Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM64902
Volume:
10 |
6 |
Jun 2026
Pages:
1-7
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
M.Sc CYBER FORENSIC AND INFORMATION SECURITY
Dr.M.G.R. EDUCATIONAL AND RESEARCH INSTITUTE
M.Sc CYBER FORENSIC AND INFORMATION SECURITY
Dr.M.G.R. EDUCATIONAL AND RESEARCH INSTITUTE
This study proposes a multi-factor authentication system combining the selection of graphical passwords, live face recognition, and AI-enabled shoulder-surfing detection.Text-based passwords are still susceptible to being observed and guessed through a dictionary attack. This system provides a solution to these problems by implementing a security model that consists of three layers:(1) a graphical password system where the user chooses the sequence of four emoji icons, (2) face recognition biometric authentication using five different face angle images of the user with dlib HOG method to get 128-dimensional face embeddings, and (3) a continuous OpenCV Haar cascade surveillance to detect an unauthorized observer and according to the situation, the user interface gets blurred or paused is also done by the threat-intent recognition model.The system is Flask web-based with a thin-client/thick-server model architecture. It utilizes WebRTC for the camera streaming from the browser side, bcrypt for hashing the credentials, and SQLite for continuously logging the threats.From the performance evaluation carried out, it was found that the facial verification part of the system is capable of giving the authentication results within a second very reliably and accurately, whereas, the shoulder-surfing detection device is able to find several faces and make the threat level classification as caution, warning, or critical. The entire system is a quite usable and practical one that can be deployed in those places where the security of the authenticity is not guaranteed physically and it can work as an improvement over the traditional authentication methods.
KEYWORDS
This paper introduces a web-based multi-factor authentication system that leverages graphical password selection, face biometric verification in real-time, and AI-powered shoulder-surfing detection. The system is a Flask application and only needs the user to have a modern browser, no installation of client-side software is necessary. The experimental test results show face verification true acceptance rate of 94.8%, false acceptance rate of 1.2%, and shoulder-surfing detection rates of 87-96% in various threat scenario simulations. On average, the graphical password validation takes 48 ms, and the threat notification reaches the dashboard within 1.2 seconds of the detection The proposed system shows that it is possible to deploy adaptive, context-aware authentication in standard web environments using open-source libraries. In the next step of the research, we will test deep learning-based face detectors for better robustness, randomised graphical grids to resist replay attacks, and client-side embedding extraction to address privacy concerns. Besides, the system might be integrated with established identity provider protocols such as OAuth 2.0 or WebAuthn to extend its applicability in enterprise settings.
[1] A. Biddle, S. Chiasson, and P. C. van Oorschot, "Graphical passwords: Learning from the first twelve years," ACM Computing Surveys, vol. 44, no. 4, pp. 1–41, 2012. [2] R. Dhamija and A. Perrig, "Déjà Vu: A user study using images for authentication," in Proc. 9th USENIX Security Symposium, 2000. [3] A. Wiedenbeck, J. Waters, J.-C. Birget, A. Brodskiy, and N. Memon, "PassPoints: Design and longitudinal evaluation of a graphical password system," International Journal of Human-Computer Studies, vol. 63, pp. 102–127, 2005. [4] D. King, A. ageitgey/face_recognition: The world's simplest facial recognition API for Python and the command line. GitHub, 2017. [Online]. Available: https://github.com/ageitgey/face_recognition [5] P. Viola and M. Jones, "Rapid object detection using a boosted cascade of simple features," in Proc. IEEE CVPR, 2001. [6] G. Bradski, "The OpenCV Library," Dr. Dobb's Journal of Software Tools, 2000. [7] M. Oka, Y. Kato, and Y. Ukita, "Privacy-preserving face recognition using secure computation," IEICE Transactions on Information and Systems, 2012. [8] N. Hopper and M. Blum, "Secure human identification protocols," in Advances in Cryptology—ASIACRYPT 2001, Springer, 2001. [9] W. Yassin, N. I. Udzir, and A. B. Abdullah, "A shoulder-surfing resistant graphical password scheme," World Academy of Science, Engineering and Technology, 2012. [10] Django Software Foundation, "Flask: A lightweight WSGI web application framework," Pallets Projects, 2010. [Online]. Available: https://flask.palletsprojects.com
Publication Date
6/13/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
2
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...