Study Of Anti-protocol Attacks Digital Image Watermarking Based On SVD

With the widespread use of computers and Internet, access and exchange of digital data became an extremely simple task. As a result, illegal reproduction of digital information started to pose a real problem. As a potential means to solve this problem, digital watermarking becomes a common focus of industry and academic society in recent years. While new watermarking technologies are being developed, some researchers are also investigating different ways of attacking digital watermarks. At present none of the existed algorithms are immune to all the attacks especially protocol attacks.Firstly, the principle and the framework of digital watermarking are introduced, and some key techniques are also analyzed. After depicting the main ways of attacking digital watermarks, the model of protocol attacks (which including the interpretation attack, copy attack and collusion attack) is studied. Then some solutions of anti-protocol attacks are analyzed, on which cryptographic and hash-based solutions are paid more attention.Secondly, the conventional SVD-based digital watermarking algorithm is analyzed. Both the theoretical and experimental results point out SVD-based algorithm is fundamentally flawed because of the protocol attack. An improved SVD-based watermarking scheme is proposed. A binary watermark is embedded in the magnitude relationship of the largest singular values of selected DWT blocks. The embedded blocks are selected by a secret key to enhance security. No original image is needed in detecting procedure, so the practicality of the system is enhanced.Thirdly, an image watermarking algorithm is proposed based on media hash and SVD. Instead of using randomly Gaussian sequence, a meaningful text message modulated by media hash sequence is used. Theoretical analysis results show that the proposed algorithm solves the problem of protocol attack, while keeping all the advantages of SVD-based schemes. Experimental results show that the proposed algorithm is robust and secure. Comparisons with previous algorithms indicate that the performance of the proposed algorithm has been significantly improved. Then, the effectivity of the proposed algorithm with respect to anti-protocol attacks is evaluated. The results again verify that the proposed algorithm is immune to protocol attacks.Finally, conclusions are drawn and open problems and subsequent work are described.