Centre de documentation
scientifique

Land-cover mapping in the Brazilian amazon using SPOT-4 Vegetation data and machine learning classification methods / João M.B. Carreiras in Photogrammetric Engineering & Remote Sensing, PERS, vol 72 n° 8 (August 2006)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 72 n° 8 (August 2006) . - pp 897 - 910 Titre : Land-cover mapping in the Brazilian amazon using SPOT-4 Vegetation data and machine learning classification methods Type de document : Article/Communication Auteurs : João M.B. Carreiras, Auteur ; J.M.C. Pereira, Auteur ; Y.E. Shimabukuro, Auteur Année de publication : 2006 Article en page(s) : pp 897 - 910 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse diachronique [Termes IGN] apprentissage automatique [Termes IGN] carte d'occupation du sol [Termes IGN] cartographie numérique [Termes IGN] classification ascendante hiérarchique [Termes IGN] image SPOT-Végétation [Termes IGN] Mato Grosso [Termes IGN] occupation du sol Résumé : (Auteur) 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 physical-meaningful 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 savonna and semi-deciduous forest. Copyright ASPRS Numéro de notice : A2006-313 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.72.8.897 En ligne : https://doi.org/10.14358/PERS.72.8.897 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=28037 [article]