HEVC video double-compression evidence obtaining method and system based on motion vector residual error

Abstract

The invention provides an HEVC video double-compression evidence obtaining method and system based on a motion vector residual error, and relates to the technical field of video recompression detection, and the method comprises the steps: S1, decoding a to-be-detected video, obtaining a decoded video frame picture, and extracting the coding mode information of each inter-frame coding frame in the decoding process; S2, for the extracted coding mode information, constructing a feature sequence of inter-frame coding difference by adopting feature fusion on a spatial scale; S3, analyzing the feature sequence through a voting algorithm, and positioning the position of an abnormal frame; S4, performing time domain analysis on the abnormal frame sequence, and judging whether the to-be-detected video is subjected to double compression or not; and S5, if the video is subjected to double compression, further deducing the GOP length in the first compression. The method can consider an adaptive coding mode designed by HEVC for a rapid change scene, and has the characteristics of high robustness and high detection efficiency.

Description

technical field [0001] The present invention relates to the technical field of video re-compression detection, in particular, to a HEVC video double-compression forensics method and system based on motion vector residuals. Background technique [0002] With the advancement of technology and the introduction of powerful video editing software, people can edit and modify digital video very conveniently. If the content of digital video is tampered with in order to distort facts, reverse black and white, destroy social order, and affect social integrity, it will undoubtedly have an extremely bad impact on society and endanger social information security. Therefore, the research on forensics of digital video tampering becomes very necessary and has very important practical significance. Digital video passive forensics technology is to judge whether the video has been tampered with by analyzing the statistical characteristics of the video data or detecting the traces left when th...

AI Technical Summary

Problems solved by technology

In the old coding standards, such as MPEG, H.264, the double compression forensic algorithm has achieved good results, but in HEVC, there is relatively little research in this area, especially for HEVC to better encode fast In changing scenarios, an adaptive parameter encoding method is proposed, but there are very few researches on double compression forensic algorithms targeting this method.

Images