Video Forgery Detection Using Machine Learning

Video Forgery Detection Using Machine Learning

Nerella Bharath Ganesh Dept of ECE IARE

Dr. S China Venkateshwarlu Professor Dept of ECE IARE

Dr. V Siva Nagaraju Professor Dept of ECE IARE

Abstract - With the rapid advancement of digital media editing tools, video forgeries have become increasingly sophisticated and harder to detect, posing serious threats to information security, legal evidence integrity, and digital content authenticity. This project presents a machine learning-based approach for the detection of video forgeries, focusing on identifying tampered frames, splicing, and copy-move operations. The system extracts spatiotemporal features and motion inconsistencies using techniques such as optical flow analysis, frame differencing, and noise residual examination. These features are then analyzed using supervised machine learning models, including Convolutional Neural Networks (CNNs) and Support Vector Machines (SVMs), to distinguish.

Key Words:  Video Forgery Detection, Machine Learning, Digital Forensics, Frame Tampering, Copy-Move Forgery, Splicing Detection, Convolutional Neural Networks (CNN), Support Vector Machine (SVM), Optical Flow, Feature Extraction, Deep Learning, Video Manipulation, Authenticity Verification.