Dr. K. Sudhakar, Associate Professor, Computer Science and Engineering,

Dr. Velappagari Sekhar, Associate Professor & HOD, Kuppam Engineering

Dr. C Sakthi Gokul Rajan, Assistant Professor (Senior), Kuppam Engineering

Dr. G Rajesh, Teaching and Research Assistant, SENSE, VIT university,

Dr. C Kavitha, Associate Professor, Electronics and Communication

Dr. Raviprakash ML, Associate Professor, Department of CSE, Kalpataru

Dr N Vinodhkumar, Associate Professor, Electronics and Communication

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

S.NO PAPER ID PAPER TITLE

1. UWB MICROSTRIP PATCH ANTENNA FOR WIRELESS SENSOR

2. MULTILAYER PERCEPTRON ANALYZED THE AUDIENCE PERCEPTION

ISBN 978-93-81288-22-1 Prince Shri Venkateshwara Padmavathy Engineering College