A Deep Learning Approach to Masked Face Recognition for Enhanced Attendance Systems

Face recognition is the process of recognizing and identifying a face. Facial recognition uses specialized cameras to match faces in addition to unlocking phones. Face or picture recognition is useful for a variety of applications, including security phone unlocking, board control and airport operations, missing person searches, banking, retail crime reduction, marketing and advertising, healthcare, tracking employee or student attendance, and driver recognition. Biometric technology provides extremely attractive security option. Face recognition systems are essential in practically every industry in our digital age. One biometric that is frequently utilized is face recognition. It has numerous more benefits in addition to being useful for security, identity, and authentication. Due to its non-invasive and contactless nature, fingerprint and iris recognition systems are still commonly employed despite their lower accuracy. Additionally, facial recognition systems can be utilized in companies, institutions, and schools to indicate attendance. The goal of this system is to create a facial recognition-based class attendance system because the current manual approach requires a lot of time and effort to maintain. Also, there might be opportunities for proxy attendance. As a result, this mechanism becomes more necessary. The four stages of this system are face detection, face identification, database building, and attendance with a covered mask. Images of students in class are used to develop databases. The Haar-Cascade classifier and the Local Binary Pattern Histogram technique are used, respectively, for face detection and recognition. Faces are identified and detected from the classroom's live streaming footage. At the conclusion of the session, attendance will be kept on file in a server.