Detecting double H.266/VVC compression with the same coding parameters

The detection of video double compression with the same coding parameters is a difficult problem in video forensics, since the recompression traces are extremely slight in this case. Although a series of methods have been proposed, detection of Versatile Video Coding (H.266/VVC) double compression with the same coding parameters is rarely reported. To address this issue, we propose a novel algorithm for the detection of H.266/VVC double compression with the same coding parameters in this paper. We first analyze the generation of H.266/VVC double compression traces, and explore the variation of coding modes caused by multiple compressions. Then we define Minimum Unit Mapping (MUM) and Subunit Prediction Mapping (SPM) to facilitate feature construction. The strength of the variation in the number of coding unit (CU) in each compression is calculated to form CU partition modes based feature set. We also use the consistency ratios of adjacent prediction mode pairs in the horizontal, vertical, major diagonal, and minor diagonal directions after multiple compressions to obtain prediction modes based feature set. These feature sets are further concatenated into a fusion feature for detection. With the thorough comparison to the state-of-the-art algorithms, it is verified that the proposed algorithm outperforms other current works in the field. Besides, the proposed method is robust against various encoding configurations. 2022 Elsevier B.V.