基于DWT-SVD压缩量化的数字图像盲水印算法  

Compression quantitative digital image blind watermarking algorithm based on DWT-SVD

作　　者：汤永利[1] 张亚萍[1] 高玉龙[1] 叶青[1] 闫玺玺[1] TANG Yongli;ZHANG Yaping;GAO Yulong;YE Qing;YAN Xixi(College of Computer Science and Technology,Henan Polytechnic University,Jiaozuo 454000,P.R.China)

机构地区：[1]河南理工大学计算机科学与技术学院,河南焦作454000

出　　处：《重庆邮电大学学报（自然科学版）》2018年第2期265-271,共7页Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)

基　　金：国家自然科学基金(61300216);河南省科技厅项目(152102410048;142300410147);河南省教育厅项目(12A520021;16A520013);河南理工大学博士基金(B2013-043;B2014-044)~~

摘　　要：为了解决奇异值分解(singular value decomposition,SVD)算法提取水印时需要原始载体图像的缺陷,以及量化索引调制(quantization index modulation,QIM)均匀量化不适用于非均匀信号的问题,通过引入μ律压缩技术,提出一种新的基于DWT-SVD压缩量化的数字图像盲水印算法。该算法对载体图像进行分块,对每一分块实施离散小波变换(discrete wavelet transform,DWT),以及对变换后的近似部分系数进行SVD分解,使用μ律压缩函数压缩分解后的最大奇异值,用QIM的方法嵌入二值水印。算法只用到了最大奇异值,可以盲提取水印,消除因传输原始载体图像产生的不安全性,μ律压缩技术也减小了嵌入水印对原始载体图像的扰乱。仿真实验结果表明,该算法保持了较高的透明性,并对高斯噪声、中值滤波、联合图像专家小组(joint photographic experts group,JPEG)压缩、缩放等常见攻击具有更强的鲁棒性。In order to address the needs of the original image when SVD algorithms extracts the watermark,and to solve the problem of not being applicable to non-uniform signal for QIM uniform quantization,a compression quantitative digital image blind watermarking algorithm based on DWT-SVD is proposed by introducingμ-law technology.The original image is firstly divided into blocks and conducted with discrete wavelet transform,and then SVD is applied to the approximate part coefficients of original image.In this regard the largest singular value is compressed and quantized byμ-law technology.Finally,the algorithm embeds binary watermark data by QIM.The algorithm can blindly extract the watermark and eliminate the insecurity caused by the transmission of original image.Theμ-law compression technique also reduces the turbulence of the embedded watermark to the original cover image.Experimental results show that this algorithm solves the problems and maintains high transparency.Moreover,the proposed scheme has strong robustness against common attacks,such as Gaussian noise,median filtering,JPEG compression and image scaling attack.

关 键 词：奇异值分解(SVD) 离散小波变换 μ律压缩 量化索引调制 盲水印 数字水印 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]