Class Attendance System Using Viola-Jones Algorithm and Principal Component Analysis

Abstract

Face recognition is one of the numerous biometrics approaches that can be implemented by smart and automated attendance management systems. The individual identity can be determined by the unique representation of the face structure of each individual face and it cannot be lost, stolen, or reproduced in the same way as other types of identification. Thus, this work is motivated to propose a class attendance system based on face recognition. With current approaches such as passwords, access cards, and identification numbers, face recognition can be used to prevent theft and fraud which can significantly reduce the chances of system hacking. In this proposed work, initially, video framing has been implemented by activating the Universal Serial Bus (USB) camera through a user-friendly interface which was created with Graphical user interfaces (GUI) in the MATLAB software. The image of each student’s face that was snapped by using the USB camera will be stored in a dataset. The dataset then will be divided into the training set and testing set. In the detection process, the Viola-Jones algorithm is utilized to detect and segment the image of student’s face from the video frame. Next, the scaling of the size of the images is carried out to prevent the loss of information in the pre-processing phase. Then, the Principal Component Analysis (PCA) is utilized in the face recognition process in order to extract the features from facial images.