Facial Recognition Based Attendance Management System Implementation: A Deep Learning Approach for the Auchi Polytechnic Institution, Edo State

Abstract

Taking attendance in a large class can be time-consuming and tiresome, wasting important instructional time. To address this issue, this research introduces the Auchi Poly-Convolutional Neural Network (AP-CNN), a deep learning-based automatic attendance system. The system is image recognition-based, with two key processes: face detection and face recognition. The system’s goal is to recognize students in the School of Information and Communication Technology (ICT), tag them with their names, and then match those names with faces in a database using real-time video feeds from a designated learning environment. According to the study’s findings, the small student dataset had a 96% accuracy rate. Using real-time attendance records of the students present, this technology can help instructors manage efficient classes.  It can also be useful for monitoring student participation, spotting absenteeism trends, and comprehending the class’s overall attendance. The information can also be utilized to identify any attendance-related problems and to modify the class schedule or teaching strategies as needed.