Development of Attendance Monitoring System using Facial Recognition

With the increasing need for automated and efficient attendance monitoring, traditional methods, such as human roll calls and RFID-based systems have limitations in terms of scalability and security. In this paper, a Face Recognition-Based Attendance Management System is presented that reliably recognizes faces using the local binary pattern histogram (LBPH) approach. The three main design components of the system are an attendance management graphical user interface, a training module, and a real-time recognition module. The training module builds an LBPH-based classifier by preprocessing images using OpenCV and extracting the features from the face and storing them in histogram format. The identification module uses a face detection method based on the Haar cascade to identify individuals in real time and map their identities against a recorded dataset. The system, which was created using Python, OpenCV, Tkinter, and Pandas, offers a secure, contactless, and efficient alternative to conventional attendance methods. Experimental results demonstrate the usability of the system and its applicability in the real world in academic environments.