期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

基于改进增强特征选择算法的特征融合图像隐写分析 被引量：4

Image steganalysis based on feature fusion by improved boosting feature selection algorithm
原文传递
导出
摘要 针对现有的基于特征融合的JPEG隐写分析方法特征冗余度高、通用性较低的问题,提出了一种基于改进的增强特征选择(BFS,boosting feature selection)算法的通用JPEG隐写分析方法。从线性相关度和非线性相关度两方面降低特征冗余,将特征自相关系数和互信息这两种统计性能引入到特征的评价准则中,重新设计了特征权重计算方法,改进了BFS算法的特征评价函数。通过改进的BFS特征选择算法将3组互补性较强且准确率高的特征进行融合降维,得到最优特征子集训练分类器。对3种高隐蔽性隐写算法F5、Outguess和MME3,在不同嵌入率下进行了大量实验。结果表明,本文方法的分析准确率高于现有的检测率较高的JPEG隐写分析方法和典型的融合分析方法,融合后的特征相关性明显下降,并且具有更强的通用性。 In view of the problems in the existing feature fusion based JPEG steganalysis schemes, such as high redundancy in selected features and weak universality, a universal JPEG steganalysis approach based on improved boosting feature selection （BFS） method is presented, Feature redundancy is reduced in as- pects of linear and nonlinear correlations. Statistical performanee including auto-correlation coefficients and mutual information is introduced in feature evaluation rules. The algorithm of computing feature weighting is redesigned. The feature evaluation function of BFS is improved. Three complementary sets of features that have high detection accuracy are fused using the improved BFS algorithm. The selected optimal feature subset is used for training classifiers. Experiments are done in various embedding rates for three steganographic schemes with high concealment,including F5, Outguess and MMEd. The results show that the detection accuracy of the proposed scheme is higher than that of some existing JPEG steganalysis approaches and some classical fusion methods. The fused features by improved BFS have lower correlation and this scheme has greater universality.
作者 时晨曦 张敏情
机构地区 武警工程大学电子技术系
出处 《光电子．激光》 EI CAS CSCD 北大核心 2014年第3期551-557,共7页 Journal of Optoelectronics·Laser
基金 国家自然科学基金(61379152) 陕西省自然科学基金基础研究(2012JIM8014)资助项目
关键词 隐写分析 特征融合 特征选择 改进的增强特征选择(BFS)算法 特征相关性 steganalysis feature fusion feature selection improved boosting feature selection （BFS） al- gorithm feature relevance
分类号 TP391 [自动化与计算机技术—计算机应用技术]
  • 相关文献

参考文献15

二级参考文献58

  • 1董火明,高隽,汪荣贵.多分类器融合的人脸识别与身份认证[J].系统仿真学报,2004,16(8):1849-1853. 被引量：17
  • 2Farid H,Siwei L.Detecting hidden messages using higher-order statistics and support vector machines[C].Proc of 5th Int Workshop on Information Hiding.Heidelberg:Springer,2002:340-354.
  • 3Fridrich J.Feature-based steganalysis for JPEG images and its implications for future design of steganographic schemes[C].Proc of 6th Int Workshop on Information Hiding.Heidelberg:Springer,2004:67-81.
  • 4Sullivan K,Madhow U,Chandrasekaran S,et al.Steganalysis of spread spectrum data hiding exploiting cover memory[C].Proc of Security,Steganography and Watermarking of Multimedia Contents-Ⅶ.San Jose:SPIE,2005:38-46.
  • 5Shi Y Q,Chen C,Chen W.A Markov process based approach to effective attacking JPEG steganography[C].Information Hiding Workshop 2006.Heidelberg:Springer,2006:249-264.
  • 6Ekenel H K,Sankur B.Muhiresolution face recognition[J].Image and Vision Computing,2005,23(5):469-477.
  • 7Cai H,Agaian S S.Spatial-frequency feature vector fusion based steganalysis[C].IEEE Int Conf on Systems,Man and Cybernetics.Taipei:IEEE Press,2006:1866-1870.
  • 8Rodriguez B M,Peterson G L,Agaian S S.Multi-class classification averaging fusion for detecting steganography[C].Proc of IEEE Int Conf on Systems Engineering.San Antonio:IEEE Press,2007:1-5.
  • 9Kharrazi M,Sencar T H,Memon N.Improving steganalysis by fusion techniques:A case study with image steganography[C].Proc of Security,Steganography and Watermarking of Multimedia Contents-Ⅷ.San Jose:SPIE,2006:123-137.
  • 10Solanki K, Sarkar A, Manjunath B S. YASS: Yet another steganographic scheme that resists blind steganalysis[C]. Lecture Notes in Computer Science. Berlin: Springer, 2007, 4567: 16-31.

共引文献33

同被引文献27

  • 1Kodovsky J, Fridrich J, Holub V. Ensemble classifiers for steganalysis of digital media[J]. IEEE Transactions on In- formation Forensics and Security, 2012,7 (2): 432-444.
  • 2Cogranne R, Zitzmann C, Retraint F, et al. A local adap- tive model of natural images for almost optimal detection of hidden data[J]. Signal Processing, 2014,100(1) : 169- 185.
  • 3Thai T,Cogranne R,Retraint F. Statistical model of quan- tized DOT coefficients: Application in the steganalysis of JSteg algorithm[J]. IEEE Transactions on Image Process- ing, 2014,23(5) : 1-14.
  • 4Cogranne R, Retraint F. An asymptotically uniformly most powerful test for Isb matching detection[J]., tEEE Trans- actions on Information Forensics and Security, 2013, 8 (3) :464-476.
  • 5Holub V, Fridrich J. Random projections of residuals for digital image steganalysis[J]. IEEE Transactions on Infor- mation Forensics and Security,2013,8(12) :1996-2006.
  • 6Kodovsky J,Sedighi V, Fridrich J. Study of cover source mismatch in steganalysis and ways to mitigate its impact[A]. Prec. of SPIE[C]. 2013,9028 : 10-22.
  • 7Ker A, Bas P, Bohme R, et el. Moving steganography and steganalysis from the laboratory into the real world[A]. Proc. of ACM Workshop on Information Hiding and Multi- media Security[C]. 2013,1 : 45-58.
  • 8Adler W, Brenning A. Ensemble classification of paired data[J]. Computational Statistics and Data Analysis, 2011,55 : 1933-1941.
  • 9Adhvaryu P, Panchal M. A review on diverse ensemble methods for classification[J]. Journal of Computer Engi- neering, 2012, l (4) : 27-32.
  • 10TIAN Jin,LI Ming-qiang, CHEN Fu-zan, et el. Coevolution- ary learning of neural network ensemble for complex classification tasks[J]. Pattern Recognition, 2012,45 (4) : 1373-1385.

引证文献4

二级引证文献5

光电子．激光
2014年 第3期

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部