2023 6th International Conference on Contemporary Computing and Informatics (IC3I)

A Novel Approach To Human Iris Recognition And

Verification Framework Using Machine Learning
Algorithm
Dr Suneetha Davuluri Srikanth Kilaru Venugopal Boppana
Professor, Dept. of CSE, Asst. Professor, Dept. of IT Assoc. Professor, Dept. of CSE
NRI Institute of Technology Vignan's Nirula Institute of Technology NRI Institute of Technology
2023 6th International Conference on Contemporary Computing and Informatics (IC3I) | 979-8-3503-0448-0/23/$31.00 ©2023 IEEE | DOI: 10.1109/IC3I59117.2023.10397886

Pothavarappadu and Science for Women Pothavarappadu, Vijayawada, AP, India

Vijayawada, AP, India Guntur, AP, India [email protected]
[email protected] [email protected]

Dr M Vekateswara Rao Kongala Nageswara Rao Sai Srinivas Vellela

Professor, Dept. of CSE, Asst. Professor, Dept. of IT Asst. Professor, Dept. of CSE
NRI Institute of Technology Vignan's Nirula Institute of Technology Chalapathi Institute of Technology
Pothavarappadu and Science for Women Guntur, AP, India
Vijayawada, AP, India Guntur, AP, India [email protected]
[email protected] [email protected]

Abstract: The component of a computer system that is in robust security. It is simple to use the biometrics
responsibility of user security is one of the most important authentication method and there is no requirement to keep
modules. It was demonstrated that simple logins and remembering of a password, cards, or pin code.
passwords are easy for hackers to obtain and cannot guarantee
high efficiency. Biometric identity recognition is the preferred
For authentication and recognition, a number of
alternative. The most popular and commonly utilised method
for system security, according to the researchers, is biometrics. biometric characteristics, including the face, iris, palm print,
Iris-based biometric systems are the most secure of biometric and fingerprint, are utilized. Biometric systems are available
systems. A valuable and effective tool for human identification, in a range of forms, such as face recognition, retina systems
the iris pattern is a different biological feature for every fingerprint recognition voice pattern recognition, [2]. The
individual. Even though a number of approaches to this iris recognition technology is the greatest biometric system
purpose have been described, features, including the high accessible. In terms of uniqueness, stability, and non-
memory consumption, long run times, and computational contact, the iris is considered to be the king of biometrics.
complexity. New iris verification algorithms have recently Biometrics have become increasingly popular over the past
been developed as a result of the advancement of machine
year in a number of fields, including border control, identity
learning technology. Hence, a novel approach to human
recognition and verification framework using Machine authentication, and forensics [6].
Learning algorithm is described in this analysis. The texture-
based features are extracted by use of the Gray Level Co- A component of the eye called the iris controls that
occurrence Matrix (GLCM) and the Hausdorff Dimension much light gets into the eye. It's the area of the eye that
(HD). The Support Vector Machine (SVM) algorithm is used to changes color depending on melatonin pigment is in the
identify and verify the human iris in this strategy. The muscle. The individual's irises patterns are one-of-a-kind
accuracy, sensitivity, and specificity of the presented approach and structurally distinct, making it suitable for use as an
are evaluated. The robustness of the system is demonstrated by attractive goal for reliable automatic recognition of
its performance.
individuals [7]. It is considered that iris texture interesting
Keywords: Biometrics, Iris Recognition, Machine Learning, patterns at randomness from the eye's fetal development and
Support Vector Machine. are invariant to age. Hence, each eye's iris pattern can be
seen as a biometric characteristic that is universally unique
I. INTRODUCTION between twins. Iris recognition is one of the most secure and
reliable biometric identification methods and is widely used
Nowadays, the most important aspect of the systems is in mobile phones, banking, border security control, and
safety. The system is protected by authentication. There are other applications [5].
three types of authentication systems: biometric-based,
password-based, and card-based. The majority of systems Iris recognition, which uses the characteristics of a
that based on passwords are present in banks, email person's iris to identify them, is one of the most important
accounts, online accounts, etc. Companies and ATMs use biometric recognition systems [3]. The iris is one of the
card-based systems for door authentication. However, most important characteristics that can guarantee high
human tendencies make it easy to broken these systems accuracy, efficiency, and recognition rate, to a number of
protection. Because of its uniqueness, biometrics offer experiments and research. There are more than 250 different

2447
979-8-3503-0448-0/22/$31.00 ©2023 IEEE

Authorized licensed use limited to: DePaul University. Downloaded on January 28,2024 at 03:51:05 UTC from IEEE Xplore. Restrictions apply.
2023 6th International Conference on Contemporary Computing and Informatics (IC3I)

components in this feature. In the form of feature vectors, practical applications because the computational costs are
each of them is utilized to describe human identity [4]. significantly lower than those of traditional processes.
Banking can make use of the authentication that iris
recognition systems can provide, transactions involving Anfal waled al-zanganawi, Sefer kurnaz et. al., [10]
money and other systems in which security is of the most uses SVM and genetic algorithms to detect and recognize
importance. human biometrics, taking the iris as an example. Biometric
identification techniques are extremely accurate due to the
The iris recognition system has seen the development of fact that even identical twins can differ. The biometric
several methods. The researchers have confirmed in earlier identification procedure conforms to the defined model.
works that accurately recognizing the iris texturing is the
primary challenge, which has a significant impact on the Rohini Shelke, S. B. Bagal et. al., [15] An innovative
level of accuracy used in identity verification. Despite the method of biometric authentication is the iris recognition
fact that these methods are appropriate for Iris Recognition system. This system's segmentation technique makes use of
Systems (IRS), these methods are time-consuming and do the pupil's localization, shape, and intensity. The
not achieve acceptable standards for accuracy. Accuracy rate normalization process makes the segmented region into a
and computation time are two important factors for rectangular region.
evaluating the IRS [12].
Md. Fahim Faysal Khan, M. A. Haque, Ahnaf Akif et.
Due to its numerous applications in security from al. [16] Iris Recognition Using Machine Learning from
airports to homeland security border control, iris recognition Images Captured by Smartphones in Visible Light is
has been active research area of study in last few years [1]. presented. This study demonstrates how various machine
In the past, various iris recognition features and algorithms learning methods can be used to recognize the iris in eye
have been described. Supervised learning systems are images are captured on by smartphones.
utilized in AI and machine learning to classify input and
desired output data. Soft computing, computer vision, and R.P. Ramkumar and Dr. S. Arumugam et. al., [19]
pattern recognition have all seen advancements due to presents Novel Iris Recognition Algorithm without
machine intelligence algorithms like Neural Networks Normalization Phase. This work describes the novel iris
(NNs), evolutionary computing, and machine learning. The recognition algorithm by extracting the iris localization and
disadvantages of the traditional iris recognition systems are normalization phase.
mostly addressed by the use of machine learning
technologies. S. Sangeetha, N. Radha et. al., [20] provides a new
framework for IRIS and fingerprint recognition that is based
Hence, a novel approach to human iris recognition and on score level fusion and makes use of extreme learning
verification framework using machine learning algorithm in machines and SVM classification. In this, the performance
this analysis. The following is the structure of the remaining of the Fingerprint.
work: The literature review is described in section II. The
human iris recognition and verification framework utilizing III. HUMAN IRIS RECOGNITION AND VERIFICATION
machine learning algorithm is demonstrated in section III. FRAMEWORK
The results analysis of the presented methodology is
described in section IV. Section V introduces the work to a In this analysis, a novel approach to human iris
conclusion. recognition and verification framework using machine
learning algorithm is presented. The Fig. 1 shows the block
II. LITERATURE SURVEY diagram of human iris recognition and verification
framework using Machine Learning.
Saša Adamović, Milan Milosavljević, Nemanja Maček,
Vladislav Miškovic, Marko Šarac, Muzafer Saračević,
Milan Gnjatović et. al., [8] presents a novel and effective
method for recognizing the iris that is based on machine
learning techniques and stylometric features. In this
analysis, an innovative machine learning-based iris
identification system is presented. The primary goals of the
presented models are to eliminate false acceptance rates,
achieve almost perfect classification accuracy, and prevent
the developing of new iris images from existing templates.
The results of experimental analysis demonstrate that the
system works as expected. In addition, the recognition
system's overall complexity is reduced, that can be used in

2448

Authorized licensed use limited to: DePaul University. Downloaded on January 28,2024 at 03:51:05 UTC from IEEE Xplore. Restrictions apply.
2023 6th International Conference on Contemporary Computing and Informatics (IC3I)

goal of the iris localization procedure: the iris-sclera

boundary on the outside and the pupil-iris boundary on the
inside. The limbic and pupil boundaries of the iris are then
located and segmented from the iris image. The segmented
image should then be normalized with Daugman's Rubber-
Sheet algorithm.

With respect to reflections, symmetric reflection is a

type of symmetry. There must be at least one line connecting
the two halves of a figure when one half is a mirror image of
the other. The term "vertical line of symmetry" refers to a
line that divides an image into two identical halves. In
addition, the image is divided into two identical halves by
the horizontal line of symmetry, which runs across.

They are a logical extension of wavelets to account for

the fact that anisotropic features like image edges, which
wavelets cannot represent as isotropic objects, frequently
control multivariate functions.

Shearlets are produced through parabolic scaling,

shearing, and translation of a few generating functions.
These are essentially supported at small sizes by thin,
directed ridges that follow the length² ≈ width scaling law.
Shearlets, which are related to wavelets in that they are also
members of the affine group, permit an accurate
implementation of the continuum and digital situation.
Despite the reality that they do not provide an orthonormal
basis for L^2 (R^2). They continue to create a framework
that enables stable expansions of any given function f∈L^2
(R^2).

|| || log , →∞ 1

Where the singularity curve's maximum curvature and the

maximum magnitudes of f, f^' and f" are the only factors
that affect the constant C. This wavelet approximation rate
Figure 1: Block Diagram of Human Iris Recognition and Verification only provides O(N-1) for this group of functions.
Framework
In order to identify a number of edges in images, the
This approach is performed in two phases namely: Canny edge detector utilizes a multi-stage method. The
Phases of enrollment and verification during the recognition image's noise is first smoothening by the canny edge
phase, the enrolled individual's iris image is captured and detection. To identify areas with high spatial derivatives, it
his or her iris patterns or features are verified. As a system then locates the image gradients. Following these regions,
input, the iris image is first captured. Both the enrolling and the algorithm suppresses any pixel that is not at the
verifying phases use the iris images as inputs. During the maximum value (non maximum suppression). The Canny
image acquisition phase, the register's iris image is Operator's goals were clearly explained. Good Detection:
extracted. To produce the images appropriate for further the capability of finding and marking all actual edges. Good
processing, it is necessary to eliminate the noise from the Localisation: minimum separation between the real edge
acquired images. The iris image's salt and pepper noise is and the detected edge One response per edge is the clear
removed with the median filter. The cilia, iris, and pupil response.
make up the captured image of the eye.
Gray Level Co-occurrence Matrix (GLCM): By setting
An iris recognition system's accuracy and reliability can offset, the images matrix is used in the calculation of the
only be improved by precisely locating the iris region, its GLCM matrix. The GLCM matrix is created by taking into
quality directly affects how well the system's subsequent considering the directions and distances. The GLCM is used
stages perform. The two boundaries of the iris region are the

2449

Authorized licensed use limited to: DePaul University. Downloaded on January 28,2024 at 03:51:05 UTC from IEEE Xplore. Restrictions apply.
2023 6th International Conference on Contemporary Computing and Informatics (IC3I)

7
to extract texture characteristics including contrast, energy, 1
homogeneity, entropy, correlation, and etc. 3= 5 0 8
!89
1. Contrast: In the GLCM matrix, it is utilized to
calculate the local variations. c. Area: The area is determined by counting the
number of non-zero elements in the iris image.
= | | # , 2
d. Perimeter: The perimeter of an iris image is
!,"
determined by measuring the separation between
the two different boundary pixels.
2. Homogeneity: The measure to which the GLCM
element distribution matches its diagonals is
e. Solidity: The solidity is proportional to the convex
referred to as homogeneity.
area of an iris image.
1 (
% & '( ( ) = # , 3 ; /< )= 9
1+ 1 = >(? (
!,"

3. Energy: The level of uniformity between the pixels f. Eccentricity: Eccentricity is measured as the
is measured by energy. difference between an elliptical foci' major axis and
minor axis.
, ( ') = # , 4
A B
!," ,= 10
4. Entropy: An image's entropy is a statistical measure
of the randomness of each pixel. Where the comparable ellipse that represents the iris
region has a major axis of a and a minor axis of b. All these
features are extracted for iris recognition. The process of
, #) = # , log # , 5 determining the degree of similarity (often expressed as a
!," match score) between two iris signatures, typically one
obtained at the enrolment stage and the other typically
5. Correlation: The correlation between the pixels can gathered at the verification or identification step is known as
be used to measure a pixel element's linear BGM (Biometric GLCM, Graph Matching).
dependence.
The collection of images that at the very minimum
, # , 01 02
(/ = 6 features the iris region of the eye is known as a "iris
31 32 database." Sensors that operated in the visible spectrum
!,"
(380-750 nm) or Near InfraRed spectrum (NIR), which
The roughness features in an image can be extracted ranges from 700-900 nm, are commonly used to capture the
using the feature extraction technique known as Hausdorff images. During the enrolment phase, the enrolled iris images
dimension (HD). The complexity of an image is measured are saved in the database for future authentication and
by its fractal dimension. HD measures the features of the iris verification of human iris.
image region as well as a sets of features for every objects in
the binary image of the iris. The features in the iris image The SVM method gets its name from these extreme
must be logical arrays, but they can have any dimensions. examples, which are known as support vectors. The SVM is
The following are the features of shape and color: used to recognize and verify the iris images of human. The
linear kernel function is used to perform classification in
a. Mean: The mean is the ratio of iris images this system, and the results are as follows:
considered to all the iris images in the database.
? = D E 5 + B 11
1 7
0= 5! 7 Where W is the hyperplane's normal vector, b is the
!89
hyperplane's offset, and X are the training samples. This
approach recognizes the iris and it verifies the iris as it is
b. Standard deviation: The measure of the amount
matched with registered iris or not. In this manner, this
measured data differ from the mean is called
approach recognizes and verifies the human iris very
standard deviation.
effectively and accurately.

2450

Authorized licensed use limited to: DePaul University. Downloaded on January 28,2024 at 03:51:05 UTC from IEEE Xplore. Restrictions apply.
2023 6th International Conference on Contemporary Computing and Informatics (IC3I)

IV. RESULT AND ANALYSIS for human iris recognition and verification. The Fig. 2
shows the performance comparison in terms of specificity
In this analysis, a novel approach to human iris and sensitivity.
recognition and verification framework using machine
learning algorithm is implemented using python. In this
analysis, The IRIS patterns in each eye are unique to each
person, making it easier to identify them.

To categorise and identify objects in an IRIS image in

accordance with a database that has been stored, it is
possible to use IRIS segmentation. In this analysis, SVM
classifier is used to recognize and verify the human iris.
Utilizing the True Positive (TP), True Negative (TN), False
Positive (FP), and False Negative (FN) properties of the
confusion matrix, the performance of the presented
approach is evaluated. The specificity, sensitivity, and
recognition accuracy of the system is implementing test are
measured and defined as follows: Figure 2: Performance Metrics Comparison

Specificity:
F
;#(= = )= I 100 12
F + GH

Sensitivity:
FH
FHJ = I 100 13
FH + G

Accuracy:

FH + F
K==L =) = I 100 14
FH + GH + F + G

The table 1 shows the performance evaluation of

different approaches.
Figure 3: Accuracy Comparison
TABLE I: PERFORMANCE METRICS EVALUATION
Specificity Sensitivity Accuracy In Fig. 3, the x-axis represents different approaches and
Methodologies
(%) (%) (%) y-axis represents performance values. The SVM algorithm
An approach to has high accuracy for iris recognition and verification.
human iris
recognition using
Hence presented novel approach to human iris recognition
quantitative and verification framework using machine learning
94 95 94.3
analysis of image algorithm has effectively recognized and verified the human
features and iris.
machine
learning
Presented novel V. CONCLUSION
approach to human
iris recognition and In this analysis, a novel approach to human iris
verification 96.7 97.4 98.2
recognition and verification framework using machine
framework using
machine learning learning algorithm is implemented. This approach is
algorithm performed in two phases namely: Enrolment phase and
verification phase. During enrolment phase, the human iris
The performance of presented SVM algorithm is are registered and stored in the database whereas in
compared with earlier approach (i.e. An approach to human verification phase, the iris images are recognized and
iris recognition using quantitative analysis of image features verified. The texture-based features are extracted using the
and ML) to know the effectiveness of presented approach Hausdorff Dimension (HD) and the Gray Level Co-
for human iris recognition and verification. Compared to occurrence Matrix (GLCM). The algorithm Support vector
earlier approach, presented SVM algorithm has better results Machine is used to identify and validate the human iris. The

2451

Authorized licensed use limited to: DePaul University. Downloaded on January 28,2024 at 03:51:05 UTC from IEEE Xplore. Restrictions apply.
2023 6th International Conference on Contemporary Computing and Informatics (IC3I)

performance of presented approach is evaluated in terms of [13] Praveen, S. P., Sarala, P., Kumar, T. K. M., Manuri, S. G., Srinivas, V.
S., &Swapna, D. (2022, November). An Adaptive Load Balancing
recognition accuracy, specificity and sensitivity. The
Technique for Multi SDN Controllers. In 2022 International
effectiveness of presented approach is compared with earlier Conference on Augmented Intelligence and Sustainable Systems
human iris recognition approach. However, presented SVM (ICAISS) (pp. 1403-1409). IEEE.
algorithm has better results in terms of recognition accuracy, [14] Vellela, S. S., BashaSk, K., &Yakubreddy, K. (2023). Cloud-hosted
concept-hierarchy flex-based infringement checking
specificity and sensitivity than earlier approach. Hence
system. International Advanced Research Journal in Science,
presented system has recognized and verified the human iris Engineering and Technology, 10(3). Vellela, S. S., &Balamanigandan,
very effectively and accurately. R. (2023).
[15] Sk, K. B., Roja, D., Priya, S. S., Dalavi, L., Vellela, S. S., & Reddy, V.
(2023, March). Coronary Heart Disease Prediction and Classification
REFERENCES
using Hybrid Machine Learning Algorithms. In 2023 International
Conference on Innovative Data Communication Technologies and
[1] Reihan Alinia Lat, Sebelan Danishvar, Hamed Heravi and Morad Application (ICIDCA) (pp. 1-7). IEEE.
Danishvar, “Boosting Iris Recognition by Margin-Based Loss [16] VenkateswaraRao, M., Vellela, S., Reddy, V., Vullam, N., Sk, K. B.,
Functions”, Algorithms 2022, 15, 118, doi:10.3390/a15040118. &Roja, D. (2023, March). Credit Investigation and Comprehensive
[2] Smita Khade, Shilpa Gite, Sudeep D. Thepade, Biswajeet Pradhan, Risk Management System based Big Data Analytics in Commercial
And Abdullah Alamri, “Detection of Iris Presentation Attacks Using Banking. In 2023 9th International Conference on Advanced
Hybridization of Discrete Cosine Transform and Haar Transform with Computing and Communication Systems (ICACCS) (Vol. 1, pp.
Machine Learning Classifiers and Ensembles”, IEEE Access, 2021, 2387-2391). IEEE.
Volume 9, doi: 10.1109/ACCESS.2021.3138455. [17] Vellela, S.S., Balamanigandan, R. Optimized clustering routing
[3] C. D. Divya, H. L. Gururaj, R. Rohan, V. Bhagyalakshmi, H. A. framework to maintain the optimal energy status in the wsn mobile
Rashmi, A. Domnick, Francesco Flammini, “An efficient machine cloud environment. Multimed Tools Appl (2023).
learning approach to nephrology through iris recognition”, Discover https://doi.org/10.1007/s11042- 023- 15926-5.
Artificial Intelligence 2021, Springer, 1:10, doi:10.1007/s44163-021- [18] Vullam, N., Vellela, S. S., Reddy, V., Rao, M. V., SK, K. B., &Roja,
00010-4. D. (2023, May). Multi-Agent Personalized Recommendation System
[4] Maciej Szymkowski, Piotr Jasiński, Khalid Saeed, “Iris‑based human in E-Commerce based on User. In 2023 2nd International Conference
identity recognition with machine learning methods and discrete fast on Applied Artificial Intelligence and Computing (ICAAIC) (pp.
Fourier transform”, Innovations in Systems and Software Engineering 1194-1199). IEEE.
(2021) 17:309–317, doi:10.1007/s11334-021-00392-9. [19] Rao, K. N., Gandhi, B. R., Rao, M. V., Javvadi, S., Vellela, S. S., &
[5] S. S. Vellela, V. L. Reddy, R. D, G. R. Rao, K. B. Sk and K. K. Basha, S. K. (2023, June). Prediction and Classification of
Kumar, "A Cloud-Based Smart IoT Platform for Personalized Alzheimer’s Disease using Machine Learning Techniques in 3D MR
Healthcare Data Gathering and Monitoring System," 2023 3rd Asian Images. In 2023 International Conference on Sustainable Computing
Conference on Innovation in Technology (ASIANCON), Ravet IN, and Smart Systems (ICSCSS) (pp. 85-90). IEEE.
India, 2023, pp. 1-5, doi: 10.1109/ASIANCON58793.2023.10270407. [20] Rao, D. M. V., Vellela, S. S., Sk, K. B., & Dalavai, L. (2023).
[6] Reddy, N.V.R.S., Chitteti, C., Yesupadam, S., Desanamukula, V.S., Stematic Review on Software Application Under-distributed Denial
Vellela, S.S., Bommagani, N.J. (2023). Enhanced speckle noise of Service Attacks for Group Website. Dogo Rangsang Research
reduction in breast cancer ultrasound imagery using a hybrid deep Journal, UGC Care Group I Journal, 13.
learning model. Ingénierie des Systèmesd’Information, Vol. 28, No. 4, [21] Venkateswara Reddy, B., &KhaderBashaSk, R. D. Qos-Aware Video
pp. 1063-1071. https://doi.org/10.18280/isi.280426 Streaming Based Admission Control And Scheduling For Video
[7] Abolfazl Zargari Khuzani, Najmeh Mashhadi, Morteza Heidari, Transcoding In Cloud Computing. In International Conference on
4Donya Khaledyan, “An approach to human iris recognition using Automation, Computing and Renewable Systems (ICACRS 2022).
quantitative analysis of image features and machine learning”, 2020 [22] Madhuri, A., Praveen, S. P., Kumar, D. L. S., Sindhura, S., &Vellela,
IEEE Global Humanitarian Technology Conference (GHTC), 978-1- S. S. (2021). Challenges and issues of data analytics in emerging
7281-7388-7/20, 2020 IEEE, DOI: scenarios for big data, cloud and image mining. Annals of the
10.1109/GHTC46280.2020.9342935. Romanian Society for Cell Biology, 412-423.
[8] Vellela, S.S., Balamanigandan, R. An intelligent sleep-awake energy [23] Vellela, S. S., Balamanigandan, R., & Praveen, S. P. (2022). Strategic
management system for wireless sensor network. Peer-to-Peer Netw. Survey on Security and Privacy Methods of Cloud Computing
Appl. (2023). https://doi.org/10.1007/s12083-023-01558-x Environment. Journal of Next Generation Technology (ISSN: 2583-
[9] Vellela, S. S., &Balamanigandan, R. (2022, December). Design of 021X), 2(1).
Hybrid Authentication Protocol for High Secure Applications in [24] Venkateswara Reddy, B., Vellela, S. S., Sk, K. B., Roja, D.,
Cloud Environments. In 2022 International Conference on Yakubreddy, K., & Rao, M. V. Conceptual Hierarchies for Efficient
Automation, Computing and Renewable Systems (ICACRS) (pp. Query Results Navigation. International Journal of All Research
408-414). IEEE. Education and Scientific Methods (IJARESM), ISSN, 2455-6211.
[10] Madhuri, A., Jyothi, V. E., Praveen, S. P., Sindhura, S., Srinivas, V. S., [25] Sk, K. B., Vellela, S. S., Yakubreddy, K., & Rao, M. V. (2023). Novel
& Kumar, D. L. S. (2022). A New Multi-Level Semi-Supervised and Secure Protocol for Trusted Wireless Ad-hoc Network
Learning Approach for Network Intrusion Detection System Based on Creation. Khader Basha Sk, Venkateswara Reddy B, Sai Srinivas
the ‘GOA’. Journal of Interconnection Networks, 2143047. Vellela, Kancharakunt Yakub Reddy, M Venkateswara Rao, Novel and
[11] Vellela, S. S., Reddy, B. V., Chaitanya, K. K., &Rao, M. V. (2023, Secure Protocol for Trusted Wireless Ad-hoc Network
January). An Integrated Approach to Improve E-Healthcare System Creation, 10(3).
using Dynamic Cloud Computing Platform. In 2023 5th International [26] Sk, K. B., & Vellela, S. S. (2019). Diamond Search by Using Block
Conference on Smart Systems and Inventive Technology Matching Algorithm. DIAMOND SEARCH BY USING BLOCK
(ICSSIT) (pp. 776-782). IEEE. MATCHING ALGORITHM", International Journal of Emerging
[12] S Phani Praveen, RajeswariNakka, AnuradhaChokka, Technologies and Innovative Research (www. jetir. org), ISSN, 2349-
VenkataNagarajuThatha, SaiSrinivasVellela and UddagiriSirisha, “A 5162.
Novel Classification Approach for Grape Leaf Disease Detection [27] S. S. Priya, S. Srinivas Vellela, V. R. B, S. Javvadi, K. B. Sk and R. D,
Based on Different Attention Deep Learning Techniques” "Design And Implementation of An Integrated IOT Blockchain
International Journal of Advanced Computer Science and Framework for Drone Communication," 2023 3rd International
Applications(IJACSA), 14(6), 2023. Conference on Intelligent Technologies (CONIT), Hubli, India, 2023,
http://dx.doi.org/10.14569/IJACSA.2023.01406128. pp. 1-5, doi: 10.1109/CONIT59222.2023.10205659.

2452

Authorized licensed use limited to: DePaul University. Downloaded on January 28,2024 at 03:51:05 UTC from IEEE Xplore. Restrictions apply.
2023 6th International Conference on Contemporary Computing and Informatics (IC3I)

[28] N. Vullam, K. Yakubreddy, S. S. Vellela, K. Basha Sk, V. R. B and S.

Santhi Priya, "Prediction And Analysis Using A Hybrid Model For
Stock Market," 2023 3rd International Conference on Intelligent
Technologies (CONIT), Hubli, India, 2023, pp. 1-5, doi:
10.1109/CONIT59222.2023.10205638.
[29] Vellela, S. S., & Krishna, A. M. (2020). On Board Artificial
Intelligence With Service Aggregation for Edge Computing in
Industrial Applications. Journal of Critical Reviews, 7(07).

2453

Authorized licensed use limited to: DePaul University. Downloaded on January 28,2024 at 03:51:05 UTC from IEEE Xplore. Restrictions apply.