基于分块FRIT-SVD的鲁棒零水印算法  

A robust zero-watermarking algorithm based on the block FRIT-SVD

作　　者：曲长波[1] 于智龙 李栋栋[1] QU Chang-bo;YU Zhi-long;LI Dong-dong(College of Software,Liaoning Technical University,Huludao 125105,China)

机构地区：[1]辽宁工程技术大学软件学院,辽宁葫芦岛125105

出　　处：《计算机工程与科学》2018年第6期1005-1016,共12页Computer Engineering & Science

基　　金：国家自然科学基金(61540056;61172144)

摘　　要：针对水印图像轮廓特征的提取,使用小波变换并不是最佳方式,脊波变换较小波变换有着更好的逼近精度和稀疏性能。结合二维混沌系统、SVD和位平面技术共同构造零水印信息,提出了一种基于脊波变换域的鲁棒零水印算法。该算法首先对载体图像进行二维一级小波变换提取低频域,利用分块策略对其分块并进行有限脊波变换(FRIT),经SVD提取最大奇异值构造特征矩阵后,使用二维混沌系统对特征矩阵进行置乱加密,提取特征矩阵中的重要位平面与加密水印信号结合生成注册零水印信息,不仅提高了水印鲁棒性,双重加密后的水印安全性能也得到提升。实验表明,该算法具有较好的鲁棒性、安全性和易操作性,能够有效抵抗各类图像攻击。Wavelet transform is not the best way for extracting watermarking image contour features.Compared with the wavelet transform,the ridgelet transform has better accuracy and sparse approximation performance.We propose a robust zero-watermarking algorithm based on ridgelet transform domain,which combines the two-dimensional chaotic system,SVD and bit-plane to construct the watermarking.Firstly,the algorithm takes advantage of the wavelet transform to extract the low-frequency domain in the carrier image.Secondly,we decompose the low-frequency domain according to the blocking strategy.When all the blocks are transformed by the FRIT,the characteristic matrix can be constructed by the SVD.Moreover,we apply the two-dimensional chaotic system to encrypt the characteristic matrix.Finally,we extract the most significant bit-plane in the feature matrix,and generate registration zero-watermarking information by combining the encrypted meaningful watermarking,which not only improves the robustness of the watermarking,but also greatly enhances the security of watermarking through double encryption.Experimental results demonstrate that the algorithm has good robustness,security and easy operation,and that it can effectively resist various types of attacks.

关 键 词：脊波变换 奇异值分解 二维混沌系统 重要位平面 零水印算法 

分 类 号：TP309.2[自动化与计算机技术—计算机系统结构]