Autoria: CECHIM JUNIOR, C.; JOHANN, J. A.; ANTUNES, J. F. G.

Resumo: ABSTRACT. The knowledge on reliable estimates of areas under sugarcane cultivation is essential for the Brazilian agribusiness, since it helps in the development of public policies, in determining prices by sugar mills to producers and allows establishing the logistics of production disposal. The objective of this work was to develop a methodology for mapping the sugarcane crop area in the state of Paraná, Brazil, using images from the Landsat/TM/OLI and IRS/LISS-3 satellites, for the crop years from 2010/2011 to 2013/2014. The mappings were conducted through the supervised Maximum likelihood classification (Maxver) achieving, on average, an overall accuracy of 94.13% and kappa index of 0.82. The correlation with the official data of the IBGE ranged from moderate to strong (0.64 ≤ rs ≤ 0.80) with average agreement (dr) of 0.81. There was an increase of 2.73% (18,630 ha) in the area with sugarcane in Paraná between 2010/2011 and 2013/2014.

Ano de publicação: 2017

Tipo de publicação: Artigo de periódico

Unidade: Embrapa Agricultura Digital

Palavras-chave: Agricultural statistics, Cana de açúcar, Classificação supervisionada, Digital image processing, Estatística agrícola, Image analysis, Maxver, Processamento de imagem digital, Remote sensing, Sensoriamento remoto, Sugarcane, Supervised classification