METHODS OF SUPERVISED CLASSIFICATION OF IMAGES OF SATELLITE APPLIED IN THE MAPPING OF THE USE OF SOIL IN HYDROGRAPHIC BASIN OF SANTO ANTONIO CREEK, SAO MANUEL/SP

Abstract

This study aimed to compare the efficiency of methods of classification of orbital images using GIS techniques.. The study area is located in the watershed Ribeirao Santo Antnio in Sao Manuel, Sao Paulo, Brazil. For this, we used the GIS techniques, and the integration of information held in the Geographic Information System (GIS) - IDRISI, coupled with the use of digital maps, published by the Brazilian Institute of Geography and Statistics - IBGE, scale 1:50,000 and satellite images LANDSAT - 8 (2014) Operational Land Imager (OLI), provided by Glovis. It noted the evaluation results of the classification accuracy were satisfactory, which the PARALELEPIPEDO classification had law quality and good the MAXVER algorithm ratings, with kappa 0.2242 and 0.3322 for respectively. The methods used in the discrimination of areas cultivated with sugar cane howed different efficiencies in the images classification than influenced by the reflectance of other classes.