基于粗糙集的图像隐藏信息检测改进方法  

An Improved Image Hidden Information Detection Method Based on Rough Set Theory

作　　者：邹劲松[1] 唐旭[2] 

机构地区：[1]重庆水利电力职业技术学院机电与电子信息工程系,重庆402160 [2]重庆大学科技处,重庆400044

出　　处：《测控技术》2015年第4期131-134,共4页Measurement & Control Technology

摘　　要：针对图像统计特征高维且相关性具有较大缺陷等问题,同时为了提高信息隐藏盲检测系统的检测效率,基于粗糙集理论,提出了一种改进的图像信息隐藏盲检测方法,并进行了实验研究。首先,提出一个改进的通用隐写分析系统框架,给出实现步骤和方法;然后利用粗糙集理论设计算法,降低特征维数,减小分类计算复杂度,消除统计特征间的相关性;最后改用支持向量机构造分类器,对两种典型的Cox和Piva扩频隐秘术进行实验。结果表明,该方法的检测正确率和时间效率等检测性能都有较大的提高,用于图像隐藏信息检测是可行、有效的。To improve the detection efficiency of blind steganalysis system for hidden information,and solve the problems of high dimension and relevancy of image statistical features,an improved blind steganalysis method for hidden information is proposed based on the rough set theory.Firstly,an improved common steganalysis system framework is proposed,and the implementation procedures and methods are provided.Then the algorithm is designed based on the rough set theory to reduce the dimension of feature,decrease the complexity of classified counting,and eliminate the relevancy between statistical features.At last,the support vector machine is used to construct the classifier,and experiments on the two typical Cox and Piva spread spectrum stealth techniques are carried out.The results show that this method can get better results in the detection accuracy and time efficiency.

关 键 词：隐写分析 盲检测 检测效率 粗糙集理论 D约简 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]