Журнал

About Editorial Board Information for authors and reviewers Publication ethics Contacts User Account: send / review a manuscript
Archive
Volume 21, 2024
Number 1 Number 2 Number 3 Number 4 Number 5
Volume 20, 2023
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 19, 2022
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 18, 2021
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 17, 2020
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 16, 2019
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 15, 2018
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 14, 2017
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 13, 2016
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 12, 2015
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 11, 2014
Number 1 Number 2 Number 3 Number 4
Volume 10, 2013
Number 1 Number 2 Number 3 Number 4
Volume 9, 2012
Number 1 Number 2 Number 3 Number 4 Number 5
Volume 8, 2011
Number 1 Number 2 Number 3 Number 4
Volume 7, 2010
Number 1 Number 2 Number 3 Number 4
Issue 6, 2009
Volume 1 Volume 2
Issue 5, 2008
Volume 1 Volume 2
Issue 4, 2007
Volume 1 Volume 2
Issue 3, 2006
Volume 1 Volume 2
Issue 2, 2005
Volume 1 Volume 2
Issue 1, 2004
Volume 1
Search
Find:
русский
ISSN 2070-7401 (Print), ISSN 2411-0280 (Online)
Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa
CURRENT PROBLEMS IN REMOTE SENSING OF THE EARTH FROM SPACE

  

Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2024, Vol. 21, No. 5, pp. 116-129

Performance assessment of multi-season machine learning models for large-scale in-season winter crops mapping

D.E. Plotnikov 1 , Yu.Sh. Boimatov 2 , E.S. Elkina 1 , E.V. Shcherbenko 3 , A.S. Plotnikova 4 
1 Space Research Institute RAS, Moscow, Russia
2 Moscow Institute of Physics and Technology, Dolgoprudny, Moscow Region, Russia
3 OOO Space Research Institute for the Earth, Moscow, Russia
4 Center for Forest Ecology and Productivity RAS, Moscow, Russia
Accepted: 30.09.2024
DOI: 10.21046/2070-7401-2024-21-5-116-129
In this study, we have evaluated the potential of ensemble machine learning methods Random Forest and XGBoost for winter crops mapping over large areas when training on multi-season datasets for five consecutive years. Two experiments were conducted: accuracy intercomparison of multi-season models of two architectures and intercomparison of multi-season and a series of single-season models of optimal architecture. The results of the experiments were provided for the study region covering more than 94 % of winter crops areas sown in Russia, based on comparison with the Rosstat (Federal State Statistics Service) data and the results of the automatic benchmark method using RMSE (root mean squared error) and coefficient of determination R 2, as well as the F-score metric for the winter crop class. In the first experiment, the potential of multi-season models was demonstrated, and the Random Forest multi-season model proved to be more efficient and more stable than XGBoost, the RMSE improvement compared to the reference maps averaged 28 th. ha and was observed for all of the five seasons; the relative improvement was 15 % on average, reaching 35 %. The results of the second experiment which compared the five single-season models and the multi-season model also indicated the effectiveness of the multi-season model: the cumulative RMSE of the multi-season model for 5 seasons is 1.32–2.02 times less than of single-season models, and the average F-score for any multi-season model falls in the range of 0.75–0.79, while for the multi-season model, it is as high as 0.84.
Keywords: multi-season model, Random Forest, XGBoost, satellite mapping, winter crops, LOWESS, time series, grid search, machine learning
Full text

References:

  1. Denisov P. V., Troshko K. A., Loupian E. A. et al., Satellite remote sensing for boundaries determination of agricultural lands, Land management, cadastre and land monitoring, 2024, Vol. 19, No. 4(231), pp. 227–232 (in Russian), DOI: 10.33920/sel-4-2404-04.
  2. Kashnitskii A. V., Loupian E. A., Plotnikov D. E., Tolpin V. A., Analysis of the possibility of using different spatial resolution data for objects monitoring, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2023, Vol. 20, No. 2, pp. 60–74 (in Russian), DOI: 10.21046/2070-7401-2023-20-2-60-74.
  3. Loupian E. A., Proshin A. A., Bourtsev M. A. et al., Experience of development and operation of the IKI-Monitoring center for collective use of systems for archiving, processing and analyzing satellite data, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2019, Vol. 16, No. 3, pp. 151–170 (in Russian), DOI: 10.21046/2070-7401-2019-16-3-151-170.
  4. Loupian E. A., Sereda I. I., Denisov P. V. et al., Remote monitoring of winter crops condition in the winter of 2020/2021 in the European territory of Russia, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2021, Vol. 18, No. 1, pp. 165–172 (in Russian), DOI: 10.21046/2070-7401-2021-18-1-165-172.
  5. Plotnikov D. E., Bartalev S. A., Zharko V. O. Mihailov V. V., Prosyannikova O. I., An experimental assessment of crop types recognisability using time-series of intra-seasonal spectral reflectance measurements by satellite sensor, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2011, Vol. 8, No. 1, pp. 199–208 (in Russian).
  6. Plotnikov D. E., Bartalev S. A., Loupian E. A., Tolpin V. A., Accuracy assessment for winter crops mapping in spring-summer growing season with MODIS data, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2017, Vol. 14, No. 4, pp. 132–145 (in Russian), DOI: 10.21046/2070-7401-2017-14-4-132-145.
  7. Plotnikov D. E., Khvostikov S. A., Bartalev S. A., Method for automated crop types mapping using remote sensing data and a plant growth simulation model, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2018, Vol. 15, No. 4, pp. 131–141 (in Russian), DOI: 10.21046/2070-7401-2018-15-4-131-141.
  8. Plotnikov D. E., Elkina E. S., Dunaieva Ie. A. et al. (2020a), Development of the method for automatic winter crops mapping by means of remote sensing aimed at crops state assessment over the Republic of Crimea, Taurida Herald of the Agrarian Sciences, 2020, No. 1(21), pp. 64–82 (in Russian), DOI: 10.33952/2542-0720-2020-1-21-64-83.
  9. Plotnikov D. E., Kolbudaev P. A., Zhukov B. S. et al. (2020b), The collection of multispectral KMSS-M (Meteor-M No. 2) satellite data aimed at quantitative assessment of the Earth surface, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2020, Vol. 17, No. 7, pp. 276–282 (in Russian), DOI: 10.21046/2070-7401-2020-17-7-276-282.
  10. Plotnikov D. E., Kolbudaev P. A., Loupian E. A., An automatic method for subpixel registration of KMSS-M imagery based on coarse-resolution actualized reference, Computer Optics, 2022, Vol. 46, No. 5, pp. 818–827 (in Russian), DOI: 10.18287/2412-6179-CO-1098.
  11. Povkh V. I., Shlyakhova L. A., Boeva I. N., Assessment of the structure of agricultural crop areas using high resolution satellite data, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2013, Vol. 10, No. 4, pp. 224–228 (in Russian).
  12. Savin I. Yu., Bartalev S. A., Loupian E. A., Tolpin V. A., Medvedeva M. A., Plotnikov D. E., Satellite monitoring of vegetation affected by drought (using drought 2010 in Russia as an example), Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2011, Vol. 8, No. 1, pp. 150–162 (in Russian).
  13. Terekhin E. A., Application of discriminant analysis for recognition of agricultural crops, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2017, Vol. 14, No. 7, pp. 89–99 (in Russian), DOI: 10.21046/2070-7401-2017-14-7-89-99.
  14. Troshko K. A., Denisov P. V., Dunaeva E. A. et al., Features of grain crops development in Russia in 2022 based on remote sensing data, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2022, Vol. 19, No. 6, pp. 301–307, DOI: 10.21046/2070-7401-2022-19-6-301-307 (in Russian).
  15. Bégué A., Arvor D., Bellon B. et al., Remote sensing and cropping practices: A review, Remote Sensing, 2018, Vol. 10, No. 1, Article  99, https://doi.org/10.3390/rs10010099.
  16. Blickensdörfer L., Schwieder M., Pflugmacher D. et al., Mapping of crop types and crop sequences with combined time series of Sentinel-1, Sentinel-2 and Landsat-8 data for Germany, Remote Sensing of Environment, Vol. 269, 2022, Article 112831, https://doi.org/10.1016/j.rse.2021.112831.
  17. Breiman L., Random Forests, Machine Learning, 2001, Vol. 45, No. 1, pp. 5–32, DOI: 10.1023/A:1010933404324.
  18. Chen T., Guestrin C., XGBoost: A scalable tree boosting system, KDD ’16: Proc. 22nd ACM SIGKDD Intern. Conf. Knowledge Discovery and Data Mining, 2016, pp. 785–794, DOI: 10.1145/2939672.2939785.
  19. Fischer G., Nachtergaele F., Prieler S., Teixeira E., Tòth G., Velthuizen H., Verelst L., Wiberg D., Global AgroEcological Zones (GAEZ ver3.0). Model documentation, IIASA/FAO, 2012, 196 p.
  20. Foerster S., Kaden K., Förster M., Itzerott S., Crop type mapping using spectral-temporal profiles and phenological information, Computers and Electronics in Agriculture, 2012, Vol. 89, pp. 30–40, DOI: 10.1016/j.compag.2012.07.015.
  21. Garland M., Le Grand S., Nickolls J., et al., Parallel Computing Experiences with CUDA, IEEE Micro, 2008, V. 28(4), pp. 13–27, https://doi:10.1109/mm.2008.57.
  22. Kolbudaev P. A., Plotnikov D. E., Loupian E. A. et al., The methods and automatic technology aimed at imagery georeferencing, cloud screening, atmospheric and radiometric correction of KMSS-M satellite data, E3S Web of Conf., 2021, Vol. 333, Article 01006, DOI: doi.org/10.1051/e3sconf/202133301006.
  23. Kussul N., Skakun S., Shelestov A. et al., Regional scale crop mapping using multi-temporal satellite imagery, Photogrammetry, Remote Sensing and Spatial Information Sciences, 2015, Vol. XL-7/W3, pp. 45–52, https://doi.org/10.5194/isprsarchives-XL-7-W3-45-2015.
  24. Lebourgeois V., Dupuy S., Vintrou E. et al., A combined random forest and OBIA classification scheme for mapping smallholder agriculture at different nomenclature levels using multisource data (simulated Sentinel-2 time series, VHRS and DEM), Remote Sensing, 2017, Vol. 9, Article 259, https://doi.org/10.3390/rs9030259.
  25. Massey R., Sankey T., Congalton R. et al., MODIS phenology-derived, multi-year distribution of conterminous U. S. crop types, Remote Sensing of Environment, 2017, Vol. 198, pp. 490–503, DOI: 10.1016/j.rse.2017.06.033.
  26. Palchoudhuri Y., Valcarce-Diñeiro R., King P., Sanabria-Soto M., Classification of multi-temporal spectral indices for crop type mapping: a case study in Coalville, UK, J. Agricultural Science, 2018, pp. 1–13, DOI: 10.1017/S0021859617000879.
  27. Plotnikov D. E., Loupian E. A., Kolbudaev P. A. et al., Daily surface reflectance reconstruction using LOWESS on the example of various satellite systems, IEEE Xplore. 8 th Intern. Conf. Information Technology and Nanotechnology (ITNT), 2022, DOI: 10.1109/ITNT55410.2022.9848630.
  28. Plotnikov D. E., Kolbudaev P. A., Matveev A. M. et al., Accuracy assessment of atmospheric correction of KMSS-2 Meteor-M No. 2.2 data over Northern Eurasia, Remote Sensing, 2023, Vol. 15, Issue 18, Article 4395, DOI: doi.org/10.3390/rs15184395.
  29. Probst P., Wright M. N., Boulesteix A. L., Hyperparameters and tuning strategies for random forest, Wiley interdisciplinary reviews: data mining and knowledge discovery, 2019, Vol. 9, Issue 3, Article e1301, https://doi.org/10.1002/widm.1301.
  30. Skakun S., Vermote E., Franch B. et al., Winter wheat yield assessment from Landsat-8 and Sentinel-2 data: incorporating surface reflectance, through Phenological fitting, into regression yield models, Remote Sens., 2019, Vol. 11. Issue 15, Article 1768, https://doi.org/10.3390/rs11151768.
  31. Valero S., Morin D., Inglada J. et al., Production of a dynamic cropland mask by processing remote sensing image series at high temporal and spatial resolutions, Remote Sensing, 2016, Vol. 8. Issue 1. Article 55, https://doi.org/10.3390/rs8010055.
  32. Waldner F., Schucknecht A., Lesiv M. et al., Conflation of expert and crowd reference data to validate global binary thematic maps, Remote Sensing of Environment, 2019, Vol. 221, pp. 235–246, DOI: 10.1016/j.rse.2018.10.039.
  33. Wang S., Azzari G., Lobell D. B., Crop type mapping without field-level labels: random forest transfer and unsupervised clustering techniques, Remote Sensing of Environment, 2019, Vol. 222, pp. 303–317, https://doi.org/10.1016/j.rse.2018.12.026.
  34. Weiss M., Jacob F., Duveiller G., Remote sensing for agricultural applications: A meta-review, Remote Sensing of Environment, 2020, Vol. 236, Article 111402, https://doi.org/10.1016/j.rse.2019.111402.
  35. Wu B., Zhang M., Zeng H. et al., Challenges and opportunities in remote sensing-based crop monitoring: A review, National Science Review, 2023, Article nwac290, https://doi.org/10.1093/nsr/nwac290.
  36. Xu J., Zhu Y., Zhong R. et al., DeepCropMapping: A multi-temporal deep learning approach with improved spatial generalizability for dynamic corn and soybean mapping, Remote Sensing of Environment, 2020, Vol. 247, Article 111946, https://doi.org/10.1016/j.rse.2020.111946.
  37. Zhang C., Di L., Lin L. et al., Towards automation of in-season crop type mapping using spatiotemporal crop information and remote sensing data, Agricultural Systems, 2022, Vol. 201, Article 103462, https://doi.org/10.1016/j.agsy.2022.103462.
  38. Zhong L., Hu L., Yu L., Gong P., Biging G. S., Automated mapping of soybean and corn using phenology, J. Photogrammetry and Remote Sensing, 2016, Vol. 119, pp. 151–164, https://doi.org/10.1016/j.isprsjprs.2016.05.014.