Avaliação do desempenho da classificação do uso e cobertura da terra a partir de imagens Landsat 8 e RapidEye na região central do Rio Grande do Sul

Abstract

Digital images with different spatial resolutions can be used to understand the dynamics of use and coverage of the Earth’s surface. The Southern Plateau Sul-rio-grandense presents diversified land use, thus, images obtained from that region may contain distinct spectral behavior. Therefore, this study aimed to evaluate the thematic classification of land use and cover in the central region of Rio Grande do Sul through images of medium (Landsat 8/OLI) and high (REIS/RapidEye) spatial resolution. The thematic classes (CT) were identified and quantified with software Spring v. 5.1.8 and evaluated taking into account the distribution of 200 random points using QGis v. 2.10.1. In the studied area the CT’s most representative in both sensors were agriculture and native forest. The classes less representative for the Landsat 8 were water and exposed soil and for the RapidEye were water and planted forest. The classification system resulted in Kappa coefficient of 0,59% (good) and 0,73% (very good), to Landsat 8 and RapidEye images, respectively. The use of both images has been satisfactory for the land use and land cover classification employed in this study, however, RapidEye images are more suitable for classifications that require greater detail of the area.