基于统计分析的音频隐写分析研究
|
摘要
研究隐写分析对监视非法信息、防止机密信息泄露,保障国家安全和维持社会的稳定有着重要的意义。隐写分析的核心问题是选取检测率高,适用性广的特征向量。以往的文献缺乏对频域统计特征的深入研究,不能有效地对频域隐写方法进行检测。本文研究了音频DCT系数的统计分布以及隐写对DCT域统计特征的影响,在此基础上设计了两种新颖的隐写分析方法。为了提高检测率和检测尽可能多的隐写方法,提出了一种基于融合的隐写分析方法。
本文的主要工作和创新点如下:
1)针对音频DCT系数统计分布研究的不足,对音频DCT系数的统计分布进行了深入研究,为音频DCT系数建立了一个由广义高斯分布和α稳定分布组成的混合模型。研究结果表明,该模型比广义高斯分布和α稳定分布更接近于音频DCT系数的统计分布。
2)提出了一种基于DCT直方图统计矩的音频隐写分析方法,采用的特征是DCT系数直方图的高阶统计量,实验结果表明该算法获得了较好的检测率。
3)根据隐写会改变音频统计特征的假设,提出一种基于统计分布模型参数统计矩的音频隐写分析方法。实验数据表明,相比于同类算法,检测率有了较大的提高。
4)针对现有隐写分析方法实用性不强的问题,基于融合思想提出了一种通用的隐写分析方法。该分析方法能够继承已有隐写分析方法的优点,提高隐写分析算法的检测性能。
The research of steganalysis is significant to prevent disclosure of confidential information, monitor illegal information, protect national security and maintain social stability. Designing high detecting accuracy and wide applibality feature vectors is the core of steganalysis. Intensive research on statistical analysis of frequency domain in the literature is lack, so that the steanography which insert the secret message in frenquency domain can not be detected. In this paper an attempt has been made to analysis statistical distribution of audio DCT coefficients and the effect of steganography on the statistical nature. Based on the study, two novel steganalysis is designed. In order to improve the detection accuracy and enlarge the range of detected steganography, a universal steganalysis based on fusion technology is proposed.
The main work and innovation are as follows:
1) Aiming at the scarcity of statistical distribution of audio DCT coefficients, the statistical distribution of audio DCT coefficients is deliberated. A mixed model which is a linearly weighted average of generalized Gaussian distribution and alpha stable distribution is proposed. The experimental results show that the mixed model is near to the true distribution of DCT coefficients than generalized Gaussian distribution and alpha stable distribution.
2) An audio steganalysis based on statistical moments of DCT histogram is proposed. The statistical moments of the histogram in DCT domain and the statistical moments of the histogram of the wavelet coefficients of every level in frequency domain are calculated as the features. Experimental results show that the proposed technique gain high detection accuracy.
3) On the Assumption that steganography change the statistics of audio, a steganalysis based on the statistics of the parameters of statistical distribution is proposed. Experimental results show that compared to similar algorithms, detection rate has improved.
4) For the defect of practicability of existing steganalysis, a universal steganalysis based on fusion theory is proposed. This method can inherit the advantages of exiting steganalysis algorithm, and then improve the detection performance.
本文的主要工作和创新点如下:
1)针对音频DCT系数统计分布研究的不足,对音频DCT系数的统计分布进行了深入研究,为音频DCT系数建立了一个由广义高斯分布和α稳定分布组成的混合模型。研究结果表明,该模型比广义高斯分布和α稳定分布更接近于音频DCT系数的统计分布。
2)提出了一种基于DCT直方图统计矩的音频隐写分析方法,采用的特征是DCT系数直方图的高阶统计量,实验结果表明该算法获得了较好的检测率。
3)根据隐写会改变音频统计特征的假设,提出一种基于统计分布模型参数统计矩的音频隐写分析方法。实验数据表明,相比于同类算法,检测率有了较大的提高。
4)针对现有隐写分析方法实用性不强的问题,基于融合思想提出了一种通用的隐写分析方法。该分析方法能够继承已有隐写分析方法的优点,提高隐写分析算法的检测性能。
The research of steganalysis is significant to prevent disclosure of confidential information, monitor illegal information, protect national security and maintain social stability. Designing high detecting accuracy and wide applibality feature vectors is the core of steganalysis. Intensive research on statistical analysis of frequency domain in the literature is lack, so that the steanography which insert the secret message in frenquency domain can not be detected. In this paper an attempt has been made to analysis statistical distribution of audio DCT coefficients and the effect of steganography on the statistical nature. Based on the study, two novel steganalysis is designed. In order to improve the detection accuracy and enlarge the range of detected steganography, a universal steganalysis based on fusion technology is proposed.
The main work and innovation are as follows:
1) Aiming at the scarcity of statistical distribution of audio DCT coefficients, the statistical distribution of audio DCT coefficients is deliberated. A mixed model which is a linearly weighted average of generalized Gaussian distribution and alpha stable distribution is proposed. The experimental results show that the mixed model is near to the true distribution of DCT coefficients than generalized Gaussian distribution and alpha stable distribution.
2) An audio steganalysis based on statistical moments of DCT histogram is proposed. The statistical moments of the histogram in DCT domain and the statistical moments of the histogram of the wavelet coefficients of every level in frequency domain are calculated as the features. Experimental results show that the proposed technique gain high detection accuracy.
3) On the Assumption that steganography change the statistics of audio, a steganalysis based on the statistics of the parameters of statistical distribution is proposed. Experimental results show that compared to similar algorithms, detection rate has improved.
4) For the defect of practicability of existing steganalysis, a universal steganalysis based on fusion theory is proposed. This method can inherit the advantages of exiting steganalysis algorithm, and then improve the detection performance.
引文
[1] Simmons G J. The prisoners’problem and the subliminal channel; proceedings of the Cryptology, proc of CRYPTO, 1984 [C]. Plenum Press.
[2] Fridrich J. Applications of data hiding in digital images; proceedings of the ISSPA '99 Proceedings of the Fifth International Symposium on Signal Processing and Its Applications, Brisbane, Qld. , Australia 1999 [C].
[3] Bender W, Gruhl D, Morimoto N, et al. Techniques for data hiding [J]. Ibm Syst J, 1996, 35(3/4): 313-36.
[4] Kirovski D, Malvar H. Robust spread-spectrum audio watermarking, 2001 [C]. IEEE.
[5] http://www.petitcolas.net/fabien/steganography/mp3stego/
[6] Tachibana R. Two-dimensional audio watermark for MPEG AAC audio [J]. Security, Steganography, and Watermaking of Multimedia Contents VI, Jan, 2004, 19-22.
[7] Neubauer C, Herre J. Audio watermarking of MPEG-2 AAC bit streams; proceedings of the AES Convention:108 (February 2000), 2000 [C].
[8] Cheng S, Yu H, Xiong Z. Enhanced spread spectrum watermarking of MPEG-2 AAC audio, 2002 [C].
[9]王鹏军,徐淑正,张鹏, et al. MEPG-4 AAC中信息隐藏的研究[J].东南大学学报:自然科学版, 2007, 37(A01): 149-53.
[10]唐步天,郭立,刘振华.高级音频编码(AAC)的一种信息隐藏方法[J].声学技术, 2008, 27(004): 533-8.
[11] Adli A, Nakao Z. Three steganography algorithms for MIDI files, 2005 [C].
[12] Moulin P, O'sullivan J. Information-theoretic analysis of information hiding [J]. Ieee T Inform Theory, 2003, 49(3): 563-93.
[13] Costa M. Writing on dirty paper [J]. IEEE TRANS INFO THEORY, 1983, 29(3): 439-41.
[14] Cohen A, Lapidoth A. The Gaussian watermarking game [J]. Ieee T Inform Theory, 2002, 48(6): 1639-67.
[15] Somekh-Baruch A, Merhav N. On the capacity game of public watermarking systems [J]. Ieee T Inform Theory, 2004, 50(3): 511-24.
[16] Moulin P, Mihcak M K. The parallel-Gaussian watermarking game [J]. Information Theory, IEEE Transactions on, 2004, 50(2): 272-89.
[17] Johnson N F, Jajodia S. Steganalysis of images created using current steganography software [J]. Lect Notes Comput Sc, 1998, 1525(273-89.
[18] Johnson N F, Jajodia S. Steganalysis: The investigation of hidden information, 1998 [C]. IEEE.
[19] Westfeld A, Pfitzmann A. Attacks on steganographic systems [J]. Lecture notes in computer science, 2000, 61-76.
[20] Westfeld A. Detecting low embedding rates [J]. Lecture notes in computer science, 2003, 324-39.
[21] Fridrich J, Long M. Steganalysis of LSB encoding in color images, 2000 [C]. IEEE.
[22] Fridrich J, Goljan M, Du R. Detecting LSB steganography in color and gray-scale images [J]. IEEE Multimedia, 2001, 8(22-8.
[23] Fridrich J, Goljan M. Practical steganalysis of digital images-state of the art, 2002 [C].
[24] Hetzl S, Mutzel P. A graph-theoretic approach to steganography [J]. Lect Notes Comput Sc, 2005,3677(119-28.
[25] Ozer H, Avcibas I, Sankur B, et al. Steganalysis of audio based on audio quality metrics [M]. Proc SPIE. Santa Clara, CA, USA 2003: 55–66.
[26] Avcibas I. Audio steganalysis with content-independent distortion measures [J]. Ieee Signal Proc Let, 2006, 13(2): 92-5.
[27] Liu Y, Chiang K, Corbett C, et al. A novel audio steganalysis based on high-order statistics of a distortion measure with hausdorff distance [J]. Information Security, 2008, 487-501.
[28] Zeng W, Hu R, Ai H. Audio steganalysis of spread spectrum information hiding based on statistical moment and distance metric [J]. Multimed Tools Appl, 1-32.
[29] Kocal O, Avcibas E. Chaotic-Type Features for Speech Steganalysis [J]. Ieee T Inf Foren Sec, 2008, 3(4): 651-61.
[30] Altun O, Sharma G, Celik M, et al. Morphological steganalysis of audio signals and the principle of diminishing marginal distortions; proceedings of the Acoustics, Speech, and Signal Processing, 2005 Proceedings (ICASSP '05) Philadelphia, Pennsylvania, USA, March 18-23, 2005, 2005 [C].
[31]汝学民,庄越挺,吴飞.基于隐写工具的自相关特性进行音频隐写分析[J].通信学报, 2006, 27(04): 101-6.
[32] Johnson M, Lyu S, Farid H. Steganalysis of recorded speech, 2005 [C].
[33] Farid H. Detecting hidden messages using higher-order statistical models; proceedings of the International Conference on Image Processing 2002, Rochester, NY, 2002 [C].
[34] Sallee P. Model-based steganography [J]. Lect Notes Comput Sc, 2004, 2939(154-67.
[35] Sallee P. Model-based methods for steganography and steganalysis [J]. International Journal of Image and Graphics, 2005, 5(1): 167-89.
[36] Harmsen J J, Pearlman W A. Steganalysis of additive noise modelable information hiding; proceedings of the 5th Conference on Security and Watermarking of Multimedia Contents, Santa Clara, Ca, Jan 21-24, 2003 [C].
[37]张涛,平西建.基于差分直方图实现LSB信息伪装的可靠检测[J].软件学报, 2004, 15(001): 151-8.
[38] Petitcolas F, Anderson R, Kuhn M. Attacks on copyright marking systems, 1998 [C]. Springer.
[39] Zeng W, Ai H, Hu R. An algorithm of echo steganalysis based on power cepstrum and pattern classification]. IEEE.
[40]杨榆,雷敏,钮心忻, et al.基于回声隐藏的VDSC隐写分析算法[J].通信学报, 2009, 002): 83-8.
[41] Xie C, Cheng Y, Chen Y. An active steganalysis approach for echo hiding based on Sliding Windowed Cepstrum [J]. Signal Process,
[42] Zeng W, Ai H, Hu R. A novel steganalysis algorithm of phase coding in audio signal [J]. alpit, 2007, 261-4.
[43] Paraskevas I, Chilton E. Combination of magnitude and phase statistical features for audio classification [J]. Acoustics Research Letters Online, 2004, 5(111.
[44]黄昊,郭立,李琳.基于失真测度的直接扩频音频隐写分析[J].中国科学院研究生院学报, 2008, 02):
[45] Gao H Y. A blind steganalysis algorithm of QIM-based hiding; proceedings of the 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, Harbin, PEOPLES R CHINA, Aug 15-17, 2008 [C]. Ieee Computer Soc.
[46] Ru X M, Zhang H J, Huang X, et al. Steganalysis of audio: Attacking the Steghide; proceedings ofthe 4th International Conference on Machine Learning and Cybernetics, Canton, PEOPLES R CHINA, Aug 18-21, 2005 [C]. Ieee.
[47] Sieffert M, Forbes R, Green C, et al. Stego intrusion detection system, 2004 [C].
[48] Terhardt E. Calculating virtual pitch [J]. Hearing research, 1979, 1(2): 155.
[49] Painter T, Spanias A. Perceptual coding of digital audio [J]. P Ieee, 2000, 88(4): 451-513.
[50] Loizou P. Mimicking the human ear [J]. IEEE Signal Processing Magazine, 1998, 15(5): 101-30.
[51] Zwicker E, Fastl H, Hartmann W. Psychoacoustics-facts and models [J]. Physics Today, 2001, 54(6): 64-5.
[52] Johnston J D. Estimation of perceptual entropy using noise masking criteria, 1988 [C]. IEEE.
[53] Johnston J D. Transform coding of audio signals using perceptual noise criteria [J]. Selected Areas in Communications, IEEE Journal on, 1988, 6(2): 314-23.
[54] Brandenburg K. OCF--A new coding algorithm for high quality sound signals, 1987 [C]. IEEE.
[55]姜楠,王健,杨义先, et al. Midi数字水印算法[J].计算机应用研究, 2005, 22(6): 36-7.
[56]高海英,吕锐. MIDI数字水印算法[J].微电子学与计算机, 2007, 24(011): 83-4.
[57] http://www.nic.funet.fi/pub/crypt/steganography/
[58] Hide4PGP [J].
[59] http://steghide.sourceforge.net/
[60] Chou J, Ramchandran K, Sachs D, et al. Audio data hiding with application to surround sound, 2003 [C]. IEEE.
[61] Jackson T, Li F, Yates K. Hidden auxiliary media channels in audio signals by perceptually insignificant component replacement, 2005 [C]. IEEE.
[62] Malik H, Khokhar A, Ansari R. Robust data-hiding in audio, 2004 [C]. IEEE.
[63] Smith J, Comiskey B. Modulation and information hiding in images, 1996 [C]. Springer.
[64] Boney L, Tewfik a H, Hamdy K N. Digital watermarks for audio signals, 1996 [C]. Published by the IEEE Computer Society.
[65]白剑,杨榆,徐迎晖, et al. GSM移动通信系统中语音隐藏算法研究[J].中山大学学报:自然科学版, 2004, 43(A02): 156-9.
[66] Sterling M, Edwardl, Titlebaum, et al. an adaptive spread-spectrum hiding technique for digital audio [J]. ICASSP, 2005, 685-8.
[67] Gruhl D, Lu A, Bender W. Echo hiding, 1996 [C]. Springer.
[68] Kim H J, Choi Y H. A novel echo-hiding scheme with backward and forward kernels [J]. Circuits and Systems for Video Technology, IEEE Transactions on, 2003, 13(8): 885-9.
[69] Yardimci Y, Cetin a E, Ansari R. Data hiding in speech using phase coding, 1997 [C].
[70] Dong X, Bocko M F, Ignjatovic Z. Data hiding via phase manipulation of audio signals]. IEEE.
[71] Ho Y L, Wang H C. An audio watermarking algorithm based on significant component modulation, 2003 [C]. IEEE.
[72] Nelder J, Mead R. A simplex method for function minimization [J]. The computer journal, 1965, 7(4): 308.
[73] Takens F. Dynamical systems and turbulence [J]. Lecture notes in mathematics, 1981, 898(366.
[74] Cheng Q, Huang T. Robust optimum detection of transform domain multiplicative watermarks [J]. Ieee T Signal Proces, 2003, 51(4): 906-24.
[75] Solachidis V, Pitas I. Optimal detector for multiplicative watermarks embedded in the DFT domain of non-white signals [J]. Eurasip J Appl Sig P, 2004, 2004(16): 2522-32.
[76] Dumitrescu S, Wu X, Wang Z. Detection of LSB steganography via sample pair analysis, 2003[C]. Springer.
[77] echo watermarking in sub-band domain [J].
[78] Piva A, Barni M, Bartolini F, et al. DCT-based watermark recovering without resorting to the uncorruptedoriginal image, 1997 [C].
[79] Zeng W, Liu B. On resolving rightful ownerships of digital images by invisible watermarks, 1997 [C].
[80] Zeng W, Liu B. A statistical watermark detection technique without using originalimages for resolving rightful ownerships of digital images [J]. Ieee T Image Process, 1999, 8(11): 1534-48.
[81] Hernandez J, Amado M, Perez-Gonzalez F. DCT-domain watermarking techniques for still images: detector performance analysis and a new structure [J]. Ieee T Image Process, 2000, 9(1): 55-68.
[82] Cheng Q, Huang T S. An additive approach to transform-domain information hiding and optimum detection structure [J]. Ieee T Multimedia, 2001, 3(3): 273-84.
[83] Nikolaidis A, Pitas I. Asymptotically optimal detection for additive watermarking in the DCT and DWT domains [J]. Ieee T Image Process, 2003, 12(5): 563-71.
[84] Briassouli A, Strintzis M. Locally optimum nonlinearities for DCT watermark detection [J]. Ieee T Image Process, 2004, 13(12): 1604-17.
[85] Briassouli A, Tsakalides P, Stouraitis A. Hidden messages in heavy-tails: DCT-domain watermark detection using alpha-stable models [J]. Ieee T Multimedia, 2005, 7(4): 700-15.
[86] Oostveen J, Kalker T, Linnartz J. Optimal detection of multiplicative watermarks [J].
[87] Barni M, Bartolini F, De Rosa A, et al. A new decoder for the optimum recovery of nonadditive watermarks [J]. Ieee T Image Process, 2001, 10(5): 755-66.
[88] Barni M, Bartolini F, De Rosa A, et al. Optimum decoding and detection of multiplicative watermarks [J]. Ieee T Signal Proces, 2003, 51(4): 1118-23.
[89] Ng T, Garg H. Maximum-likelihood detection in DWT domain image watermarking using Laplacian modeling [J]. Ieee Signal Proc Let, 2005, 12(4): 285-8.
[90] Briassouli A, Strintzis M. Optimal watermark detection under quantization in the transform domain [J]. Ieee T Circ Syst Vid, 2004, 14(12): 1308-19.
[91] Kirovski D, Malvar H. Spread-spectrum watermarking of audio signals [J]. Ieee T Signal Proces, 2003, 51(4): 1020-33.
[92] Kirovski D, Malvar H, Yacobi Y. A dual watermarking and fingerprinting system [J]. Microsoft Research[Online] Available: http://research microsoft com, 2001,
[93] Chang C, Lin C. LIBSVM: a library for support vector machines [M]. 2001.
[94]马翼平,韩纪庆. DCT域音频水印:嵌入对策和算法[J].电子学报, 2006, 34(07): 1260-4.
[95]王秋生,孙圣和.一种在数字音频信号中嵌入水印的新算法[J].声学学报, 2001, 26(05): 464-7.
[96]马翼平,韩纪庆. DCT域音频水印算法的嵌入位置选择策略[J].计算机科学, 2005, 32(011): 139-41.
[97] Zhou Z, Zhou L. A novel algorithm for robust audio watermarking based on quantification DCT domain, 2007 [C]. IEEE Computer Society.
[98] Reininger R, Gibson J. Distributions of the two-dimensional DCT coefficients for images [J]. Communications, IEEE Transactions on [legacy, pre-1988], 1983, 31(6): 835-9.
[99] Bellifemine. F C a, Chimienti.R., and Ponti. R. Statistical analysis of the 2D-DCT coefficients of the differential signal of images [J]. Signal Process Image Commun, 1992, 4(477-88.
[100] Lam E, Goodman J. A mathematical analysis of the DCT coefficient distributions for images [J]. Ieee T Image Process, 2000, 9(10): 1661-6.
[101] F.Muller. Distributions shape of two-dimensional DCT coefficients for nature images [J]. Electron Lett, 1993, 29(22): 1935-6.
[102] Tanabe N, Farvardin N. Subband image coding using entropy-coded quantization over noisychannels [J]. Ieee J Sel Area Comm, 1992, 10(5): 926-43.
[103] Eude T, Grisel R, Cherifi H, et al. On the distribution of the DCT coefficients; proceedings of the International Conference on Acoustics, Speech, and Signal Processing, ICASSP-94, Adelaide 1994 [C].
[104] Yu R, Lin X, Rahardja S, et al. A statistics study of the MDCT coefficient distribution for audio, 2004 [C].
[105] Yaroslavsky L, Wang Y. DFT, DCT, MDCT, DST and signal fourier spectrum analysis; proceedings of the EUPSICO 2000: European signal processing, 2000 [C].
[106] Sharifi K, Leon-Garcia A. Estimation of shape parameter for generalized Gaussiandistributions in subband decompositions of video [J]. Ieee T Circ Syst Vid, 1995, 5(1): 52-6.
[107] Do M, Vetterli M. Wavelet-based texture retrieval using generalized Gaussian density and Kullback-Leibler distance [J]. Ieee T Image Process, 2002, 11(2): 146-58.
[108] Nolan J. Maximum likelihood estimation and diagnostics for stable distributions [J]. Lévy Processes: Theory and Applications, 2001, 379–400.
[109] Nolan J. Parameterizations and modes of stable distributions [J]. Statistics & Probability Letters, 1998, 38(2): 187-95.
[110] Bilmes J. A gentle tutorial of the EM algorithm and its application to parameter estimation for Gaussian mixture and hidden Markov models [J]. International Computer Science Institute, 1998, 4(
[111] Houck C, Joines J, Kay M. A genetic algorithm for function optimization: a Matlab implementation [M]. Genetic Algorithms + Data Structures = Evolution Programs. Springer-Verlag, New York. 1995.
[112]王秋生,孙圣和.基于量化数字音频信号频域参数的水印嵌入算法[J].声学学报(中文版), 2002, 04):
[113]温泉,王树勋,年桂君. DCT域音频水印:水印算法和不可感知性测度[J].电子学报, 2007, 35(09): 1702-5.
[114] Lie W N, Lin G S. A feature-based classification technique for blind image steganalysis [J]. Multimedia, IEEE Transactions on, 2005, 7(6): 1007-20.
[115]胡良梅,高隽.基于信息融合的图像理解方法研究[D];合肥工业大学出版社, 2007.
[116] Bebis G, Gyaourova A, Singh S, et al. Face recognition by fusing thermal infrared and visible imagery [J]. Image and Vision Computing, 2006, 24(7): 727-42.
[117] Scalzo F, Bebis G, Nicolescu M, et al. Evolutionary Learning of Feature Fusion Hierarchies; proceedings of the IEEE ICPR2008, Dec. 2008 [C].
[118] Vasilescu M a O, Terzopoulos D. A Tensor Algebraic Approach to Image Synthesis, Analysis and Recognition; proceedings of the sixth international Conference on 3-D Digital Imaging and Modeling,3DIM'07, Montreal, QC 21-23 Aug. 2007 2007 [C].
[119] Kittler J, Hatef M, Duin R P W, et al. On combining classifiers [J]. Ieee T Pattern Anal, 1998, 20(3): 226-39.
[120] Kraetzer C, Dittmann J. The impact of information fusion in steganalysis on the example ofaudio steganalysis; proceedings of the Proceedings of the SPIE Media Forensics and Security 2009 [C].
[121] Kraetzer C, Dittmann J. Improvement of information fusion-based audio steganalysis; proceedings of the Proceedings of SPIE, Multimedia on Mobile Devices 2010, San Jose, California, USA, 2010 [C].
[122] Bayram S, Dirik a E, Sencar H T, et al. An Ensemble of Classifiers Approach to Steganalysis].
[2] Fridrich J. Applications of data hiding in digital images; proceedings of the ISSPA '99 Proceedings of the Fifth International Symposium on Signal Processing and Its Applications, Brisbane, Qld. , Australia 1999 [C].
[3] Bender W, Gruhl D, Morimoto N, et al. Techniques for data hiding [J]. Ibm Syst J, 1996, 35(3/4): 313-36.
[4] Kirovski D, Malvar H. Robust spread-spectrum audio watermarking, 2001 [C]. IEEE.
[5] http://www.petitcolas.net/fabien/steganography/mp3stego/
[6] Tachibana R. Two-dimensional audio watermark for MPEG AAC audio [J]. Security, Steganography, and Watermaking of Multimedia Contents VI, Jan, 2004, 19-22.
[7] Neubauer C, Herre J. Audio watermarking of MPEG-2 AAC bit streams; proceedings of the AES Convention:108 (February 2000), 2000 [C].
[8] Cheng S, Yu H, Xiong Z. Enhanced spread spectrum watermarking of MPEG-2 AAC audio, 2002 [C].
[9]王鹏军,徐淑正,张鹏, et al. MEPG-4 AAC中信息隐藏的研究[J].东南大学学报:自然科学版, 2007, 37(A01): 149-53.
[10]唐步天,郭立,刘振华.高级音频编码(AAC)的一种信息隐藏方法[J].声学技术, 2008, 27(004): 533-8.
[11] Adli A, Nakao Z. Three steganography algorithms for MIDI files, 2005 [C].
[12] Moulin P, O'sullivan J. Information-theoretic analysis of information hiding [J]. Ieee T Inform Theory, 2003, 49(3): 563-93.
[13] Costa M. Writing on dirty paper [J]. IEEE TRANS INFO THEORY, 1983, 29(3): 439-41.
[14] Cohen A, Lapidoth A. The Gaussian watermarking game [J]. Ieee T Inform Theory, 2002, 48(6): 1639-67.
[15] Somekh-Baruch A, Merhav N. On the capacity game of public watermarking systems [J]. Ieee T Inform Theory, 2004, 50(3): 511-24.
[16] Moulin P, Mihcak M K. The parallel-Gaussian watermarking game [J]. Information Theory, IEEE Transactions on, 2004, 50(2): 272-89.
[17] Johnson N F, Jajodia S. Steganalysis of images created using current steganography software [J]. Lect Notes Comput Sc, 1998, 1525(273-89.
[18] Johnson N F, Jajodia S. Steganalysis: The investigation of hidden information, 1998 [C]. IEEE.
[19] Westfeld A, Pfitzmann A. Attacks on steganographic systems [J]. Lecture notes in computer science, 2000, 61-76.
[20] Westfeld A. Detecting low embedding rates [J]. Lecture notes in computer science, 2003, 324-39.
[21] Fridrich J, Long M. Steganalysis of LSB encoding in color images, 2000 [C]. IEEE.
[22] Fridrich J, Goljan M, Du R. Detecting LSB steganography in color and gray-scale images [J]. IEEE Multimedia, 2001, 8(22-8.
[23] Fridrich J, Goljan M. Practical steganalysis of digital images-state of the art, 2002 [C].
[24] Hetzl S, Mutzel P. A graph-theoretic approach to steganography [J]. Lect Notes Comput Sc, 2005,3677(119-28.
[25] Ozer H, Avcibas I, Sankur B, et al. Steganalysis of audio based on audio quality metrics [M]. Proc SPIE. Santa Clara, CA, USA 2003: 55–66.
[26] Avcibas I. Audio steganalysis with content-independent distortion measures [J]. Ieee Signal Proc Let, 2006, 13(2): 92-5.
[27] Liu Y, Chiang K, Corbett C, et al. A novel audio steganalysis based on high-order statistics of a distortion measure with hausdorff distance [J]. Information Security, 2008, 487-501.
[28] Zeng W, Hu R, Ai H. Audio steganalysis of spread spectrum information hiding based on statistical moment and distance metric [J]. Multimed Tools Appl, 1-32.
[29] Kocal O, Avcibas E. Chaotic-Type Features for Speech Steganalysis [J]. Ieee T Inf Foren Sec, 2008, 3(4): 651-61.
[30] Altun O, Sharma G, Celik M, et al. Morphological steganalysis of audio signals and the principle of diminishing marginal distortions; proceedings of the Acoustics, Speech, and Signal Processing, 2005 Proceedings (ICASSP '05) Philadelphia, Pennsylvania, USA, March 18-23, 2005, 2005 [C].
[31]汝学民,庄越挺,吴飞.基于隐写工具的自相关特性进行音频隐写分析[J].通信学报, 2006, 27(04): 101-6.
[32] Johnson M, Lyu S, Farid H. Steganalysis of recorded speech, 2005 [C].
[33] Farid H. Detecting hidden messages using higher-order statistical models; proceedings of the International Conference on Image Processing 2002, Rochester, NY, 2002 [C].
[34] Sallee P. Model-based steganography [J]. Lect Notes Comput Sc, 2004, 2939(154-67.
[35] Sallee P. Model-based methods for steganography and steganalysis [J]. International Journal of Image and Graphics, 2005, 5(1): 167-89.
[36] Harmsen J J, Pearlman W A. Steganalysis of additive noise modelable information hiding; proceedings of the 5th Conference on Security and Watermarking of Multimedia Contents, Santa Clara, Ca, Jan 21-24, 2003 [C].
[37]张涛,平西建.基于差分直方图实现LSB信息伪装的可靠检测[J].软件学报, 2004, 15(001): 151-8.
[38] Petitcolas F, Anderson R, Kuhn M. Attacks on copyright marking systems, 1998 [C]. Springer.
[39] Zeng W, Ai H, Hu R. An algorithm of echo steganalysis based on power cepstrum and pattern classification]. IEEE.
[40]杨榆,雷敏,钮心忻, et al.基于回声隐藏的VDSC隐写分析算法[J].通信学报, 2009, 002): 83-8.
[41] Xie C, Cheng Y, Chen Y. An active steganalysis approach for echo hiding based on Sliding Windowed Cepstrum [J]. Signal Process,
[42] Zeng W, Ai H, Hu R. A novel steganalysis algorithm of phase coding in audio signal [J]. alpit, 2007, 261-4.
[43] Paraskevas I, Chilton E. Combination of magnitude and phase statistical features for audio classification [J]. Acoustics Research Letters Online, 2004, 5(111.
[44]黄昊,郭立,李琳.基于失真测度的直接扩频音频隐写分析[J].中国科学院研究生院学报, 2008, 02):
[45] Gao H Y. A blind steganalysis algorithm of QIM-based hiding; proceedings of the 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, Harbin, PEOPLES R CHINA, Aug 15-17, 2008 [C]. Ieee Computer Soc.
[46] Ru X M, Zhang H J, Huang X, et al. Steganalysis of audio: Attacking the Steghide; proceedings ofthe 4th International Conference on Machine Learning and Cybernetics, Canton, PEOPLES R CHINA, Aug 18-21, 2005 [C]. Ieee.
[47] Sieffert M, Forbes R, Green C, et al. Stego intrusion detection system, 2004 [C].
[48] Terhardt E. Calculating virtual pitch [J]. Hearing research, 1979, 1(2): 155.
[49] Painter T, Spanias A. Perceptual coding of digital audio [J]. P Ieee, 2000, 88(4): 451-513.
[50] Loizou P. Mimicking the human ear [J]. IEEE Signal Processing Magazine, 1998, 15(5): 101-30.
[51] Zwicker E, Fastl H, Hartmann W. Psychoacoustics-facts and models [J]. Physics Today, 2001, 54(6): 64-5.
[52] Johnston J D. Estimation of perceptual entropy using noise masking criteria, 1988 [C]. IEEE.
[53] Johnston J D. Transform coding of audio signals using perceptual noise criteria [J]. Selected Areas in Communications, IEEE Journal on, 1988, 6(2): 314-23.
[54] Brandenburg K. OCF--A new coding algorithm for high quality sound signals, 1987 [C]. IEEE.
[55]姜楠,王健,杨义先, et al. Midi数字水印算法[J].计算机应用研究, 2005, 22(6): 36-7.
[56]高海英,吕锐. MIDI数字水印算法[J].微电子学与计算机, 2007, 24(011): 83-4.
[57] http://www.nic.funet.fi/pub/crypt/steganography/
[58] Hide4PGP [J].
[59] http://steghide.sourceforge.net/
[60] Chou J, Ramchandran K, Sachs D, et al. Audio data hiding with application to surround sound, 2003 [C]. IEEE.
[61] Jackson T, Li F, Yates K. Hidden auxiliary media channels in audio signals by perceptually insignificant component replacement, 2005 [C]. IEEE.
[62] Malik H, Khokhar A, Ansari R. Robust data-hiding in audio, 2004 [C]. IEEE.
[63] Smith J, Comiskey B. Modulation and information hiding in images, 1996 [C]. Springer.
[64] Boney L, Tewfik a H, Hamdy K N. Digital watermarks for audio signals, 1996 [C]. Published by the IEEE Computer Society.
[65]白剑,杨榆,徐迎晖, et al. GSM移动通信系统中语音隐藏算法研究[J].中山大学学报:自然科学版, 2004, 43(A02): 156-9.
[66] Sterling M, Edwardl, Titlebaum, et al. an adaptive spread-spectrum hiding technique for digital audio [J]. ICASSP, 2005, 685-8.
[67] Gruhl D, Lu A, Bender W. Echo hiding, 1996 [C]. Springer.
[68] Kim H J, Choi Y H. A novel echo-hiding scheme with backward and forward kernels [J]. Circuits and Systems for Video Technology, IEEE Transactions on, 2003, 13(8): 885-9.
[69] Yardimci Y, Cetin a E, Ansari R. Data hiding in speech using phase coding, 1997 [C].
[70] Dong X, Bocko M F, Ignjatovic Z. Data hiding via phase manipulation of audio signals]. IEEE.
[71] Ho Y L, Wang H C. An audio watermarking algorithm based on significant component modulation, 2003 [C]. IEEE.
[72] Nelder J, Mead R. A simplex method for function minimization [J]. The computer journal, 1965, 7(4): 308.
[73] Takens F. Dynamical systems and turbulence [J]. Lecture notes in mathematics, 1981, 898(366.
[74] Cheng Q, Huang T. Robust optimum detection of transform domain multiplicative watermarks [J]. Ieee T Signal Proces, 2003, 51(4): 906-24.
[75] Solachidis V, Pitas I. Optimal detector for multiplicative watermarks embedded in the DFT domain of non-white signals [J]. Eurasip J Appl Sig P, 2004, 2004(16): 2522-32.
[76] Dumitrescu S, Wu X, Wang Z. Detection of LSB steganography via sample pair analysis, 2003[C]. Springer.
[77] echo watermarking in sub-band domain [J].
[78] Piva A, Barni M, Bartolini F, et al. DCT-based watermark recovering without resorting to the uncorruptedoriginal image, 1997 [C].
[79] Zeng W, Liu B. On resolving rightful ownerships of digital images by invisible watermarks, 1997 [C].
[80] Zeng W, Liu B. A statistical watermark detection technique without using originalimages for resolving rightful ownerships of digital images [J]. Ieee T Image Process, 1999, 8(11): 1534-48.
[81] Hernandez J, Amado M, Perez-Gonzalez F. DCT-domain watermarking techniques for still images: detector performance analysis and a new structure [J]. Ieee T Image Process, 2000, 9(1): 55-68.
[82] Cheng Q, Huang T S. An additive approach to transform-domain information hiding and optimum detection structure [J]. Ieee T Multimedia, 2001, 3(3): 273-84.
[83] Nikolaidis A, Pitas I. Asymptotically optimal detection for additive watermarking in the DCT and DWT domains [J]. Ieee T Image Process, 2003, 12(5): 563-71.
[84] Briassouli A, Strintzis M. Locally optimum nonlinearities for DCT watermark detection [J]. Ieee T Image Process, 2004, 13(12): 1604-17.
[85] Briassouli A, Tsakalides P, Stouraitis A. Hidden messages in heavy-tails: DCT-domain watermark detection using alpha-stable models [J]. Ieee T Multimedia, 2005, 7(4): 700-15.
[86] Oostveen J, Kalker T, Linnartz J. Optimal detection of multiplicative watermarks [J].
[87] Barni M, Bartolini F, De Rosa A, et al. A new decoder for the optimum recovery of nonadditive watermarks [J]. Ieee T Image Process, 2001, 10(5): 755-66.
[88] Barni M, Bartolini F, De Rosa A, et al. Optimum decoding and detection of multiplicative watermarks [J]. Ieee T Signal Proces, 2003, 51(4): 1118-23.
[89] Ng T, Garg H. Maximum-likelihood detection in DWT domain image watermarking using Laplacian modeling [J]. Ieee Signal Proc Let, 2005, 12(4): 285-8.
[90] Briassouli A, Strintzis M. Optimal watermark detection under quantization in the transform domain [J]. Ieee T Circ Syst Vid, 2004, 14(12): 1308-19.
[91] Kirovski D, Malvar H. Spread-spectrum watermarking of audio signals [J]. Ieee T Signal Proces, 2003, 51(4): 1020-33.
[92] Kirovski D, Malvar H, Yacobi Y. A dual watermarking and fingerprinting system [J]. Microsoft Research[Online] Available: http://research microsoft com, 2001,
[93] Chang C, Lin C. LIBSVM: a library for support vector machines [M]. 2001.
[94]马翼平,韩纪庆. DCT域音频水印:嵌入对策和算法[J].电子学报, 2006, 34(07): 1260-4.
[95]王秋生,孙圣和.一种在数字音频信号中嵌入水印的新算法[J].声学学报, 2001, 26(05): 464-7.
[96]马翼平,韩纪庆. DCT域音频水印算法的嵌入位置选择策略[J].计算机科学, 2005, 32(011): 139-41.
[97] Zhou Z, Zhou L. A novel algorithm for robust audio watermarking based on quantification DCT domain, 2007 [C]. IEEE Computer Society.
[98] Reininger R, Gibson J. Distributions of the two-dimensional DCT coefficients for images [J]. Communications, IEEE Transactions on [legacy, pre-1988], 1983, 31(6): 835-9.
[99] Bellifemine. F C a, Chimienti.R., and Ponti. R. Statistical analysis of the 2D-DCT coefficients of the differential signal of images [J]. Signal Process Image Commun, 1992, 4(477-88.
[100] Lam E, Goodman J. A mathematical analysis of the DCT coefficient distributions for images [J]. Ieee T Image Process, 2000, 9(10): 1661-6.
[101] F.Muller. Distributions shape of two-dimensional DCT coefficients for nature images [J]. Electron Lett, 1993, 29(22): 1935-6.
[102] Tanabe N, Farvardin N. Subband image coding using entropy-coded quantization over noisychannels [J]. Ieee J Sel Area Comm, 1992, 10(5): 926-43.
[103] Eude T, Grisel R, Cherifi H, et al. On the distribution of the DCT coefficients; proceedings of the International Conference on Acoustics, Speech, and Signal Processing, ICASSP-94, Adelaide 1994 [C].
[104] Yu R, Lin X, Rahardja S, et al. A statistics study of the MDCT coefficient distribution for audio, 2004 [C].
[105] Yaroslavsky L, Wang Y. DFT, DCT, MDCT, DST and signal fourier spectrum analysis; proceedings of the EUPSICO 2000: European signal processing, 2000 [C].
[106] Sharifi K, Leon-Garcia A. Estimation of shape parameter for generalized Gaussiandistributions in subband decompositions of video [J]. Ieee T Circ Syst Vid, 1995, 5(1): 52-6.
[107] Do M, Vetterli M. Wavelet-based texture retrieval using generalized Gaussian density and Kullback-Leibler distance [J]. Ieee T Image Process, 2002, 11(2): 146-58.
[108] Nolan J. Maximum likelihood estimation and diagnostics for stable distributions [J]. Lévy Processes: Theory and Applications, 2001, 379–400.
[109] Nolan J. Parameterizations and modes of stable distributions [J]. Statistics & Probability Letters, 1998, 38(2): 187-95.
[110] Bilmes J. A gentle tutorial of the EM algorithm and its application to parameter estimation for Gaussian mixture and hidden Markov models [J]. International Computer Science Institute, 1998, 4(
[111] Houck C, Joines J, Kay M. A genetic algorithm for function optimization: a Matlab implementation [M]. Genetic Algorithms + Data Structures = Evolution Programs. Springer-Verlag, New York. 1995.
[112]王秋生,孙圣和.基于量化数字音频信号频域参数的水印嵌入算法[J].声学学报(中文版), 2002, 04):
[113]温泉,王树勋,年桂君. DCT域音频水印:水印算法和不可感知性测度[J].电子学报, 2007, 35(09): 1702-5.
[114] Lie W N, Lin G S. A feature-based classification technique for blind image steganalysis [J]. Multimedia, IEEE Transactions on, 2005, 7(6): 1007-20.
[115]胡良梅,高隽.基于信息融合的图像理解方法研究[D];合肥工业大学出版社, 2007.
[116] Bebis G, Gyaourova A, Singh S, et al. Face recognition by fusing thermal infrared and visible imagery [J]. Image and Vision Computing, 2006, 24(7): 727-42.
[117] Scalzo F, Bebis G, Nicolescu M, et al. Evolutionary Learning of Feature Fusion Hierarchies; proceedings of the IEEE ICPR2008, Dec. 2008 [C].
[118] Vasilescu M a O, Terzopoulos D. A Tensor Algebraic Approach to Image Synthesis, Analysis and Recognition; proceedings of the sixth international Conference on 3-D Digital Imaging and Modeling,3DIM'07, Montreal, QC 21-23 Aug. 2007 2007 [C].
[119] Kittler J, Hatef M, Duin R P W, et al. On combining classifiers [J]. Ieee T Pattern Anal, 1998, 20(3): 226-39.
[120] Kraetzer C, Dittmann J. The impact of information fusion in steganalysis on the example ofaudio steganalysis; proceedings of the Proceedings of the SPIE Media Forensics and Security 2009 [C].
[121] Kraetzer C, Dittmann J. Improvement of information fusion-based audio steganalysis; proceedings of the Proceedings of SPIE, Multimedia on Mobile Devices 2010, San Jose, California, USA, 2010 [C].
[122] Bayram S, Dirik a E, Sencar H T, et al. An Ensemble of Classifiers Approach to Steganalysis].