摘要

This paper proposes a novel self-adaptation differential energy watermarking based on the Watson visual model, which inserts robust watermark into video streaming according to the differential energy theory. This algorithm can control the watermark's embedding intensity of sub-low AC coefficients in the video streaming adaptively based on the Watson visual model. And it also can be self-adaptive cheesed that the region should be embed watermarks according to the relationship between the energy adjustable threshold and their differential energy. So watermark not only meets the non-visual perception, but also has the better robustness. Experiments show that this algorithm has strong robustness and security against the usual video attacks such as noise, filter and compression attack etc with low complexity of energy computation and high capacity.

  • 出版日期2009

全文