Steganalysis Based on Awareness of Selection-Channel and Deep Learning

  • Jianhua Yang
  • Kai Liu
  • Xiangui KangEmail author
  • Edward Wong
  • Yunqing Shi
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10431)


Recently, deep learning has been used in steganalysis based on convolutional neural networks (CNN). In this work, we propose a CNN architecture (the so-called maxCNN) to use the selection channel. It is the first time that the knowledge of the selection channel has been incorporated into CNN for steganalysis. The proposed method assigns large weights to features learned from complex texture regions while assigns small weights to features learned from smooth regions. Experimental results on the well-known dataset BOSSbase have demonstrated that the proposed scheme is able to improve detection performance, especially for low embedding payloads. The results have shown that with the ensemble of maxCNN and maxSRMd2+EC, the proposed method can obtain better performance compared with the reported state-of-the-art on detecting WOW embedding algorithm.


Adaptive steganography Steganalysis Convolutional neural networks (CNN) Selection-channel 



This work was supported by NSFC (Grant nos. U1536204, 61379155), Special funding for basic scientific research of Sun Yat-sen University (6177060230).


  1. 1.
    Filler, T., Judas, J., Fridrich, J.: Minimizing additive distortion in steganography using syndrome-trellis codes. IEEE Trans. Inf. Forensics Secur. 6(3), 920–935 (2011)CrossRefGoogle Scholar
  2. 2.
    Pevný, T., Filler, T., Bas, P.: Using high-dimensional image models to perform highly undetectable steganography. In: Böhme, R., Fong, P.W.L., Safavi-Naini, R. (eds.) IH 2010. LNCS, vol. 6387, pp. 161–177. Springer, Heidelberg (2010). doi: 10.1007/978-3-642-16435-4_13 CrossRefGoogle Scholar
  3. 3.
    Holub, V., Fridrich, J.: Designing steganographic distortion using directional filters. In: 2012 IEEE International Workshop on Information Forensics and Security (WIFS), pp. 234–239 (2012)Google Scholar
  4. 4.
    Holub, V., Fridrich, J., Denemark, T.: Universal distortion function for steganography in an arbitrary domain. EURASIP J. Inf. Secur. 2014(1), 1 (2014)CrossRefGoogle Scholar
  5. 5.
    Li, B., Wang, M., Huang, J., Li, X.: A new cost function for spatial image steganography. In: Proceedings of IEEE ICIP, Paris, France, pp. 4206–4210, October 2014Google Scholar
  6. 6.
    Sedighi, V., Cogranne, R., Fridrich, J.: Content-adaptive steganography by minimizing statistical detectability. IEEE Trans. Inf. Forensics Secur. 11(2), 221–234 (2016)CrossRefGoogle Scholar
  7. 7.
    Fridrich, J., Kodovsky, J.: Rich models for steganalysis of digital images. IEEE Trans. Inf. Forensics Secur. 7(3), 868–882 (2012)CrossRefGoogle Scholar
  8. 8.
    Tang, W., Li, H., Luo, W., Huang, J.: Adaptive steganalysis against WOW embedding algorithm. In: Proceedings of ACM IH&MMSec, pp. 91–96 (2014)Google Scholar
  9. 9.
    Tang, W., Li, H., Luo, W., Huang, J.: Adaptive steganalysis based on embedding probabilities of pixels. IEEE Trans. Inf. Forensics Secur. 11(4), 734–745 (2016)Google Scholar
  10. 10.
    Denemark, T., Sedighi, V., Holub, V., Cogranne, R., Fridrich, J.: Selection-channel-aware rich model for steganalysis of digital images. In: Proceedings of Information Forensics Security (WIFS), pp. 48–53, December 2014Google Scholar
  11. 11.
    Denemark, T., Boroumand, M., Fridrich, J.: Steganalysis features for content-adaptive JPEG steganography. IEEE Trans. Inf. Forensics Secur. 11(8), 1736–1746 (2016)CrossRefGoogle Scholar
  12. 12.
    Holub, V., Fridrich, J.: Low-complexity features for JPEG steganalysis using undecimated DCT. IEEE Trans. Inf. Forensics Secur. 10(2), 219–228 (2015)CrossRefGoogle Scholar
  13. 13.
    Holub, V., Fridrich, J.: Phase-aware projection model for steganalysis of JPEG images. In: Proceedings of SPIE, vol. 9409, p. 94090T, February 2015Google Scholar
  14. 14.
    Song, X., Liu, F., Yang, C., Luo, X., Zhang, Y.: Steganalysis of adaptive JPEG steganography using 2D Gabor filters. In: Proceedings of the 3rd ACM IH&MMSec. Workshop, pp. 15–23, June 2015Google Scholar
  15. 15.
    Tan, S., Li, B.: Stacked convolutional auto-encoders for steganalysis of digital images. In: Proceedings of APSIPA, pp. 1–4, December 2014Google Scholar
  16. 16.
    Qian, Y., Dong, J., Wang, W., Tan, T.: Deep learning for steganalysis via convolutional neural networks. In: Proceedings of SPIE Electronic Imaging, p. 94090J, March 2015Google Scholar
  17. 17.
    Qian, Y., Dong, J., Wang, W., Tan, T.: Learning and transferring representations for image steganalysis using convolutional neural network. In: 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, pp. 2752–2756, September 2016Google Scholar
  18. 18.
    Xu, G., Wu, H., Shi, Y.: Structural design of convolutional neural networks for steganalysis. IEEE Signal Process. Lett. 23(5), 708–712 (2016)CrossRefGoogle Scholar
  19. 19.
    Xu, G., Wu, H.-Z., Shi, Y.Q.: Ensemble of CNNs for Steganalysis: an empirical study. In: Proceedings of the 4th ACM Workshop on Information Hiding and Multimedia Security. ACM (2016)Google Scholar
  20. 20.
    Kodovsky, J., Fridrich, J., Holub, V.: Ensemble classifiers for steganalysis of digital media. IEEE Trans. Inf. Forensics Secur. 7(2), 432–444 (2012)CrossRefGoogle Scholar
  21. 21.
    Sedighi, V., Fridrich, J.: Histogram layer, moving convolutional neural networks towards feature-based steganalysis. In: IS&T/SPIE Electronic Imaging, Burlingame, California, 29 January - 2 February 2017Google Scholar
  22. 22.
    Holub, V., Fridrich, J.: Random projections of residuals for digital image steganalysis. IEEE Trans. Inf. Forensics Secur. 8(12), 1996–2006 (2013)CrossRefGoogle Scholar
  23. 23.
    Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift, February 2015. arXiv:1502.03167
  24. 24.
    Bas, P., Filler, T., Pevný, T.: “Break our steganographic system”: the ins and outs of organizing BOSS. In: Filler, T., Pevný, T., Craver, S., Ker, A. (eds.) IH 2011. LNCS, vol. 6958, pp. 59–70. Springer, Heidelberg (2011). doi: 10.1007/978-3-642-24178-9_5 CrossRefGoogle Scholar
  25. 25.
    Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., Guadarrama, S., Darrell, T.: Caffe: convolutional architecture for fast feature embedding. In: Proceedings of ACM International Conference Multimedia, pp. 675–678 (2014)Google Scholar
  26. 26.
    Liu, K., Yang, J., Kang, X.: Ensemble of CNN and rich model for steganalysis. In: 2017 IEEE International Conference on Systems, Signals and Image Processing (IWSSIP), pp. 1–5 (2017)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Jianhua Yang
    • 1
  • Kai Liu
    • 1
  • Xiangui Kang
    • 1
    Email author
  • Edward Wong
    • 2
  • Yunqing Shi
    • 3
  1. 1.Guangdong Key Lab of Information Security, Data and Computer ScienceSun Yat-Sen UniversityGuangzhouChina
  2. 2.Computer Science and EngineeringNew York UniversityNew YorkUSA
  3. 3.Electrical and Computer EngineeringNew Jersey Institute of TechnologyNewarkUSA

Personalised recommendations