Binary Classification for Video Forgery Detection Using REWIND Data Set
摘要
Altering a video sequence such that objects within the frame are either added to or removed from the scene is one of the malicious video forgery techniques that are most often utilized. In the present article, we trained logistic regression model, which predict the copy-move forgeries in the videos. These model is trained on the REWIND video copy-move forgeries dataset. In this study, we trained and tested the data using Logistic regression model. Simulation results show that the Logistic Regression model achieves the 86% accuracy with REWIND dataset.