| 基于支持向量机同步的自适应水印检测方法 |
| |
| 引用本文： | 吴健珍,谢剑英.基于支持向量机同步的自适应水印检测方法[J].上海交通大学学报,2006,40(3):480-484,489. |
| |
| 作者姓名： | 吴健珍 谢剑英 |
| |
| 作者单位： | 上海交通大学,电子信息与电气工程学院,上海,200240 |
| |
| 摘 要： | 提出了一种基于支持向量机(SVM)同步的自适应水印检测方法.该方法利用模糊聚类分析图像纹理、亮度和对比度掩蔽特性,在离散小波变换域自适应嵌入水印信号.为抵抗几何攻击,利用6个图像组合矩表示原始图像的几何特征,通过SVM学习图像旋转、缩放和平移的变换参数,同步后再进行水印的检测和提取.仿真实验表明,该算法在一定的视觉强度下可嵌入鲁棒性更强的水印,能有效抵抗噪声、JPEG压缩和几何攻击,水印检测无需原始图像,具有实用性.
|
| 关 键 词： | 数字图像水印 图像配准 几何攻击 离散小波变换 支持向量机 |
| 文章编号： | 1006-2467（2006）03-0480-05 |
| 收稿时间： | 2005-03-05 |
| 修稿时间： | 2005-03-05 |
An Adaptive Watermarking Scheme Utilizing Support Vector Machine for Synchronization |
| |
| WU Jian-zhen,XIE Jian-ying.An Adaptive Watermarking Scheme Utilizing Support Vector Machine for Synchronization[J].Journal of Shanghai Jiaotong University,2006,40(3):480-484,489. |
| |
| Authors: | WU Jian-zhen XIE Jian-ying |
| |
| Institution: | School of Electronic, Information and Electrical Eng. , Shanghai Jiaotong Univ. , Shanghai 200240, China |
| |
| Abstract: | Through the classification of image's brightness,texture and contrast sensitivity,watermark is adaptively embedded in DWT domain.In order to register rotation,scale and translation(RST) transform parameters,support vector machine is utilized to learn image geometric pattern represented by six combined low order image moments.The watermark extraction is carried out after watermarked image has been synchronized without original image.The experiments show that it can embed more robust watermark under certain visual distance,(effectively) resist JPEG compression,noise and geometric attacks. |
| |
| Keywords: | digital image watermarking image registration geometric attack discrete wavelet transform(DWT) support vector machine(SVM) |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
