Currently, various augmentation methods, i.e. artificial multiplication of original data based on various transformations to increase sampling in machine learning, are being actively developed. The paper investigates the change in the structure of images after transformations. Images from satellite images are considered as the initial data. The structure of the original images is compared with the sets of images formed by different filters. Numerical results of image comparison based on topological properties of objects and structural similarity index are shown.

Sergey V. Eremeev
Candidate of Technical Sciences, Senior Lecturer
Associate Professor Department of Information Technology
Murom Institute (branch), Vladimir State University named after Alexander and Nikolay Stoletovs
23, Orlovskaya str., Murom, Vladimirskaya reg., 602264, Russia
e-mail: sv-eremeev@yandex.ru
ORCID: 0000-0001-8482-1479
SPIN-код: 3020-9020
AuthorID: 618264

Artyom V. Abakumov
Graduate Student Department of Information Technology
Murom Institute (branch), Vladimir State University named after Alexander and Nikolay Stoletovs
23, Orlovskaya str., Murom, Vladimirskaya reg., 602264, Russia
e-mail: artem210966@yandex.ru
ORCID: 0000-0001-5784-7147
SPIN-код: 3267-3100
AuthorID: 1082668

Denis A. Pankratov
Graduate Student Department of Information Technology
Murom Institute (branch), Vladimir State University named after Alexander and Nikolay Stoletovs
23, Orlovskaya str., Murom, Vladimirskaya reg., 602264, Russia
e-mail: denis_pankratov2000@mail.ru

Bogdan A. Khavronin
Student Department of Information Technology
Murom Institute (branch), Vladimir State University named after Alexander and Nikolay Stoletovs
23, Orlovskaya str., Murom, Vladimirskaya reg., 602264, Russia
e-mail: ev3nt.official@gmail.com

1. Sirota A., Akimov A., Otyrba R. Image Warping and Its Application for Data Augmentation when Training Deep Neural Networks. Informatics and Automation. 2024;23(2):407–435. DOI: 10.15622/ia.23.2.4.
2. Emelyanov S.O., Ivanova A.A., Shvets E.A., Nikolaev D.P. Methods of training data augmentation in the task of image classification. Sensory Systems. 2018;32(3):236–245. DOI: 10.1134/S0235009218030058.
3. Chupikova S.А., Chuldum A.F. Geoinformation and remote methods for assessing snow cover pollution (Kyzyl). Geoinformatika. 2023;(1):63–70. DOI: 10.47148/1609-364X-2023-1-63-70.
4. Poryvai M.V. A Comparative Study for Natural Image Augmentation Methods. International Journal of Open Information Technologies. 2024;12(10):26–33.
5. Alomar K., Aysel H.I., Cai X. Data Augmentation in Classification and Segmentation: A Survey and New Strategies. Journal of Imaging. 2023;9(2):46. DOI: 10.3390/jimaging9020046.
6. Xu M., Yoon S., Fuentes A., Park D.S. A Comprehensive Survey of Image Augmentation Techniques for Deep Learning. Pattern Recognition. 2023;137:109347. DOI: 10.1016/j.patcog.2023.109347.
7. Eremeev S.V., Abakumov A.V., Andrianov D.E., Titov D.V. Image decomposition method by topological features. Computer Optics. 2022;46(6):939–947. DOI: 10.18287/2412-6179-CO-1080.
8. Eremeev S.V. Abakumov A.V., Krainov S.A., Kozlov A.S. Using the method of image decomposition based on topological features for processing satellite images. Geoinformatika. 2023;(4):74–80. DOI: 10.47148/1609-364X-2023-4-74-80.
9. Eremeev S.V., Abakumov A.V., Andrianov D.E. Bogoyavlensky I.V., Nikonov R.A. Detection of perennial heaving mounds in digital elevation model images using decomposition by topological features. Sovremennye problemy distantsionnogo zondirovaniya zemli iz kosmosa. 2023;20(6):129–143. DOI: 10.21046/2070-7401-2023-20-6-129-143.
10. EuroSAT: Land Use and Land Cover Classification with Sentinel-2. Available at: https://github.com/phelber/EuroSAT (accessed 10.11.2024).
11. Satellite Image Segmentation Using PyTorch. Available at: https://github.com/obravo7/satellite-segmentation-pytorch (accessed 10.11.2024).
12. Simard P.Y., Steinkraus D., Platt J.C. Best practices for convolutional neural networks applied to visual document analysis. In: Seventh International Conference on Document Analysis and Recognition, 2003: Proceedings (Edinburgh, 06 August 2003). 2003;3:958–963. DOI: 10.1109/ICDAR.2003.1227801.