- Página Inicial
- Publicações
- Artigos
- Utilização de Séries Temporais de Imagens AVHRR/NOAA no apoio à Estimativa Operacional da Produção da cana-de-açúcar no Estado de São Paulo
Publicações
Utilização de Séries Temporais de Imagens AVHRR/NOAA no apoio à Estimativa Operacional da Produção da cana-de-açúcar no Estado de São Paulo
Brazil is the first world producer of sugar cane. The State of Sao Paulo is the first national producer of sugar cane, contributing with more than 60% of national production. Due to the high relevance of production, it is important to have insurance estimates and surveys of areas cultivated with the crop. The progress of the different remote sensing techniques has allowed to use satellite images to monitor and assist the estimation of these areas. There are numerous options, including images from the AVHRR/NOAA. Combining the need to obtain more precise estimates of the yields of cane sugar, with the potential to acquire agricultural information culture through the NDVI, this study explored the time series analysis of NDVI/AVHRR images and the identification of areas with sugar cane in the State of Sao Paulo. The results indicate that areas with sugar cane are well modeled, from the harmonic analysis, the five crops examined, allowing them to differentiate them from other cultures, the RGB compositions used according to their growing seasons. From the supervised classification using the maximum likelihood algorithm of image amplitude, phase and aditive term was generated a map representing the spatial distribution of culture in the state in the crop-season 03/04, 04/05, 05/06, 06/07 and 07/08. The results obtained in each crop-season were compared with as reference, provided by the project CANASAT/INPE. It was found that the smaller relative errors, around 10% for the season 07/08, were found from the estimate based on the error matrix.
Download