Video Inter-Frame Forgery Identification Based on Consistency of Correlation Coefficients of Gray Values

Abstract

Identifying inter-frame forgery is a hot topic in video forensics. In this paper, we propose a method based on the assumption that the correlation coefficients of gray values is consistent in an original video, while in forgeries the consistency will be destroyed. We first extract the consistency of correlation coefficients of gray values (CCCoGV for short) after normalization and quantization as distinguishing feature to identify inter-frame forgeries. Then we test the CCCoGV in a large database with the help of SVM (Support Vector Machine). Experimental results show that the proposed method is efficient in classifying original videos and forgeries. Furthermore, the proposed method performs also pretty well in classifying frame insertion and frame deletion forgeries.

Digital video tampering detection: An overview of passive techniques

“…Differences of correlation coefficients of gray values between sequential frames (Wang et al, 2014a) Increase in compression leads to decrease in performance; frame deletion accuracy is less VPF (Gironi et al, 2014) Works with fixed GOP size fails when G1=G2; localization of cutting and insertion point is not as precise as in MV based schemes Quotients of consecutive correlation coefficients of local binary pattern coded frames (Zhang et al, 2015) Method can only report if forgeries exist or not and cannot distinguish between frame insertion and deletion; performance decreases if the number of frames deleted is less. Optical flow (Chao et al, 2013) Works well on frame insertion when the inserted frame have a different background scene Block-wise Brightness Variance Descriptor (BBVD) (Zheng et al, 2015) Detection accuracy decreases when number of frames inserted or deleted is less than 25; work on videos recorded by stationary cameras and forged videos having only one type of forgery where insertion or deletion is performed once Sum of absolute differences between video frames before and after applying deblocking filter (Su et al, 2015b) Deblocking filter and intra prediction in H.264/AVC will degrade the performance; fails when same rate control method is used to transcode the video Frame deletion & duplication Consistency of velocity field intensity (Wu et al, 2014) As compression increases, frame duplication detection accuracy decreases; works in videos recorded by static surveillance cameras M A N U S C R I P T…”

“…Thresholding approach is used to locate the deletion point. In (Wang et al, 2014a), the differences of correlation coefficients of gray values are used to detect forgery. For original sequences, they are consistent where as it is abnormal for inter-frame forgeries.…”

Video content authentication techniques: a comprehensive survey

“…An inter-frame forgery detection scheme was suggested in [76], where the authors used Consistency of Correlation Coefficients of Gray Values (CCCoGV) as a forensic feature, and stated that while for original videos, CCCoGV remained consistent, any post-production disturbance in the frame sequence caused this value to demonstrate abnormalities. An SVM was used to perform the final classification.…”

Exposing video inter-frame forgery by Zernike opponent chromaticity moments and coarseness analysis

“…Frame-to-frame optical flows and double adaptive thresholds are applied to detect frame deletion forgery. Qi et al [19] extracted the consistency of correlation coefficients of gray values after normalization and quantization as distinguishing feature to identify inter-frame forgeries, then they tested the feature in a large database with the help of SVM.…”

“…The method in [18] has a high requirement of the video background because of the assumption that the optical flows are consistent in the original video sequences. The authors of [19] tested video sequences with the help of an SVM based on the consistency of the correlation coefficients of the gray values. This method can only classify original video sequences and forgeries and cannot localize the juggled points, and the precision in detecting a frame deletion is lower than for a frame insertion.…”