Neural network analysis of hyperspectral images of soil

Abstract

The article approaches to the classification of high-resolution hyperspectral images in the problem of classification of soil species is proposed. A spectral-spatial convolutional neural network with compensation for lighting variations is used as a classifier. The effectiveness of the proposed approach in the problem of classification of hyperspectral images of soils obtained by a scanning hyperspectral camera is shown. The essence of the developed method is to use binary classification together with multiclass, thereby improving the result of the latter.