Class Attendance Tracking System Using Face Recognition

Khor, Bon Han (2019) Class Attendance Tracking System Using Face Recognition. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text Khor Bon Han.pdf Restricted to Registered users only Download (4MB)

Abstract

As the world human population increases, so do the quantity of students in school, college and university. It has become a difficulty for all the lecturers and tutors who are teaching in an enormous class, for them to take the attendance. It is very time consuming while taking the attendance on a piece of paper and may affect the teaching progress. Such disadvantage may cause the students just come to class and sign the attendance and leave while some may help their friends to sign fake signatures. Therefore, this paper will be introducing a class attendance tracking system will be built with the help of Viola-Jones method will be planted in the face detection phase and as for face recognition phase will be using Eigenfaces method. There were six objectives which could be seen in chapter 1 and all of them had been achieved by various testing in chapter 4. The first objective had achieved by using the Matlab code such as “vision.CascadeObjectDetector” for face detection, built in Eigenfaces function for face recognition and directory path folder as database to store the students’ information. Second objective could be achieved by generating the Matlab code to create the report for the absenteeism rate. As for the third objective, the developed system would then compare with the traditional method and record the time taken. The next objective could be achieved by giving the students to sign their attendance on a piece of paper first then after that used the developed system to check whether the number of attendance was same as the number of signatures on the paper. As for the fifth objective, it could achieve by setting different number of training data set to the system to calculate the accuracy of the system. Last but not least, the last objective could be determined by using the same location with same posture of the student but with different devices taking the photo.