# 22 134-140 View full article

УДК 528.88:004.032.26:004.93’1:631.147

DOI: https://doi.org/10.36887/2415-8453-2024-2-22

Vynohradenko Serhii,
PhD in Economics, Associate Professor of the Department Land Management, Geodesy and Cadastre,
State Biotechnological University,
https://orcid.org/0000-0002-8520-6504
Kulbaka Olesia,
PhD in Economics, Associate Professor of the Department of Highways, Geodesy and Land
Management,
Ukrainian State University of Science and Technologies,
https://orcid.org/0000-0002-6066-8112
Kulbaka Viktor,
PhD in Economics, Associate Professor of the Department of Highways, Geodesy and Land Management,
Ukrainian State University of Science and Technologies,
https://orcid.org/0000-0003-3634-4334
Hrek Mariia,
PhD in Technical Sciences, Assistant of the Department Land Management, Geodesy and Cadastre,
State Biotechnological University,
https://orcid.org/0000-0001-8243-8273

This study presents a detailed comparative analysis of the performance of advanced algorithms across three neural network models for mapping agricultural lands and classifying land cover based on multispectral satellite imagery from Sentinel-2 and Landsat-8, acquired for the Berestyn district of Kharkiv region. The primary objective was to determine the optimal methodology for automated processing and interpretation of remote sensing data for agricultural land mapping, with further applications in monitoring, environmental assessment, and land use planning. Within the study, land cover was classified into four categories: vegetation, agricultural lands, water bodies, and built-up areas. Three classification approaches were applied: a Deep Neural Network (DNN), a Convolutional Neural Network (CNN), and the Random Forest (RF) algorithm. The DNN combines mechanisms for extracting both spectral and spatial features, allowing for efficient consideration of local textural characteristics of objects as well as their spectral differences. Experimental results confirmed its superiority over the other methods: classification accuracy reached 95,90 % for Sentinel-2 and 90,90 % for Landsat-8. By comparison, CNN achieved 93,8 % and 87,5 %, while RF attained 90,3 % and 87,2 %, respectively. The advantage of DNN was particularly evident in the analysis of areas with complex mosaic land use structures, where traditional algorithms often misinterpret class boundaries. At the same time, certain limitations were identified: high computational complexity and substantial memory requirements make it less suitable for real-time analysis in resource-constrained or field conditions. The practical value of the obtained results lies in the potential to implement this methodology for automated monitoring of crop conditions, early detection of stress factors and plant diseases, assessment of crop growth dynamics, and control of land use changes under the influence of anthropogenic and natural factors. Future research will focus on optimizing the DNN architecture to reduce resource consumption without compromising accuracy, integrating additional data types, including hyperspectral and radar imagery, and developing hybrid solutions capable of combining the advantages of deep learning with classical image processing methods to enhance the versatility and adaptability of remote monitoring systems.

Keywords: mapping, Sentinel-2, Landsat-8, Google Earth Engine, neural networks, agricultural land, multispectral data.

Rеferences

  1. Colliander, A., et al. (2017). Spatial downscaling of SMAP soil moisture using MODIS land surface temperature and NDVI during SMAPVEX15. IEEE Geoscience and Remote Sensing Letters, vol. 14, no. 11, pp. 2107–2111. https://doi.org/10.1109/LGRS.2017.2753203.
  2. Zhang, Y., Cao, D., Yang, J., Lu, F., Wang, D., Liu, R., Zhang, H., Liu, D., Chen, Z., Lyu, H., Cai, W., Bao, S., & Qie, X. (2023). A parallax shift effect correction based on cloud top height for FY-4A lightning mapping imager (LMI). Remote Sensing, vol. 15, no. 19, 4856. https://doi.org/10.3390/rs15194856.
  3. Opara, V. M., Buzina, I. M., Khainus, D. D., Vynohradenko, S. O., & Kovalenko, L. M. (2020). Teoretychni y metodychni osnovy vykorystannia HIS-tekhnolohii ta stvorennia elektronnykh kart pry provedenni zemleustroiu [Theoretical and methodological foundations of using GIS technologies and creating electronic maps in land management]. Problemy bezperervnoi heohrafichnoi osvity i kartohrafii, no. 31, pp. 50–59. https://doi.org/10.26565/2075-1893-2020-31-06.
  4. Zhao, Q., Yu, L., Li, X., Peng, D., Zhang, Y., & Gong, P. (2021). Progress and trends in the application of Google Earth and Google Earth Engine. Remote Sensing, vol. 13, no. 18, 3778. https://doi.org/10.3390/rs13183778.
  5. Waldner, F., Diakogiannis, F. I., Batchelor, K. L., Ciccotosto-Camp, M. M., Cooper-Williams, E., Herrmann, C., Mata, G., & Toovey, A. F. (2021). Detect, consolidate, delineate: Scalable mapping of field boundaries using satellite images. Remote Sensing, vol. 13, no. 11, 2197. https://doi.org/10.3390/rs13112197.
  6. Teyou, D., & Kamdem, G. (2020). Deep neural networks for automatic extraction of features in time series satellite images. arXiv preprint. https://doi.org/10.48550/arXiv.2008.08432.
  7. Mohammadi, S., Belgiu, B., & Stein, A. (2023). Improvement in crop mapping from satellite image time series by effectively supervising deep neural networks. ISPRS Journal of Photogrammetry and Remote Sensing, vol. 198, pp. 272–283. https://doi.org/10.1016/j.isprsjprs.2023.03.007.
  8. Tzepkenlis, A., Marthoglou, K., & Grammalidis, N. (2023). Efficient deep semantic segmentation for land cover classification using Sentinel imagery. Remote Sensing, vol. 15, no. 8, 2027. https://doi.org/10.3390/rs15082027.
  9. Balraj, A. (2020). DeepGlobe land cover classification dataset. Kagglehub. Available at: https://www.kaggle.com/datasets/balraj98/deepglobe-land-cover-classification-dataset.
  10. Mahdi, A. S. (2022). The land use and land cover classification on the urban area. Iraqi Journal of Science, vol. 63, no. 10, pp. 4609–4619. https://doi.org/10.24996/ijs.2022.63.10.42.
  11. Shelestov, A., Lavreniuk, M., Kussul, N., Novikov, A., & Skakun, S. (2017). Exploring Google Earth Engine platform for big data processing: Classification of multi-temporal satellite imagery for crop mapping. Frontiers in Earth Science, vol. 5, 17.

The article was received 12.04.2024

Quote article, APA style

Vynohradenko Serhii, Kulbaka Olesia, Kulbaka Viktor, Hrek Mariia. 12.04.2024 . Mapping of agricultural land based on neural networks using sentinel-2 and landsat-8 data. The journal "Ukrainian Journal of Applied Economics and Technology". 2024 / #2. 134-140pp. https://doi.org/10.36887/2415-8453-2024-2-22

Quote article, MLA style

Vynohradenko Serhii, Kulbaka Olesia, Kulbaka Viktor, Hrek Mariia. "Mapping of agricultural land based on neural networks using sentinel-2 and landsat-8 data". The journal "Ukrainian Journal of Applied Economics and Technology". 12.04.2024 . https://doi.org/10.36887/2415-8453-2024-2-22

Quote article, translit