摘要
文中提出了一种基于离散小波变换和奇异值分解的新的数字水印方法. 首先宿主图像被均匀分割成四块,对每一块进行一层小波分解,得到每一块的低频子带,再对每一低频子带进行奇异值分解,得到各低频子带的奇异值. 水印图像经过Arnold变换预处理后也进行奇异值分解,将得到的奇异值分别以一定的强度嵌入到宿主图像每一块低频子带的奇异值中,再进行反变换、拼接操作即可得到嵌入水印后的图像,实验结果表明,本文提出的方法,对噪声、剪切、滤波、JPEG压缩等常见的信号处理和攻击均具有较强的鲁棒性.
A new watermarking algorithm for digital image based on the combination of DWT and SVD is proposed in this paper. The original image is split into four blocks equally, and then DWT is applied to each block. The watermarking image is decomposed with SVD after being Arnold scrambled. The singular values of scrambled watermarking are embedded into the singular values of each subband LL of each block. Experiment results have demonstrated that the proposed algorithm can keep good quality for watermarking image and has good robustness under the condition of the common image processing.
出处
《西南民族大学学报(自然科学版)》
CAS
2007年第5期1029-1034,共6页
Journal of Southwest Minzu University(Natural Science Edition)
基金
Supported by National Ethnic Affairs Commission Foundation(05YN06).
关键词
数字水印
离散小波变换
奇异值分解
ARNOLD变换
digital watermarking
discrete wavelet transform
singular value decomposition
Arnold transform