基于Tanimoto系数的JPEG碎片数据识别方法  

A JPEG DATA FRAGMENT IDENTIFICATION METHOD BASED ON TANIMOTO COEFFICIENT

作　　者：汤燕彬[1] 许榕生[2] 

机构地区：[1]福州大学数学与计算机学院,福建福州350000 [2]中国科学院高能物理研究所,北京100049

出　　处：《计算机应用与软件》2011年第9期80-81,92,共3页Computer Applications and Software

基　　金：中国科学院知识创新工程重点方向项目(YYYJ-1013)

摘　　要：当前市面上存在多种功能强大的电子取证工具,但这些取证工具缺乏对特定碎片数据类型的识别。提出满足当前实际需求的一种碎片数据类型识别方法。该方法以数据的字节频率分布为基础,首次将Tanimoto系数应用于JPEG碎片数据类型识别,利用文件类型特有的结构特征以及相邻碎片数据的关联性作为优化参数,提高识别率。实验结果表明,该方法能较好地识别出JPEG碎片数据类型。Several powerful digital forensics tools have been employed in the progress of obtaining judicial evidences.However,there is lack of identification of specific data fragments.In this article,a method that meets actual demands is proposed,which,based on the byte frequency distribution of data fragments,applies tanimoto coefficient to the identification of JPEG data fragments for the first time and promotes the chances to identifying those fragments by using special structural features and relationship between adjacent fragments as optimization parameters.Experimental results show that this method is able to efficiently identify the JPEG data fragments.

关 键 词：计算机取证 JPEG碎片数据 字节频率分布 Tanimoto系数 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]