Abstract
In recent years, research has been conducted in scientific institutions of the Ministry of Agriculture of the Russian Federation and the Russian Academy of Sciences on introducing new technologies for the use of aerospace information in agriculture. This article, using the example of Stavropol krai, considers the possibility of using cloud services such as Google Earth Engine (GEE) and Kaggle machine learning systems for mapping agricultural fields using deep learning methods based on remote sensing data. Median images of the Sentinel 2 space system for the 2022 growing season are used as data for the selection of training and validation samples. The total volume of the prepared training samples is 3998 images. One problem for researchers and manufacturers in the field of agriculture is a lack of centralized and verified sources of geospatial data. Deep learning methods are able to solve this problem by automating the task of digitizing the geometries of agricultural fields based on remote sensing data. One of the limitations in the widespread use of deep learning is its high demand for computing resources, which are not always available to a researcher or manufacturer in the field of agriculture. This paper describes the process of preparing the necessary data for working with a neural network, including correcting and obtaining satellite images using GEE, their standardization for training a neural network in Kaggle, and further use locally. A neural network of the U-net architecture is used as part of the study. The final classification quality is 97%. The threshold of division into classes according to the classification results is established empirically and amounts to 0.62. The proposed approach makes it possible to significantly reduce the requirements for the local use of PC computing power. All the most resource-intensive processes related to the processing of satellite images are performed in the GEE system, and the learning process is transferred to the resources of the Kaggle system. The proposed combination of cloud services and deep learning methods can contribute to a wider spread of the use of modern technologies in agricultural production and scientific research.
This is a preview of subscription content, log in via an institution to check access.
REFERENCES
Bagaev, S.M. and Medvedeva, E.V., Experimental assessment of the accuracy of multiclass segmentation of objects from satellite images based on a modified convolutional neural network U-net, Sovrem. Probl. Distantsionnogo Zondirovaniya Zemli Kosmosa, 2021, vol. 18, no. 6, pp. 35–45.
Ermolaev, N.R., Yudin, S.A., Belobrov, V.P., Dridiger, V.K., and Gadzhiumarov, R.G., No-till identification by crop residues on the soil surface using the multi-temporal integral index minNDTI, AgroEkoInfo, 2021, vol. 4, no. 46, pp. 1–14.
Kulintsev, V., Sistema zemledeliya novogo pokoleniya Stavropol’skogo kraya (New Generation Agricultural System of the Stavropol Territory), Stavropol: AGRUS Stavropol’skogo gos. agrarnogo univ., 2013.
Kuprichenkov, M.T., Antonova, T.N., Simbirev, N.F., and Tsygankov, A.S., Zemel’nye resursy Stavropol’ya i ikh plodorodie (Stavropol Land resources and Fertility), 2002. Kussul’, N.N., Lupyan, E.A., Shelestov, A.Yu., Skakun, S.V., Savorskii, V.P., and Tishchenko, Yu.G., Neural network method of monitoring flooded territories using radar satellite data, Issled. Zemli Kosmosa, 2008, pp. 29–35.
Linkov, S.A., Akinchin, A.V., Melent’ev, A.A., Chuprynina, N.S., and Kuznetsova, A.E., Application of GIS technologies in agriculture, Innovatsii APK: Probl. Perspekt., 2018, vol. 1, no. 17.
Pavlov, V.A., Khryashchev, V.V., Ostrovskaya, A.A., and Kokuitseva, T., Comparative analysis of neural network algorithms for segmentation of objects in satellite images, in Tsifrovaya obrabotka signalov i ee primenenie. Sbornik materialov konferentsii (Digital Signal Processing and Its Application: Proceedings of Conference), 2019, pp. 399–403.
Ronnenberg, O., Fisher, P., and Brox, T., U-Net: Convolutional networks for biomedical image segmentation, IEEE Access, 2021, vol. 9, pp. 16591–16603.
Sengupta, S., Basak, S., Saikia, P., Paul, S., Tsalavoutis, V., Atiah, F., Ravi, V., and Peters, A., A review of deep learning with special emphasis on architectures, applications and recent trends, Knowledge-Based Syst., 2020, vol. 194, p. 105596.
Shokin, Yu.I. and Potapov, V.P., GIS today: State, prospects, and solutions, Vychisl. Tekhnol., 2015, pp. 175–213.
Skvortsov, E.A., Nabokov, V.I., Nekrasov, K.V., Skvortsova, E.G., and Krotov, M.I., Application of artificial intelligence technologies in agriculture, Agrar. Vestn. Urala, 2019, vol. 08, no. 8, pp. 91–98.
Taravat, A., Wagner, M.P., Bonifacio, R., and Petit, D., Advanced fully convolutional networks for agricultural field boundary detection, Remote Sens., 2021, vol. 13, no. 4, pp. 1–12.
Tokarev, K.E., Rudenko, A.Yu., Kuzmin, V.A., and Chernyavskii, A.N., Theory and digital technologies of intellectual decision support for increasing the bio-productivity of agroecosystems based on neural network models, Izv. Nizhnevolzh. Agrouniv. Kompleksa, 2021, vol. 4, pp. 421–440.
Upreti, A., Machine learning application in GIS and remote sensing: An overview, Int. J. Multidiscip. Res. Growth Eval., 2022, vol. 3, no. 4, pp. 546–553.
Vedeshin, L.A. and Shapovalov, D.A., The first scientific and technical experiments in space Earth sciences (on the 60th anniversary of satellite images of the Earth from manned spacecraft), Izv., Atmos. Ocean. Phys., 2022, vol. 58, no. 12, pp. 1689–1692.
Zhang, H., Liu, M., Wang, Y., Shang, J., Liu, X., Li, B., Song, A., and Li, Q., Automated delineation of agricultural field boundaries from Sentinel-2 images using recurrent residual U-Net, Int. J. Appl. Earth Obs. Geoinf., 2021, vol. 105, p. 102557.
Funding
This study was carried out with financial support from the Russian Federation (agreement with the Ministry of Education and Science of Russia no. 075-15-2022-321 dated April 21, 2022) and the Ministry of Science and Higher Education of the Russian Federation (theme “Monitoring,” state registration no. 122042500031-8).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
The authors of this work declare that they have no conflicts of interest.
Additional information
Publisher’s Note.
Pleiades Publishing remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Ermolaev, N.R., Yudin, S.A., Belobrov, V.P. et al. Using Deep Learning and Cloud Services for Mapping Agricultural Fields on the Basis of Remote Sensing Data of the Earth. Izv. Atmos. Ocean. Phys. 59, 1301–1306 (2023). https://doi.org/10.1134/S0001433823120083
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0001433823120083