一种基于支持向量机和遗传算法的自适应图像水印方法  

An Adaptive Image Watermarking Scheme Based on Support Vector Machine and Genetic Algorithm

作　　者：孟凡满[1] 彭宏[1,2] 裴峥[1] 王军[3] 

机构地区：[1]西华大学数学与计算机学院,成都610039 [2]电子科技大学电子工程学院,成都610054 [3]西华大学电气信息学院,成都610039

出　　处：《模式识别与人工智能》2009年第2期312-317,共6页Pattern Recognition and Artificial Intelligence

基　　金：四川省教育厅重点项目(No.07ZA112);四川省学术和技术带头人培养基金项目(No.08209057)资助

摘　　要：提出一种基于支持向量机(SVM)和遗传算法(GA)的离散余弦变换(DCT)域盲数字图像水印方法.该方法能自适应于图像的局部特征.依据图像块的局部特性,利用SVM对图像块分类,自适应地确定水印嵌入强度,GA用来优化水印嵌入位置.实验结果表明该方法有较好的不可见性和较强对抗攻击的鲁棒性.An adaptive blind image watermarking scheme in DCT domain is proposed based on support vector machine （SVM） and genetic algorithm （GA）. The original image is divided into small image blocks, and then SVM is used to classify image blocks into several classes based on their local characteristics. The embedding strength of each block is adaptively determined according to the types of the image block, and GA is used to seek optimal embedding positions. Experimental results demonstrate that the p scheme has good invisibility and strong robustness against several attacks.

关 键 词：数字水印 支持向量机(SVM) 遗传算法(GA) 人类视觉系统 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]