Statistical Correlations and Machine Learning for Steganalysis

  • Qingzhong Liu
  • Andrew H. Sung
  • Bernardete M. Ribeiro


In this paper, we present a scheme for steganalysis based on statistical correlations and machine learning. In general, digital images are highly correlated in the spatial domain and the wavelet domain; hiding data in images will affect the correlations. Different correlation features are chosen based on ANOVA (analysis of variance) in different steganographic systems. Several machine learning methods are applied to classify the extracted feature vectors. Experimental results indicate that our scheme in detecting the presence of hidden messages in several steganographic systems is highly effective.


Support Vector Machine Wavelet Domain Hide Message Hiding Method Fisher Linear Discriminant 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    A. Westfeld, A. Pfitzmann (2000) Attacks on Steganographic Systems. LNCS 1768, Springer-Verlag, Berlin: 61–75Google Scholar
  2. [2]
    N. Provos (2001) Defending against Statistical Steganalysis. Proceedings of the 10th USENIX Security Symposium: 323–335Google Scholar
  3. [3]
    J. Fridrich, M. Goljan, and R. Du (2001) Detecting LSB steganography in color and gray-scale image, IEEE Multimedia: 22–28Google Scholar
  4. [4]
    H. Farid (2002) Detecting hidden messages using higher-order statistical models. In ICIP 2002, Rochester, New YorkGoogle Scholar
  5. [5]
    S. Lyu, H. Farid (2004) Steganalysis Using Color Wavelet Statistics and One-Class Support Vector Machines. Proc. SPIE: Security, Steganography, and Watermarking of Multimedia Contents VI, vol. 5306Google Scholar
  6. [6]
    I. Avcibas, N. Memon and B. Sankur (2003) Steganalysis using image quality metrics. IEEE transactions on Image Processing, Vol. 12, No. 2Google Scholar
  7. [7]
    M. U. Celik, G. Sharma, and A. M. Tekalp, (2004) Universal image steganalysis using rate-distortion curves. Proc. SPIE: Security, Steganography, and Watermarking of Multimedia Contents VI, vol. 5306Google Scholar
  8. [8]
    E. Kawaguchi, R. O. Eason (1998) Principle and Applications of BPCS-Steganography. Proc. Of SPIE, Vol.3528: 464–473CrossRefGoogle Scholar
  9. [9]
    A. Westfeld (2001) High Capacity Despite Better Steganalysis (F5-A Steganographic Alogrithm). LNCS 2137, Springer-Verlag: 289–302Google Scholar
  10. [10]
    J. Yu, J. Han, K. Lee, D. Ryu, S. Lee (2003) SES (Steganography Evading Statistical analyses). Pacific Rim Workshop on Digital Steganography 2003Google Scholar
  11. [11] Scholar
  12. [12] Scholar
  13. [13] Scholar
  14. [14] Scholar
  15. [15]
    Rencher A. C. (1995) Methods of Multivariate Analysis, John Wiley, New YorkGoogle Scholar
  16. [16]
    J. Huang, D. Mumford (1999) Statistics of Natural Images and Models. Computer Vision and Pattern Recognition, Vol. 1Google Scholar
  17. [17]
    A. Srivastava, A. Lee, E.P Simoncelli and S. Zhu (2003) on advances in statistical modeling of natural images, Journal of Mathematical Imaging and Vision 18(1): 17–33MathSciNetCrossRefGoogle Scholar
  18. [18]
    G. Winkler (1995) Image Analysis, Random Fields and Dynamic Monte Carlo Methods. Springer: BerlinGoogle Scholar
  19. [19]
    M. J. Wainwright, E. P. Simoncelli (2000) Scale mixtures of Gaussians and the statistics of natural images. In: Advances in Neural Information Processing Systems, S. A. Solla, T. K. Leen and K.-R. Muller (Eds.): 855–861Google Scholar
  20. [20]
    M. I. Schlesinger and V. Hlavac (2002) Ten lectures on statistical and structural pattern recognition. Kluwer Academic PublishersGoogle Scholar
  21. [21]
    J. A. K. Suykens, T. Van Gestel, J. De Brabanter, B. De Moor, J. Vandewalle (2002) Least Squares Support Vector Machines, World Scientific, SingaporeGoogle Scholar
  22. [22]
    R. O. Duda, P. E. Hart, D. G. Stork (2001) Pattern Classification. John Wiley & Sons, 2nd editionGoogle Scholar
  23. [23]
    B. Scholkopf, A. J. Smola (2002) Learning with Kernels, MIT PressGoogle Scholar
  24. [24]
    S. Raudys (2001) Statistical and Neural Classifiers: An Integrated Approach to Design, Springer-VerlagGoogle Scholar

Copyright information

© Springer-Verlag/Wien 2005

Authors and Affiliations

  • Qingzhong Liu
    • 1
  • Andrew H. Sung
    • 1
  • Bernardete M. Ribeiro
    • 2
  1. 1.Department of Computer ScienceNew Mexico TechUSA
  2. 2.Department of Informatics EngineeringUniversity of CoimbraPortugal

Personalised recommendations