Keywords
How to Cite
Abstract
This topic analyzes the application of deep learning algorithms for the automatic classification of land cover types using Sentinel-2 satellite imagery. The study examines the structures of convolutional neural networks (CNNs), encoder-decoder architectures (U-Net, SegNet), densely connected networks (DenseNet), and modern transformer-based models, as well as data preparation stages (atmospheric correction, normalization, spectral index calculation) and evaluation metrics (overall accuracy, Kappa coefficient, F1-score, IoU). Prospects for their application in the context of Uzbekistan—specifically for monitoring agricultural crops, assessing soil salinity levels in irrigated lands, and tracking landscape changes in the Aral Sea basin—are substantiated. The research results demonstrate the superiority of deep learning-based methods over traditional pixel-based classification techniques in terms of accuracy and scalability.
References
1. Ochilov Sh.Sh. Yer kadastri ishlarini samarali yuritishda monitoring kartalaridan foydalanishni takomillashtirish: geografiya fanlari bo‘yicha falsafa doktori (PhD) dissertatsiyasi avtoreferati. Toshkent: O‘zbekiston Milliy universiteti, 2026. 44 b.
2. Husanov Sh.A. Sun’iy intellekt texnologiyalari asosida tarmoq ma’lumotlarini tahlil qilish model va algoritmlari: texnika fanlari bo‘yicha falsafa doktori (PhD) dissertatsiyasi avtoreferati. Toshkent: Muhammad al-Xorazmiy nomidagi TATU, 2026. 50 b.
3. Sharipov Sh.M., Fedorko V.N., Safarova N.I., Rafiqov V.A. Geografiya (amaliy geografiya): 10-sinf uchun darslik. Toshkent: «O‘zbekiston milliy ensiklopediyasi» Davlat ilmiy nashriyoti, 2017. 176 b.
4. Maxsudov B.Yu. Geoaxborot tizimlari menejmenti: o‘quv-uslubiy majmua. Toshkent: Toshkent arxitektura-qurilish instituti, 2022. 96 b.
5. Tojimamatov I.N. Sun’iy neyron tarmoqlarini o‘rganishning nazariy va amaliy asoslari // Oriens: ilmiy-uslubiy jurnal. 2025. B. 191–203.
6. Murodullayev D.M. Sun’iy neyron tarmoqlarning boshqaruv tizimlarida qo‘llanilish istiqbollari // Oriens Renaissance: Innovative, educational, natural and social sciences. 2026. №6(5). B. 542–547.
7. Remelgado R., Zaitov Sh., Kenjabaev Sh., Stulina G., Sultanov M., Ibrakhimov M., Akhmedov M., Dukhovny V., Conrad C. A crop type dataset for consistent land cover classification in Central Asia // Scientific Data. 2020. Vol. 7. Article 250. P. 1–6.
8. Ben Hamida A., Benoit A., Lambert P., Klein L., Ben Amar C., Audebert N., Lefèvre S. Deep learning for semantic segmentation of remote sensing images with rich spectral content // arXiv preprint arXiv:1712.01600. 2017.
9. Ronneberger O., Fischer P., Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation // Medical Image Computing and Computer-Assisted Intervention (MICCAI), Lecture Notes in Computer Science. Springer, 2015. Vol. 9351. P. 234–241.
10. Ulmas P., Liiv I. Segmentation of Satellite Imagery using U-Net Models for Land Cover Classification // arXiv preprint arXiv:2003.02899. 2020.
11. Kroupi E., Kesa M., Navarro-Sánchez V.D., Saeed S., Pelloquin C., Alhaddad B., Moreno L., Soria-Frisch A., Ruffini G. Deep convolutional neural networks for land-cover classification with Sentinel-2 images // Journal of Applied Remote Sensing. 2019. Vol. 13, No. 2. Article 024525.
12. Kotaridis I., Lazaridou M. Semantic segmentation using a UNET architecture on Sentinel-2 data // The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 2022. Vol. XLIII-B3-2022. P. 119–126.
13. LeCun Y., Bengio Y., Hinton G. Deep learning // Nature. 2015. Vol. 521. P. 436–444.
14. European Space Agency (ESA). Sentinel-2 User Handbook. – Paris: ESA Standard Document, 2015. – 64 p.