See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/371503819

AN AADHAR AUTHENTICATION APPLICATION USING A FACE RECOGNITION

SYSTEM AND VERIFICATION FOR IDENTIFYING

Article · June 2023

CITATIONS READS

0 659

1 author:

Britto Raj S
RRASE COLLEGE OF ENGINEERING
32 PUBLICATIONS 48 CITATIONS

SEE PROFILE

All content following this page was uploaded by Britto Raj S on 12 June 2023.

The user has requested enhancement of the downloaded file.

Education and Society (शिक्षण आिण समाज) ISSN: 2278-6864
(UGC Care Journal) Vol-47, Issue-2, No.6S, April-June: 2023

AN AADHAR AUTHENTICATION APPLICATION

USING A FACE RECOGNITION SYSTEM AND
VERIFICATION FOR IDENTIFYING
DR.G.SIMI MARGARAT, ASSOCIATE PROFESSOR, DEPT OF COMPUTER SCIENCE AND
ENGINEERING, NEW PRINCE SHRI BHAVANI COLLEGE OF ENGINEERING AND
TECHNOLOGY, CHENNAI, TAMILNADU, INDIA
simimargaratphd@gmail.com

DR.K.RAVIKUMAR,
ASSOCIATE PROFESSOR,
DEPT OF COMPUTER SCIENCE AND ENGINEERING,
RRASE COLLEGE OF ENGINEERING CHENNAI, TAMILNADU, INDIA.
ravikumarcsephd@gmail.com

DR. S. BRITTORAJ
PROFESSOR, DEPT OF CSE,
RRASE COLLEGE OF ENGINEERING, CHENNAI, TAMILNADU, INDIA.
brittorajs@gmail.com

DR.G.B.SANTHI,
ASSISTANT PROFESSOR,
DEPT OF COMPUTER SCIENCE AND ENGINEERING.NEW PRINCE SHRI BHAVANI
COLLEGE OF ENGINEERING AND TECHNOLOGY,
CHENNAI, TAMIL NADU, INDIA,
santhi.bhavani@gmail.com

Abstract
Face recognition technology within biometrics is one of many methods that may be used to
verify and authenticate persons. All three branches of government, as well as law enforcement
and the private sector, make use of this surveillance and security technology to monitor and
control access. It has many applications presently, and no doubt more are being discovered all
the time. In this paper, we develop a face recognition system that enables the identification and
verification of the individuals using Aadhaar authentication.

Keywords:
Face Recognition, Verification, Security Technology, Aadhaar Authentication

1. Introduction
The number of fields that could gain from FRT is expanding as a result of significant scientific
and industrial breakthroughs. In order to analyse the FRT market, numerous publications have
segmented the information they gathered [1] [3]. It has been determined that healthcare,
marketing/retail, and security/law enforcement are the three most crucial FRT application areas
at the present time [2].
To begin, facial recognition CCTV systems, which are used to identify and track criminals or
find missing children, and identity authentication security systems, which are used in both
public and private settings (including mobile phones, homes, airports, and government
buildings), are essential components of security and law enforcement (such as airports, banks,
schools, hotels, etc.) [3]-[5].

235
Education and Society (शिक्षण आिण समाज) ISSN: 2278-6864
(UGC Care Journal) Vol-47, Issue-2, No.6S, April-June: 2023

Second, in the realm of medicine, some of the most crucial applications are tracking patient
drug use, identifying genetic disorders like DiGeorge syndrome, and refining methods of pain
management [6]. In the case of COVID-19 in particular, FRT has been used for both case
detection and contact tracking [7].
Third, in the realm of marketing and retail, the most widely used applications include facial
recognition payments, consumer behaviour analysis to enhance shopping experiences, and data
collection to deliver individualised offerings [8] [9]. However, it is expected that the FRT
market will grow to encompass additional industries in the near future as the number of
technical applications grows. Some markets that fit this description are the automotive sector
and the smart home industry [10].

2. Background
Early FRT perception polls, on the other hand, only compared the public acceptance of a
selection of FRT applications, rather than isolating the FRT apps that are thought to be more
prone to misuse [11]. Members of the general public may have a vastly different understanding
of the programme internal examples than the programmers themselves [12].
As an example, experts have pinpointed three common pillars of confidence in automated
systems [13]. The three pillars are effectiveness, efficiency, and meaning. The term process is
used to describe the way automation is run, whereas the term performance describes what the
automation is capable of doing [14]. In this context, purpose alludes to the motivations behind
the rise of mechanisation (the reasons for automation development). Therefore, comparing
public viewpoints on multiple FRT applications is not sufficient, as the automation trust model
demonstrates that the public may observe these traits in the internal instances of FRT apps and
produce varying levels of trust [15]. This is because these characteristics may be seen in internal
FRT app instances, leading to varied degrees of trust among the general audience. Law
enforcement agencies can employ FRT for a variety of purposes, such as tracking down missing
children or analysing data from widespread surveillance operations [16]. These two cases may
be perceived differently by the general public since they were designed for different purposes
and are applied in different ways [17].
On the other hand, there are constraints to analysing how the public views FRT by only listing
examples of FRT apps, as some characteristics may be shared by several FRT applications.
Because of progress in both technology and infrastructure, the FRT sector will also continue to
grow and penetrate new markets. To address the issues the public has encountered in some
FRT scenarios and to benefit FRT long-term growth, it is required to break FRT application
examples into unique scenarios based on certain principles in order to compare the public
reactions to each scenario. This is necessary to solve problems people have run across when
using FRT in specific situations. [18]-[25].

3. Proposed Method
An important goal of this research was to create a public perception model that may be used
by FRT creators and legal experts to spot scenarios in which the technology could be misused.
First, we compiled a list of features included in several FRT applications that the average
person would recognise. Next, we used these broad categories to divide the possible and
existing FRT applications into a wide variety of narrower ones. Finally, we chose three major
characteristics that have been shown to have an effect on FRT adoption from previous research
in order to characterise public opinion which is shown in Fig 1.
After a face has been identified in a given frame, the image undergoes preprocessing, which
includes noise reduction and the removal of unnecessary information. The feature extraction
process follows, and it is at this point that the classifier becomes useful. When the real-time
image has been processed, it is compared against archived images that have also been

236
Education and Society (शिक्षण आिण समाज) ISSN: 2278-6864
(UGC Care Journal) Vol-47, Issue-2, No.6S, April-June: 2023

processed. When a possible match is made, the photos are checked against a local watch list
database to determine the person criminal or suspicious history. The length of time he was
captured on video will be detailed in the case that he is a criminal or suspect. If it turns out the
person is not a citizen, the photo is checked against a global database of criminals who are
wanted by authorities. If a match is made, the amount of time the suspect was subject to camera
surveillance will be recorded. He is judged innocent because the two lists contain no common
information.
Images

Pre-processing

Feature Extraction

Template
Matching

Criminal Innocent

Figure 1: Proposed Model

In this study, we examine two distinct methods for identifying rotating objects. In the first
approach, multiple iterations of the first XML file are used to feed the image into the initial
classifier from different vantage points. This is followed by the second approach. The
alternative method entails duplicating the original XML file and using that duplicate to create
new XML files. Multiple recently developed classifiers use the XML files that were generated
to identify certain objects inside an image.

4. Results and Discussions

Although equivalent services can be obtained from other software packages, this one offers a
number of benefits that could ultimately be good for society as a whole. It feasible that, given
the complexity of its development and the revelations it has made, automated facial
identification technology will be challenging to recognise. A user-friendly interface that
necessitates minimal effort on the part of the user is likewise feasible to give. The user need
only provide an input in the form of a photograph that is updated in real time for this system to
carry out the remaining arduous image processing activities. Any skilled programmer can make
changes and add new features to the project because it is open source. This paradigm
fundamental simplicity makes it applicable in a wide variety of contexts which shown in Fig 2
and Fig 3.

237
Education and Society (शिक्षण आिण समाज) ISSN: 2278-6864
(UGC Care Journal) Vol-47, Issue-2, No.6S, April-June: 2023

90
80
70
60

Accuracy (%)
50
40
30
20
10
0
10 20 30 40
Images

FRT FRT-Aadhaar

Figure 2: Accuracy of FRT

18
16
14
12
Error Rate

10
8
6
4
2
0
10 20 30 40
Images

FRT FRT-Aadhaar

Figure 3: Error Rate

5. Conclusions

The proposed facial recognition technology will be put into use. The system can correctly
recognise faces even if the input image does not match the set of photographs of the individual
that are kept in the database. The algorithm needs to extract and calculate the main components
of the image before it can tell how unlike the input image is to the saved photos. This means
that the newly generated face image used for recognition can be tweaked somewhat before
being put into use. Since smaller images that contain the essential features require the least
processing to train the wavelets, this has a positive effect on computing cost, recognition
accuracy, and discriminatory power. The most notable benefit is that we leverage an already-
existing database of citizens.

238
Education and Society (शिक्षण आिण समाज) ISSN: 2278-6864
(UGC Care Journal) Vol-47, Issue-2, No.6S, April-June: 2023

References
[1] Srikrishnaswetha, K., Kumar, S., & Rashid Mahmood, M. (2019). A study on smart
electronics voting machine using face recognition and aadhar verification with iot.
In Innovations in electronics and communication engineering (pp. 87-95). Springer,
Singapore.
[2] Singh, B., Ranjan, K. S., & Aggarwal, D. (2020). Smart voting web based application
using face recognition, Aadhar and OTP verification. International Journal of Research
in Industrial Engineering, 9(3), 260-270.
[3] Kannan, S., Dhiman, G., Natarajan, Y., Sharma, A., Mohanty, S. N., Soni, M., ... &
Gheisari, M. (2021). Ubiquitous vehicular ad-hoc network computing using deep neural
network with iot-based bat agents for traffic management. Electronics, 10(7), 785.
[4] Parmar, A., Gada, S., Loke, T., Jain, Y., Pathak, S., & Patil, S. (2021, July). Secure E-
Voting System using Blockchain technology and authentication via Face recognition
and Mobile OTP. In 2021 12th International Conference on Computing
Communication and Networking Technologies (ICCCNT) (pp. 1-5). IEEE.
[5] Kanimozhi, S., Arun, N., Singh, A. M., Kumar, S. S., & Tharanidharan, T. (2022, May).
Identification of Non-Vaccinated People Using Face Recognition Based on CNN.
In 2022 International Conference on Applied Artificial Intelligence and Computing
(ICAAIC) (pp. 225-231). IEEE.
[6] Gupta, S., Jain, D., & Themalil, M. T. (2021, April). Electronic Voting Mechanism
using Microcontroller ATmega328P with Face Recognition. In 2021 5th International
Conference on Computing Methodologies and Communication (ICCMC) (pp. 1471-
1476). IEEE.
[7] Natarajan, Y., Srihari, K., Dhiman, G., Chandragandhi, S., Gheisari, M., Liu, Y., ... &
Alharbi, H. F. (2022). An IoT and machine learning‐based routing protocol for
reconfigurable engineering application. IET Communications, 16(5), 464-475.
[8] Hannah, S., Deepa, A. J., Chooralil, V. S., BrillySangeetha, S., Yuvaraj, N., Arshath
Raja, R., ... & Alene, A. (2022). Blockchain-based deep learning to process IoT data
acquisition in cognitive data. BioMed Research International, 2022.
[9] Praveenbalaji, D., Srinivas, R., Roopa, S., Suresh, M., & Gayathri, A. (2020, March).
ID photo verification by face recognition. In 2020 6th International Conference on
Advanced Computing and Communication Systems (ICACCS) (pp. 1449-1453). IEEE.
[10] Varshitha, K. S., & Shivanand, N. (2021). Android Mobile Based Voting System
through Facial Recognition. JNNCE Journal of Engineering & Management
(JJEM), 4(2), 1.
[11] Srikrishnaswetha, K., Kumar, S., & Johri, P. (2018, December). Comparision Study on
Various Face Detection Techniques. In 2018 4th International Conference on
Computing Communication and Automation (ICCCA) (pp. 1-5). IEEE.
[12] Poirson, C. (2021). The legal regulation of facial recognition. In The Fourth Industrial
Revolution and Its Impact on Ethics (pp. 283-302). Springer, Cham.
[13] Rizwan, S., Zubair, M., Ghani, A., & Ahmed, S. (2022). Decentralized Voting System
Based on Regions Using Facial Recognition. Journal of Independent Studies and
Research Computing, 20(1), 63-71.
[14] Gowrishankar, J., Kumar, P. S., Narmadha, T., & Yuvaraj, N. (2020). A Trust Based
Protocol For Manets In Iot Environment. International Journal of Advanced Science
and Technology, 29(7), 2770-2775.
[15] Boutros, F., Damer, N., Kirchbuchner, F., & Kuijper, A. (2022). Elasticface: Elastic
margin loss for deep face recognition. In Proceedings of the IEEE/CVF Conference on
Computer Vision and Pattern Recognition (pp. 1578-1587).

239
 Education and Society (शिक्षण आिण समाज) ISSN: 2278-6864
(UGC Care Journal) Vol-47, Issue-2, No.6S, April-June: 2023

[16] Annapurna, K., Chandrani, V., Mounika, P., & Sree, P. T. (2021, January). Design of
Authenticated Radio Frequency Identification based Electronic Voting Machine.
In 2021 6th International Conference on Inventive Computation Technologies
(ICICT) (pp. 658-665). IEEE.
[17] Godswill, O., Osas, O., Anderson, O., Oseikhuemen, I., & Etse, O. (2018). Automated
student attendance management system using face recognition. International Journal
of Educational Research and Information Science, 5(4), 31-37.
[18] Pooja, S., Raju, L. K., & Chhapekar, U. (2021). Face detection using deep learning to
ensure a coercion resistant blockchain-based electronic voting. Engineered Science, 16,
341-353.
[19] Lai, X., & Rau, P. L. P. (2021). Has facial recognition technology been misused? A
public perception model of facial recognition scenarios. Computers in Human
Behavior, 124, 106894.
[20] Zheng, S. (2021). The Application of Facial Recognition in China and India: Potential
Risks and Regulation Suggestions. Journal of US-China Public Administration, 18(3),
115-132.
[21] Osorio-Roig, D., Rathgeb, C., Drozdowski, P., & Busch, C. (2021). Stable hash
generation for efficient privacy-preserving face identification. IEEE Transactions on
Biometrics, Behavior, and Identity Science, 1-1.
[22] Singhal, S. K. (2021, December). Security Analysis of Aadhaar Authentication Process
and Way Forward. In 2021 3rd International Conference on Advances in Computing,
Communication Control and Networking (ICAC3N) (pp. 1304-1307). IEEE.
[23] Cheng, K., & Kumar, A. (2012, September). Contactless finger knuckle identification
using smartphones. In 2012 BIOSIG-proceedings of the international conference of
biometrics special interest group (BIOSIG) (pp. 1-6). IEEE.
[24] Gupta, S., Modgil, S., Lee, C. K., & Sivarajah, U. (2022). The future is yesterday: Use
of AI-driven facial recognition to enhance value in the travel and tourism
industry. Information Systems Frontiers, 1-17.
[25] Kumar, T. D., Tejaswini, M. N. D., & Kumar, O. C. Smart Electronic Voting Machine
with Face Recognition using Raspberry PI. Journal homepage: www. ijrpr. com
ISSN, 2582, 7421.

240

View publication stats