基于时空混沌和小波域HVS模型的图像扩频水印  

Spread-Spectrum Watermarking for Images Based on Spatiotemporal Chaotic Binary Sequences and HVS Model in Wavelet Domain

作　　者：贾淑芸[1] 余乐军[2] 

机构地区：[1]太原重型机械学院,太原030024 [2]中国科学院计算所,北京100085

出　　处：《太原重型机械学院学报》2004年第2期130-134,141,共6页Journal of Taiyuan Heavy Machinery Institute

摘　　要：数字水印为数字媒体版权保护的有效方法。本文利用小波分析的优良性质,结合基于小波变换域中的视觉门限模型,选择在图像的较低频子带的重要系数中加入一定幅值的水印,水印为基于时空混沌的二值序列,它具有良好的随机性、不可见性和健壮性。检测时,根据水印图像提取的待测水印和水印序列的相似性来判断水印是否存在。模拟实骀结果表明,该方案具有良好的效果。Digital watermarking is an effective method of copyright protection.By using fine wavelet analysis and HVS model,the robustness of the watermarking can be enhanced.Further more,it is proposed to realize spread-spectrum watermarking for still images with binary sequences produced by spatiotemporal chaotic systems.The binary sequences arc sensitive to the initial values of spatiotemporal chaotic systems.The statistical properties of the binary spread-spectrum sequences.such as the run lengths distribution properties,auto and cross correlation,are very excellent.The simulation results show that tire watermarking is invisibleand robust.

关 键 词：数字水印 时空混沌 小波自适应变换 人类视觉系统 小波分析 二值序列 版权保护 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]