Tampering detection of double-compression with the same quantization parameter in HEVC video streams
Ali A. Elrowayati, M. F. L. Abdullah, Azizah Abd Manaf, Abdalrahman Alfagi · 2017
High-Efficiency Video Coding (HEVC) is a new video codec scheme, which has been published lately in 2013 by the ITU-T and ISO/IEC. The HEVC codec standards is a compressed video format; thus, its content cannot intentionally tamper unless decoding the video streams into to the spatial domain. In order for attackers to tamper any HEVC video content, the video stream should be initially decompressed into frames, then re-compressed. Recently, the video tampering detection has been used as a tool for supporting forensics and authentication process. By detecting the second compression, the integrity of video contents can be authenticated. In this paper, a novel method for distinguishing double-compression in HEVC is proposed, even if the same quantization parameter value is used in both compression processes. Our method is based on analyzing changes in the quantized residual DST coefficients and the values of intra-prediction modes. This approach increases the ability to discriminate the single and double compression in HEVC sequence streams, which can be utilized as an indication of double compression tampering. Several HEVC video classes are used to carry out the experiments, where the results show a high level of accuracy and efficiency for most quantization parameter values.