基于视觉感知模型的大容量视频隐写算法  

High capacity video steganographic algorithm based on visual perception model

作　　者：梁惠[1] 郭立[1] 陈运必[1] 

机构地区：[1]中国科学技术大学电子科学与技术系,合肥230027

出　　处：《中国科学院研究生院学报》2010年第1期55-62,共8页Journal of the Graduate School of the Chinese Academy of Sciences

基　　金：国家自然科学基金项目(60772032)资助

摘　　要：为了获取较大隐写容量和较好不可感知性,结合人眼视觉感知特性,提出一种视觉感知指导数据嵌入的大容量视频隐写算法.该算法综合了多参数空间对比灵敏度函数、基于DCT(discrete cosine transform)块分类的纹理掩蔽效应及运动注意机制,以确定各DCT系数的失真掩蔽测度,进而控制隐写容量.实验结果表明,该隐写算法具有较大隐写容量,同时较好地保持了含密视频的不可感知性.In order to provide large capacity of steganography and maintain a good visual quality of stego-video,a novel high capacity video steganographic algorithm based on visual perception model is presented. This algorithm integrates multi-parameter spatial contrast sensitivity function,contrast masking based on classification of DCT blocks,and motion attention. Then secret information is embedded under control of distortion masking measure of each DCT coefficient. Experimental results show that the new algorithm has ability to hide secret data of large capacity without noticeable distortion.

关 键 词：隐写 视频 视觉感知模型 失真掩蔽测度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]