-
oa Remote identification of winter wheat crops in monitoring tasks
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, XIV International Scientific Conference “Monitoring of Geological Processes and Ecological Condition of the Environment”, Nov 2020, Volume 2020, p.1 - 5
Abstract
In this article, satellite images of Landsat-8 and terrestrial imagery were used as a source of information for the automated recognition of winter wheat sowing fields. The land of the Lichanka territorial community, Kyiv-Svyatoshinsky district of Kyiv region was selected as a research object. During the work, a correlation was established between the brightness values of terrestrial snapshots RGB of winter wheat, obtained from terrestrial field surveys, with the pixel brightness values of remote sensing images of different spectral channels and NDVI values. The most significant correlation coefficients of magnitude (RGB) with the values of brightness are as follows, r: channel 5 - 0.77, channel 6 - 0.69, channel 7 - 0.53, channel 10 - 0.74, channel 11-0, 59, NDVI - 0.91. Given the significant correlation between magnitudes (RGB) and NDVI, high informativeness of the 5th channel near infrared (0.845–0.885 μm), chosen to build the identification model should be considered natural. The use of the developed model of culture recognition allowed to territorially identifi winter wheat crops and distinguish them from other vegetation. The results of comparing the model’s identification ability and the NDVI’s vegetation index indicate its reliability and the feasibility of further improvement.