Video copyright protection has become important as it is possible to make unlimited copies of digital video without quality loss. Video watermark is the main method of video copyright protection. A novel blind video watermarking scheme is proposed in this paper, in which independent component analysis (ICA) is used to extract the common static component and relative dynamic component from the consecutive frames. And then a designed watermark is embedded into the wavelet coefficients of the obtained dynamic component according to the 4-neighboring-mean based algorithm. The binarization of the 3-level-approximate sub-image of the I-frame in each GOP (GOP-Group of Picture) is selected as the watermark for this GOP. Given such watermark, the scheme can easily resist copy attack and temporal synchronization. Besides, because the watermarks of different GOPs and their embedding positions in a GOP are different, the intra-video collusion attack can be resisted effectively and there is no possibility for inter-video collusion attack to succeed. The simulation shows the feasibility of this scheme, which can resist well temporal synchronization, copy attack, and collusion attack.
Blind detection Collusion resistance Copy attack resistance Copyright protection Digital watermark Independent component analysis (ICA) Temporal synchronization Video watermark