Autoria: MORAES, D. A. de C.; SARTORI, A. A. da C.; NOSSACK, F. A.; SPADOTTO, C. A.; ZIMBACK, C. R. L.

Resumo: This study aims to estimate, by using the NDVI index and the unsupervised classification method CLUSTER, the areas covered by aquatic plants in Salto Grande dam, Americana - SP, comparing 2008 and 2010. The NDVI is sensitive to capture the spectral response of vegetation, which has high reflectance in the near infrared band. The classification, supervised or unsupervised, of images obtained by the index of vegetation contributes to the elaboration of thematic maps. In the process of unsupervised classification, the algorithm CLUSTER was used, which is based on a composite image, to obtain satisfactory groupings. This classifier selects the pixels at the peak of the histogram. It was observed that the area mapped with aquatic plants in 2008 is more representative due to a higher volume of them, compared to 2010. With the use of images from the satellite CBERS-2B was possible to estimate the area covered by aquatic plants on the surface of Salto Grande Dam, in Americana (SP), supported by the unsupervised classification CLUSTER, obtained after NDVI process. The maps offer an opportunity to work with a better interpreting of the different targets in the study area images, supporting the monitoring of aquatic plants in the dams and other water bodies.

Ano de publicação: 2011

Tipo de publicação: Artigo em anais e proceedings

Unidade: Embrapa Territorial

Palavras-chave: Classificação não supervisionada, Mapas temáticos, Método CLUSTER, Índice NDVI