Recording of Class Attendance Using DL-Based Face Recognition Method

In today’s rapidly evolving world, technology plays a vital role in simplifying various aspects of our lives. Numerous daily tasks have shifted into digital processes, as more individuals prefer electronic methods for accomplishing their work. However, despite these advancements, the process of recording student attendance at the universities level still relies on manual methods. Lecturers typically rely on handwritten attendance sheets and signed papers to document attendance, which is a slow, inefficient, and time-consuming process. Therefore, it is necessary to develop an efficient automated system that can track and record student attendance accurately. This paper aims to propose a system that leverages deep learning (DL)-based facial recognition approach to assist and enhance the attendance process. Facial recognition is one of the biometric processes that involves identifying and verifying an individual based on digital images or videos of their face. The proposed method performs better in non-cooperative environment too. The achieved recognition accuracy on the own created dataset is 96.8% and on LFW dataset is 97.5%. This approach can be used in various applications, such as universities, banks, and airports.