EVALUATING THE IMPACT OF IMAGE SEMANTICS ON THE EFFECTIVENESS OF STEGANOGRAPHIC METHODS

DOI: 10.31673/2409-7292.2025.041208

Abstract

The article investigates the issue of increasing the efficiency of steganographic systems by analyzing the results of
information hiding when using different embedding algorithms. The aim of the work is to assess the impact of the steganographic
transformation method on the visual invisibility of the container and the resistance of hidden data to destructive influences. A
comparison of classical LSB modification algorithms with frequency methods based on the discrete cosine transform (DCT)
and wavelet transform (DWT), as well as with combined approaches that take into account the spatial-frequency structure of
the image, is carried out. A set of efficiency criteria is proposed, which includes objective indicators of visual quality (PSNR,
SSIM, MSE), normalized mean square error, as well as indicators of resistance to various types of attacks. The paper presents
the results of experimental studies of the efficiency of steganographic algorithms using sets of test images with different
semantic content. It is shown that the structural and content features of the image significantly affect the hiding efficiency:
images with a complex texture provide higher capacity and visual invisibility, while smooth homogeneous areas are more
sensitive to changes. Additionally, the resistance of the hidden information to JPEG compression attacks, impulse noise, filtering
and geometric transformations was assessed. The results of the study confirm that increasing the efficiency of the steganographic
system is possible due to the adaptive choice of the embedding location, which takes into account the type, semantics and statistical characteristics of the container image. The conclusions obtained can be used to develop new intelligent
steganosystems that automatically determine the optimal embedding parameters in order to achieve a balance between visual
invisibility and resistance to attacks.
Keywords: cyber threats, steganographic algorithm, information protection, neural networks, spatial and frequency
domain.

References
1. Fridrich, J. Steganography in Digital Media: Principles, Algorithms, and Applications. Cambridge: Cambridge
University Press, 2009. 437 p. DOI: 10.1017/CBO9780511762800.
2. Zhang, Y., Luo, X., Guo, Y., Qin, C., Liu, F. Multiple robustness enhancements for image adaptive
steganography in lossy channels // IEEE Transactions on Circuits and Systems for Video Technology. – 2020. – Vol. 30,
No. 8. pp. 2750–2764. DOI: 10.1109/TCSVT.2019.2939315.
3. Журавель, Ю. І., Мичуда, Л. З. Методи підвищення непомітності та стійкості стеганоповідомлень із
застосування фрактальних розмірностей // Сучасний захист інформації. 2025. № 1(61). С. 74–81. DOI: 10.31673/
2409-7292.2025.011576.
4. Журавель, І. М., Мичуда, Л. З., Журавель, Ю. І. Підвищення ефективності стеганографічного методу
приховування даних із застосуванням ітераційних функцій та додаванням шуму // Український журнал
інформаційних технологій. 2021. Т. 3, № 2. С. 66–73. Режим доступу: https://ujit.undip.org.ua/index.php/ujit/
article/view/182.
5. Hu, Y., Yang, Z., Cao, H., Huang, Y. Multi-modal steganography based on semantic relevancy // International
Workshop on Digital Watermarking. Cham: Springer International Publishing, 2020. pp. 3-14. DOI: 10.1007/978-3-030-
69325-6_1.
6. Zeng, L., Yang, N., Li, X., Chen, A., Jing, H., Zhang, J. Advanced image steganography using a U-Net-based
architecture with multi-scale fusion and perceptual loss // Electronics. 2023. Vol. 12, No. 18. Article 3808. DOI:
10.3390/electronics12183808.
7. Liu, L., Meng, L., Wang, X., Peng, Y. An image steganography scheme based on ResNet // Multimedia Tools
and Applications. 2022. Vol. 81, No. 27. pp. 39803-39820. DOI: 10.1007/s11042-022-13254-2.
8. Lu, H., Zhang, T. RobustStegFormer: A Cross-Attention Guided and Noise-Aware Transformer for Image
Steganography // 6th International Conference on Computer Vision, Image and Deep Learning (CVIDL). IEEE, 2025. p.
617-621. DOI: (уточнюється після публікації).
9. Hu, D., Wang, L., Jiang, W., Zheng, S., Li, B. A novel image steganography method via deep convolutional
generative adversarial networks // IEEE Access. 2018. Vol. 6. P. 38303–38314. DOI: 10.1109/ACCESS.2018.2852771.
10. Xie, G. Adaptive spatial image steganography and steganalysis using perceptual modelling and machine
learning: Dissertation for the degree of Doctor of Philosophy. Guildford: University of Surrey, 2022. DOI:
10.15126/thesis.2022.001.
11. Ponomarenko, N., Lukin, V., Zelensky, A., Egiazarian, K., Carli, M., Battisti, F. TID2013 Image quality
assessment database [Електронний ресурс]. 2013. Режим доступу: https://www.ponomarenko.info/tid2013.htm.