Coffee crop, an important economic and social activity to Brazil, is a segment of the agro-business that has not updated information about its agroecosystem and the productive profile. The present work comes to corroborate in the sense to develop and/or adequate fundamental methods in geotechnologies that may significantly contribute to the delimitation and characterization of the coffee crop in Minas Gerais State. As a pilot study area the municipalities of Aguanil, Boa Esperança, Campo Belo and Cristais were selected. In the development of the work images from CCD sensor on board of CBERS satellite and images from TM sensor on board of Landsat-5 were used. Images were classified using the supervised Maxver classifier followed by a manual interpretation on the computer screen. The results showed that it was possible to identify and map coffee crop areas through the use of remote sensing satellite images.
Sensoriamento remoto; Mapeamento da cafeicultura; CBERS; LANDSAT; Agricultura; Multitemporal; Multisensor
