摘要
文本信息隐藏(TIH)是把隐秘信息嵌入文本载体中的技术,探讨其逆向攻击手段对于信息安全(IS)领域有着重要的意义.本文提出一种基于词间相关度统计特性的文本信息隐藏检测算法,并利用该算法对文本信息隐藏系统NICETEXT做了检测实验,取得了很好的效果.在对115个正常的文本段和212个由NICETEXT生成的文本段进行检测的实验中,成功地检测出106个正常的文本段和209个由NICETEXT生成的文本段.其中仅有9个正常文本段和3个生成文本段发生了误判,检测正确率高达96.3%.
Text Information Hiding （TIH） is a technique for embedding secret messages into text carriers. Research on attacking methods against TIH plays an important role in Information Security （IS） area. In this paper, a text information hiding detecting algorithm based on statistical characteristics of correlations between words is presented. An experiment to detect text segments generated by text information hiding system NICETEXT using the algorithm is carried out and a pretty good result is obtained. In the experiment, there are 115 normal text segments and 212 generated text segments detected, and 106 normal segments and 209 generated segments are successfully picked out. Only 9 normal segments and 3 generated segments are mistakenly examined. The accuracy of the detecting reaches as high as 96.3%.
出处
《小型微型计算机系统》
CSCD
北大核心
2008年第12期2199-2201,共3页
Journal of Chinese Computer Systems
基金
国家自然基金项目(60773032
60703071)资助
高校博士点基金(20060358014)
江苏省自然科学基金(BK2007060)项目的资助
关键词
信息隐藏
检测
词间相关度
N窗互信息
information hiding
detecting
correlations between words
N-window mutual information
