ML-Driven Automated Professor Presence Monitoring Using GPS Geo-Fencing and Face Liveness Detection

Abstract

This project proposes a Machine Learning-Driven Automated Professor Presence Monitoring System that combines GPS geofencing and face liveness detection to ensure accurate and fraud-proof attendance. When a professor enters a predefined geofenced area, their presence is detected automatically, eliminating manual check-ins. To prevent spoofing through photos, the system uses real-time liveness detection (e.g., blink or head movement) powered by machine learning. Additionally, models like Random Forest and Linear Regression are used to predict arrival times based on past behavior and traffic patterns. A web-based dashboard provides real-time data visualization, attendance logs, and automated reporting. This mobile-first solution enhances security, improves efficiency, and offers a smart, scalable approach to attendance monitoring in academic environments.

Introduction

1. Problem Statement

Traditional attendance tracking in educational institutions is manual, time-consuming, error-prone, and vulnerable to proxy attendance. There’s a need for a secure, automated, and scalable solution.

2. Proposed Solution

The project introduces an automated professor attendance system using:

• GPS Geofencing to detect presence within a designated area.

• Face Liveness Detection using real-time gestures (e.g., blinking, head movement) to confirm authenticity.

• Machine Learning (ML), primarily CNN-based models, for facial verification.

Upon verification:

• Attendance is logged with timestamp, GPS coordinates, and professor ID.

• Data is stored securely in the Firebase Realtime Database.

• Professors receive real-time confirmations; admins access updates via a web dashboard (built using React + Firebase).

3. Key Features

• Contactless, real-time attendance.

• Prevents spoofing using photos/videos/masks.

• Scalable to multiple classrooms and institutions.

• Low user effort with mobile integration.

• Admin dashboard for monitoring, alerts, and auditing.

4. Methodology Overview

5. Literature Survey – Key Insights

• Past systems using facial recognition, RFID, QR codes, or GPS alone are either vulnerable to spoofing, resource-intensive, or not identity-verified.

• Face liveness detection using gestures like blinking improves spoof protection but requires optimization for mobile deployment.

• Geofencing alone is inaccurate in dense buildings and lacks identity verification.

• Hybrid approaches (e.g., combining GPS, facial recognition, and ML) show potential but often lack real-time performance or liveness validation.

6. Contributions & Advantages

• Combines GPS, facial recognition, and liveness detection in a single, automated system.

• Eliminates manual attendance and proxy issues.

• Offers a secure, accurate, and scalable alternative for educational institutions.

• Designed for mobile deployment, with low latency and intuitive interfaces.

Conclusion

This survey research highlights the growing potential of integrating geospatial technologies with machine learning to build robust, real-time professor attendance monitoring systems. By combining GPS-based geofencing with face liveness detection using mobile-optimized ML models, the proposed solution addresses the key limitations of traditional attendance systems namely, manual errors, proxy attendance, and lack of real-time verification. Through an in-depth analysis of existing literature, technologies, and system design strategies, it is evident that a mobile-only solution is not only feasible but also scalable and user-friendly for academic institutions. Geofencing ensures accurate location-based triggers, while liveness detection using facial cues like blinking or head movement ensures that the presence is genuine and not spoofed. The survey also demonstrates that leveraging real-time databases like Firebase enhances the system\'s responsiveness and security. Overall, the convergence of GPS, facial recognition, and AI presents a reliable framework for automated, secure, and efficient professor attendance systems. Future improvements could include voice verification, multi-modal biometrics, or predictive analytics to further enhance performance.

References

[1] Google Developers, “ML Kit: Face Detection,” Google Machine Learning Documentation, 2024. [Online]. Available: https://developers.google.com/ml-kit/vision/face-detection [2] Android Developers, “Location and Context: Geofencing,” Android Developer Guide, 2024. [Online]. Available: https://developer.android.com/training/location/geofencing [3] A. Mohan and S. Thomas, “A Secure Mobile Biometric Attendance System with Face Liveness Detection,” Journal of Mobile Computing, vol. 18, no. 2, pp. 88–95, 2021. [4] M. Singh and K. Raj, “Anti-spoofing techniques for face liveness detection: A comprehensive survey,” ACM Computing Surveys, vol. 55, no. 1, pp. 1–38, 2022. [5] Y. Zuo, W. Gao and J. Wang, “Face Liveness Detection Algorithm based on Lightweight Neural Network,” 2020 Int. Conf. on High Performance Big Data and Intelligent Systems (HPBDIS), China, pp. 1–5, 2020, doi: 10.1109/HPBDIS49115.2020.9130568. [6] K. Dong, C. Zhou, Y. Ruan and Y. Li, “MobileNetV2 Model for Image Classification,” 2020 2nd Int. Conf. on Information Technology and Computer Application (ITCA), Guangzhou, China, pp. 476–480, 2020, doi: 10.1109/ITCA52113.2020.00106. [7] V. Patel and D. Dhruiti, “Face Mask Recognition Using MobileNetV2,” International Journal of Scientific Research in Computer Science and Engineering, vol. 13, pp. 35–42, 2021, doi: 10.32628/CSEIT1217519. [8] A. S. Shaibah and R. Sarno, “Android Application for Presence Recognition based on Face and Geofencing,” 2020 International Seminar on Application for Technology of Information and Communication (iSemantic), Semarang, Indonesia, 2020, pp. 208–213, doi: 10.1109/iSemantic50169.2020.9234253. [9] Firebase, “Realtime Database,” Firebase Documentation, 2024. [Online]. Available: https://firebase.google.com/docs/database [10] A. Nugroho, S. Sumarsono, and E. H. Gunawan, “Framework of the Employee Attendance System with QR Code in the Pandemic Covid-19,” 2021 Int. Conf. on Software Engineering and Information Management (ICSECS-ICOCSIM), Malaysia, pp. 503–506, 2021, doi: 10.1109/ICSECS52883.2021.00098. [11] Gupta, A., Kundu, A., and Das, R., “Automated Attendance System for Efficient Employee Management: A Biometry Based Approach,” 2019. [12] N. Barve, P. Madhani, Y. Ghule, P. Potdukhe, K. Pawar and N. Bhandare, “Synchronized Speech and Video Synthesis,” 2023 International Conference on Smart Computing and Application (ICSCA), Saudi Arabia, 2023, pp. 1–7, doi: 10.1109/ICSCA57840.2023.10087508.

Copyright

Copyright © 2025 Sheela Rani C M, Sumukha M D, Hariprasad M, Manu kumar N, Srinivas J R. 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.