AI Based Online Exam Proctoring System

Abstract

Artificial Intelligence (AI)-based online exam proctoring systems have emerged as an effective solution for ensuring 
academic integrity in remote learning environments. The increasing adoption of online education has created a need 
for secure, scalable, and automated examination monitoring systems capable of detecting suspicious activities without 
constant human supervision. This project presents an AI-Based Online Exam Proctoring System that utilizes computer 
vision, machine learning, and facial recognition techniques to monitor candidates during online examinations. The 
proposed system continuously analyzes webcam video streams to detect face presence, multiple-person detection, head 
movements, eye gaze, and unauthorized object usage such as mobile phones. It also monitors browser activities and 
generates alerts whenever suspicious behavior is identified. The system authenticates candidates before the 
examination using facial recognition and records examination sessions for future review. By integrating real-time 
monitoring, automated violation detection, and detailed reporting, the proposed solution minimizes cheating attempts 
while reducing the workload of human invigilators. The system provides an efficient, accurate, and cost-effective 
approach for conducting secure online examinations while maintaining fairness, transparency, and academic 
credibility.