Attendance monitoring system with face recognition technologies

Abstract

In recent years, spending a lot of time to check  students' attendance becomes a problem for lecturers. In order to solve this problem, many researchers proposed automated student attendance systems with face recognition technologies. However, those proposed systems still have the problems of recognition accuracy.  Also, they do not have the system for students to check their attendance results and correct the results of false detection. In this paper, the attendance monitoring system with face recognition technologies has been proposed in order to improve the accuracy of face detection and solve the conventional problems. The experiments were conducted on the accuracy of Eigenface recognition, Fisherface recognition, and Local Binary Pattern Histogram (LBPH) recognition in order to select the best face recognition technology for the proposed system. Experimental results show that LBPH recognition has the highest accuracy of 94.21%. Moreover, the proposed system also allows  students to confirm their attendance results on web application,thereby asking the lecturer to correct any false detection results.