Adaptation of Face Recognition for Student Attendance in Distance Education

The abrupt transition from traditional in-person education to remote online learning as a result of the COVID-19 pandemic had a significant impact on all parties involved, with students being particularly affected. A challenge arises in accurately assessing the levels and rates of student participation. There are limited methods available for monitoring student attendance particularly in the context of distance education. This study focuses on the utilisation of facial recognition technology to automate the process of attendance-taking, hence facilitating the participation of students in online classes offered through distance education. The present study employs Support Vector Machine (SVM) capacity to categorise photos into three distinct classes. The research commences with acquiring images and subsequently performing segmentation through the Graph-Based Segmentation technique. The Viola-Jones technique is employed for face detection, which is subsequently followed by feature extraction via the Local Binary Pattern (LBP) method. Ultimately, the process of face identification is achieved by the utilisation of the Support Vector Machine (SVM) methodology. The proposed methodology demonstrates a commendable level of recognition accuracy, with an 80.303% success rate when employing the SVM approach. Based on the obtained findings, it can be inferred that the implementation of facial recognition technology in the context of long-distance education holds promise as a viable solution to support the educational sector.