Face Recognition Based Attendance System for Students

Abstract

The traditional method of scoring student attendance is often fraught with difficulties. As in UTHM, lecturers need to check the attendance by calling out student name and tick the attendance on paper or pass the paper to student to sign the attendance or login to the system, register date to tick the attendance and student scan the QR code for attendance or student need to pass the paper during lecture., these ways may cause some human error, proxy attendance and waste time. The study of this project is to address issues with the previous attendance system or using face recognition as a new attendance system.  The proposed system is implemented in OpenCV (Open Source Computer Vision Library), Python and Viola Jones algorithm was chosen as face detection method because of faster rate for face detection and feature extraction using LBPH (Local Binary Patterns Histograms). System will identify a student's face segment from a video frame and record his or her attendance. Recognition effect will be greatly reduced depend on the environment, expression and position but project can expect that will benefit the college to prevent proxy attendance and save time.