Detection of Malpractice in Offline Examination Using Deep Learning

Abstract

Exam proctoring is a hectic task i.e.; the monitoring of students' activities becomes difficult for supervisors in the examination rooms. It is a costly approach that requires much labor and difficult task for supervisors to keep an eye on all students at a time. Automatic exam activities recognition is therefore necessitating and a demanding field of research. In this research work, categorization of students' activities during the exam is performed using a deep learning approach. Adeep CNN architecture a kernel size of 7 * 7 and 64 different kernels all with a stride of size 2 givingus 1 layer. After that, the model is validated upon ImageNet. In this paper, we present amultimedia analytics system which performs automatic offline exam proctoring. The system hardwareincludes one webcam for the purpose of monitoring the visual environment of the testing location. Toevaluate our proposed system, we collect multimedia (visual) data from many exam centers performing various types of activities while taking exams. Extensive experimental results demonstratethe accuracy, robustness, and efficiency of our offline exam proctoring system.