Detekce změn vegetačního pokryvu lesních porostů pomocí dat DPZ velmi vysokého rozlišení

Abstract

The aim of this thesis is to detect changes in vegetation cover of forest using remote senzing data of very high resolution. Images were acquired from satellite SPOT and Pleiades for two areas such as area in Jeseníky and part of area of Přírodní park Sovinecko. The issue is handled by supervised classification by classifier Parallelepiped with Maximum Likelihood as Tie Breaker and by computaion of vegetation indices NDVI. Filter SIEVE was used too. After all of this the post-classification makeover was done. At the end of work results of classification were compared between each other. The result of work are map outputs such as classification or NDVI and statistical tables.