基于Radon变换与Schur分解的彩色图像水印算法  

Color Image Watermarking Algorithm Based on Radon Transform and Schur Decomposition

作　　者：孙彦飞 陈青[1] SUN Yan-fei;CHEN Qing(University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学,上海200093

出　　处：《包装工程》2018年第21期183-188,共6页Packaging Engineering

基　　金：国家自然科学基金(61672354)

摘　　要：目的针对彩色图像盲水印算法较为复杂且抗旋转能力差等问题,结合矩阵Schur分解和Radon变换提出一种抗旋转攻击的彩色图像盲水印算法。方法该算法首先提取彩色宿主图像的U通道图像,其次对U通道图像经NSCT变换后的低频分量进行分块处理,将每个像素块分解为对称矩阵和反对称矩阵,最后对每块对称矩阵进行Schur分解,将二值水印嵌入至分解后上三角矩阵的最大特征值中。提取水印信息时,利用Randon变换对含水印图像进行几何校正后再进行水印提取操作。结果文中算法提取水印相似度较高,其NC值达到0.9414;结构相似性较高,其SSIM值达到0.9183。结论该水印算法在常规信号处理攻击、抗旋转攻击下均具有极强的鲁棒性。The work aims to propose a color image blind watermarking algorithm resistant to rotation attacks in view of such problems as the complex color image blind watermarking algorithm with poor anti-rotation ability,and combined with Schur decomposition and Radon transform.Firstly,the proposed algorithm extracted the U channel image of color host image,and then divided the low frequency components of U channel image after NSCT transform into blocks to decompose each pixel block into symmetric and anti-symmetric matrix.Finally,the Schur decomposition of each symmetric matrix was conducted,and the binary watermark was embedded in the maximum eigenvalue of the decomposed upper triangular matrix.For the extraction of watermark information,Radon transform was used to extract the watermark after the geometric correction of watermark image.The proposed algorithm extracted watermarks with high similarity and its NC value reached 0.9414.The structural similarity was high and its SSIM value reached 0.9183.The watermarking algorithm is highly robust against conventional signal processing attacks and anti-rotation attacks.

关 键 词：重构彩色图像 非下采样CONTOURLET变换 SCHUR分解 RADON变换 几何校正 

分 类 号：TP309.7[自动化与计算机技术—计算机系统结构]