小波域中基于支持向量机的鲁棒性数字水印算法  

Robust watermarking scheme based on support vector machine in DWT domain

作　　者：付永钢[1] 王惠蓉[2] 

机构地区：[1]集美大学计算机工程学院,福建厦门361021 [2]集美大学文学院,福建厦门361021

出　　处：《计算机工程与设计》2008年第17期4584-4588,共5页Computer Engineering and Design

基　　金：福建省自然科学基金计划项目(2007J0029);福建省青年创新基金项目(2006F3090)

摘　　要：提出一种在小波域中新的基于支持向量机的鲁棒性水印嵌入和检测算法。通过对彩色图像中蓝色分量的小波变换,同时在小波域中计算低频子带的JND(just noticeable distortions)值,水印信号被自适应地嵌入到小波变换后的低频子带系数中。当图像经历各种攻击以后,通过支持向量机的训练,它仍然可以记住添加水印信号的位置同其周围像素之间的相对关系,利用支持向量机强的学习泛化能力,可以有效实现对水印信号的鲁棒性检测。实验结果表明,该方法对多种攻击具有较好的鲁棒性。A novel watermarking scheme based on support vector machine （SVM） in discrete wavelet transform （DWT） domain is proposed. By applying the DWT on blue channel of the color image, and the computation of the JND value in the DWT domain according to the human visual system （HVS） model, the watermark is embedded into the lower frequency subband at its adaptive strength. By the training of the SVM, it can memorize the relationship between the pixel at the embedding position and its＇ neighbors when the image is undergone some attacks. Since the SVM have good learning and generalization ability, the proposed method can achieve good robust- ness. Experimental results show good robustness of proposed method against different attacks.

关 键 词：数字水印 鲁棒性 小波变换 支持向量机 信息隐藏 

分 类 号：TP309.2[自动化与计算机技术—计算机系统结构]