一种基于扫描相关度的LSB算法  

Improved LSB steganography based on scanning correlative degree

作　　者：任孝平[1] 蔡自兴[1] 卢薇薇[1] 

机构地区：[1]中南大学信息科学与工程学院,长沙410083

出　　处：《计算机应用》2009年第5期1349-1352,共4页journal of Computer Applications

基　　金：国家自然科学基金资助项目(9082030260805027);国家博士点基金资助项目(200805330005)

摘　　要：自然图像相邻像素序列具有一定的相关性。基于图像的这种统计特性,提出了扫描相关度的概念。隐藏信息的非均匀嵌入使得载体图像的相邻像素序列相关度降低,而且图像局部有相关度突变的现象,增大了隐藏信息被检测到的风险。进一步提出了一种LSB改进算法,该算法将隐藏信息均匀嵌入到载体图像中,并在嵌入过程中,根据相邻像素期望值对扫描相关度有突变的区域进行平滑。实验结果表明,该算法明显降低了图像相关度突变的现象。The adjacent pixels sequence of natural images has the correlation. Based on the statistical character, the concept of scanning correlative degree was proposed. Secret information embedded asymmetrically made the correlative degree of adjacent pixels sequence of carrier image reduced, especially the correlative degree of some local image may saltate. This phenomenon enhanced the risk of secret information being detected. Then an improved method of Least Significant Bit （LSB） was presented. Secret information could be embedded into carrier images symmetrically, and some areas of image which sahated were smoothed through expectation of adjacent pixels. Experiment results show that the new algorithm significantly reduces the saltation of correlative degree.

关 键 词：信息隐藏 扫描相关度 最低有效位算法 相邻像素期望 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]