拟合盲隐写分析结果的隐写组合测评方法

doi:10.3969/j.issn.1001-2400.2015.04.025

Abstract

Abstract:

KL divergence gives a precise estimation of the difference between cover and stego mediums, but the high computational complexity makes it impropriate for steganography benchmarking. The existing benchmarking methods use other statistics to evaluate the divergence between cover and stego features, but the performance is relatively poor. To solve this problem, we propose an assorted benchmarking for steganography based on blind steganalyzer accuracy fitting. The two complementary basic statistics, i.e., the mean value of single dimensional mutual information and the maximum mean discrepancy, are combined to obtain a better estimate of the divergence between cover and stego features.

Key words: steganography benchmarking, mutual information, maximum mean discrepancy, regression analysis

XIA Bingbing;ZHAO Xianfeng;ZHANG Hong. Assorted benchmarking for steganography based on blind steganalyzer accuracy fitting[J].J4, 2015, 42(4): 153-158.

References

［1］ Pevny T, Fridrich J. Benchmarking for Steganography ［C］//Lecture Notes in Computer Science: 5284. Berlin: Springer, 2008: 251-267.
［2］ Cachin C. An information-theoretic Model For Steganography ［C］//Lecture Notes in Computer Science: 1525. Heidelberg: Springer, 1998: 306-318.
［3］ Haussler D, Opper M. Mutual Information, Metric Entropy and Cumulative Relative Entropy Risk ［J］. The Annals of Statistics, 1997, 25(6): 2451-2492.
［4］ Gretton A, Borgwardt K M, Rasch M, et al. A Kernel Method for the Two-sample-problem ［C］//Proceedings of the Conference on Advances in Neural Information Processing Systems: 19. Canada: Neural Information Proceeding System Foundation, 2007: 513-520.
［5］ Chang C C, Lin C J. LIBSVM: a Library for Support Vector Machines ［J］. Transactions on Intelligent Systems and Technology, 2011, 2(3): 171-197.
［6］ Bas P, Filler T, Pevny T. Break Our Steganographic System: The Ins and Outs of Organizing BOSS ［C］//Lecture Notes in Computer Science: 6958. Berlin: Springer, 2011: 59-70.
［7］ Westfield A. F5—a Steganographic Algrithm High Capacity Despite Better Steganalysis ［C］//Lecture Notes in Computer Science: 2137. Berlin: Springer, 2001: 289-302.
［8］ Latham A. JPHIDE and JPSEEK ［CP／OL］. ［2014-01-12］. http://linux01.gwdg.de／～alatham／stego.html.
［9］ Upham D. JSteg ［CP／OL］. ［2014-01-12］. http://zooid.org／～paul／crypto／jsteg／.
［10］ Sallee P. Model-based Steganography ［C］//Lecture Notes in Computer Science: 2939. Berlin: Springer, 2004: 154-167.
［11］ Kim Y, Duric Z, Richards D. Modified Matrix Encoding Technique for Minimal Distortion Steganography ［C］//Lecture Notes in Computer Science: 4437. Berlin: Springer, 2007: 314-327.
［12］ Provos N. Defending Against Statistical Steganalysis ［C］//Proceedings of the 10th USENIX Security Symposium: 10. Berkeley: USENIX Association , 2001: 323-335.
［13］ Fridrich J, Goljan M, Soukal D. Perturbed Quantization Steganography ［J］. Multimedia Systems, 2005, 11(2): 98-107.
［14］ Hetzl S, Mutzel P. A Graph Theoretic Approach to Steganography ［C］//Lecture Notes in Computer Science: 3677. Berlin: Springer, 2005: 119-128.
［15］ Filler T, Judas J, Fridrich J. Minimizing Additive Distortion in Steganography Using Syndrome-Trellis Codes ［J］. IEEE Transactions on Information Forensics and Security, 2011, 6(3): 920-935.
［16］ Sachnev V, Kim H J, Zhang R. Less Detectable JPEG Steganography Method Based on Heuristic Optimization and BCH Syndrome Coding ［C］//Proceedings of the 11th ACM Workshop on Multimedia and Security. New York: ACM, 2009: 131-140.
［17］ Chen B, Wornell G W. Quantization Index Modulation Methods for Digital Watermarking and Information Embedding of Multimedia ［J］. Journal of VLSI Signal Processing, 2001, 27(1-2): 7-33.
［18］ Solanki K, Sarkar A, Manjunath B S. YASS: Yet Another Steganographic Scheme That Resists Blind Steganalysis ［C］//Lecture Notes in Computer Science: 4567. Berlin: Springer, 2007: 16-31.
［19］ Fridrich J, Kodovsky J. Rich Models for Steganalysis of Digital Images ［J］. IEEE Transactions on Information Forensics and Security, 2012, 7(3): 868-882.
［20］ Kodovsky J, Fridrich J. Steganalysis of JPEG Images Using Rich Models ［C］//Proceedings of SPIE: 8303. Bellingham: SPIE, 2012: 1-13.
［21］ Kodovsky J, Fridrich J. Steganalysis in High Dimensions: Fusing Classifiers Built on Random Subspaces ［C］//Proceedings of SPIE: 7880. Bellingham: SPIE, 2011: 1-13.
［22］ Kodovsky J, Fridrich J, Holub V. Ensemble Classifiers for Steganalysis of Digital Media ［J］. IEEE Transactions on Information Forensics and Security, 2012, 7(2): 432-444.
［23］ Pevny T, Bas P, Fridrich J. Steganalysis by Subtractive Pixel Adjacency Matrix ［J］. IEEE Transactions on Information Forensics and Security, 2010, 5(2): 215-224.

[1]	WANG Yuxi;HUANG Guoce;LI Wei;WANG Yequn. MIMO radar waveform design in the presence of clutter and colored noise [J]. Journal of Xidian University, 2017, 44(4): 125-131.
[2]	LI Qi;JI Hongbing;TONG Ming. Robust registration of multimodality medical images based on the principal ordinal feature and hybrid entropy [J]. J4, 2011, 38(6): 30-36+145.
[3]	DU Zhe;LIU San-yang. Research on variations of least square support vector machine [J]. J4, 2009, 36(2): 331-337.
[4]	JIU Bo;LIU Hong-wei;LI Li-ya;WU Shun-jun. Feature mutual information method for radar waveform optimization [J]. J4, 2009, 36(1): 139-144.
[5]	JIU Bo;LIU Hong-wei;LI Li-ya;WU Shun-jun. Method for waveform design based on mutual information [J]. J4, 2008, 35(4): 678-684.

Assorted benchmarking for steganography based on blind steganalyzer accuracy fitting

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 5

Metrics

Comments

Recommended 0