Proficient Exam Monitoring System Using Deep Learning Techniques

Recently, online education and remote learning have grown in popularity. With this increase comes the necessity for efficient safeguards to guarantee academic integrity during online exams. This research paper includes a thorough investigation into the design of a deep learning-based automated online exam monitoring system. By monitoring and identifying instances of misconduct or cheating during online exams, the suggested method attempts to create a safe and impartial evaluation environment. The research begins by examining the many difficulties and restrictions related to online tests, such as identity verification, content security, and student behavior monitoring. The literature study examines current techniques and tools for proctoring systems, highlighting their advantages and disadvantages. This research paper has suggested a novel method that uses deep learning algorithms for reliable and accurate monitoring based on this research. Face detection, face recognition, face spoofing detection, head-pose estimation, eye tracking, mouth ratio analysis, facial landmark identification, object detection, audio recording, plagiarism checking, and speech-to-text conversion are the main elements of the automated online exam monitoring system. The system offers a multi-modal strategy to monitor and identify probable instances of cheating or unauthorized behavior during exams by merging various components. We have made an effort to suggest a method that is efficient, improves academic integrity in online exams, and has the potential to be included in a number of e-learning platforms and educational institutions.