A blind video watermarking scheme robust to frame attacks combined with MPEG2 compression

In this paper, we propose a robust digital video watermarking scheme with completely blind extraction process where the original video data, original watermark or any other information derivative of them are not required in order to retrieve the embedded watermark. The proposed algorithm embeds 2D binary visually recognizable patterns such as company trademarks and owner's logotype, etc., in the DWT domain of the video frames for copyright protection. Before the embedding process, only two numerical keys are required to transform the watermark data into a noise-like pattern using the chaotic mixing method which helps to increase the security. The main advantages of the proposed scheme are its completely blind detection scheme, robustness against common video attacks, combined attacks and its low complexity implementation. The combined attacks consist of MPEG-2 compression and common video attacks such as noise contamination, collusion attacks, frame dropping and swapping. Extensive simulation results also show that the watermark imperceptibility and robustness outperform other previously reported methods. The extracted watermark data from the watermarked video sequences is clear enough even after the watermarked video had suffered from several attacks.