TWO-FACTOR USER AUTHENTICATION SYSTEM USING FACIAL RECOGNITION

DOI: 10.31673/2412-4338.2023.039699

Abstract

In today's world of information technology, data security is becoming a paramount concern. One of the most effective protection methodologies is two-factor authentication. This article delves into a cutting-edge method of two-factor authentication based on the combination of neural networks and facial recognition. Deep learning, employed in neural networks, allows the system to adapt to minor changes in a user's appearance, such as a new hairstyle, the presence or absence of makeup, wearing glasses, and so on. This makes the system flexible and capable of recognizing the user even with slight alterations in their look. The core idea of the method is to analyze the unique features of the user's face. The neural network "learns" the characteristics of each user, creating their unique "portrait". This "portrait" is then used for identity verification upon attempting to access the system. In addition to facial recognition, the system may require password input or another form of authentication, making the login process even more secure. The combination of these two methods ensures a high level of protection against unauthorized access. A significant advantage of such a system is its convenience for the user. The user's face becomes the "key" to the system, making the login process quick and seamless. It's also worth noting that the advancement of facial recognition technology opens new horizons for data security. Using neural networks in conjunction with two-factor authentication may become the standard in the near future.

Keywords: Two-Factor Authentication (2FA), facial recognition, biometrics, security, authentication, identification, account, token, biometric data, privacy, recognition error, security system, technology, camera, algorithm, neural network.

References
1. Alzubaidi L., Zhang J., Humaidi A.J., Al-Dujaili A., Duan Y. Development of a Secure Access Control System Based on Two-Factor Authentication Using Face Recognition and OTP SMS-Token. [Electronic resource] // URL: https://dx.doi.org/10.1109/ICIMCIS51567.2020.9354328.
2. Zhang B., Wang Y., Wang J. Convolutional Neural Network Face Recognition Method Using Fisher’s Criterion. [Electronic resource] // URL: https://www.hindawi.com/journals/misy/2022/1101282/.
3. Nelson D. Image Recognition and Classification in Python with TensorFlow and Keras. [Electronic resource] // URL: https://stackabuse.com/image-recognition-in-python-with-tensorflow-and-keras/.
4. Naoyuki K. Applying Artificial Neural Networks for Face Recognition. [Electronic resource] // URL: https://www.hindawi.com/journals/aans/2011/673016/.
5. Zhang B., Wang Y. Capsule Network-Based Deep Transfer Learning Model for Face Recognition. [Electronic resource] // URL: https://www.hindawi.com/journals/wcmc/2022/2086613/.
6. Banumalar K., Manikandan B.V., Vanaja N. Face Recognition using Deep Learning. [Electronic resource] // URL: https://www.e3s-conferences.org/articles/e3sconf/pdf/2023/24/e3sconf_icseret2023_05001.pdf.
7. Wenting L., Zhou L., Chen J. Face Recognition Based on Lightweight Convolutional Neural Networks. [Electronic resource] // URL: https://www.mdpi.com/2078-2489/12/5/191.
8. Wei-Meng L. Implementing Face Recognition Using Deep Learning and Support Vector Machines. [Electronic resource] // URL: https://www.codemag.com/Article/2205081/Implementing-Face-Recognition-Using-Deep-Learning-and-Support-Vector-Machines.
9. Manisha M. K., Bhattacharyya D., Tai-hoon K. Face Recognition Using Neural Network: A Review. [Electronic resource] // URL: http://article.nadiapub.com/IJSIA/vol10_no3/8.pdf.
10. Noor T. Facial Recognition Using Deep Learning. [Electronic resource] // URL: https://towardsdatascience.com/facial-recognition-using-deep-learning-a74e9059a150.
11. Zhebka V., Gertsiuk M., Sokolov V., Malinov V.Sablina M. Optimization of Machine Learning Method to Improve the Management Efficiency of Heterogeneous Telecommunication Network // CEUR Workshop Proceedings, 2022, 3288, p. 149–155.