CNN Based Approach for Copy-Move Video Forgery Detection

  • Niraj Ramakant Chaudhari, Omkar Vijay Kulkarni, Siddhant Manohar Patil
Keywords: CNN, video forgery detection, video to frame conversion, deeplearning.


The utilization of videos as a means to exchange information has increased on a vast scale. The advancement and the revolutionary changes taking place in the video editing tools and technologies have made it of greater importance to assure the authenticity of video content and to ensure thatthe right information has been circulated across the globe. Video as an entity is used in surveillance, medical, forensics, and various other fields. To use this as a proof for any crime scene its authenticity must be scrutinized, which demands AdvancedForgery Detection techniques. This paper presents CNN (Convolutional Neural Network) to find whether the video is a real one ornot. CNN is used here to overcome the shortcomings of the traditional methods such as time and computational complexity. The traditional method uses handcraft feature analysis which has high resource consumption and yields less accurate results. The proposed deep learning approach accurately detects forged and original videos with 97% accuracy.The method is compared with existing deep learning methodsand perform superior for copy-move forgery detection.

How to Cite
Siddhant Manohar Patil, N. R. C. O. V. K. (2021). CNN Based Approach for Copy-Move Video Forgery Detection. Design Engineering, 3553-3561. Retrieved from