Multi-level Wavelet Decomposition Of Complex Digital Watermarking Technology,

Internet has become a main way for us to get information along with the widely use of network and the applications of multimedia technology. Nowadays, digital information products can be distributed via the World Wide Web to a large number of people in a cost efficient way and people can easily get these through Internet. The digital information can be easily duplicated and revised. These characters of digital media, however, bring a new challenge in applications. How to protect the benefits of the owner of the digital information products becomes the first imperative problem to resolve. Digital watermarking technology plays an important role for copyright protection; it allows embedding an imperceptible or perceptible mark (for example, embedded fingerprinting and digital signature information) in multimedia data to identify the owner copyright.Since the traditional coding technology for image and video is using of DCT, the blocked and whole DCT technology in frequency domain was used in order to embed watermarking. However the wavelet transform has some advantages better than DCT, it can be used as both image compression and watermarking embedding. Because of the standardization process of JPEG2000, we think it is imperative to consider the wavelet transform domain for watermarking embedding.After studying and analyzing the disadvantages of the previous algorithms in digital watermarking, we focus our work on the researching and implementation of the combining watermarking technology. The main idea of the combining watermarking is that the watermarking information is embedded into the high frequency section and the low frequency section of the wavelet domain respectively. We will embed a small logo watermarking using image fusion technology in low frequency and a Gaussian distribution sequence with mean 0 and variance 1 in high frequency section by using of spread spectrum technology, so we can achieve a more robust watermarking.There are totally five chapters in this paper. In chapter one, we will try to motivate the research effort in watermarking and discuss some applications. The "watermarking problem" is also introduced in chapter one and we analyses why it is hard to design watermarking algorithms that it can fulfill the requirements derived in this chapter. We also investigate the relationship between watermarking technology and other information hiding technology. In chapter two, the required background about wavelet theory and perceptual coding techniques is briefly introduced. In this section, we explain why we choose the technology based on discrete wavelet transformation rather than other transformation methods to embed watermarking. Furthermore, we introduced the Human Visual System that was used in watermarking algorithm widely. In chapter three, we illuminate a general model of digital watermarking in discrete wavelettransformation domain, and then several typical algorithms are introduced. Our owncontributions are presented in chapter four. We presented a new idea about digital watermarking that it is called a combining watermarking in frequency domain for static image. The presented proposal takes advantage of high frequency and low frequency section watermarking. We present a new idea in order to decide the dividing location between the low frequency and the high frequency according to energy distribution characters. In chapter five, we simulate and implement our algorithm by computer, and we compare it with the algorithm presented by Cox in some performance. Finally, we propose some suggestions for how to establish a more robust watermarking algorithm in future researching.