Skip to Main Content
Digital watermarking can be used to protect the intellectual property for multimedia data. In this paper, a robust content-adaptive watermarking algorithm based on human visual system (HVS) is presented for video copyright protection. By comparing the direct current discrete cosine transform (DCT) coefficients of the intra frame and its neighbor frame of a given video, a fast moving image block set was chosen firstly. Then watermarking bits were adaptively embedded into the DCT domain of image blocks in the fast-moving block set, which were decided based on HVS and the characteristics of middle frequency DCT coefficients of these blocks. The experimental results indicate that the proposed algorithm is highly robust against MPEG-2 compression and frame's geometrical distortions, such as rotation, aspect ratio and bending.