Classification of very high resolution SAR images of urban areas by dictionary-based mixture models, copulas and Markov random fields using textural features - Archive ouverte HAL Access content directly
Conference Papers Year : 2010

Classification of very high resolution SAR images of urban areas by dictionary-based mixture models, copulas and Markov random fields using textural features

Abstract

This paper addresses the problem of the classification of very high resolution (VHR) SAR amplitude images of urban areas. The proposed supervised method combines a finite mixture technique to estimate class-conditional probability density functions, Bayesian classification, and Markov random fields (MRFs). Textural features, such as those extracted by the greylevel co-occurrency method, are also integrated in the technique, as they allow to improve the discrimination of urban areas. Copulas are applied to estimate bivariate joint class-conditional statistics, merging the marginal distributions of both textural and SAR amplitude features. The resulting joint distribution estimates are plugged into a hidden MRF model, endowed with a modified Metropolis dynamics scheme for energy minimization. Experimental results with COSMO-SkyMed and TerraSAR-X images point out the accuracy of the proposed method, also as compared with previous contextual classifiers.
Nous nous intéressons au problème de la classification d'images d'amplitude SAR très haute résolution, qui contiennent des zones urbaines. La méthode de classification supervisée proposée ici combine une estimation des fonctions de densité de probabilité, correspondant aux statistiques de chacune des classes envisagées, avec des champs de Markov. L'extraction de textures (e.g. GLCM) à partir de l'image SAR permet d'améliorer la classification par la discrimination des zones urbaines. L'introduction de copules permet le calcul d'une fonction de densité de probabilité conjointe pour chacune des classes à partir des densités marginales de l'image d'amplitude SAR et de sa texture, obtenues par calculs préalables. Ces estimations des densités conjointes, utiles pour l'apprentissage, sont introduites dans un modèle de Markov caché en vue d'établir la classification.
Fichier principal
Vignette du fichier
Classification_of_VHR_SAR_SPIE_sept2010_Toulouse_Voisin.pdf (2.53 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

inria-00516333 , version 1 (09-09-2010)

Identifiers

Cite

Aurélie Voisin, Gabriele Moser, Vladimir Krylov, Sebastiano B. Serpico, Josiane Zerubia. Classification of very high resolution SAR images of urban areas by dictionary-based mixture models, copulas and Markov random fields using textural features. SPIE Remote Sensing, Sep 2010, Toulouse, France. ⟨10.1117/12.865023⟩. ⟨inria-00516333⟩
484 View
631 Download

Altmetric

Share

Gmail Facebook X LinkedIn More