Using the qualitative characteristics of an image for comprehensive steganalysis

Aim. The problem of image steganalysis is especially relevant given the use of steganographical concealment in graphic files for delivering malicious code and information as part of cyber attacks. That requires improvements to the existing methods of detecting steganographically embedded information. One method is to use a comprehensive steganalysis technique that involves concluding on the detection of embedded information based on the findings of a group of steganalysis methods, as well as auxiliary calculations. Methods. It is proposed improving the accuracy of hidden information detection by using qualitative image estimation. The paper demonstrates the relationship between the estimates and the increased rate of steganalysis errors. The method of comprehensive steganalysis that involves accounting for the qualitative characteristics of images allows improving the accuracy of estimation by reducing the rate of false positives. The paper uses statistical methods for calculating the qualitative characteristics of images, Spearman correlation, and machine learning. Results. A software package has been developed that integrates elements of the comprehensive steganalysis method described in the paper that includes both a group of steganalysis methods, and a set of evaluated qualitative characteristics of an image. The author evaluates the relationship between the qualitative characteristics of an image and the steganalysis errors in the case of empty containers. Test samples have been defined and machine learning models have been built that generate a conclusion as regards the detection of hidden information in an image. Conclusion. The proposed method enables improved accuracy of hidden information detection, while taking into account the estimates of the qualitative characteristics of an image as part of steganalysis, which is confirmed experimentally.

1. Grachev Y.L., Sidorenko V.G. Tasks of stegoanalysis automation. In: Proceedings of the II International Science and Practice Conference Intelligent Transportation Systems (May 25, 2023). Moscow: RUT MIIT; 2023. Pp. 450-457. (in Russ.)

2. Konakhovich G.F., Puzyrenko A.Y. [Computer steganography. Theory and practice]. Kiev: MK-Press; 2006. (in Russ.)

3. Fridrich J. Steganography in Digital Media. Principles and Applications. Cambridge: Cambridge University Press; 2010.

4. Grachev Ya.L., Sidorenko V.G. [Using stegoanalysis to ensure the integrity of information in intelligent transport systems]. In: [Proceedings of the International Science and Practice Conference Intelligent Transportation Systems] (May 26, 2022). Moscow: RUT MIIT; 2023. Pp. 389-396. (in Russ.)

5. Grachev Ya.L., Sidorenko V.G. Steganalysis of the methods of concealing information in graphic containers. Dependability 2021;21(3):39-46.

6. Westfeld A., Pfitzmann A. Attacks on Steganographic Systems. Dresden University of Technology, Department of Computer Science. Dresden (Germany); 1999. DOI: 10.1007/10719724_5.

7. Fridrich J., Goljan M., Du R. Reliable Detection of LSB Steganography in Color and Grayscale Images. Binghamton University, New York (USA); 2001. DOI: 10.1145/1232454.1232466.

8. Belim S.V., Vilkhovsky D.E. Koch-Zhao algoritm stganalysis. Mathematical Structures and Modeling 2018;4(48):113- 119. DOI: 10.25513/2222-8772.2018.4.139-119. (in Russ.)

9. Monich Yu.I., Starovoitov V.V. [Quality assessments for the analysis of digital images]. Iskusstvenny intellekt 2008;4:376-386. (in Russ.)

10. Grachev Ya.L., Sidorenko V.G. [Using statistical characteristics of an image to improve the results of steganalysis]. In: [Proceedings of the III International Science and Practice Conference Intelligent Transportation Systems] (May 30, 2024). Pp. 563-569. (in Russ.)

11. Novikov A.I., Pronkin A.V. Methods for image noise level estimation. Computer Optics 2021;45(5): 713-720. (in Russ.)

12. Wang X., Tian B., Liang C. et al. Blind Image Quality Assessment for Measuring Image Blur. In: Proceedings of 2008 Congress on Image and Signal Processing (27-30 May 2008) 2008;1:467-470.

13. Monich Y.I., Starovoitov V.V. Measure of digital image blur evaluation. Doklady BGUIR 2011;1(55):80–84. (in Russ.)

14. Golub Yu.I., Starovoitov F.V. Study of local assessments of contrast for digital images. Sistemny analiz 2019;2:4-11. (in Russ.)

15. Timoshenko A.V., Koshkarov A.S. Comparative analysis of entropic metrics of space objects optical images informativity. Trudy MAI 2020;112:1-18. (in Russ.)

16. Kendall M.G., Stuart A. The advanced theory of statistics. London: Griffin; 1969.

17. Chaddock R.E. Principles and methods of statistics. Boston: Houghton Mifflin Company; 1925.