Analysis of Antropic Intervention from Satellital Imagery Processing, NDVI, NDBI and Tasseled Cap Transformation

Abstract

This research intends to describe the results obtained in a study area from the use of a Sentinel2A L1C image of the metropolitan area of the city of Rio de Janeiro, Brazil (known as “El Gran Rio”), after applying some classification methods like Vector Support Machines (SVM) and decision trees (DT), as well as the Tasseled Cap Transform (TTC), the NDVI (Normalized Differential Vegetation Index) and the NDBI (Normalized Differential Build-up Index). The main purpose is to estimate the classification of land cover and determine the anthropic affectation generated by urbanization processes. Better results were obtained when implemented the classification with DTs than with a classification based on SVM, However, better values of percentage of agreement and kappa coefficient were obtained for SVM