Real-Time Face Recognition for Automated Student Attendance Monitoring

Conventional approaches to student attendance management, chiefly oral roll calls and handwritten registers, suffer from inherent inefficiencies, susceptibility to proxy fraud, and significant administrative overhead. This paper proposes an automated attendance monitoring framework that leverages real-time biometric face recognition to address these shortcomings. The system integrates the Haar Cascade algorithm for face detection with the Local Binary Pattern Histogram (LBPH) method for identity recognition, implemented entirely in Python with the OpenCV library. Live video acquired through a standard webcam is processed on a frame-by-frame basis: detected faces are recognized against a trained model and confirmed attendance events persisted in a date-stamped CSV repository without requiring specialized hardware. Experimental evaluation on a 60-student cohort demonstrated strong recognition accuracy across varied illumination environments. The proposed solution substantially reduces clerical burden, curtails fraudulent attendance practices, and delivers a contactless, cost-effective mechanism for institutional attendance governance.