基于图像属性的Coiflet域数字水印算法

水印是由具有良好伪随机性的二值混沌序列构成的,为了实现水印图像在稳健性和不可见性之间的折衷,嵌入因子是由3个分因子构成的:信噪比(SNR)因子α,活动度因子β和兴趣因子γ.α是从SNR的要求算出的最佳水印强度因子,β表示了人类视觉系统(HVS)对图像的平滑区、边缘区和纹理区的不同感知特性,β是在l³空间计算的.考虑到人眼视觉的选择性,通过γ对图像中感兴趣区(ROI)嵌入的水印强度进行适当的调整,以提高水印图像的视觉效果.实验结果表明,本文算法是有效的.

Image Characteristics-Based Coiflet Domain Digital Watermarking Algorithm

The watermark is constructed by dyadic value chaotic sequence,which has very good pseudorandom properties.In order to perform the tradeoff between robustness and imperceptibility,the embedding factor consists of three sub-factors:signal-to-noise ratio(SNR)factor α ,activity measure factor β and interest factor γ . α is obtained through determining the optimal watermarking intensity,based on SNR. β represents different perceptual significances of human visual system(HVS)to smooth,edge and textured regions of image. β is calculated in the l³ space.Due to the HVS selectivity,the inserted watermark intensity is adjusted suitably by γ in the region of interest(ROI),so that the visual effect of the watermarked image can be enhanced.The experimental results show that the algorithm is effective.