Automated Attendance Monitoring System using Real-Time Facial Recognition and Python based Computer Vision Techniques

This research focuses on building an attendance system using face detection. The existing form of biometric attendance tracking is time consuming and inefficient as users generally have to use one or two machines and have to wait for their turn to log their attendance. A face recognition system is capable of detecting faces as they enter a building, room, or other designated areas, thus eliminating the need to $\log$ in every time. This also serves as physical proof of presence is recorded in order to mark the attendance. Further, this project aims to track the user’s movement throughout the environment, and if they are not in their assigned place during duty hours, their attendance will be frozen until they return back to their place. If they do not return within the permitted time period, their attendance will be unmarked for that time period. This ensures that the user simply doesn’t mark attendance and leave; instead they must remain present to retain it. The attendance log is stored in an excel sheet file to be easily accessible to administrators for further processing by the organization. Daily logs are generated for all users and weekly logs are generated for each individual user. This helps to cross verify and infer errors in the system in case the attendance is not accurate. This type of attendance system is useful for a number of organizations, such as schools, colleges, hospitals, public offices, factories etc., and provides a more secure system than bio-metrics. Strategic positioning of the camera ensures that the user will have to show their face while moving in or out of the environment. Overall, this system is more efficient than other form of attendance tracking.