Open AccessDETECTION OF FERTILE SOILS BASED ON SATELLITE IMAGERY PROCESSING
Author(s) -
Valery Grishkin,
E. Zhivulin,
A. Khokhriakova,
С Г Каримов
Publication year - 2021
Publication title -
9th international conference "distributed computing and grid technologies in science and education"
Language(s) - English
Resource type - Conference proceedings
DOI - 10.54546/mlit.2021.13.12.001
Subject(s) - soil water , normalized difference vegetation index , desertification , support vector machine , pixel , satellite , vegetation (pathology) , computer science , feature extraction , remote sensing , satellite imagery , environmental science , pattern recognition (psychology) , artificial intelligence , soil science , geology , engineering , ecology , medicine , oceanography , pathology , climate change , aerospace engineering , biology
The paper proposes a method for detecting fertile soils based on the processing of satellite images. Asa result of its application, a map of the location of fertile and infertile soils for a given region of theearth's surface is formed and the corresponding areas are calculated. The method for detecting fertilesoils is based on the fact that fertile soil includes areas covered with vegetation in the spring-summerperiod. Therefore, by measuring the spectral characteristics of these areas in the late autumn period,when there is no vegetation on them, it is possible to obtain objective parameters of fertile soils. Fordetection, a number of classifiers are being built that recognize two classes - fertile soil and sand,which is especially important when monitoring areas prone to desertification. The feature vector usedfor classification is a set of indices similar to the well-known NDVI index. This set of indices iscalculated for each pixel of the image by its values in different spectral channels. Classifiers areimplemented using CUDA parallel computing technology on a GPU. Based on the results of theexperimental study, a classifier is selected that has shown the best characteristics of the recognitionquality.