Face Recognition-Based Automated Attendance System

Abstract

Traditionally student attendance in institutes is taken manually on attendance sheets. This is not a very efficient method because it is susceptible to proxy attendance of absent students. This paper implements an Automatic Attendance System based on face recognition that marks the presence of students by detecting their faces against its trained set, from the image of everyone sitting in the classroom. Camera fixed in the classroom captures the faces of everyone present there and submits those images to this system, which then processes them to detect and identify individual faces and mark their attendance for that class in the master database. Haar Cascade classifier has been used to detect faces by capturing features on the images like eye, nose etc. This system is accurate and prevents any fake attendance. The proposed method is evaluated against principal component analysis (PCA) and other techniques like the local binary patterns histogram (LBPH). It is observed that the accuracy of the proposed system is better than those of the latter ones.