Real-Time Student Emotion and Drowsiness Detection Using YOLOv5 and CNN for Enhanced Learning

Recognition of student's behaviour in class based on video has a significant impact on both teacher quality and student attention. Previously, student behaviour recognition focused only on single students with low performance and efficiency. This study proposes a system for detecting student behaviour and attention in the classroom using computer vision and machine learning techniques and it uses a webcam to capture video of the students and analyse their facial expressions and eye movements to determine whether they are paying attention or not. This can also detect the emotions of students. The final output of our proposed system is, it can detect different kinds of student behaviours in classrooms, such as attention in class, drowsiness, emotion of students and finally provide the rating for staff.