Logo image
基於最佳摻入位址的高抗壓縮浮水印系統
Thesis

基於最佳摻入位址的高抗壓縮浮水印系統

范志鴻
Masters, National Tsing Hua University
2005

Abstract

數位浮水印最佳摻入位址最佳摻入數值壓縮零值JPEG壓縮 digital watermarkoptimal embedded pointoptimal embedded valuecompressive zero numberJPEG compression
With the popularity of high speed internet, the transmissionof digital media need to be more faster and easier. Therefore, people usually compress the media data to keep the size as small as possible. Hence, there exist an important issue for copyright protection to be solved. In order to avoid the piracy from illegal behavior, several kinds of technologies are developed to solve this problem. Digital watermarking is a novel and developed technology that can overcome above disadvantage. Whether the digital data is modified or not, we can still judge the legitimacy by checking its watermarking. The unique property of the the watermark is that it cannot be separated from embedded media. It means that the watermark can survive in any arbitrary attack. Therefore, it is important that how to detect rightful ownerships under various attacks. Accordingly, in this thesis we pay more attention on compressive attack and propose a new watermark system to avoid it.In this thesis, we present a new method of digital watermarking system. We want to design the system not only to resist high compressive attack but also to keep the image quality as good as possible. My proposed method mainly concentrates on high JPEG compression, which has been established the international compressive standard for still image. There are two main proposed methods, one is the proposed analysis method, and the other is the proposed embedded method. In the proposed analysis method, owing towe particularity analyze the distortion of host image on JPEG compression. Therefore, we can individually find out the optimal embedded points and values for every image. This process is the most important factor to achieve our high performance of experiments. The experiments shows that my proposed method can fit theoretical anticipation under JPEG compression. Also we will compare our method with others' to indirectly prove the performance of our experiment .

Metrics

1 Record Views

Details

Logo image