Centre de documentation
scientifique

Interpretation of remotely sensed data using guided techniques for land cover analysis / Julien Flack (1995)

Public Titre : Interpretation of remotely sensed data using guided techniques for land cover analysis Type de document : Thèse/HDR Auteurs : Julien Flack, Auteur Editeur : Perth [Australia] : Curtin University of Technology Année de publication : 1995 Importance : 248 p. Format : 21 x 30 cm Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] Australie occidentale (Australie) [Termes IGN] blé (céréale) [Termes IGN] classification contextuelle [Termes IGN] classification pixellaire [Termes IGN] extraction de traits caractéristiques [Termes IGN] méthode du maximum de vraisemblance (estimation) [Termes IGN] segmentation d'image [Termes IGN] système d'information géographique [Termes IGN] théorie de Dempster-Shafer Résumé : (auteur) An automated method for extracting information from remotely sensed data using guided segmentation is described. A task oriented approach to interpretation enables a greater flexibility for exploratory analysis, and is essential for providing a dynamic interface between remotely sensed data and other forms of spatial data within a Geographic Information Systems (GIS) environment. Segmentation is used as an effective means of investigating the relationship between spatial and spectral properties of the data with respect to contextual classification. A fundamental difficulty is the combination of evidence from multiple disparate sources. A method, based on the Dempster-Shafer evidential theory, is presented which combines evidence to effectively guide the segmentation towards a suitable solution. This combination is influenced using weights which are configured to provide a top-down or model based component to the data-driven segmentation. The results of this approach are evaluated with respect to the more traditional per pixel classifications such as the maximum likelihood and decision tree classifiers using ground reference data of an area of the West Australian wheat belt in order to quantify relative accuracies. Note de contenu : Introduction 1- Contextual classification 2- Preliminary data analysis 3- binding remotely sensed data to a GIS 4- Combining data sources 5- Guided segmentation 6- Results 7- Summary and conclusions Numéro de notice : 24568 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Thèse étrangère Note de thèse : PhD Thesis : : Perth - Australie : 1995 Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=92111

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
24568-01	THESE	Livre	Centre de documentation	Thèses	Disponible