基于图像归一化的Contourlet域数字水印方案  

Image Normalization Based Robust Digital Watermarking Scheme in Contourlet Domain

作　　者：王向阳[1,2] 朱丹丹[1] 杨红颖[1] 

机构地区：[1]辽宁师范大学计算机与信息技术学院,辽宁大连116029 [2]中国科学院软件研究所信息安全国家重点实验室,北京100039

出　　处：《小型微型计算机系统》2009年第11期2272-2276,共5页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(60773031;60873222)资助;信息安全国家重点实验室开放基金项目(03-06))资助;计算机软件新技术国家重点实验室开放基金项目(A200702)资助;大连市科技基金项目(2006J23JH020)资助;"图像处理与图像通信"江苏省重点实验室开放基金项目(ZK205014)资助;江苏省计算机信息处理技术重点实验室开放课题基金项目(KJS0602)资助;近代声学教育部重点实验室开放课题基金项目(0802)资助;辽宁省教育厅科技研究项目(2008351)资助

摘　　要：Contourlet变换是一种全新的高维信号奇异性分析工具,其不仅具有良好的方向性和各向异性,而且能够高效率捕获图像几何结构.本文以Contourlet变换理论及图像归一化技术为基础,提出一种基于图像归一化的Contourlet域数字水印方案.该方案首先利用图像归一化技术将原始载体映射到几何不变空间内,并提取出归一化图像的重要区域;然后对重要区域实施Contourlet变换;最后根据人眼视觉特性及系数相关性,利用量化调制策略将水印信息嵌入到Contourlet域低频子带内.仿真实验结果表明,该数字图像水印方案不仅具有良好的透明性,而且具有较强的抵抗常规信号处理及几何攻击的能力.Contoudet transform is an efficient anisotropy analysis tool in high dimensions, which not only possess the main features of directionality and anisotropy, but also can effectively capture the intrinsic geometrical structures such as smooth contours （ at different scales and directions） in natural images. In this paper, an image normalization based robust digital watermarking scheme in Contourlet domain is proposed. Firstly, the geometrically invariant space is constructed by using image normalization and the significant region is obtained from the normalized image by utilizing the invariant centroid theory. Then, the Contourlet transform is performed on the significant region. Finally, the digital watermark is adaptively embedded into the significant region by quantizing the low-frequency Contourlet coefficients according to the Human Visual System （HVS ）. Especially, the predistortion compensation technique is used to reduce image distortion generated by image normalization. Experimental results show that the proposed scheme is not only invisible and robust against common signals processing such as sharpening, noise adding, and JPEG2000 compression, but also robust against the geometric distortion such as rotation, translation, scaling, row or column removal, cropping, and local random bending.

关 键 词：图像水印 几何攻击 CONTOURLET变换 图像归一化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]