An Intelligent System For Online Exam Monitoring

For the past few years, e-learning has become popular across countries because of its flexibility, availability and user friendliness. As far as online examinations are concerned; the major challenge faced by the research community is the proctoring techniques used. In this paper, we present a method to avoid the physical presence of a proctor throughout the exam by creating a comprehensive multi modal system. We have used hardware such as web-cam to capture audio and video along with active window capture. This combination forms the inputs to an intelligent rule based inference system which has the capability to decide whether any malpractices have happened. Examinee's face is detected and is used to extract feature points thereby estimating a head pose. Misconduct is detected based on yaw angle variations, audio presence and active window capture. Our system has been tested in an e-learning scenario and we were able to make exam monitoring easy. Experiment results proved that our system performed well than the existing systems.