IMAGE IMPROVEMENT USING DEEP LEARNING: ANALYSIS AND PROSPECTIVE DEVELOPMENTS

I. V. Chernyadev, T. V. Yermolenko

About the authors

I. V. Chernyadev, T. V. Yermolenko

Чернядьев Иван Валерьевич – инженер-исследователь ФГБНУ "Институт проблем и скусственного интеллекта", бакалавр физико -технического факультета ФГБОУ ВО "ДонГУ". Область научных интересов: компьютерное зрение, машинное обучение, нейронные сети.

Ермоленко Татьяна Владимировна – к.т.н. научный сотрудник отдела системного анализа и интеллектуальных интерфейсов ФГБНУ "Институт проблем искусственного интеллекта", кандидат технических наук, доцент. Область научных интересов: машинное обучение, обработка естественного языка, распознавание образов, интеллектуальный анализ данных.

Download PDF

UDC: 004.932.4
DOI: 10.24412/2413-7383-2025-4-39-254-265
Language: Russian

Abstract: The article discusses the current problem of improving imag e quality using deep learning methods, with a focus on super-resolution, noise reduction, and artifact removal. It analyzes key architectures based on transformers and convolutional neural networks, their advantages and disadvantages, such as high resource consumption and narrow specialization. The article proposes improvements in the form of randomized data augmentation for synthesizing the training set and replacing the standard convolution with Fast Fourier Convolution to capture global context. The goal of the work is to improve the quality of image restoration without high computational costs.

Keywords: super-resolution, neural networks, computer vision, deep learning, image processing.

References

1. Sai S.V. A method for assessing photorealistic image quality with high resolution [Text] // Computer Optics. – 2022. – Vol. 46, No. 1. – P. 121-129. – ISSN 2412-6179. – DOI: 10.18287/2412-6179-CO-899.
2. Konovalov, V.F. Unified Neural Network Method for Enhancing Resolution for Heterogeneous Digital Images of Earth Remote Sensing / V.F. Konovalov, V.V. Myasnikov, and V.V. Sergeev // Computer Optics. - 2024. - Vol. 48, No. 6. - Pp. 944-955. - DOI: I0.18287/2412-6179-C0-I610.
3. Singh P. Ganotra D. Brain MRT image super resolution using discrete cosine transform and convolutional neural network. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2023, vol. 23, no. 4, pp. 734-742. doi: 10.17586/2226-1494-2023-23-4-734-742.
4. Sai S.V. Kudyashov A.A. Assessment of the quality of natural landscape images with increased resolution based on GAN [Text] // Bulletin of TOGU. – 2024. – No. 1 (72). – P. 45-52. – UDC 004.896.
5. Shelkovnikova TE, Egorov SF, Gulyaev PV. Application of generative adversarial neural networks for the formation of databases in scanning tunneling microscopy. Computer Optics 2023; 47(2): 314-322. DOI: 10.18287/2412-6179-CO-1144.
6. Yang F. Yang H. Fu J. Lu H. Guo B. Learning Texture Transformer Network for Image Super- Resolution / F. Yang, H. Yang, J. Fu, H. Lu, B. Guo. – 2020. – P. 5790-5799. – DOI: 10.1109/CVPR42600.2020.00583.
7. Liu, Z. Swin Transformer: Hierarchical Vision Transformer using Shifted Windows [Text] / Z. Liu, Y. Lin, Y. Cao, H. Hu, Y. Wei, Z. Zhang, S. Lin, B. Guo // arxiv preprint. – 2021. – arxiv:2103.14030.
8. Liang, J. SwinIR: Image Restoration Using Swin Transformer [Text] / J. Liang, J. Cao, G. Sun, K. Zhang, L. Van Gool, R. Timofte // arxiv preprint. – 2021. – arxiv:2108.10257.
9. Berezhnov NI, Sirota AA. Improving attention mechanisms in transformer architecture in image restoration. Computer Optics 2024; 48(5): 726-733. DOI: 10.18287/2412-6179-CO-1393.
10. Yermolenko T.V. Samorodsky I.Ye. Analysis of the efficiency of deep neural networks architectures for the classification of products images [Text]//Yermolenko T.V. Samorodsky I.Ye. Journal "Problems of Artificial Intelligence". – 2022. – № 1(24). – С. 54-64. – ISSN 2413-7383.
11. Denisov A.K. Bykovskii S.V. Kustarev P.V. Reference-based diffusion model for super-resolution. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2025, vol. 25, no. 2, pp. 321-327 (in Russian). doi: 10.17586/2226-1494-2025-25-2-321-327.
12. Yoo, J.S. Kim, M. & Kim, S.J. (2020). Rethinking Data Augmentation for Image Super-Resolution. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 13-19, 2020, Seattle, WA, USA, pp. 10735-10744. DOI: 10.1109/CVPR42600.2020.01075.
13. Chaganova O.B. Grigoryev A.S. Nikolaev D.P. Nikolaev I.P. Applied aspects of modern non-blind image deconvolution methods [Text] // Computer Optics. – 2024. – Vol. 48, no. 4. – P. 562–572. – ISSN 2412-6179. – DOI: 10.18287/2412-6179-CO-1409.
14. Zhang X. Wu X. Dual-layer Image Compression via Adaptive Downsampling and Spatially Varying Upconversion / X. Zhang, X. Wu. – 2023. – P. 1-10. – DOI: arXiv:2302.06096. Ч Чернядьев И. В. Ермоленко Т. В.
15. Kotenko, I. V. A Method for Countering Competitive Attacks on Image Classification Systems [Text] / I. V. Kotenko, I. B. Saenko, O. S. Lauta, N. A. Vasilyev, and V. E. Sadovnikov // Cybersecurity Issues. – 2025. – No. 2(66). – Pp. 114-123. – ISSN 2311-3456. – DOI: 10.21681/2311-3456-2025-2-114-123.
16. Pikalyov Ya.S. Yermolenko T.V. About Neural Architectures of Feature Extraction for the Problem of Object Recognition on Devices with Limited Computing Power [Text] // Journal "Problems of Artificial Intelligence". – 2023. – № 3. – С. 44-54. – ISSN 2413-7383. – DOI 10.34757/2413-7383.2023.30.3.004.
17. Chernyadev, I. V. System Analysis of the Software of the Conceptual Neural Network Model of the Emotion Classification System in Images [Text] / I. V. Chernyadyev, V. V. Bondarchuk // Eurasian Union of Scientists. Series: Technical and Physical and Mathematical Sciences. – 2024. – No. 08(123). – Pp. 6-
18. – ISSN 2413-9335. – DOI: 10.31618/ESU.2413-9335.
19. Rao, Y. Global Filter Networks for Image Classification [Text] / Y. Rao, W. Zhao, Z. Zhu, J. Lu, J. Zhou // arxiv preprint. – 2021. – arxiv:2107.00645.
20. Quattrini, F. Volumetric Fast Fourier Convolution for Detecting Ink on the Carbonized Herculaneum Papyri [Text] / F. Quattrini, V. Pippi, S. Cascianelli, R. Cucchiara // arxiv preprint. – 2023. – arxiv:2308.05070.
21. Gu, L. Fast Fourier Convolution Based Remote Sensor Image Object Detection for Earth Observation [Text] / L. Gu, R. Qiao, X. Chen // arxiv preprint. – 2022. – arxiv:2209.00551.
22. Miao W. Shen J. Xu Q. Hamalainen T. Xu Y. Cong F. SpikingYOLOX: Improved YOLOX Object

Issue: 4(39)'2025
Section: SYSTEM ANALYSIS, CONTROL AND INFORMATION PROCESSING, STATISTICS
How to cite: I. V. Chernyadev, T. V. Yermolenko. IMAGE IMPROVEMENT USING DEEP LEARNING: ANALYSIS AND PROSPECTIVE DEVELOPMENTS // Problems of Artificial Intelligence. - 2025. - № 4 (39). - P. 254-265. - https://paijournal.guiaidn.ru/en/2025/4(39)-22.html

Back to issue 4(39)'2025