Integration of FaceNet and DeepFace Algorithms for Optimized Attendance Management in Educational Systems
Keywords:
- Facial Recognition, Classroom Attendance, Video Surveillance, FaceNet, DeepFace, Real-time Tracking, Educational TechnologyAbstract
The objective of this study is to enhance classroom attendance management systems through the implementation of advanced facial recognition technologies, addressing the shortcomings of existing methods. Traditional attendance systems, predominantly manual or biometric-based, suffer from several issues, including reliance on active participation, vulnerability to proxy attendance, lack of real-time tracking, dependency on technological infrastructure, and high costs. Our proposed method involves the application of cutting-edge facial recognition algorithms, specifically FaceNet and DeepFace, integrated into a video surveillance system for automatic and accurate attendance tracking. FaceNet and DeepFace algorithms are chosen for their robustness and high accuracy in facial recognition tasks. FaceNet demonstrated an accuracy of 96%, precision of 95%, recall of 94%, and an F1 score of 94.5%, whereas DeepFace showed a slightly lower performance with 94% accuracy, 93% precision, 92% recall, and a 92.5% F1 score. These metrics underscore the potential of these technologies in effectively replacing traditional attendance systems. The study's findings reveal that the integration of these algorithms into classroom settings not only simplifies the attendance process but also significantly enhances accuracy and efficiency. This integration eliminates the need for manual intervention and reduces the likelihood of attendance fraud. It also offers real-time tracking capabilities, which are essential in large educational settings. In conclusion, the implementation of FaceNet and DeepFace algorithms in classroom attendance systems marks a significant advancement in educational technology. It achieves the dual goals of automating the attendance process and ensuring higher accuracy and reliability. Future work will focus on optimizing these algorithms for diverse and challenging scenarios, ensuring privacy and security, and expanding their application to larger educational environments.