Research Of Encryption Image Digital Watermark Algorithm Based On Transform Domain

With the rapid development of computer and network technology, digital multimedia information resources are widely used in people's work and life. Computer formidable information-handling capacity provides the enormous convenience for the multimedia information's deposits. At the same time, the information copyright protection and series of related questions also become a hot topic. As a new copyright protection means of digital products, digital watermark is the important branch in the international information hiding domain.The basic method of digital watermark is the copyright owner inserts own copyright information into the product, the inserting information is literal showing, special indication or image form. When the product is found has been infringed in the spreading process, the copyright owner depends upon the extraction inserting information to prove that he has the copyright. Thus protect copyright or integrity of digital works. The digital watermark technology is an emerging multi-disciplinary cross-application technology, involves the different discipline theory and the technology, such as signal processing, image processing, and information theory, coding theory, cryptography, detection theory, and digital communications and networking technology and so on. Watermark technology development also led to the further development of related fields. Therefore, regardless of from theory angle or application angle, development digital watermark technology research, not only has the vital academic significance, but also has the great importance practical value. The digital watermark already involved the image watermark, video watermark, audio watermark, text watermark and three-dimensional grid data watermark and so on. Most of the watermark research and thesis research has focused on the image research.The paper mainly conducts the research to the image digital watermark algorithm, use of discrete wavelet transform implementation of digital watermarking embedding and extraction. Because of the wavelet transformation can achieve signal or the image the multi-resolution decomposition. Wavelet transform has many functions of mathematical microscope, such as zoom, deflation and shift. Using wavelet decomposition characteristic can embeds information into different regions and resists each kind of signal processing operation well.The main research work includes:First, the paper makes a comprehensive introduction about the digital watermark technology research background, the present development situation and the application domain. Introduce the basic concept of digital watermarking features, generic model, classification, major applications, typical algorithm, watermark attack analysis and system evaluation, etc.The paper is divided two parts carries on the introduction and the analysis to the watermark image and the carrier image processing method. The watermark image processing mainly introduced watermark encryption method such as chaos, scramble. Unifies each kind of encryption technology to propose one chaos scrambling encryption, thus guarantee watermark information security. In the carrier image processing aspect mainly introduced and analyzed model LSB, DCT, the DWT algorithm, human visual system and the singular value decomposition. Quantitative analysis through the experiment to the watermark image's each kind of encryption algorithm. Using MATLAB tools experiment three watermark inserting algorithms. Compare to three algorithm anti-attack capabilities, watermark image quality under attack. Realize the digital watermark system through software programming.Second, analysis the carrier image's complexity, propose two measure methods of image complexity based on image characteristic analysis and singular value decomposition technology, realize the complexity quantitative analysis to the carrier image. Choose high complexity image block as watermark embedding block to solve watermark system's balanced question between invisibility and robustness.Third, Combination of human visual system and the singular value decomposition technology proposes two kind of digital watermark algorithms. The first algorithm uses the human visual model computation watermark embedment strength on the high complexity image block. After the encryption, the watermark data embedded according to the different embedment strength to the carrier image. The second algorithm combines multi-resolution analysis in wavelet domain as well as the singular value decomposition technology. Propose one kind of digital watermark algorithm revision the wavelet coefficient and the singular value coefficient in the sub band wavelet domain. The algorithm carries on the wavelet decomposition first to the carrier image, then chooses high complexity block as the watermark embedding block, modifies the watermark embedding block singular value decomposition coefficient to achieve the watermark embedded. Use MATLAB tools to experiment on the two algorithms. The experiment indicated that two algorithms have the very strong robustness in the resistance common attack. Use MATLAB tools to do programming experiments on the two algorithms.Finally, summary the entire work, and make the forecast to present work and the development direction of digital watermark.