基于压缩传感的灰度图像水印算法  

A grayscale image watermarking algorithm based on compressed sensing

作　　者：詹旭[1] 马将[1] 付克兰[2] 

机构地区：[1]四川理工学院自动化与电子信息学院,四川自贡643000 [2]成都农业科技职业学院信息技术分院,四川成都610000

出　　处：《西北师范大学学报（自然科学版）》2016年第1期47-52,共6页Journal of Northwest Normal University(Natural Science)

基　　金：国家自然科学基金资助项目(61178068);四川省教育厅资助项目(14ZB0223)

摘　　要：将水印图像采用压缩传感OMP算法进行一维小波逐行观测,生成观测矩阵,以观测矩阵做为嵌入水印,将原始载体灰度图像进行DCT变换后,选取低频信息段作为水印嵌入位置,然后将水印信息进行变换图像置乱处理,并采用基于奇异值的分解算法嵌入到原始载体灰度图像中.实验结果表明,此算法能够抵抗一定的几何攻击,对JPEG压缩、噪声攻击的抵抗能力有待进一步提高.This paper progressively observes watermark image compressed sensing algorithm of OMP to generate observation matrix which is used as the embedding watermark image.Then,the paper scrambles the embedding watermark image by arnold transform algorithm and embeds it into the original grayscale image by the singular value decomposition algorithm in the low-frequency of DCT domain.In order to test the robustness of this algorithm,the experiment is carried out.The results show that this algorithm can resist certain geometric attack and can improve anti-attack capability when suffering JPEG compression attack and noise attack.

关 键 词：压缩传感 奇异值 ARNOLD变换 

分 类 号：TN919.8[电子电信—通信与信息系统]