%0 Book Section
%@holdercode {isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S}
%3 MONDOZA, Procesamiento de imágenes ASTER.pdf
%X El objetivo de este estudio es demostrar Ia aplicabilidad del algoritmo de síntesis genética basado en una red neuronal artificial no supervisada ART2 (Adaptative Resonance Theory) en el tratamiento, de imágenes dei ASTERITERRA, para Ia clasificacióii temática del uso y cobertura de Ia tierra para los aflos 2002 y 2003. El área de estudio está situada cri Ia región norte dei Estado de Mato Grosso - Brasil, siendo caracterizada por un fuerte proceso de ocupacióti humana, que ha causado cambios intensivos en el paisaje por Ia deforestacióti, Ia tala selectiva de árboles y por el incremento de Ia frontera agrícola. EI resultado final. de Ia clasificaciótt fue evaluado posteriormente con datos levantados cri campo mediante el método de estadística Kappa. La cornbinación de bandas 2(630-690 rim), 3(760-860 rim) y 4(1600-1700 nm) del ASTER permiticron una diferenciación creciente de clases temáticas; por otro lado, Ias bandas 8 (2295-2365 rim) y 6 (2185-2225 rim) proporcionaron informaciones complementarias importantes para Ia identificación de Ias clases. La comparación de Ia elasificación obtenida por Ia red neuronal con Ias informaciones obtenidas en campo, fue considerada satisfactoria (Kappa = 0,64) para Ias elases i emáticas investigadas: bosque primario, bosque con explotación maderera reciente, bosque degradado, regeneración, cultivos, pastos y suelos expuestos. El uso de los datos ASTER contribuyó significativamente para alcanzar un conocimiento adecuado sobre Ia dinâmica del proceso de ocupación en esta área tropical. ABSTRACT: The objective of this study is to show the applicability of the genetic synthesis of the unsupervised artificial neural network ART2 (Adaptive Resonance Theory) in the classification of ASTER image for land uselland cover mapping. The area under study is located in northem Mato Grosso State, Brazil, and is characterized by a strong human occupation process, which caused intensive changes ai the landscape, by deforestation, selective logging and agriculture. Field data were acquired in May/June 2003. The use of ASTER data allowed an improvement of the analysis of the occupation process in tropical forest areas. ASTER images have adequate spatial and spectral resolution and are an alternative to the remaining remote sensing data available. The data had a correction of the cross-taik problem, after realized a resampling from SWIR bands (spatial resolution 30 to 15 m), an atmospherie correction and rectification of ASTER images from both data sets 2002 e 2003. The input parameters for the neural network ART2 were optimized by genetic algorithm and the neural network was evaluated by a comparison of classification results with field data. The evaluation of accuracy was done using Kappa statistics. The results of the classification were of satisfactory quality. ASTER bands 2 (630-690 rim), 3 (760-860 rim) and 4 (1600-1700 rim) allowed an increased différentiation of classes, while bands 8 (2295-2365 rim) and 6 (2185-2225 rim) were complementary for the identification of classes. The main land use changes that occurred between 2002 and 2003 were related to deforestation, since many areas of tropical forest were replaced by agriculture and pastures.
%E Santos, João Roberto dos,
%E Disperati, Attilio Antonio,
%T Procesamiento de imágenes ASTER con red neuronal ART2 para el análisis temporal del uso y cobertura de la tierra en la Amazônia brasilenã
%@isbn 85-904724-1-8
%@secondarytype PRE LI
%K VEGETAÇÃO, Mato Grosso (Estado), Amazônia (Região), uso da terra, Adaptative Resonance Theory (ART2), Advanced Spaceborne Thermal Emission Reflection Radiometer (ASTER), artificial neural network,  Amazon, land use, neural network.
%@nexthigherunit 8JMKD3MGPCW/3ER446E
%B Aplicações de geotecnologias na engenharia florestal.
%@usergroup administrator
%@usergroup jefferson
%@group
%@group DSR-INPE-MCT-BR
%@e-mailaddress jroberto@ltid.inpe.br
%C Curitiba
%@copyholder SID/SCD
%@secondarykey INPE-12188-PRE/7523
%2 sid.inpe.br/sergio/2005/02.21.11.17.22
%@affiliation WWF - Perú, Calle Trinidad Moran 895, Lima, Perú.
%@project Sensoriamento Remoto Aplicado à Ecossistemas Terrestres
%I Copiadora Gabardo Ltda
%P 78-85
%4 sid.inpe.br/sergio/2005/02.21.11.17
%D 2004
%@documentstage not transferred
%V v.1
%O Comissão Organizadora do VI Seminário de Atualização em SensoriamentoRemoto e Sistemas de Informações Geográficas Aplicados à Engenharia Florestal.
%A Mendoza, Eddy H. R.,
%A Santos, João Roberto dos,
%A Rosa, A. N. C. Santa,
%A Silva, N. C. da,
%@dissemination NTRSNASA; BNDEPOSITOLEGAL.
%@ 85-904724-1-8
%@area SRE