Research on Blind Watermarking Technique of Digital Images Based on Contourlet Transform
|Course||Applied Computer Technology|
|Keywords||Digital Watermarking Contourlet Transform Singular Value Decomposition(SVD) Human Visual System(HVS) Copyright Protection|
With the development of multimedia technology and the popularization of the Internet, the digital products are used more widely than before. The ways of how to release, acquire, and change information are more and more simple and varied. As a kind of important means in the domain of copyright protection, digital watermarking technology becomes a hot spot in the academic field. In this thesis, we do a lot of research on digital watermarking based on Contourlet domain. The main works of this paper are as follows:1. First of all, we introduce the background significance history and development of the digital watermarking both at home and abroad. Then analyze the digital watermarking technology and Contourlet transformation theory.2. A blind digital watermarking algorithm is proposed based on the analysis of characteristic of Contourlet domain in this thesis. Firstly, perfrom Contourlet transform on the host image, choosing the low frequency sub-band as the embedding domain; then find the position of digital watermarking in the embedding domain combining with 2D linear mapping. At last, embed digital watermarking to the host image with the method of parity quantitative.3. Combining with the character of Contourlet, an algorithm of quantitative is proposed in this thesis. We first compute the threshold of HVS(Human Visual System) and perform Contourlet transform to the host image, dividing the transformed host image according to the size of digital watermarking. Then choose the sub-block to embed watermark combined with HVS. The digital watermark can be extracted without original information.4. A blind watermarking algorithm with SVD based on Contourlet domain is proposed;1) select the low frequency sub-band of Contourlet transform as the embedding domain, and divide it into sub-block as the size of 8×8 ; 2) calculate the variance of each block, selecting the step length of the quantization adaptively according to the value of the variance , and perform SVD transfrom on each block; 3) choose the proper point to embed the watermark with the method of quantization. The robustness of the proposed algorithm has been improved.