Published Paper Details:

The rapid proliferation of digital education platforms and remote learning environments has fundamentally transformed the way academic examinations are conducted worldwide. While online examinations offer unparalleled flexibility and accessibility, they simultaneously introduce a range of pressing challenges related to academic integrity, fairness, and institutional accountability. Traditional invigilation methods, which depend on physical presence and human observation, are wholly inadequate in virtual settings, necessitating the development of sophisticated, automated monitoring systems capable of operating at scale. This research paper presents a comprehensive Intelligent Examination Monitoring System (IEMS) that employs artificial intelligence, computer vision, and machine learning techniques to ensure secure, reliable, and tamper-resistant online examinations. The proposed system integrates facial recognition, multi-object detection, gaze tracking, and behavioral pattern analysis to monitor candidates in real time, identifying and flagging anomalies such as multiple faces in frame, prolonged absence from the camera view, presence of unauthorized devices or materials, and irregular movement patterns. The system architecture follows a modular, layered design comprising a responsive frontend interface, a robust backend server, and a dedicated AI processing engine. Experimental evaluations conducted across diverse testing scenarios demonstrate that the system consistently achieves accuracy rates between 93% and 96%, with low false positive rates and real-time response capabilities. These results underscore the system's viability as a scalable and practically deployable solution for modern educational institutions seeking to uphold examination integrity without disproportionate reliance on human invigilators.