Centre de documentation
scientifique

Model-based prediction error uncertainty estimation for K-NN method / H.J. Kim in Remote sensing of environment, vol 104 n° 3 (15/10/2006)  

Public [article]  inRemote sensing of environment > vol 104 n° 3 (15/10/2006) . - pp 257 - 263 Titre : Model-based prediction error uncertainty estimation for K-NN method Type de document : Article/Communication Auteurs : H.J. Kim, Auteur ; Erkki Tomppo, Auteur Année de publication : 2006 Article en page(s) : pp 257 - 263 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] Betula (genre) [Termes IGN] classification barycentrique [Termes IGN] erreur moyenne quadratique [Termes IGN] Finlande [Termes IGN] image Landsat-TM [Termes IGN] inventaire forestier étranger (données) [Termes IGN] Picea abies [Termes IGN] Pinus (genre) [Termes IGN] Populus (genre) [Termes IGN] variogramme Résumé : (Auteur) The k-nearest neighbour estimation method is one of the main tools used in multi-source forest inventories. It is a powerful non-parametric method for which estimates are easy to compute and relatively accurate. One downside of this method is that it lacks an uncertainty measure for predicted values and for areas of an arbitrary size. We present a method to estimate the prediction uncertainty based on the variogram model which derives the necessary formula for the k-nn method. A data application is illustrated for multi-source forest inventory data, and the results are compared at pixel level to the conventional RMSE method. We find that the variogram model-based method which is analytic, is competitive with the RMSE method. Numéro de notice : A2006-414 Affiliation des auteurs : non IGN Thématique : FORET/IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.rse.2006.04.009 En ligne : https://doi.org/10.1016/j.rse.2006.04.009 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=28138 [article]