A Novel Approach to Identify Student's Attentiveness Based on Drowsiness Detection During Online/ Live Classes

This paper proposes a novel and comprehensive approach to identifying student's attentiveness based on drowsiness detection during online/ live classes.

The trend and necessity of online and live classes in education have got popularity in the last three months' dues pandemic of COVID-19.

Teachers are putting a lot of effort into taking online classes. Still, at the same time, they should also monitor the vigilance level of students to help them not to lose any critical material delivered by a teacher.

The proposed system uses time-efficient image processing techniques to identify nap detection and yawning as the parameters to conclude student drowsiness.

The proposed system continuously captures the subject's image on-site using the web camera and detects the face region, then focuses on eyes and lips using efficient image processing techniques to monitor their behavior. If abnormality either in the behavior of eyes or mouth is detected, it indicates that the student is falling asleep or having a state of drowsiness; therefore, drowsiness is detected, and a warning alarm is generated which can be listened to by a teacher so that he/she may ask the student to be vigilant.
The system is developed using the DIP library of Python and tested in different scenarios and gave satisfactory results.