(47-3) 16 * << * >> * Russian * English * Content * All Issues

On classification of Sentinel-2 satellite images by a neural network ResNet-50
I.V. Bychkov 1, G.M. Ruzhnikov 1, R.K. Fedorov 1, A.K. Popova 1, Y.V. Avramenko 1

1 ISDCT SB RAS – Matrosov Institute for System Dynamics and Control Theory of the Siberian Branch of the RAS,
664033, Irkuts, Russia, Lermontova 134

 PDF, 830 kB

DOI: 10.18287/2412-6179-CO-1216

Pages: 474-481.

Full text of article: Russian language.

Abstract:
Various combinations of neural network parameters and sets of input data for satellite image classification are considered in the article. The training set is completed with a NDVI (normalized difference vegetation index) and local binary patterns. Testing of classifiers created on a different number of epochs and samples is carried out. Values of the neural network hyperparameters are determined that allow a classification accuracy of 0.70 and an F-measure of 0.65 to be achieved. Separation into classes with similar spectral characteristics is shown to offer low classification quality at different parameters and input data sets. Additional information is required. For example, for forests to be divided into more detailed classes, one needs to employ classifiers that use images from different seasons and vegetation periods. In addition, the training set needs to be extended to take into account various natural zones, soils, etc.

Keywords:
neural networks, classification, Sentinel-2, remote sensing, image processing.

Citation:
Bychkov IV, Ruzhnikov GM, Fedorov RK, Popova AK, Avramenko YV. On classification of Sentinel-2 satellite images by a neural network ResNet-50. Computer Optics 2023; 47(3): 474-481. DOI: 10.18287/2412-6179-CO-1216.

Acknowledgements:
The work was supported by grant No. 075-15-2020-787 of the Ministry of Science and Higher Education of the Russian Federation for the implementation of a large scientific project in priority areas of scientific and technological development (the project "Fundamentals, methods and technologies for digital monitoring and forecasting of the ecological situation of the Baikal natural territory").

References:
  1. Talukdar S, Mahato S, Shahfahad Pal S, Liou YA, Rahman A. Land-use land-cover classification by machine learning classifiers for satellite observations–A review. Remote Sens 2020; 12(7): 1135. DOI: 10.3390/rs12071135.
  2. Keshtkar H, Voigt W, Alizadeh E. Land-cover classification and analysis of change using machine-learning classifiers and multi-temporal remote sensing imagery. Arab J Geosci 2017; 10: 154. DOI: 10.1007/s12517-017-2899-y.
  3. Lastovicka J, Svec P, Paluba D, Kobliuk N, Svoboda J, Hladky R, Stych P. Sentinel-2 data in an evaluation of the impact of the disturbances on forest vegetation. Remote Sens 2020; 12(12): 1914. DOI: 10.3390/rs12121914.
  4. Feng Q, Liu J, Gong J. UAV remote sensing for urban vegetation mapping using random forest and texture analysis. Remote Sens 2015; 7: 1074-1094. DOI: 10.3390/rs70101074.
  5. Liu Y, Gong W, Hu X, Gong J. Forest type identification with random forest using Sentinel-1A, Sentinel-2A, multi-temporal Landsat-8 and DEM data. Remote Sens 2018; 10: 946. DOI: 10.3390/rs10060946.
  6. Immitzer M, Neuwirth M, Böck S, Brenner H, Vuolo F, Atzberger C. Optimal input features for tree species classification in Central Europe based on multi-temporal Sentinel-2 data. Remote Sens 2019; 11: 22. DOI: 10.3390/rs11222599.
  7. Axelsson A, Lindberg E, Reese H, Olsson H. Tree species classification using Sentinel-2 imagery and Bayesian inference. Int J Appl Earth Obs Geoinf 2021; 100: 102318.
  8. Chambon T. Fighting hunger through open satellite data: a new state of the art for land use classification. 2019. Source: <https://medium.com/omdena/fighting-hunger-through-open-satellite-data-a-new-state-of-the-art-for-land-use-classification-f57f20b7294b>.
  9. Wang D, Wan B, Qiu P, Su Y, Guo Q, Wang R, Sun F, Wu X. Evaluating the performance of Sentinel-2, Landsat 8 and Pléiades-1 in mapping mangrove extent and species. Remote Sens 2018; 10: 9.
  10. Abdi AM. Land cover and land use classification performance of machine learning algorithms in a boreal landscape using Sentinel-2 data. GIScience Remote Sens 2020; 57(1): 1-20.
  11. Paringer RA, Mukhin AV, Kupriyanov AV. Formation of an informative index for recognizing specified objects in hyperspectral data. Computer Optics 2021; 45(6): 873-878. DOI: 10.18287/2412-6179-CO-930.
  12. Carranza-García M, García-Gutiérrez J, Riquelme JC. A framework for evaluating land use and land cover classification using convolutional neural networks. Remote Sens 2019; 11(30): 274. DOI: 10.3390/rs11030274.
  13. Campos-Taberner M, García-Haro FJ, Martínez B, Izquierdo-Verdiguier E, Atzberger C, Camps-Valls G, Gilabert MA. Understanding deep learning in land use classification based on Sentinel-2 time series. Sci Rep 2020; 10(1): 17188.
  14. Ojala T, Pietikäinen M, Hawood D. A comparative study of texture measures with classification based on featured distributions. Pattern Recogn 1996; 29(1): 51-59.
  15. Huang D, Shan C, Ardabilian M, Wang Y, Chen L. Local binary patterns and its applications on facial image: A survey. IEEE Trans Syst Man Cybern Syst 2011; 41(6): 765-781.
  16. Myasnikov VV. Description of images using a configuration equivalence relation. Computer Optics 2018; 42(6): 998-1007. DOI: 10.18287/2412-6179-2018-42-6-998-1007.
  17. Wei X, Yu X, Liu B, Zhi L. Convolutional neural networks and local binary patterns for hyperspectral image classification. Eur J Remote Sens 2019; 52(1): 448–462. DOI: 10.1080/22797254.2019.1634980.
  18. Zhao F, Sun R, Zhong L, Meng R, Huang C, Zeng X, Wang M, Li Y, Wang Z. Monthly mapping of forest harvesting using dense time series Sentinel-1 SAR imagery and deep learning. Remote Sens Environ 2022; 269: 112822.
  19. Bychkov IV, Ruzhnikov GM, Fedorov RK, Popova AK, Avramenko YV. Classification of Sentinel-2 satellite images of the Baikal Natural Territory. Computer Optics 2022; 46(1): 90-96. DOI: 10.18287/2412-6179-CO-1022.
  20. Gitelson AA, Merzlyak MN, Lichtenthaler HK. Detection of red edge position and chlorophyll content by reflectance measurements near 700 nm. J Plant Physiol 1996; 148(3-4): 501-508. DOI: 10.1016/S0176-1617(96)80285-9.

© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: journal@computeroptics.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20