一种基于DWT和SVD的彩色图像零水印算法  

A zero-watermark algorithm of color image based on DWT and SVD

作　　者：潘黄琼[1] 陈光喜[1] 丁勇[1] 

机构地区：[1]桂林电子科技大学数学与计算科学学院,广西桂林541004

出　　处：《桂林电子科技大学学报》2011年第5期399-402,共4页Journal of Guilin University of Electronic Technology

基　　金：国家自然科学基金(60963024);广西教育厅科研基金(200807LX112);广西学位与研究生教育改革和发展专项课题(200910595R06)

摘　　要：为了提高水印技术的鲁棒性,提出了彩色图像在离散小波变换(DWT)和奇异值分解(SVD)下的零水印算法。对原始载体图像的R、G、B这3个空间分别进行离散小波变换,分别对每个空间的低频子带进行分块,且对分块后的低频子带进行奇异值分解;通过提取3个空间中最大的前若干项来构造水印。实验结果表明,该算法对抗各种攻击有较强的鲁棒性。In order to enhance the robustness of watermark technology,using discrete wavelet transformation（DWT） and the block singular value decomposition（SVD）, a zero-watermark algorithm based on color image is proposed. It conducts DWT with the original image of the R, G, B space, then divides its low frequency band into blocks respec- tively and conducts SVD with every block. Zero-watermark is generated by several former largest singular values of the three spaces. Experimental results show that the proposed algorithm has strong robustness against various attacks.

关 键 词：奇异值分解 离散小波变换 零水印 鲁棒性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]