C&SA | Contents

Volume 57 >>> № 3 MAY — JUNE 2021

-->

UDC 004.056; 004.415.24

N.V. Koshkina¹

¹ V.M. Glushkov Institute of Cybernetics, National Academy
of Sciences of Ukraine, Kyiv, Ukraine

J-UNIWARD STEGANOANALYSIS

Abstract. The author analyzes the problem of detecting adaptive steganography by the J-UNIWARD method by steganoanalytical systems based on machine learning. A comparative analysis of the accuracy has determined that statistical models of constructing characteristic vectors that are calculated in the spatial domain, such as GFR, PHARM and DCTR, are most sensitive to J-UNIWARD. Two ways to improve the accuracy of steganoanalysis based on these models are proposed: via the analysis of the most probable embedding locations and via the balanced vote on the three models. Significant degradation of the accuracy of steganoanalysis without preliminary classification of images according to their parameters is demonstrated. The obtained results can be used to generate efficient steganoanalysis systems for JPEG images.

Keywords: information security, steganography, J-UNIWARD, steganalysis, machine learning methods, detection accuracy.

FULL TEXT

REFERENCES

Holub V., Fridrich J., Denemark T. Universal distortion function for steganography in an arbitrary domain. EURASIP Journal on Information Security. 2014. N 1. P. 1–13. https://doi.org/10.1186/1687-417X-2014-1.

Filler T., Judas J., Fridrich J. Minimizing embedding impact in steganography using trellis-coded quantization. Proc. SPIE, Electronic Imaging, Media Forensics and Security II (27 Jan 2010, San Jose, USA). San Jose, USA, 2010. Vol. 7541. P. 1–14. https://doi.org/10.1117/12.838002.

Sidorenko V., Zyablov V. Decoding of convolutional codes using a syndrome trellis. IEEE Trans. on Information Theory. 1994. Vol. 40, N 5. P. 1663–1666. https://doi.org/10.1109/18.333887.

Chen C., Shi Y.Q. JPEG image steganalysis utilizing both intrablock and interblock correlations. IEEE ISCAS, Intern. Symp. on Circuits and Systems (18–21 May 2008, Seattle, USA). Seattle, USA: IEEE, 2008. P. 3029–3032. https://doi.org/10.1109/ISCAS.2008.4542096.

Kodovsky J., Fridrich J. Calibration revisited. Proc. of the 11th ACM Multimedia and Security Workshop (Sept 2009, Princeton, USA). New York: ACM, 2009. P. 63–74. https://doi.org/10.1145/1597817.1597830.

Liu Q. Steganalysis of DCT-embedding based adaptive steganography and YASS. Proc. of the 13th ACM Multimedia & Security Workshop (Sept 2011, Buffalo, USA). New York: ACM, 2011. P. 77–86. https://doi.org/10.1145/2037252.2037267.

Pevny T., Fridrich J. Merging Markov and DCT features for multiclass JPEG steganalysisю Proc. SPIE, Electronic Imaging, Security, Steganography, and Watermarking of Multimedia Contents IX (2 March 2007, San Jose, USA). San Jose, USA, 2007. Vol. 6505. P. 301–314. https://doi.org/ 10.1117/12.696774.

Kodovsky J., Fridrich J. Steganalysis in high dimensions: fusing classifiers built on random subspaces. 8th SPIE Electronic Imaging, Media, Watermarking, Security and Forensics (23–27 Jan 2011, San Francisco, USA). San Francisco, USA, 2011. Vol. 7880, P. 1–13. https://doi.org/10.1117/ 12.872279.

Song X., Liu F., Yang C., Luo X., Zhang Y. Steganalysis of adaptive JPEG steganography using 2D Gabor filters. Proc. of the 3rd ACM Workshop on Information Hiding and Multimedia Security. ACM (June 2015, Portland, USA). New York: ACM, 2015. P. 15–23. https://doi.org/10.1145/2756601.2756608.

Holub V., Fridrich J. Low complexity features for JPEG steganalysis using undecimated DCT. IEEE Trans. on Information Forensics and Security. 2015. Vol. 10, N 2. P. 219–228. https://doi.org/ 10.1109/TIFS.2014.2364918.

Holub V., Fridrich J. Phase-aware projection model for steganalysis of JPEG images. Proc. SPIE, Electronic Imaging, Media Watermarking, Security, and Forensics XVII (8–10 Feb 2015, San Francisco, USA). San Francisco, USA, 2015. Vol. 9409. https://doi.org/10.1117/12.2075239.

KodovskН J., Fridrich J., Holub V. Ensemble classifiers for steganalysis of digital media. IEEE Trans. on Information Forensics and Security. 2012. Vol. 7, N 2. P. 432–444. https://doi.org/ 10.1109/TIFS.2011.2175919.

Koshkina N.V. Comparison of efficiency of statistical models used for formation of feature vectors by JPEG images steganalysis. Theoretical and Applied Cybersecurity. 2020. Vol. 2, N 1. P. 22–28. https://doi.org/10.20535/tacs.2664-29132020.1.

Koshkina N.V. Investigation of the main components of JPEG-steganoanalysis systems based on machine learning. Information protection. 2020. Vol. 22, N 2. P. 97–108. https://doi.org/10.18372/2410 -7840.22.14801.

Song X., Liu F., Chen L., Yang C., Luo X. Optimal Gabor filters for steganalysis of content-adaptive JPEG steganography. KSII Trans. on Internet and Information Systems. 2017. Vol. 11, N 1. P. 552–569. https://doi.org/10.3837/tiis.2017.01.029.