How to Cite:

Nithya, J., Vignesh, S. R., Devadarsan, A., & Venkatesh, S. (2022). AI based contactless
attendance monitoring and management system. International Journal of Health
Sciences, 6(S3), 10600–10614. https://doi.org/10.53730/ijhs.v6nS3.8365

AI based contactless attendance monitoring

and management system

Nithya J.
Professor, Department of Information Technology, K.S.Rangasamy College of
Technology, Tiruchengode – 637215
*Corresponding author email: [email protected]

Vignesh S. R.
UG Students, Department of Information Technology, K. S. Rangasamy College of
Technology, Tiruchengode – 637215

Devadarsan A.
UG Students, Department of Information Technology, K.S.Rangasamy College of
Technology, Tiruchengode – 637215

Venkatesh S.
UG Students, Department of Information Technology, K. S. Rangasamy College of
Technology, Tiruchengode – 637215

Abstract---The traditional methods of attendance marking need to be

replaced with an Artificial Intelligence based attendance monitoring
system. The process of checking attendance using the existing
methods is time consuming and open to easy fraud. The usage of Bio-
metrics poses the threat of viruses. A contact-less Attendance
Management System (CAMS) is proposed to overcome the above
challenges. AI based attendance monitoring system will record the
students attendance automatically while they enter the class, by
recognizing their faces. The in-time and out-time of the students are
also recorded on a regular basis. The Holistic Face Matching algorithm
is used to compare the facial images. The total time duration, the
students were present inside the class can be easily calculated. There
is no chance of fake attendance as each student has to record his or
her facial identity. The main objective is to achieve reliability,
accessibility and security in a very efficient way. The results showed
that by interfacing a face recognition system with a server, a real-time
attendance system can be built and be monitored automatically.
Finally, the student’s attendance details can be updated in the
academic management system.

Keywords---face recognition, RFID, IoT.

International Journal of Health Sciences ISSN 2550-6978 E-ISSN 2550-696X © 2022.

Manuscript submitted: 18 Feb 2022, Manuscript revised: 27 April 2022, Accepted for publication: 9 June 2022
10600
 10601

Introduction

The (IoT) Internet of Things refers to the system of interrelated, internet-connected

objects that are able to collect and also transfer data over a wireless network
without any human intervention. Artificial intelligence (AI), is a branch of
computer science that focuses on building and managing technology that can
learn to autonomously make decisions and carry out actions on behalf of a
human being. In any organization, the management of attendance will involve an
important role. The duration in which an employee is inside an organization, is to
be noted to calculate pay rolls. The difficulties in the traditional system are time-
consuming, higher cost, and insecurity almost a teacher spending more quality
hours for getting and storing student’s attendance.

This issue must be approached in a new and sophisticated technological manner

than a traditional way. There are various softwares and advanced technologies
that they even allow blurred images to be rendered enough and investigated to
know the personality of the individual. Facial recognition technology is capable
enough to verify the identity of an individual by analyzing a picture or video
footage. The objective of this paper is to make a face recognition based automated
attendance system. For obtaining a better performance, the test and training
images of this project are limited to frontal and upright facial images which
consist of single face only. Both the test and the training images need to be
captured using the same device to ensure there is no quality difference. If
possible, the owner or the person having the rights to access the database can
add high quality images captured from a high quality camera and later that image
to the database, but as mentioned only the administrator or the person having
the rights to access the database can enroll or remove the students or faculty
data from it. In educational institutions, managing attendance of
students/candidates is a tedious task, as there happen to be a large number of
students and keeping track of all of them is hard. There are certain situations
where students can act as proxies for their friends even though they are not
present.

Face Recognition

A Facial Recognition System is capable of comparing and matching a digital image

or a video frame against a database that consists of facial images. It is usually
used to authenticate users by ID verification services. It functions by pinpointing
and measuring the facial features from the given image. When the RFID card is
recognized by the RFID reader, the Pi camera module scans the face of the user.
The scanned image is compared with the pre existing image in the database using
Holistic Face Matching Algorithm. Face authentication allows users to unlock
their device just by staring at the front of their device. In the initial stage, user is
authenticated by capturing his/her image and compares with those already
stored. If there's a match then attendance process starts, else the system
terminates. The primary priority is to test the lighting condition and noise within
the background before capturing the image. Then, the image detection of faces are
done using haar cascade algorithm. Finally, the face is compared or recognized
with face of student already stored. If match is found, attendance is written or
updated.
10602

A methodology is used to avoid fake attendance and proxies, there by reducing

the light intensity problem and head pose problem. A portable device has been
developed to monitor the attendance of the students using face recognition
technology. This is aimed to avoid human errors and to reduce time. A RFID
based attendance system is designed to operate as a standalone system by
sending E-mail notification to the users. Face recognition system is established
for marking the attendance of the students. The clustering of the facial images is
done with the help of PCA. RFID tag and RFID reader is used by another system
to track the attendance. It focuses on privacy and security related issues.

Literature Review

Facial recognition technology could be a framework or software which is capable

enough to verify the identity of the person by analyzing an image or video footage.
RFID uses radio waves to identify and track objects and track tags within the
following Literature review , various methods of existing attendance marking
systems are clearly analysed. Attendance is an important part of daily classroom
ascertainment for the teacher for his or her smooth running of class. At the
beginning and ending of the class, usually teacher check the attendance, but the
manual attendance system may lead to appear that a teacher may miss someone
or some students may answer multiple times. Now a days, Machine Learning has
been highly explored for computer vision applications.[1] So, we use the concept
of machine learning in Face – recognition for automatic attendance systems. In
this paper, we perform the face recognition and face detection algorithms, to
provide the computer systems the ability of finding and recognizing human faces
fast and precisely in images or videos so that the systems can be used in giving
attendance.

The conventional methods of attendance management such as calling by name is

time consuming with a chance of proxy attendance. The following face recognition
system is used to maintain attendance record of students.[2] The daily attendance
of students is recorded individually by subject which happens to be already stored
by the administrator. When the corresponding subject time arrives the system
automatically starts taking snaps and also applies face detection and recognition
technique to the given image and the students that are recognised are marked as
present and their attendance is updated with corresponding time and subject id.
Various machine learning techniques have been used to develop this system,
histogram of oriented gradient method is used to detect faces in images and the
Deep learning method is used for computing and comparing the facial features of
the students to recognise them. The system is capable of identifying multiple
faces in real time.

The Face can be defined as the representation of one’s identity. An automated

student attendance system is created based on face recognition technique. This
system happens to be very useful in day to day life applications such as in the
security and surveillance systems. Security systems in airports use face
recognition to identify suspects. CBI (Central Bureau of Investigation) and FBI
(Federal Bureau of Investigation) use face recognition for the purpose of criminal
investigations.[3] In this paper, video framing is also performed by accessing the
camera through an user friendly interface. The face gets detected and is
 10603

segmented from the video frame by means of using HOG (Histogram of Oriented
Gradient) algorithm.

Attendance by means of calling names or passing around sign-in sheets is time-

consuming and receptive easy fraud. This paper presents the detailed
implementation of a real-time roll call system supported face recognition and its
results. The system must first take and save an image of the scholar as a
reference in a very database. During the attendance checking process, the online
camera takes pictures of the scholars face to be recognise then the face is
automatically detected by a computer and therefore the student name who
presumably matches the image is identified. An excel file is updated for
attendance record supported the face recognition results. [5]In this method, a
pretrained Haar Cascade model is employed to detect faces from web camera
video. A Face-Net, trained by minimizing the triplet loss generates a 128-
dimensional encoding for a face image. The similarity between the encoding of two
facial images determines whether the two facial images are coming from the
identical students or not. this method has been used for a category, and therefore
the results are very satisfactory. A survey has been conducted to research the
pros and cons of the attendance system on college education management.

In the modern meeting room, smart systems to make attendance quickly are
mandatory. The existing systems perform manual method of attendance such as
registration and fingerprint. Despite the fact that fingerprint method can reject
the unknown person and give the grant access to the known person, it takes time
to register a person one-by-one. It is possible to create long queues for fingerprint
checking before the meeting room entrance. Machine learning, with the Internet
of Things (IoT) technology is the best solution as it offers various advantages when
getting applied in the meeting rooms. This method used is to create a presence by
means of detecting the faces. In the paper, a facial recognition authentication
based on machine learning technology for connection to the meeting rooms is
provided. Also a specific website for displaying the detection result and data
storage design testing is developed. The method uses the Dlib library for deep
learning purposes, OpenCV for video camera processing, and Face Recognition
for Dlib processing. The proposed system allows placing as much cameras as
needed in a meeting room. The various tests conducted are one known person,
identification of one unknown person, identification of two people, and three
people. The parameter that needs to be focused on is the required time for
detecting the number of faces recorded by the camera. These results reveal that
the face can be recognised or not recognised, after which it is displayed on the
website.

In most of the organizations, attendance is taken manually by calling their

register numbers and names. This in turn is noted in attendance registers issued
by the heads of departments as proof and in certain organisations the students
are supposed to sign in these sheets which are in turn stored for future
references. [8] This method additionally makes it more complex to track all the
students attendance and also makes it difficult to monitor individual student
attendance in big classroom atmospheres. In this article, face detection and
recognition framework are used to continuously recognize students going to class
and by comparing their faces with database to match and mark attendance. The
10604

facial biometric framework takes a picture of the person using a camera and
contrasts that image and compares it with the image that is stored at the time of
enrolment and if it matches attendance is marked and student performance is
monitored continuously. The concept of artificial intelligence concept is used to
find out how much time the student presents in class.

Automated attendance system records the status of the particular student;

whether he or she is present in the class. This system, detects the student using
web cam when entering the class room and marks the attendance by recognition.
The system architecture and algorithms used at each stage are described in this
paper. Various real time scenarios are considered to evaluate the performance of
various face recognition systems. The paper proposes the techniques to be used
in order to handle the spoofing threats. When compared with the traditional
attendance marking, this system saves time and monitor the students effectively.
Therefore from the above conducted literature review a common thing we notice in
almost all the works is that 'Face Recognition' plays a major role in creating an
attendance based management system and that any kind of attendance
management system created would have to rely on a Face recognising algorithm.
It is understandable that there is a lot of scope in the topic Face recognition for a
future research.

Methodology

The implementation of the AI based contact less attendance monitoring system is

explained below. An RFID reader and a Raspberry Pi camera are placed at the
entrance of the building. When the person passes through the RFID reader, their
ID (RFID) card is detected and the unique number is sent via WIFI to the server.
The facial image of the person is also captured simultaneously by the Raspberry
Pi camera. The server stores the entry time and the captured facial image. The
captured image is compared with the image of the particular ID card holder
(RFID) in the MySQL Database using the Holistic Face Matching Algorithm. The
entire process for attendance monitoring system is depicted in the figure 1.

Then there are 4 conditions to be analyzed,

Condition 1
If the image and the RFID matches, the attendance is marked.

Condition 2
If the person doesn't carry an ID card when going inside the building, an alert is
raised and attendance is not marked.

Condition 3
If the person carries an ID card (RFID) and the picture taken doesn't match the
image in the MySQL database, an alert is raised and an intimation is sent via mail
to the ID card holder along with a picture captured at the entry.

Condition 4
If someone enter with unauthorized card or other card it gives an alert and it is
noted in the cloud.
 10605

At the beginning when entering the building Unique ID is read by the RFID
reader. If ID is valid, Face is captured and stored in the database. The time at
which ID is detected and facial image is captured is noted.If ID isn't valid, Face is
captured and stored in a seperate part of the database. The time at which ID is
detected and facial image is captured is noted. The captured image is compared
with the image in the existing database. If both images match, Photo and ID are
valid. Attendance is marked and the process comes to an end.If Both Images
Don't Match, Photo and ID card are invalid. An alert notification mail is sent with
captured photo. Then the process comes to an end. The system architecture is
represented in figure 2, and the components involved in implementation is
explained below.

• RFID Reader
The RFID Reader could be a device that has one or more antennas that emit
radio waves and receive signals back from the RFID tag.
• PI Camera Module
The Pi camera module could be a portable light weight camera that supports
Raspberry Pi. Communication with Pi is done using the MIPI camera serial
interface protocol. it's normally utilized in image processing, machine
learning or in surveillance projects.
• Micro SD Card
A type of very small memory card typically utilized in mobile phones and
other portable devices.
• NodeMCU ESP8266
NodeMCU may be a low-cost open source IoT platform. It initially included
firmware which runs on the ESP8266 Wi-Fi SoC from Espressif Systems, and
hardware which was supported the ESP-12 module.
• MySQL Database
MySQL is an open-source computer database management system. like other
relational databases, MySQL stores data in tables made of rows and
columns. Users can define, manipulate, control, and query data using
Structured command language, more commonly called SQL.
10606

Figure 1. Flowchart for the Attendance Monitoring System

Figure 2. System Architecture

 10607

GPIO Pins

A General-purpose input/output is an uncommitted digital signal pin on an

microcircuit or electronic card which can be used as an input or output, or both,
and is controllable by the user at runtime. The Raspberry Pi 3 module is
connected to the Power supply (5v). The Micro SD card is it's memory. The
NODEMCU ESP8266 is working as it's WIFI server. The Monitor is used to display
what's happening inside the Raspberry Pi Module. It uses a MySQL Database.
They all are connected using GPIO (General Purpose Input Output Pins). When
the RFID card is recognized by the RFID reader, the Pi camera module scans the
user-face. The scanned image is compared with the pre-existing image in the
database using Holistic Face Matching Algorithm.

Result and Discussion

Figure 3. Attendance Monitoring System

It is seen that improvement in the process of taking attendance using AI happens.

An improvement in the accuracy of taking attendance is noted. After applying the
Holistic Face Matching Algorithm the time utilised on taking attendance is
drastically reduced and this results in us focusing on other important activities.
The usage of this form of attendance management helps us in avoiding the use of
bio-metrics which have a higher percentage of resulting in spread of diseases
such as COVID. The usage of this form of Attendance Management eliminates the
chances of malpractice which are a major flaw of Manual Attendance Marking.

Report Generation

Attendance report status of the persons is given by total time noted in records
according to the conditions mentioned below: If a person is present for a time
equal to or more than 7 hours, Their attendance is marked as Present. If a person
is present for equal to or above 3 hours 30 minutes and below 7 hours, their
attendance is marked as Half-Day Present. If a person is present below 3 hours
30 minutes, their attendance is marked as absent. We can calculate the total time
by using the record of clock in and out. If we need details of a particular person
10608

we can find them by using the search bar by typing any information from the
column. The report generation is performed for a day, week or even a month.

Student Registration

A new student can register for any course from any department. The chances of
success registration depends on the availability of seats in the department. The
student can enter their fundamental data in the registration form. Once the
fundamental data are filled and the form is submitted a pop-up message
indicating that the registration is successful is displayed along with a Reference
Number. Inside the staff login, the concerned authorized staff can view the list of
people who have applied using the registration form. The student is authorized
once the registration form is viewed and the register number is allotted to the
particular student. The allotment of register number and admission of student
can be performed by The Principal, Head of Department, Coordinator and Admin.
Those who are mentioned above can also preview the fundamental data of the
students filled in the registration form before assigning the register number. Since
some students join by lateral entry/transfer, the option for viewing the semester
of joining is also included. Once authorized a pop up message is received on the
staff side declaring the message that the student has been activated successfully.
It is possible for the respective staff to download the namelist of a particular class
or department in the form of a “PDF”.

Student OD/Leave Application

A student can apply for OD/Leave using the student login. The following three
step process is involved in the procedure,

Step 1: Application of OD/Leave Form by Student

In this form the student must specify whether it is OD or leave , specify the time
duration using From/To option, Specify the reason for applying under the reason
tab and attach documents/reference ( if necessary) and submit the form. A pop-
up message is displayed saying "Successfully applied for OD".The student can
view if the OD/Leave is approved or cancelled.

Application of OD/Leave Form by Student

 10609

Step 2: Approval of OD/Leave by Class Advisor

Once the student fills and submits the OD/Leave form the next step in this
process is approval of Class Advisor. The form will move into Step 3 only if
approved by the Class Advisor. Inside the staff login the Class advisor can see the
details of the student who applied for OD/Leave. In case the Class Advisor
approves the request, a message is displayed asking for confirmation to approve
OD. Once OK, is selected a pop-up message is displayed saying OD/Leave is
granted successfully.

Step 3: Approval of OD/Leave by HOD

The next step in this process is approval by HOD. Once the HOD approves the
request OD/Leave will be generated successfully.

Once all these three steps are completed student can view the status of
OD/Leave.
10610

Screenshots

Figure 1
Attendance: ( Today attendance )

Generate Report :(Current Month Report)

 10611

Entry Time : (Current Month Entry Time )

Benefits

• Payroll Integration : Assuming that the payroll process has been automated,
time and attendance module should also be able to integrate seamlessly into
payroll functionality. It can provide accurate attendance data for the payroll
cycle that can be used to rapidly and accurately calculate salary and other
benefits.
• Clocking In & Out: It can be considered as the most important feature.
Needless to say, the main purpose of a time and attendance module is to allow
employees to clock in and out easily and keep a record of their working hours.
• Reliability & Accuracy: In the traditional method, employees report their work
hours manually. This process has many loopholes. Employees forget to log in
correct timings and there is also a chance of employees misusing the system by
giving incorrect inputs. However, having an automated attendance
10612

management removes all these problems and helps us to keep the data safe
and accurate.
• Cost Effectiveness : Getting automated attendance management system helps
in cost cutting and it is one time investment and you don’t have to pay a
human for doing this task.
• Increased productivity :Automated time & attendance management systems
save time & human effort in keeping track of employees. Now work schedules
are easily created based on job requirements, employee availability and
eligibility, and they are easily duplicated. This will free managers from routine,
manual scheduling activities so they can focus strategically on their business.
• Regular Alerts : This function of the attendance management system allows
the parents of school children to get alerts if they're children haven't reached
school and gotten their attendance marked. It is easier than ever before for
parents to be updated about their children's daily school attendance.
• Advanced Analytics : Attendance management systems powered with
advanced analytics offer granular visibility into attendance patterns,
highlighting those who frequently put in overtime or diverge from scheduled
hours.
• Leave Management : Attendance management system plays an important role
in leave management. A missed day shouldn't simply be marked as absent
instead an alert should be sent to the concerned employee with options to
apply for leave or make a special request ( such as half day).
• Online OD form : We can apply the OD through online in facilitated space ,
view the status of the OD applied of particular student by themselves.
• No long queues : It is not necessary to stand in long queue to punch the smart
card or thumb for fingerprint reader.
• No biometric issues : No biometric issues such as finger printings can't be
determined. Biometric/punch system can pose a threat to the spread of
various flu and virus.
• Identifying multiple faces : It can recognize multiple faces simultaneously in
the photo captured and give attendance.

Conclusions and Future Work

This study in particular focuses on the double confirmation of the automatic

attendance system. A two step verification RFID attendance system along with AI
will result in an efficient system. Any IoT application must be done with higher
security, reliability, faster performance and cost-effectiveness. In that respect, all
the characteristics are obeyed by the developed prototype and it works without
any glitches. Practically all academic institutions require attendance record of
students and maintaining attendance physically can be hectic as well as time
consuming task. Hence automatic maintaining of attendance with the help of face
recognition will be exceptionally useful and less prone to mistakes or errors as
compared to manual procedure. This will also reduce the manipulation of
attendance record done by students and reduces time consumption too. The
future extent of the proposed work can be, catching numerous definite pictures of
the students and utilizing any cloud innovation to store these pictures. This
framework can be designed and utilized in ATM machines to identify frauds. Also,
 10613

the framework can be utilized at the time of elections where the voters can be
distinguished by perceiving the face.

References

1. S. S. Nawaz and H. A. Saththar, “The Implications of the Industry 4.0

Concept: A Review of the Fourth Industrial Revolution,” J. Inf. Syst. Inf.
Technol., vol. 4, no. 2, pp. 8– 14, 2019.
2. S. S. Nawaz, “Embracing Industry 4.0-Guidelines for Managers.,” Int. J.
Psychosoc. Rehabil., vol. 24, no. 2, 2020.
3. R. K. A. R. Kariapper and M. S. Suhail Razeeth, “A RFID BASED (IoT)
AUTOMATIC ATTENDANCE SYSTEM: A SURVEY ANALYSIS,” Appl. Comput.
eJOURNAL, vol. 2, no. 61, 2019, doi: 10.11603/2311-9624.2014.2.3210.
4. R. K. A. R. Kariapper, A. C. M. Nafrees, M. S. S. Razeeth, and P. Pirapuraj,
“Emerging Smart University using various Technologies: A Survey Analysis,”
Test Eng. Manag. Mattingley Publ. Co., Inc. ISSN, vol. 83
5. Bhattacharya S, Nainala G S, Das P and Routray A 2018 Smart Attendance
Monitoring System (SAMS): A Face Recognition Based Attendance System for
Classroom Environment 2018 IEEE 18th International Conference on
Advanced Learning Technologies (ICALT) pp 358-360
6. Lodha R, Gupta S, Jain H and Narula H 2015 Bluetooth Smart Based
Attendance Management System Procedia Computer Science 45 pp 524–527
7. Mohamed, B. K. P and Raghu C V 2012 Fingerprint attendance system for
classroom needs 2012 Annual IEEE India Conference (INDICON) pp 433-438
8. Raghuwanshi A and Swami P D 2017 An automated classroom attendance
system using video based face recognition 2017 2nd IEEE International
Conference on Recent Trends in Electronics, Information & Communication
Technology (RTEICT) pp 719-724
9. Chen H 2014 Intelligent Classroom Attendance Checking System Based on
RFID and GSM Advanced Materials Research 989–994 pp 5532–5535
10. Poornima S, Sripriya N, Vijayalakshmi B and Vishnupriya P 2017 Attendance
monitoring system using facial recognition with audio output and gender
classification 2017 International Conference on Computer, Communication
and Signal Processing (ICCCSP)
11. Pireva K. R, Siqeca J and Berisha S 2013 RFID: Management System for
students’ attendance IFAC Proceedings Volumes 46(8) pp 137–140
12. Alghamdi S 2019 Monitoring Student Attendance Using A Smart System at
Taif University SSRN Electronic Journal pp 107-115
13. Rajan Patel, Nimisha Patel, Mona Gajjar 2012 Online Students' Attendance
Monitoring System in Classroom Using Radio Frequency Identification
Technology: A Proposed System Framework International Journal of Emerging
Technology and Advanced Engineering 2, Issue 2 pp 61-66.
14. O.K. Oyetola, A.A. Okubanjo, O.O Olaluwoye 2017 A Secure Students’
Attendance Monitoring System Journal of Engineering Technology 2, Issue 1,
pp 14-25
15. M. Olagunju, A. E. Adeniyi, T. O. Oladele 2018 Staff Attendance Monitoring
System using Fingerprint Biometrics International Journal of Computer
Applications 179 , Issue No.21
10614

16. Suryasa, W. (2019). Historical Religion Dynamics: Phenomenon in Bali

Island. Journal of Advanced Research in Dynamical and Control
Systems, 11(6), 1679-1685.
17. Suryasa, W., Sudipa, I. N., Puspani, I. A. M., & Netra, I. (2019). Towards a
Change of Emotion in Translation of Kṛṣṇa Text. Journal of Advanced Research
in Dynamical and Control Systems, 11(2), 1221-1231.