A Novel Watermarking Algorithm in Entropy Coding Based on Image Complexity Analysis

This paper presents a novel blind video watermarking method based on image complexion analysis. The image complexion information is extracted from the DC coefficient in sub blocks transformed by discrete cosine transform (DCT) to make sure the macroblock of image adapt to be embedded watermark information. Subsequently, the algorithm presents DCT coefficients space layered mode (DCSLM), and embeds watermark in the intermediate frequency AC coefficients according to DCSLM and some rules. Watermark detection can be done without the original video. Therefore, it is blind detection. Ultimately, image which has been embedded watermark show the proposed algorithm has favorable imperceptible. Simulation results show the proposed watermarking algorithm has strong robustness against some attacks such as JPEG compression, noise attack, frame statistical average and so on.