Privacy-Preserving On-Screen Activity Tracking in E- Learning

Abstract

The rapid growth of online education has made effective proctoring systems essential to maintaining academic integrity and ensuring fair exams and online learning. Conventional proctoring methods like in-person invigilation are not feasible for online exams and e-learning. To address this issue, an automated proctoring method based on computer vision is proposed. This system tracks and analyzes applicant behavior throughout online learning and testing using advanced computer vision algorithms. A webcam and microphone are used by the Automated Proctoring System to continuously monitor the test environment. The Automated Proctoring System uses a webcam and screen recorder to monitor the test environment in real time. Computer vision algorithms are used to identify and monitor the examinee\'s face, gaze direction, eye movements, body posture, and suspicious activities. Machine learning models identify unusual patterns of behavior, such as continually turning away from the screen or showing many faces in the camera frame. These abnormalities cause the examiners to receive alerts for further inquiry. The proposed method aims to enhance exam security and e-learning by mitigating academic dishonesty, including cheating, impersonation, and unauthorized aids. It uses computer vision to provide a scalable and non-intrusive method of monitoring online assessments and e-learning.

Introduction

The text addresses the urgent need for academic integrity in online exams, especially given the rapid growth of remote learning. Traditional in-person proctoring is impractical online due to logistical and privacy concerns, prompting the rise of automated proctoring systems that leverage computer vision and machine learning to monitor students. These systems detect suspicious behaviors such as abnormal gaze patterns, unusual movements, or potential communication, providing a non-intrusive, scalable solution to uphold fairness in e-learning.

Literature Review:

• Multiple studies have explored vision-based, audio-assisted, and semi-automated proctoring systems to prevent cheating.

• Challenges include detecting multiple students simultaneously, avoiding impersonation, mitigating false positives, and addressing privacy concerns.

• Some works suggest 360-degree cameras or multi-camera setups for better monitoring, while others focus on gaze tracking, head pose estimation, and behavior classification using machine learning algorithms.

• Low-cost solutions using standard webcams and microphones have also been proposed to make proctoring accessible.

Existing System:

• COVID-19 accelerated the shift to online learning through platforms like Zoom, Google Meet, and Microsoft Teams.

• While online exams became widespread, fair and secure assessment remains a challenge due to opportunities for academic dishonesty.

Proposed System:

• Introduces a privacy-preserving, AI-powered monitoring system for online exams.

• Uses computer vision to track facial expressions, head and eye movements, and on-screen activity.

• Audio-to-text analysis detects potential communication during exams.

• Combines face detection, gaze tracking, and activity recognition to identify cheating without human invigilators.

• Ensures data privacy by processing information locally and only sharing essential results securely.

• Offers a secure, scalable, and non-intrusive approach to maintain academic integrity in online learning environments.

In essence, the system aims to provide fair, efficient, and ethical automated proctoring to support the legitimacy of online education.

Conclusion

The study concludes that automated proctoring systems using AI and computer vision are effective in maintaining fairness during online exams and E-learning. These systems can detect cheating by analyzing a student’s face, eye movement, and behavior in real time. They provide a smart, scalable, and privacy-friendly way to monitor exams without human invigilators. Overall, such technology helps improve academic integrity and ensures trust in online education.

References

[1] Jay Mayekar1, Shubham Pal2, Aditya Pandey3, Bikra Pani4, Prof. Preeti Mishra5, \"AUTOMATED PROCTORING SYSTEM\", [2] International Research Journal of Engineering and Technology (IRJET) e- ISSN: 2395-0056 Volume:10Issue:04|Apr2023. [3] SimonWenig,MichaelSuriyah, Freiber Rojas, Kevin Schonleber, Thomas Leibfried, \"Simulation Framework for DC Grid Control and ACDC Interaction Studies Basedon Modular Multilevel Converters\", @ IEEE 2016. [4] Aiman A Turani, Jawad H Alkhateeb, Abdul Rahman A. Alsewari, \"Students Online Exam Proctoring: A Case Study Using 360 Degree Security Cameras\",2020EmergingTechnology in Computing, Communication and Electronics (ETCCE) [5] Asep Hadian Sudrajat Ganidisastra, Yoanes Bandung, \"An Incremental Training on Deep Learning Face Recognition for M-Learning Online Exam Proctoring\", The 2021 IEEE Asia Pacific Conference on Wireless and Mobile (APWiMob) [6] \"Automated Proctoring System using Computer Vision Techniques\", Conference Paper • July 2021.DOI: 10.1109/ICSCAN53069.2021.9526411 [7] Renuka Devi, Gowri Srinivasa, \"Detection of anomalous behavior in an examination hall towards automated proctoring\", https://www.researchgate.net/publicatio n/321260760,2017. [8] Yousef Atoum, Liping Chen, Alex X. Liu, Stephen D. H. Hsu, and Xiaoming Liu, \"Automated Online Exam Proctoring\", ACCEPTED WITH MINOR REVISION BY IEEE TRANSACTION ON MULTIMEDIA, DEC 30, 2015 [9] Yusep Rosmansyah,\"Impersonation Attack-Defense Tree\", https://doi.org/10.3991/ijet.v15i19.12699 [10] Aditya Nigam, Rhitvik Pasricha, Tarishi Singh, Prathamesh Churi, \"A Systematic Review on AI?based Proctoring Systems: Past, Present and Future\" , Education and Information Technologies(2021)26:6421– 6445,2021. [11] Tejaswi Potluri1 & Venkatramaphanikumar S1 &Venkata Krishna Kishore K1, \"An automated online proctoring system using attentive-net to assess student mischievous behavior\", Multimedia Tools and Applications (2023) 82:30375–30404,2023. [12] A. Sun and X. Chen, ‘‘Online education and its effective practice: A research review,’’ J. Inf. Technol. Educ., Res., vol. 15, pp. 157–190, May 2016. [13] I. E. Allen and J. Seaman, ‘‘Digital learning compass: Distance education enrollmentreport,’’BabsonSurveyRes. Group, Wellesley, MA, USA, Tech.Rep. 2017-1, 2017. [14] S. I. U. Rehman, H. S.Ullah, and A. Akhtar, ‘‘Consumption of social media and academic performance: A cross- sectional survey of perception of students in KP universities,’’ Global Mass Commun. Rev., vol. 4, pp. 57–71, 2020. [15] I. K. Sohail and N. A. Nabaz, ‘‘The influence of social media on student’s academic performance: A case study of Lebanese French University,’’ Mod. Manag. Theory Pract., vol.25, no.2,pp. 117–127, 2019. [16] J. Goet, ‘‘Impact of social media on academic performance of students,’’ KIC Int. J. Social Sci. Manage., vol. 1, no. 1, pp. 35–42, Dec. 2022. [17] S. I. U. Rehman, H. S. Ullah, and Akhtar, ‘‘Consumption of social media and academic performance: A cross-sectional survey of perception of students in KP universities,’’ Global Mass Commun. Rev., vol. 5, no. 4, pp. 57–71,Dec.2020.[8]C.LiandF. Lalani, The COVID-19 Pandemic has Changed Education Forever. This is How, vol. 29. Geneva, Switzerland: World Economic Forum, 2020. [18] E. Vayena, A.Blasimme, and I. G. Cohen, ‘‘Machine learning in medicine: Addressing ethical challenges,’’ PLOS Med., vol. 15, no. 11, Nov. 2018, Art. no. e1002689. [16] B.Imlerand, M.Eichelberger, ‘‘Using screen capture to study user research behavior,’’Library Hi Tech, vol. 29,no. 3, pp. 446–454, 2011. [19] P. Krieter and A. Breiter, ‘‘Track every move of your students: Log files for Learning Analytics from mobile screen recordings,’’ in Proc. Die 16. E- Learning Fachtagung Informatik (DeLFI), 2018, pp. 1–12. [20] B. J. Ferdosi, M. Sadi, N. Hasan, and M. A. Rahman, ‘‘Tracking digital device utilization from screenshot analysis using deep learning,’’ in Proc. Int.Conf. DataSci. Appl.(ICDSA),vol.1.Singapore:Springer,2023,pp.661–670. [21] C. Dwork, ‘‘Differential privacy,’’ in Proc. Int. Colloq. Automata, Lang., Program. Berlin, Germany: Springer, 2006, pp. 1–12. [22] X. Yi, R. Paulet, E. Bertino, X. Yi, R. Paulet, and E. Bertino, ‘‘Homomorphic encryption,’’ in Tutorials on the Foundations of Cryptography. Cham, Switzerland: Springer, 2014. [23] O. Goldreich, ‘‘Secure multi- party computation,’’ Manuscript. Preliminary version, vol. 78, no.110, pp.1–108, 1998. [24] J. Kone?ný, H. B. McMahan, F. X. Yu, P. Richtárik, A. T. Suresh, and D. Bacon, ‘‘Federated Learning: Strategies for improving communication efficiency,’’ 2016. [25] M. Abadi, A. Chu, I. Good fellow, H. B. Mc Mahan, I. Mironov, K. Talwar, and L. Zhang, ‘‘Deep learning with differential privacy, ’’in Proc. ACM SIGSAC Conf. Comput. Commun. Secur. New York, NY, USA: ACM, 2016, pp. 308–318. [26] W. Dai, S. Wang, H. Xiong, and X. Jiang,‘‘Privacypreservingfederatedbig data analysis,’’ inGuide to Big Data Applications. 2018, pp. 49–82. [27] H. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y. Arcas, ‘‘Communication-efficient learning of deep networks from decentralized data,’’ 2016, arXiv:1602.05629. [28] S.J.Panand, Q.Yang,‘‘Asurveyon transfer learning,’’IEEETrans.Knowl. DataEng.,vol.22,no.10,pp.1345– 1359, Oct. 2009. [29] K. R. Varshney, ‘‘Trustworthy machine learning and artificial intelligence,’’XRDS,Crossroads,ACM Mag.Students, vol. 25, no.3, pp. 26–29, Apr. 2019.

Copyright

Copyright © 2026 Dr. Dewanand Meshram, Satyam Kale, Harish Mengade, Swaraj Aghav, Sanika Ahire. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.