A Digital Watermarking Algorithm Based On Singular Value Decomposition And HVS

The rapid growth of the Internet increases access to multimedia data tremendously. This fact presents a challenge to copyright protection of multimedia data. Digital watermarking, which is inserting copyright information into the data, has been proposed to solve the problem. There are two important issues that the watermarking algorithms need to address. First, watermarking schemes are required to keep the quality of the original data and imperceptibility. Second, good watermarking schemes should satisfy the requirement of the robustness and resist distortions due to common image manipulations (such as filtering , compression, etc.) In this paper, the basic concepts and principles of digital watermarking are given first, then some typical algorithms are introduced. A novel watermarking algorithm based on singular value decomposition (SVD) and human visual system (HVS) is proposed late. The image is partitioned into 8×8 blocks first, then we perform SVD on the selected blocks. The image partition is to be agree with the standard of the JPEG compression, so the algorithm can resist some common image processing better. On the other hand, the algorithm can be more effective. The main properties of SVD in our algorithm are as follows: First, the singular values (SVs) of an image have very good stability, that is, when a small perturbation is added to an image, its SVs do not change significantly; Second, SVs represent intrinsic algebraic image properties. The properties of SVD make our algorithm more robust. To select appropriate image blocks, based on the characteristics of the human visual system (HVS), we introduce a parameter T, which is corresponding with entropy and covariance, as the referring to partition different kinds of image blocks. It can help to achieve the imperceptibility of the watermark. According to the proposed algorithm , we experimentize on the computer. Experimental results and analysis show that the new watermarking algorithm is feasible and performs well in keeping the quality of the image and robustness againstcompression, filtering, and salt & pepper noise.