@Article{CarreirasPereShim:2006:LaMaBr,
               author = "Carreiras, Jo{\~a}o M. B. and Pereira, Jos{\'e} M. C. and 
                         Shimabukuro, Yosio Edemir",
          affiliation = "Department of Forestry, Instituto Superior de Agronomic, Tapada da 
                         Ajuda, 1349-017 Lisboa, Portugal and Department of Forestry, 
                         Instituto Superior de Agronomic, Tapada da Ajuda, 1349-017 Lisboa, 
                         Portugal and {Instituto Nacional de Pesquisas Espaciais (INPE)}",
                title = "Land-cover mapping in the Brazilian Amazon using SPOT-4 vegetation 
                         data and machine learning classification methods",
              journal = "Photogrammetric Engineering and Remote Sensing",
                 year = "2006",
               volume = "72",
               number = "8",
                pages = "897--910",
                month = "Aug.",
             keywords = "VEGETA{\C{C}}{\~A}O, remotely-sensed data, decision-tree 
                         classification, resolution satellite data, spatial-resolution, 
                         accuracy assessment, avhrr data, multispectral data, tropical 
                         regions, eastern amazon, mixing models.",
             abstract = "The main objective of this study is to evaluate the feasibility of 
                         deriving a land-cover map of the state of Mato Grosso, Brazil, for 
                         the year 2000, using data from the 1 km SPOT-4 VEGETATION (VGT) 
                         sensor. For this purpose we used a VGT temporal series of 12 
                         monthly composite images, which were further transformed to 
                         physicalmeaningful fraction images of vegetation, soil, and shade. 
                         Classification of fraction images was implemented using several 
                         recent machine learning developments, namely, filtering input 
                         training data and probability bagging in a classification tree 
                         approach. A 10-fold cross validation accuracy assessment indicates 
                         that filtering and probability bagging are effective at increasing 
                         overall and class-specific accuracy. Overall accuracy and mean 
                         probability of class membership were 0.88 and 0.80, respectively. 
                         The map of probability of class membership indicates that the 
                         larger errors are associated with cerrado savanna and 
                         semi-deciduous forest.",
                 issn = "0099-1112",
             language = "en",
           targetfile = "Carreiras_etal_PERS2006.pdf",
        urlaccessdate = "12 maio 2024"
}