IRJET - Automated Fraud Detection Framework in Examination Halls

This document proposes an automated fraud detection framework to detect impersonation of candidates and possession of electronic gadgets in examination halls. It uses image processing techniques like face detection and recognition along with machine learning algorithms like Random Forest and Histogram of Oriented Gradients (HoG) for detection, classification and training. The framework is trained on datasets of images collected and labeled for anomalies. It detects impersonation and presence of gadgets during examinations by processing images using HoG and recognizing faces using a pre-trained Random Forest model for high accuracy classification.