Development of a Student Absence Monitoring System Based on Face Recognition and Geolocation

Abstract

Student absence management in schools is still largely performed manually, requiring considerable time and effort, particularly for duty teachers responsible for monitoring absent students. This study aims to develop and evaluate a student absence monitoring system based on face recognition and geolocation to improve the efficiency of attendance management in schools. The system was developed using the ADDIE model and implemented as a web-based application by integrating the Face-api.js library for facial recognition and the Haversine formula for attendance location verification. System evaluation was conducted through functional testing, face recognition accuracy testing, impostor testing, and expert validation based on the ISO/IEC 25010 software quality characteristics. The results demonstrate that the proposed system achieved a face recognition accuracy of 98.72% with an average processing time of 226.543 ms. Expert validation involving three evaluators indicated that the system was classified as highly suitable for practical implementation. These findings suggest that the proposed system can effectively support more efficient monitoring and management of student absences. However, its performance remains influenced by lighting conditions and the similarity of certain facial features.