Advertisement

Weighted Stego-Image Steganalysis for JPEG Covers

  • Rainer Böhme
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5284)

Abstract

This paper contains two new results for the quantitative detector of LSB replacement steganography based on a weighted stego-image (WS). First, for spatial domain steganalysis, a variant of the WS method is known to be highly accurate only when cover images have never been subject to lossy compression. We propose a new variant of WS which increases the accuracy for JPEG pre-processed covers by one order of magnitude, thus leaving behind the best structural detectors which were known to be more robust on JPEG pre-compressed covers than WS. Second, we explain why WS-like estimators can also detect LSB replacement steganography in the transformed domain, and derive a reduced-form estimator for JSteg steganography which has equal or slightly better performance than the currently best JSteg detectors.

Keywords

Discrete Cosine Transformation Bias Correction Cover Image JPEG Compression Local Predictor 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Fridrich, J., Goljan, M.: On estimation of secret message length in LSB steganography in spatial domain. In: Delp, E.J., Wong, P.W. (eds.) Security, Steganography and Watermarking of Multimedia Contents VI (Proc. of SPIE), San Jose, CA (2004)Google Scholar
  2. 2.
    Ker, A.D., Böhme, R.: Revisiting weighted stego-image steganalysis. In: Delp, E.J., et al. (eds.) Security, Forensics, Steganography and Watermarking of Multimedia Contents X (Proc. of SPIE), San Jose, CA, vol. 6819 (2008)Google Scholar
  3. 3.
    Ker, A.D.: A general framework for the structural steganalysis of LSB replacement. In: Barni, M., Herrera-Joancomartí, J., Katzenbeisser, S., Pérez-González, F. (eds.) IH 2005. LNCS, vol. 3727, pp. 296–311. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  4. 4.
    Böhme, R., Ker, A.D.: A two-factor error model for quantitative steganalysis. In: Delp, E.J., Wong, P.W. (eds.) Security, Steganography and Watermarking of Multimedia Contents VIII (Proc. of SPIE), San Jose, CA, vol. 6072, pp. 59–74 (2006)Google Scholar
  5. 5.
    Upham, D.: JSteg, Version 4 (1992), http://zooid.org/~paul/crypto/jsteg/
  6. 6.
    Böhme, R.: Assessment of steganalytic methods using multiple regression models. In: Barni, M., Herrera-Joancomartí, J., Katzenbeisser, S., Pérez-González, F. (eds.) IH 2005. LNCS, vol. 3727, pp. 278–295. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  7. 7.
    Dumitrescu, S., Wu, X., Wang, Z.: Detection of LSB steganography via sample pair analysis. IEEE Trans. on Signal Processing 51, 1995–2007 (2003)CrossRefzbMATHGoogle Scholar
  8. 8.
    Fridrich, J., Goljan, M., Du, R.: Steganalysis based on JPEG compatibility. In: Tescher, A.G., Vasudev, B., Bove Jr., V.M. (eds.) Multimedia Systems and Applications IV (Proc. of SPIE), Denver, CO, pp. 275–280 (2001)Google Scholar
  9. 9.
    Fridrich, J.: Feature-based steganalysis for JPEG images and its implications for future design of steganographic schemes. In: Fridrich, J. (ed.) IH 2004. LNCS, vol. 3200, pp. 67–81. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  10. 10.
    Zhang, T., Ping, X.: A fast and effective steganalytic technique against JSteg-like algorithms. In: Proc. of ACM Symposium on Applied Computing, Melbourne, Florida, pp. 307–311. ACM Press, New York (2003)Google Scholar
  11. 11.
    Lee, K., Westfeld, A., Lee, S.: Category attack for LSB steganalysis of JPEG images. In: Shi, Y.Q., Jeon, B. (eds.) IWDW 2006. LNCS, vol. 4283, pp. 35–48. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  12. 12.
    Lee, K., Westfeld, A., Lee, S.: Generalised category attack – improving histogram-based attack on JPEG LSB embedding. In: Furon, T., Cayre, F., Doërr, G., Bas, P. (eds.) IH 2007. LNCS, vol. 4567, pp. 378–391. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  13. 13.
    Ker, A.D.: Derivation of error distribution in least squares steganalysis. IEEE Trans. on Information Forensics and Security 2, 140–148 (2007)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Rainer Böhme
    • 1
  1. 1.Institute of Systems ArchitectureTechnische Universität DresdenDresdenGermany

Personalised recommendations