基于分类与选择机制的鲁棒水印算法  

Robust Watermarking Algorithm Based on Classification and Selection Mechanism

作　　者：董晶[1] 齐春[1] 阎鸿森[1] 

机构地区：[1]西安交通大学电子与信息工程学院,西安710049

出　　处：《西安交通大学学报》2006年第12期1400-1404,共5页Journal of Xi'an Jiaotong University

基　　金：陕西省自然科学基础研究计划资助项目(2003E212)

摘　　要：针对用均值滤波计算边缘信息的水印算法会引起图像某些边缘失真的问题,通过分析均值滤波和图像边缘的特性,提出了一个基于分类与选择机制的水印算法.先将图像的8×8像素块划分为边缘块或非边缘块,并根据每一分块选定的离散余弦变换(DCT)的中低频系数的均值挑选出边缘块中不适合嵌入水印信息的分块,再通过增加均值滤波窗函数的长度以尽量减少对DCT系数的改动,得到一个改进的水印嵌入规则.与利用均值滤波计算边缘信息的水印算法相比,基于分类与选择机制的水印算法可以达到更好的不可感知性,同时对多种攻击具有更好的鲁棒性.仿真实验表明,当JPEG压缩品质因子为20时,提取水印和原始水印的归一化相关值平均提高了0.205.Focusing on the problem that the watermarking algorithm using mean filtering to calculate side information can cause visible image distortions in some edge blocks, an algorithm based on classification and selection mechanism was proposed through analyzing the characteristics of mean filtering and edge blocks, in which the blocks with size 8 ×8 pixels were classified in to edge or non-edge blocks, then the blocks that were unsuitable for embedding watermark in edge blocks were excluded according to the mean values of chosen low frequency coefficients of discrete cosine transformation （DCT） for each block. Moreover, the length of mean-filter window function was increased so that the DCT coefficients were modified as less as possible, thus an improved rule for embedding watermark was obtained. Experimental results indicate that the proposed algorithm has better imperceptibility and higher robustness against many kinds of common attacks. The average normalized correlation value between the extracted and the original watermarks can be increased by 0. 205 when the quality factor of JPEG compression is equal to 20.

关 键 词：均值滤波 边缘信息 分类与选择机制 离散余弦变换 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]