| The rapid growth of the information technology, the development of e-commerce, the increasingly easy access to digital media, and the increasingly powerful digital media manipulation tools have made media security a very important issue. In recent years, some new concepts of information security have been proposed, especially in the application of intellectual property right protection, copyright protection and Tamper-proofing. Information hiding and digital watermarking are new measures of digital media protection, which are processes of embedding hidden messages within digital media such as text, images, video and audio. The hidden messages could be copyright information, user distributor identification number, company logo, personal signature, control bits, or any information considered useful for a specific application. Supported by the natural science fund of Guang Dong province, this dissertation mainly focus on some key problems, such as the capacity of still images, embedding and detection of image watermark, Tamperproofing of digital media, watermark attack and the design of a final watermarking system. Some new algorithms are presented and tested and some problems in both theory and practical applications are solved successfully. The major achievements of this dissertation are outlined as follows.In Chapter 2, the states of the art of information hiding and digital watermarking are introduced. The parallels between common paper watermarks and digital watermarks are elaborated and the category of image watermarking system is summarized. Some problems, which could be considered in design of an applied watermarking system, are mentioned. The difference among spatial domain watermarking, watermarking based on fractals and frequency domain watermarking are analyzed.Chapter 3 describes the theoretical capacity measures for information hiding based on still image. The concept of watermarking channel is presented, and the difference between watermarking channel and communication channel is discussed. From the viewpoint of information theory, a mode of watermarking channel in still image is proposed and an evaluation of the number of bits that can be hidden within an image is given. As an example, the number of bits that can be hidden within a standard 512 X 512-grayscale Lenna image is evaluated.And in Chapter 4, the process of watermarking embedding and detection are discussed. The formulas for detector threshold and detection error probability are derived. In this Chapter, the constraints that a watermarking scheme should satisfy are analyzed,and combining with the techniques of traditional encryption, spread spectrum communication and digital image processing, a novel image watermarking scheme based on the content of the still image is presented. Such watermarks show low-pass characteristics and, subsequently, are robust to lossy compression and low-pass filtering. Moreover, it can be straightforward used for other media data. Finally, experimental results are reported proving the robustness of the proposed method against some of the most common image processing techniques. At last, a method that can automate adjustment of watermark intensity based on the measurement of image texture is proposed. The results of experiments show that it is practicable to predict the watermark strength to be applied to an image based on its texture.In Chapter 5, the method of image authentication is discussed. At first, the authentication system by digital signature and the system by fragile watermark are compared. Some problems that could be considered in design of fragile watermarking system are mentioned. The states of the art of DVD protection are described, and two protocol (CPRM and DTCP), which is proposed by W09 group, are introduced. A novel fragile watermarking approach and related tamper detection are proposed. Taking still image as an example, the watermark are embedded in the discrete wavelet domain of an image, so tamper detection is possible in localized spatial and frequen...
|
PDF Full Text Request