Methodology for automated classification of farmland based on Earth remote sensing data

Abstract

Abstract The analysis of spectral characteristics from satellite images of the earth for different periods of the growing season is carried out. Using a geographic information system, a training sample was obtained for identifying agricultural land based on satellite monitoring data for the Krasnoyarsk region. The classification is carried out on the basis of spectral characteristics of agricultural land according to the data of remote sensing of the earth. Field studies were carried out to verify the classification results. Based on the results of the work, modules of a geographic information system were created, containing a cartographic database of agricultural land in the Sukhobuzimsky district of the Krasnoyarsk Territory.