ARTIFICIAL INTELLIGENCE AND COMPUTER VISION TECHNIQUES FOR ONLINE EXAMINATION MONITORING

Online examinations have become a routine part of academic life, especially after institutions across the world shifted large portions of their teaching and assessment online.While this shift has made testing more flexible and accessible, it has also opened the door to new forms of academic dishonesty that are harder to catch without a human invigilator physically present in the room.Artificial Intelligence and Computer Vision have stepped in to fill this gap, giving rise to automated proctoring systems that can watch a candidate through a webcam and flag unusual behaviour in real time.This paper reviews the major AI and computer vision techniques currently applied to online examination monitoring, including face detection and recognition, eye gaze and head pose estimation, object detection for spotting prohibited items, and audio-visual behaviour analysis.We compare these approaches in terms of accuracy, robustness, and computational cost, and we examine the practical and ethical challenges that come with deploying such systems at scale, such as privacy concerns, lighting variability, and bias in detection models.The paper closes with a discussion of open research questions and possible directions for building more reliable and fair examination monitoring systems.