Performance Analysis Of Stego Image Calibration With Usage Of Denoising Autoencoders

: pp. 46 - 54
Igor Sikorsky Kyiv Polytechnic Institute

Methods for early detection of sensitive information leakage by data transmission in open (public) communication systems have been of special interest. Reliable detection of modified (stego) cover files, like digital images, requires usage of computation-intensive methods of statistical steganalysis, namely covering rich models and deep convolutional neural networks. Necessity of fine- tuning parameters of such methods to minimize detection accuracy for each embedding methods has made fast re- train of stegdetectors in real cases impossible. Therefore, development of low-complexity methods for detection of weak alterations of cover image parameters under limited prior information about used embedding methods has been required. For solving this task, we have proposed to use special architectures of artificial neural networks, such as denoising autoencoder. Ability of such networks to estimate parameters of original (cover) image from the noisy ones under limited prior information about introduced alterations has made them an attractive alternative to state- of-the-art solutions. The results of performance evaluation for shallow denoising autoencoders showed increasing of detection accuracy (up to 0.1 for Matthews correlation coefficient) in comparison with the state-of-the-art stegdetectors by preserving low-computation complexity of network retraining.

  1. J.-P. A. Yaacoub, O. Salman, H. N. Noura, N. Kaaniche, A. Chehab, M. Malli. “Cyber-physical systems security: Limitations, issues and future trends”, Microprocessors and Microsystems, vol. 77, 2020. [Online]. DOI: 10.1016/j.micpro.2020.103201
  2. D. Legezo. “MontysThree: Industrial espionage with steganography and a Russian accent on both sides”, SecureList. Available at:: espionage/98972/  (Accessed  2022-March-30).
  3. V. Kopeytsev. “Steganograph in attacks on industrial enterprises”. Kaspersky Inc., Tech. Rep, 2020. [Online] Available at: SKY_Steganography_in_targeted_attacks_EN.pdf (Accessed: 10 November 2021)
  4. J. Fridrich, J. Kodovsky. “Rich models for steganalysis  of digital images”, IEEE Transactions on Information Forensics and Security, vol. 7, iss. 3, 2012, pp. 868-882, DOI 10.1109/TIFS.2012.2190402.
  5. M. Boroumand, M. Chen, J. Fridrich. “Deep Residual Network for Steganalysis of Digital Images”, IEEE Transactions on Information Forensics and Security, vol. 14, iss. 5, 2018, pp. 1181-1193. DOI: 10.1109/TIFS.2018.2871749.
  6. J. Fridrich. Steganography in Digital Media: Principles, Algorithms, and Applications. Cambridge: Cambridge University Press, 2009, 437 pages, ISBN 978-0-521-19019-0, DOI:  10.1017/CBO9781139192903.
  7. G. Konachovych, D. Progonov, O. Puzyrenko. Digital steganography processing and analysis of multimedia files. Kyiv, ‘Tsentr uchbovoi literatury’ publishing, 2018, 558 pages, ISBN 978-617-673-741-4, Available at: (Accessed: 17 November 2021).
  8. J. Kodovsky, J. Fridrich. “Calibration revisited”, in Multimedia and security: 11th ACM workshop, Princeton, 2009, pp. 63-74, DOI:  10.1145/1597817.1597830.
  9. R. Zhang, F. Zhu, J. Liu, and G. Liu, ‘‘Efficient feature learning and multisize image steganalysis based on CNN,’’ Jul. 2018, arXiv:1807.11428.                                                 [Online].                    Available: (Accessed: 10 November 2021)
  10. J. Butora, Y. Yousfi, J. Fridrich. “How to Pretrain for Steganalysis”, in ACM Workshop on Information Hiding and Multimedia Security, Brussels, Belgium, 2021, pp. 143-148, DOI:  10.1145/3437880.3460395.
  11. D. Progonov. “Influence of digital images preliminary noising on statistical stegdetectors performance”, Radio Electronics, Computer Science, Control, vol.  1(56), pp.  184-193, 2021, DOI:  10.15588/1607-3274-2021-1-18.
  12. D. Progonov. “Detection Of Stego Images With Adaptively Embedded Data By Component Analysis Methods”, Advances in Cyber-Physical Systems, Vol. 6, Number 2, pp. 146-154, 2021, DOI: 10.23939/acps2021.02.146.
  13. A. Cohenab, A. Cohena, N. Nissim. “ASSAF: Advanced and Slim StegAnalysis Detection Framework for JPEG images based on deep convolutional denoising autoencoder and Siamese networks”, Neural Networks, vol. 131, pp. 64-77, Nov. 2020. [Online]. DOI: 10.1016/j.neunet.2020.07.022
  14. T. Filler, J. Fridrich. “Gibbs construction in steganography”, IEEE Transactions on Information Forensics Security, vol. 5, 2010, pp. 705-720, DOI: 10.1109/TIFS.2010.2077629.
  15. T. Filler, J. Fridrich. “Design of adaptive steganographic schemes for digital images”, in Electronic Imaging, Media Watermarking, Security, and Forensics: The International Society for Optical Engineering, San Francisco, CA, 2011, DOI:  10.1117/12.872192.
  16. T. Denemark, V. Sedighi, V. Holub, R. Cogranne, J. Fridrich. “Selection-Channel-Aware Rich Model for Steganalysis of Digital Images”, in IEEE Workshop on Information Forensic and Security, Atlanta, USA, 2014, DOI 10.1109/WIFS.2014.7084302.
  17. V. Holub, J. Fridrich, T. Denemark. “Universal Distortion Function   for   Steganography   in   an   Arbitrary   Domain”,EURASIP Journal on Information Security, Vol. 1, 2014, DOI: 10.1186/1687-417X-2014-1.
  18.  V. Sedighi, J. Fridrich, R. Cogranne. “Content-adaptive pentary steganography using the multivariate generalized gaussian cover model”, in Electronic Imaging, Media Watermarking, Security, and Forensics: The International Society for Optical Engineering, San Francisco, CA, 2015, DOI: 10.1117/12.2080272.
  19. V. Sedighi, R. Cogranne, J. Fridrich. “Content adaptive steganography by minimizing statistical detectability”, IEEE Transactions on Information Forensics Security, vol. 11, 2015, pp. 221-234, DOI: 10.1109/TIFS.2015.2486744.
  20. Stan Z. Li. Markov Random Field Modeling in Image Analysis. In Advances in Computer Vision and Pattern Recognition Series, Springer, 2009, 362 pages, ISBN 978-1-84800-278-4, Available at: 84800-279-1 (Accessed: 17 November 2021).
  21. R.  Gonzalez,  R.  Woods.  Digital  Image  Processing.  4th   ed. Pearson  Press,  2017.  1192  pages,  ISBN  978-0133356724, Available             at: ge%20Processing%203rd%20ed.%20-%20R.%20Gonzalez%2C%20R.%20Woods-ilovepdf- compressed.pdf (Accessed: 17 November 2021).
  22. C. Tomasi, R. Manduchi. “Bilateral Filtering for Gray and Color Images.” IEEE International Conference on Computer Vision, 1998, pp. 839-846. DOI:10.1109/ICCV.1998.710815
  23. Jae S. Lim. Two-Dimensional Signal and Image Processing, Englewood Cliffs, NJ, Prentice Hall, 1990, p. 548. ISBN: 978- 0139353222
  24. A. Buades. A non-local algorithm for image denoising. Computer Vision and Pattern Recognition, 2005. 2. pp. 60–65. DOI:10.1109/CVPR.2005.38.
  25. I. Goodfellow, Y. Bengio, A. Courville. Deep Learning, Cambridge:  The  MIT  Press,  2016,  p.  800.  ISBN:  978-0262035613.
  26. D. Progonov. “Multi-Datasets Evaluation Of GB-Ras Network Based Stegdetectors Robustness To Domain Adaptation Problem”. Information Theories & Applications. Volume 28, Number 4, 2021. pp. 372-396.
  27. D. Progonov, M. Yarysh. “Analyzing The Accuracy Of Detecting Steganograms Formed By Adaptive Steganographic Methods When Using Artificial Neural Networks”, Eastern- European Journal of Enterprise Technologies, Vol. 1, Issue 9 (115),   pp.45-55,      2022,      DOI:      10.15587/1729-4061.2022.251350.
  28. R. Cogranne, Q. Gilboulot, P. Bas. “The alaska steganalysis challenge: A first step towards steganalysis”, in Information Hiding and Multimedia Security, Paris, 2019, ACM Press, pp. 125-137, DOI: 10.1145/3335203.3335726.
  29. T. Pevny, P. Bas, J. Fridrich. “Steganalysis by subtractive pixel adjacency matrix”, IEEE Transactions on Information Forensics Security, vol. 5, 2010, pp. 215-224, DOI: 10.1109/TIFS.2010.2045842.
  30. J. Kodovsky, J. Fridrich. “Ensemble classifiers for steganalysis of digital media”, IEEE Transactions on Information Forensics Security, vol. 7, 2012, p. 432-444, DOI: 10.1109/TIFS.2011.2175919.
  31. D. Progonov. “Performance of Statistical Stegdetectors in Case of Small Number of Stego Images in Training Set”, in IEEE Problems of Infocommunications Science and Technology, Kharkiv,  2020,  DOI:10.1109/PICST51311.2020.9467901.
  32. D. Chicco, G. Jurman. “The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation”, BMC Genomics, vol. 21, 2020.  DOI: