The Design And Realization Of Watermarking System Based On Image Edge

In recent years, the rapid development of Digital Technology and Internet, broadens the benefit of rights owner as maximum as possible, also causes the crisis of copyright at the same time. Digital Watermarking is the effective supplementary means for copyright protection of multimedia data, as one branch of Information Hiding. Digital Watermarking embeds some signs information into the multimedia content by a certain algorithm, but doesn't affect the value and use of the original content, and can't be aware or noticed by the perceptual of human, only can be extracted through a dedicated detector or reader. At present, there are many literatures focus on the algorithm and application of Digital Watermarking.According to the requirement of Digital Watermarking invisibility, after watermarking embedded, there must be no significant decline in image quality, so choose a suitable location for the digital watermarking which is embedded in is especially important. Based on the masking properties of HVS, we can know that, the more complex the texture is, the higher the threshold of the watermark embedded is, so choosing the edges of images to be the location of embedding watermark, can minimize down the affection since watermark embedding to the images quality, and can help to improve the visibility of watermark.In this paper, image digital watermarking robustness and invisibility have been studied, and one thinking of embedding digital watermarking in the edges of images is proposed. First of all, carries out wavelet transform on the original image, after that extracts edge feature points from the similar sub-band by the extracting algorithm of image edge points based on data revised, and then finishes the process of watermarking embedding in low-frequency sub-band according to the different adaptive watermark embedding strength. Simultaneously, on the basis of the characteristics of human visual system, chooses the suitable coefficients in the details sub-band after wavelet transform, and completes watermarking embedding in the high-frequency sub-band, which could constitutes complementary watermark together with low-frequency sub-band watermark, and strengthen the robustness of image watermarking. Because of choosing different watermark strength according to image feature points when embedding watermarking in the similar sub-band, the algorithm could achieve the invisibility of image watermarking very well.Finally, one watermarking embedded system is realized based on image edges, including every phase, such as Discrete Wavelet Transform, Image Feature Point Extraction, watermark strength choosing, and Brightness Factor Calculating. The experimental results displayed that the image embedded watermarking didn't drop in quality obviously ,and could resist the compression, the filter, the noise effectively, and so on. So this adaptive and complementary watermarking algorithm based on image edge is proved to be good at both robustness and invisibility.