Smart AI Based Attendance Monitoring System Using YOLOV8

The proposed system utilizes the YOLOv8 – a computer vision model for object detection, classification and instant segmentation of captured images. The proposed work helps to identify and extract facial regions from the input images or real-time videos through the fixed CCTV camera in the classrooms and the mark the attendance for the students. The current attendance system makes it difficult to track the student attendance during each lecture period and it also consumes lot of time. There is a chance for human errors also to occur. The proposed system uses YOLOv8 algorithm, that compares the registered faces from the database and identifies the individual identities. The algorithm employs a CNN that divides the captured image into grid. Hence, the proposed system has a friendly user interface, that enables the administrators to manage the record of student’s attendance easily and helps to save time and reduce human errors. It makes attendance monitoring more easier and faster. The used algorithm performs faster than Faster R-CNN in terms of speed and accuracy. Its GPU latency compared to R-CNN is 1.3ms. The rate of skipping classes and students leaving early can be reduced greatly using the proposed work. Experimental data shows that the accuracy rate of video-monitoring system is up to 80%. Hence the proposed system will help in the reduction of errors and compute the attendance data faster compared to any other real-time monitoring algorithms.