Research On Watermarking Technique Of Digital Images Based On SVD

With the development of the multimedia and the network technology, the application of the digital product is more and more wide. The security of the digital information becomes the concern in people's life. Especially, the copyright protection of digital product is more important. Of all the solution to the copyright protection, the watermarking technology is more effective one. Therefore, the study of this technology becomes a focus in the academic world.In this thesis, the background, classification, model, application and recent development of the digital watermarking are summarized first. Then the design requirement is analyzed, and the method to evaluate a watermarking system is provided. In addition, the different watermarking algorithms and watermarking attacks are analyzed together. Taking still images as example, the thesis utilizes several techniques and methods such as chaos, singular value decomposition, image scrambling, etc. The work is focused on the image watermarking technique based on singular value transform domain.In this paper a novel watermark algorithm for binary images based on SVD and DCT is proposed first. The watermark is performed by chaos encryption. Then, the original binary image is divided into blocks according to the size of the watermark; each block is transformed using DCT. We extract the largest DC coefficient from the frequency coefficients of each block, which will be constructed the new matrix. The new matrix is performed SVD transformation; the watermark is embedded into the singular value.According to studying the correlation among the coefficients of the matrix after SVD transformation, a new watermarking scheme having the ability of sharing secret based on visual cryptography and SVD is proposed in this paper. The watermark is embedded into the original image during comparing and adjusting the coefficients of the matrix after the SVD transformation.In order to test the imperceptibility and robustness of the two algorithms mentioned above, a lot of experiments are done. Experimental results demonstrate that our schemes are effective and useful. Because of the simpleness and convenience, they may be expected to have a better application prospect.