Inter-frame Forgery Video Identification Based On Optical Flow Consistency

Identifying inter-frame forgery is a hot topic in video forensics. Video is widely used in the area of legal, political and scientific research, etc. However, with the ongoing development of video editing techniques, it becomes increasingly easy to modify the digital videos. Identification the authenticity of videos has become an important field in information security.Many researches on inter-frame forgery forensics were made and the most of the existing algorithms were designed to locate the tamper point under the premise that the video has been tampered. Then a new feature was proposed to distinguish original video and forgery video. Three jobs have been accomplished in this paper:1. Optical flow consistency feature was proposed to detect video inter-frame forgery based on the assumption that the optical flows are consistent in an original video, while in forgeries the consistency will be destroyed.2. A new video inter-frame forgery detection algorithm based on optical flow consistency was proposed. The optical flow consistency features are extracted from the original videos and video forgeries based on a large video database. Some samples are used to train the Support Vector Machine and the rest are used to test the classification results of the proposed algorithm. Experimental results demonstrate that the proposed method is efficient in classifying original videos and forgeries.3. The proposed method was applied in the classification of different kinds of video forgeries. The results show that the method is also efficient in classifying video forgeries attacked by frame deletion and video forgeries attacked by frame insertion.The proposed algorithm will be further improved in the computational complexity.