Techniques for the extraction of spatial and spectral information in the supervised classification of hyperspectral imagery for land-cover applications

Abstract

The objective of this PhD thesis is the development of spatialspectral information extraction techniques for supervised classification tasks, both by means of classical models and those based on deep learning, to be used in the classification of land use or land cover (LULC) multi- and hyper-spectral images obtained by remote sensing. The main goal is the efficient application of these techniques, so that they are able to obtain satisfactory classification results with a low use of computational resources and low execution time.