An Intra-Drift-Free Robust Watermarking Algorithm in High Efficiency Video Coding Compressed Domain

The existing watermarking algorithms in H.265/high efficiency video coding (HEVC) compressed domain cannot achieve intra-frame distortion drift-free and resist the recompression and signal processing attacks simultaneously. Therefore, the application of HEVC watermarking in copyright protection is limited in practice. To solve the problem, this paper proposes an intra-drift-free robust watermarking algorithm, which uses a multi-coefficients modification method to embed the watermark into intra prediction residual pixels of
$4\times 4$
luminance transform blocks in the spatial domain. Based on the spatial-domain embedding method, watermark information can be extracted from the blocks that have been attacked and merged into
$8\times 8$
size, which improves the robustness. To further improve the performance of the watermarking algorithm, the proposed algorithm uses a combination of three thresholds according to the application requirements. Experimental results on a publicly available video database (JCT-VC) have justified the effectiveness of the proposed algorithm which has better imperceptibility and robustness than existing algorithms. In addition, the proposed algorithm has high real-time performance and is suitable for application scenarios where the video is distributed on a large scale.