Advertisement

Ensemble Steganalysis Based on Deep Residual Network

  • Qiangjie Li
  • Guorui FengEmail author
  • Hanzhou Wu
  • Xinpeng Zhang
Conference paper
  • 67 Downloads
Part of the Lecture Notes in Computer Science book series (LNCS, volume 12022)

Abstract

The performance of the steganography detector built on deep learning has been superior to the traditional feature-based methods, and more adaptive methods for steganalysis are beginning to emerge. However a single model may encounter a bottleneck in classification accuracy due to the absent diversity of training data and parameter configuration, it maybe fails to exert a strong fitting performance of the deep learning network. To make full use of the classification performance of the combination of multiple models, we first obtained multiple base learners from different snapshot and different training sets. Then two strategies to combine multiple base learners: one achieves the optimal ensemble effect by majority voting and product combination, another, in view of the insufficient performance of Softmax classifier, propose a scheme of feature extraction based on convolutional neural network. Experiments show that the ensemble scheme proposed can well fuse the output of multiple convolutional neural networks, thus effectively reduce the detection error rate of a single model.

Keywords

Steganalysis Ensemble Feature fusion Convolutional neural networks 

Notes

Acknowledgement

This work was partly supported by the Natural Science Foundation of Shanghai under Grant 19ZR1419000 and National Natural Science Foundation of China under Grants 61902235, U1936214, U1636206, 61525203.

References

  1. 1.
    Boroumand, M., Chen, M., Fridrich, J.: Deep residual network for steganalysis of digital images. IEEE Trans. Inf. Forensics Secur. 14(5), 1181–1193 (2019) CrossRefGoogle Scholar
  2. 2.
    Chen, M., Sedighi, V., Boroumand, M., Fridrich, J.: JPEG-phase-aware convolutional neural network for steganalysis of JPEG images. In: ACM Workshop on Information Hiding & Multimedia Security (2017)Google Scholar
  3. 3.
    Feng, G., Zhang, X., Ren, Y., Qian, Z., Li, S.: Diversity-based cascade filters for jpeg steganalysis. IEEE Trans. Circ. Syst. Video Technol. (2019) Google Scholar
  4. 4.
    Fridrich, J., Kodovsky, J.: Rich models for steganalysis of digital images. IEEE Trans. Inf. Forensics Secur. 7(3), 868–882 (2012)CrossRefGoogle Scholar
  5. 5.
    Gao, H., Li, Y., Pleiss, G., Zhuang, L., Hopcroft, J.E., Weinberger, K.Q.: Snapshot ensembles: Train 1, get m for free (2017)Google Scholar
  6. 6.
    He, K., Zhang, X., Ren, S., Sun, J.: Delving deep into rectifiers: surpassing human-level performance on imagenet classification. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1026–1034 (2015)Google Scholar
  7. 7.
    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778. IEEE, Las Vegas, June 2016Google Scholar
  8. 8.
    Holub, V., Fridrich, J.: Designing steganographic distortion using directional filters. In: IEEE International Workshop on Information Forensics & Security (2012)Google Scholar
  9. 9.
    Holub, V., Fridrich, J.: Digital image steganography using universal distortion. In: Proceedings of the First ACM Workshop on Information Hiding and Multimedia Security - IH&MMSec 2013, p. 59. ACM Press, Montpellier (2013)Google Scholar
  10. 10.
    Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift. In: International Conference on International Conference on Machine Learning (2015)Google Scholar
  11. 11.
    Kodovsky, J., Fridrich, J., Holub, V.: On dangers of overtraining steganography to incomplete cover model. In: Proceedings of the Thirteenth ACM Multimedia Workshop on Multimedia and Security, pp. 69–76. ACM (2011)Google Scholar
  12. 12.
    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
  13. 13.
    Loshchilov, I., Hutter, F.: SGDR: Stochastic gradient descent with restarts (2017)Google Scholar
  14. 14.
    Ni, D., Feng, G., Shen, L., Zhang, X.: Selective ensemble classification of image steganalysis via deep Q network. IEEE Signal Process. Lett. 26(7), 1065–1069 (2019)CrossRefGoogle Scholar
  15. 15.
    Ni, J., Jian, Y., Yang, Y.I.: Deep learning hierarchical representations for image steganalysis. IEEE Trans. Inf. Forensics Secur. 12(11), 1 (2017)CrossRefGoogle Scholar
  16. 16.
    Pevny, T., Bas, P., Fridrich, J.: Steganalysis by subtractive pixel adjacency matrix. IEEE Trans. Inf. Forensics Secur. 5(2), 215–224 (2010)CrossRefGoogle Scholar
  17. 17.
    Pevny, T., Fridrich, J.: Merging Markov and DCT features for multi-class JPEG steganalysis. In: Security, Steganography, and Watermarking of Multimedia Contents IX (2007)Google Scholar
  18. 18.
    Qian, Y., Dong, J., Wang, W., Tan, T.: Deep learning for steganalysis via convolutional neural networks. In: Media Watermarking, Security, and Forensics 2015, vol. 9409, p. 94090J. International Society for Optics and Photonics (2015)Google Scholar
  19. 19.
    Song, X., Liu, F., Yang, C., Luo, X., Yi, Z.: Steganalysis of adaptive JPEG steganography using 2D Gabor filters. In: ACM Workshop on Information Hiding & Multimedia Security (2015)Google Scholar
  20. 20.
    Xu, G.: Deep convolutional neural network to detect J-UNIWARD. In: ACM Workshop on Information Hiding & Multimedia Security (2017)Google Scholar
  21. 21.
    Xu, G., Wu, H.Z., Shi, Y.Q.: Ensemble of CNNS for steganalysis: an empirical study. In: ACM Workshop on Information Hiding & Multimedia Security (2016)Google Scholar
  22. 22.
    Xu, G., Wu, H.Z., Shi, Y.Q.: Structural design of convolutional neural networks for steganalysis. IEEE Signal Process. Lett. 23(5), 708–712 (2016)CrossRefGoogle Scholar
  23. 23.
    Zhong, K., Feng, G., Shen, L., Luo, J.: Deep learning for steganalysis based on filter diversity selection. Sci. China Inf. Sci. 61(12), 129105 (2018)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Qiangjie Li
    • 1
  • Guorui Feng
    • 1
    Email author
  • Hanzhou Wu
    • 1
  • Xinpeng Zhang
    • 1
  1. 1.School of Communication and Information EngineeringShanghai UniversityShanghaiChina

Personalised recommendations