基于双正交小波变换的彩色图像水印嵌入算法  

A digital watermarking embedding algorithm based on biorthogonal wavelet transform for color images

作　　者：王晨旸 李新华[1] 

机构地区：[1]安徽大学计算机智能与信号处理教育部重点实验室,安徽合肥230039

出　　处：《安徽大学学报（自然科学版）》2013年第6期65-71,共7页Journal of Anhui University(Natural Science Edition)

基　　金：国家自然科学基金资助项目(61172127);高等学校博士学科点专项科研基金资助项目(20113401110006);安徽省自然科学基金资助项目(1208085QF104);安徽大学"211工程"学术创新团队基金资助项目(KJTD007A)

摘　　要：基于传统小波变换的数字水印算法,提出一种基于双正交小波变换的彩色图像数字水印嵌入算法.算法利用双正交小波具有的良好分解和重构特性,对原始水印图像进行Arnold置乱后,将载体图像及水印图像分别进行2级和1级双正交小波变换.根据小波各频带系数之间的关系以及人类视觉系统掩蔽性,确定水印嵌入位置和强度,将置乱后的水印按照一定权重规则嵌入到载体相应的分解系数中,利用双正交小波逆变换得到含有水印的图像.仿真实验结果表明该算法不仅有较好的隐藏性,而且对于常见的操作攻击也表现出很强的鲁棒性.Based on traditional wavelet transform processing digital watermark, an algorithm which was based On multi-channel biorthogonal wavelet transform for embedding a digital color image into an original color image was presented in this paper. The algorithm combined the well characteristics of decomposition and reconstruction of biorthogonal wavelet. After the Arnold transform to the watermark image, the carrier and the watermark image were respectively into two and one level biorthogonal wavelet transform. The position and intensity of the watermark embedding were decided according to the weighting coefficients corresponding to different frequency bands and characteristics of the human visual system, the watermark with corresponding strength was embedded into the carrier image. The embedded watermark image got through the inverse biorthogonal wavelet transform. According to the experiment, the algorithm had invisibility and good robustness for common image processing operations attack.

关 键 词：双正交小波变换 数字水印 ARNOLD变换 HVS 版权保护 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]