Klasifikace lesní vegetace pomocí družicových dat vysokého a velmi vysokého prostorového rozlišení

Abstract

The bachelor thesis deals with the classification of forest vegetation from commercial data PlanetScope and RapidEye and free data from Sentinel-2 and Landsat-8 missions. Aim of this thesis is to create a classification of forest vegetations in the locality of Černý les, Groníček, and Kovářův žleb in the environment of PCI Geomatica software, and compare the results of partial classifications. The forest vegetations were classified with supervised classification method Maximum Likelihood with null class. The results of classifications from individual data and classifications from Corine Land Cover 2018 and the map of coniferous and deciduous forests from ÚHÚL were compared. The vegetation index NDVI was computed for the localities. The results of this thesis are map outputs with classified forest vegetation, NDVI raster maps, tables with classification results and comparative maps of classified data from ÚHÚL and CLC.