【摘要】 为设计更加鲁棒的彩色图像水印算法,研究重要参数对水印的影响,提出一种基于非下采样Contourlet变换(NSCT)和奇异值分解(SVD)的水印算法。选用彩色图像R分量做NSCT分解,分解后提取中频系数进行奇异值分解,生成水印模板;中频系数矩阵根据加性规则嵌入水印,嵌入水印后对图像进行重构。实验重点分析算法中参数取值对算法性能的影响,给出合适的取值范围,结论具有典型性和普遍性。与其它同类算法相比,在大多数情况下,该算法的归一化互相关函数(NCC)和峰值信噪比(PSNR)值更高,对攻击的鲁棒性更强。更多还原

【Abstract】 To design a more robust color image watermarking algorithm,and study the influence of important parameters on the watermarking,a watermarking algorithm based on non-sub sampling Contourlet transform(NSCT)and singular value decomposition(SVD)was proposed.Red component of the color images was selected for NSCT decomposition,and middle frequency coefficients were extracted by singular value decomposition after NSCT decomposition,generating a watermark template.The watermarking was embedded according to the adding rules with the coefficient matrix.After embedding,the watermarked images were reconstructed.The experiment focused on the influence of the parameter values on the performance of the algorithm,and gave the appropriate range of values.The conclusion is typical and universal.Compared with other similar algorithms,in most cases,the normalized cross-correlation(NCC)function and peak signal-to-noise ratio(PSNR)of the proposed algorithm are high,so the robustness to attacks is better.更多还原