Intelligent Invigilation Using Video Surveillance

Sakya, G.; Singh, S.; Mishra, S.; Tiwari, S. M.

At the examination center, examinees daily face difficulties in finding their seating position that consumes a lot of time. Also in COVID-19 pandemic, we need to maintain social distancing while taking examination offline. So, this paper proposes system that can help examinees in knowing their seat number in their respective examination hall with the help of online Web portal. It can help in eliminating the crowd at the notice board and can also save their time. With the help of an admit card generated to which a barcode be attached, an examinee can scan that code each day and know their seating location as per a daily basis. Another system is also proposed for face recognition that can help the examiner in identifying the identity of candidates that can also alert them if they are not found in the database using deep learning. Within the examination hall, the orientation of the head and movement of the mouth of an examinee provide us the clue of suspicious behaviors. Nowadays, most of the suspicious behaviors monitoring procedures are done manually that involves a lot of invigilators in each examination hall. In the pandemic situation, we proposed an intelligent invigilation using video surveillance that can autonomously detect and track examinee’s eye gaze, head orientation, and mouth movement to robustly detect their cheating activities. Algorithms which are implemented independently include eigen-face, fisherface, and linear binary pattern histograms. With the help of webcam installed at each examinee’s desk, if any suspicious activity is detected, the system will generate an alarm indicating such behaviors. © 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.