基于Contourlet域重要系数特性的自适应图像水印算法

Adaptive Image Watermarking Algorithm Based on the Characteristics of Important Coefficients in Contourlet Domain

对图像Contourlet变换的特性进行了分析和讨论,在此基础上提出了一种基于Contourlet域重要系数特性的自适应的水印算法,包括:(1)根据Contourlet变换域各阶方向子带的对应关系,利用低频和高频子带的树形结构确定出低频方向子带中的重要系数,以此系数作为水印嵌入位置的索引,把水印嵌入这些重要系数按树形结构映射到中频方向子带中;(2)根据嵌入点对应低频方向子带的能量和高尺度对应子带区域的纹理信息确定了一种自适应地调整嵌入水印强度的策略,使水印在嵌入位置和嵌入强度的选择上都保证使嵌入的水印具有很好的透明性和极大鲁棒性;(3)水印检测无需原始图像,通过统计图像的相关性即可验证水印的存在。仿真实验表明,所提出的水印算法不仅具有很好的透明性,而且对常见的图像处理具有很好的鲁棒性。

This paper analyzed and discussed the Contourlet coefficients＇ characteristics of natural images, and proposed an adaptive watermarking algorithm based on the characteristics of important coefficients in Contourlet domain. The proposed algorithm includes three important techniques： （1） according to the tree structure relation of Contourlet coeffi- cients between directional sub-bands in different scales, selects the important coefficients in the low frequency sub-band as the indices of embedding position, and embeds the watermarking into the middle frequency directional sub-bands mapped using those indices in terms of tree structure; （2） embeds watermarking with different strength adaptively ad- justed in terms of the energy of its parent coefficient in the lowest sub-band and the texture characteristics of its child coefficients in finer scale. The choice of embedding position and the embedding strategy effectively guarantee good transparency and maximum robustness of the proposed algorithm; （3） when detection, determines the existence of wa- termarking by the statistical correlation between the extracted watermarking and the original one without the original host image. The experimental results show the proposed algorithm has good transparency and is very robust to common image processing.