基于神经网络分类的图像水印算法  

Image watermarking algorithm based on artificial neural networks classification

作　　者：邵晓根[1] 孙天凯[1] 丁宾[1] 王兴元[2] 

机构地区：[1]徐州工程学院信电工程学院,江苏徐州221008 [2]大连理工大学计算机科学与技术学院,辽宁大连116024

出　　处：《计算机应用》2011年第6期1505-1507,共3页journal of Computer Applications

基　　金：国家自然科学基金资助项目(60973152);高等学校博士学科点专项基金资助项目(20070141014);徐州工程学院重点基金资助项目(XKY2008107;XKY2009115)

摘　　要：将人工神经网络(ANN)、广义猫映射及概率统计等知识相结合构造了一种图像空间域水印算法。采用神经网络作为载体图像的纹理分类器,突出原始图像的纹理区。使用广义猫映射对水印进行置乱预处理,提高了水印信息的安全性。水印嵌入时采用最小化像素改变的优化策略,提取时应用概率统计等知识较好地实现了水印信息不可见性和鲁棒性的统一。实验结果表明,该方法能有效地抵抗剪切攻击、噪声攻击、最低有效位(LSB)攻击、滤波攻击等。The paper,by adopting jointly Artificial Neural Network（ANN）,general Amold mapping,and statistical methods,intends to formulate an algorithm based on image spatial domain watermark.While the ANN was used as the texture analyzer of the original image,the texture zone of the original image would be given prominently.The safety of the watermark was improved by using the amold mapping rearrangement.An optimizing strategy of minimizing the changes of pixels was adopted when the watermark is embedded.The application of statistical method guaranteed the sightlessness and the robustness of the digital watermark information.The experimental results demonstrate that the watermark is robust against cut attack,salt-pepper noise attack,Least Significant Bit（LSB） attack,winner attack,and so on.

关 键 词：人工神经网络 水印 猫映射 鲁棒性 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]