%0 Conference Proceedings
%4 dpi.inpe.br/marte/2011/07.12.17.13
%2 dpi.inpe.br/marte/2011/07.12.17.13.22
%@isbn 978-85-17-00056-0 (Internet)
%@isbn 978-85-17-00057-7 (DVD)
%T Classificação de uso da terra no município de Lucas do Rio Verde: uma análise da confiabilidade na classificação automática e manual de imagem Landsat nas áreas de preservação permanente
%D 2011
%A Barros, Adriana Oliveira,
%A Castanha, Márcio,
%A Schwenk, Lunalva Moura,
%@affiliation Universidade Federal de Mato Grosso – UFMT
%@affiliation Universidade Federal de Mato Grosso – UFMT
%@affiliation Universidade Federal de Mato Grosso – UFMT
%@electronicmailaddress dricabarros1@hotmail.com
%@electronicmailaddress mcastanha1@hotmail.com
%@electronicmailaddress lunalvaschwenh104@hotmail.com
%E Epiphanio, José Carlos Neves,
%E Galvão, Lênio Soares,
%B Simpósio Brasileiro de Sensoriamento Remoto, 15 (SBSR).
%C Curitiba
%8 30 abr. - 5 maio 2011
%I Instituto Nacional de Pesquisas Espaciais (INPE)
%J São José dos Campos
%P 6089-6096
%S Anais
%1 Instituto Nacional de Pesquisas Espaciais (INPE)
%K Image classification, APP, Validation, Index Kappa.
%X The present work has as main objective the mapping of agricultural expansion in the city of Lucas do Rio Verde emphasizing the permanent preservation areas, where there will be an analysis of the reliability of automatic classification and manual as validation of Landsat data classification. Thus, for the mapping and classification of land use in that municipality became a supervised classification on Landsat-TM5 image with a resolution of 30 meters using Bhattacharya classifier using the software SPRING version 5.0.6. Subsequently the vector was done in the areas where there were the highest environmental impacts in a timely manner, in this case, the Permanent Preservation Areas, in Arc-GIS 9.3 software obtained as parameter three Google image clippings High Spatial Resolution. Finally the evaluation of data, made the comparison of manual and automatic classification of Landsat image data for the clips of vectorized images from Google. Thus the work resulted in the classification of Lucas do Rio Verde has a total area of 368,044 ha ², 66.37% of deforested areas, 29.62% of areas with natural vegetation and even 4.01% of wetlands or rivers. As for the areas of APP are degraded considerably since the three methods were classified between 40 and 60% degradation in APP within the limits of cut samples. Finally, the accuracy of the data was considered reasonable for the automatic classification and excellent for the vector image in regard to the method of index Kappa.
%9 Uso e Cobertura da Terra
%@language pt
%3 p1069.pdf