基于运动分析的DWT-SVD域视频水印算法  

DWT- SVD Domain Video Watermarking Algorithm Based on Motion Analysis

作　　者：阳溢[1] 刘博文[1] 李扬 

机构地区：[1]重庆通信学院 [2]解放军61655部队

出　　处：《重庆理工大学学报（自然科学）》2015年第10期132-138,共7页Journal of Chongqing University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(61272043);重庆市基础与前沿研究计划项目(cstc2013jj B4009);重庆通信学院2014年度院立科研项目

摘　　要：为了有效保护数字视频版权,在深入研究视频画面运动特征的基础上,提出一种基于运动分析的DWT-SVD域视频水印算法。先将视频I帧亮度分量进行离散小波变换,并对LL分量进行运动分析来确定水印嵌入强度;然后对水印混沌置乱;再将置乱水印嵌入到LL分量分块奇异值分解后的U矩阵中,实现水印盲提取。实验结果表明:该算法在不可见性和鲁棒性等方面具有良好表现。In order to protect the copyright of digital videos,on the basis of studying the video motion features,a DWT-SVD domain video watermarking algorithm based on motion analysis was proposed.Firstly,the luminance components of the I-frame were transformed by DWT and the watermarking embedding strengths were determined by motion analyzing on LL components. Then the original watermarking was scrambled by chaotic mapping. Furthermore,the scrambled watermarking was embedded into the U matrixes which were transformed by blocking SVD on LL components. And also,the blind extraction was realized. Experimental results demonstrate that the algorithm has good performance on invisibility and robustness.

关 键 词：视频水印 运动分析 混沌置乱 离散小波变换 奇异值分解 不可见性 鲁棒性 

分 类 号：TN911.7[电子电信—通信与信息系统]