Research On Algorithms Of Digital Image Watermarking Based On Contents

With the rapid development of computer, network and communication, especially the popularization of the Internet, information security and copyright protection have been a key and attractive problem. Digital watermarking scheme has been a rising network information security technology. It provides a novel way for the protection of intellectual property rights. It becomes a hot spot in the research of multimedia data security, and holds wide application prospect in several domains.In this paper, the technology of robustness digital watermarking based on content for copyright protection of digital image is mainly discussed. By the research and analysis of the existing methods, the new robustness watermarking development direction is found, which is how to make full use of the inherent characteristics of the digital image. A digital image watermark algorithm with strong robustness is proposed in this paper. The algorithm extracts image strong texture regions according to the morphological features, analyzes and verifies the correctness and accuracy; The algorithm maps the image strong texture regions to the wavelet tree structures, and embeds watermark into the corresponding wavelet coefficients, by time-frequency mapping theory; the algorithm adaptively adjusts the watermark embedding intensity to balance transparency and robustness by HVS (Human Visual System). Experimental results show the algorithm is robust to common attacks. The image strong texture regions extraction method based on morphology in this algorithm has natural parallel structures,and is simple and effective compared with the classic spatial method, adaptive to various images.In order to improve the robustness by using of the image characteristics adequately, an image double watermark algorithm based on EMD (Empirical Mode Decomposition) is presented. The algorithm can accurately extract image strong texture regions by the favorable sifting ability of EMD to the image details characteristics, the watermark is embeded into the contourlet mid-frequency coefficients, which are on the direction-subbands with the largest energy corresponding image strong texture regions; Zero-watemark is constructed by extracting image edge characteristics using anti-noise multi-scale morphological edge detector. In this paper, the double watermark algorithm makes use of the image textrue and edge characteristics, closely connects the host image with the watermark, improves the robustness and security of the algorithm. Experimental results show this double watermark algorithm is robust to noise, filtering, compression, etc; and because of the inherent excellent feature that the zero watermark does not change the host image, the algorithm has a good performance in the transparency.Finally, the summary of this paper is given and the future direction of the research is presented.