Detection of double video compression is of particular importance in video forensics, as it reveals partly\nthe video processing history. In this paper, a double compression method is proposed for HEVCââ?¬â??the latest video\ncoding standard. Firstly, four 5Ã?â??5 co-occurrence matrixes were derived from DCT coefficients along four directions\nrespectively, i.e., horizontal, vertical, main diagonal and minor diagonal. Then four 4Ã?â??4 co-occurrence matrixes were\nderived from PU types which are innovative features of HEVC and rarely been utilized by researchers. Finally, these\ntwo feature set are combined and sent to support vector machine (SVM) to detect re-compressed videos. In order to\nreduce the feature dimension, only the co-occurrence matrixes of DCT coefficients and PU types in horizontal\ndirection are adopted to identify whether the video has undergone double compression. Experimental results show the\neffectiveness and the robustness against frame deletion of the proposed scheme.
Loading....