基于直方图差异的模型隐写分析  

Histogram-Based Analysis against Model-Based Steganography

作　　者：孙文顒 张新鹏[1] 王朔中[1] 

机构地区：[1]上海大学通信与信息工程学院,上海200072

出　　处：《上海大学学报（自然科学版）》2008年第2期111-115,共5页Journal of Shanghai University:Natural Science Edition

基　　金：国家高技术研究发展计划(863计划)资助项目(2007AA01Z477);国家自然科学基金资助项目(60502039);上海市科委基础研究重点项目(04JC14037);上海市青年科技启明星计划资助项目(06QA14022)

摘　　要：基于模型(Model-Based,MB)的隐写方法可以保持JPEG载体图像的直方图在隐写前后几乎不变,使秘密信息具有良好的统计隐蔽性.根据MB隐写使得图像平坦区域和纹理区域的DCT系数分布变得相似这一线索,提出一种新的隐写分析方法,通过考察纹理较丰富的内容中的DCT系数直方图与全局DCT系数直方图之间的差异情况来分析图像是否经过MB隐写.实验结果表明,该方法可有效击破MB隐写.Model-based （MB） steganography for JPEG images can keep the histogram of DCT coefficients almost unchanged, therefore can resist the previous steganalytic methods. However, a loophole still exists. In MB steganography, a model derived from the original histogram is used so that the distribution of DCT coefficients in a smooth area becomes similar to that in a texture area. In view of this, we propose a steganalytic scheme against the MB steganography, which examines difference between DCT coefficient histogram in texture blocks and that of the whole image, and compare this difference with a threshold.Experimental results show that the proposed method can effectively detect the presence of secret data in an image embedded with the MB steganography.

关 键 词：信息隐藏 隐写分析 基于模型的隐写 JPEG图像 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]