Smart Group Attendance System Based on Face Recognition
Haoyang Wang · 2024
In order to improve student register efficiency and campus security, the smart group attendance system is proposed. The system, which is divided in to three subsystems, can use a group photo to implement student classroom register. Firstly, in order to cope with complex and changing lighting environments, the automatic exposure system based on OpenCV can adjust the input photo to normal exposure. Secondly, a YOLOv5s based face detection subsystem is conducted to detect all the faces and cropped them from input image. Thirdly, a FaceNet based face recognition subsystem is conducted to compare the cropped face images from face detection subsystem with the face information in the face library in order to do the students' register and recognize strangers. Experiments show that the proposed method has better real-time performance and can make the right decisions.