摘要
为了提高水印的综合效果,根据多小波变换后相邻系数之间具有很强的相关性的特点,提出了一种基于支持向量机的图像水印算法.由于支持向量机在小样本训练的情况下具有良好的学习和泛化能力.因此,可以首先利用支持向量回归机建立相邻系数之间的关系模型.然后,通过调整模型的输入来嵌入或提取水印.实验结果表明,用算法得到的水印不但具有很好的图象感知质量,而且鲁棒性好,实用性强.
Considering the coherence among neighborhood coefficients in an muhiwavelet transform, a kind of muhiwavelet domain watermarking scheme based on support veetor maehine is proposed. It uses support vector machine to embed the watermark and gains satisfaetory results. Due to the good learning ability and generaliza- tion ability of SVM with limited training samples, it can learn the relationship between the selected coeffieients and its neighboring coefficients well with support vector regression. Then, a bit of the watermark is embedded or extracted by adjusting the value between the seleeted eoeffieients and the actual output of the trained SVR. Experimental results show that the proposed algorithm has good image pereeptual quality, which also possesses high watermark robustness to common image processing operation and praetieability.
出处
《西安文理学院学报（自然科学版）》
2009年第4期76-80,共5页
Journal of Xi’an University(Natural Science Edition)
关键词
数字水印
支持向量回归机(SVR)
多小波变换
SVM
watermark
support vector machine（SVM）
discrete wavelet transform（DWT）
insect digital information
