Multiscale classification of very high resolution SAR images of urban areas by Markov random fields, copula functions, and texture extraction - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2012

Multiscale classification of very high resolution SAR images of urban areas by Markov random fields, copula functions, and texture extraction

Abstract

This paper addresses the problem of classifying very high resolution synthetic aperture radar (SAR) images of urban areas by using a supervised Bayesian classification method via a contextual hierarchical approach. We develop a bivariate copula-based statistical model that combines amplitude SAR data and textural information. This model is plugged into a hierarchical Markov random field based on a quadtree structure and on multiscale wavelet features. The contribution of this paper is thus the development of a novel hierarchical classification approach that uses a quadtree model based on the wavelet decomposition and an innovative statistical model. The performance of the developed approach is illustrated on a COSMO-SkyMed image of a urban area.
Fichier principal
Vignette du fichier
gtti12_submission_23.pdf (779.84 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00727404 , version 1 (03-09-2012)

Identifiers

  • HAL Id : hal-00727404 , version 1

Cite

Aurélie Voisin, Vladimir Krylov, Gabriele Moser, Sebastiano B. Serpico, Josiane Zerubia. Multiscale classification of very high resolution SAR images of urban areas by Markov random fields, copula functions, and texture extraction. GTTI - Riunione annuale dell'associazione Gruppo nazionale Telecomunicazioni e Tecnologie dell'Informazione, Jun 2012, Cagliari, Italy. ⟨hal-00727404⟩

Collections

INRIA INRIA2
175 View
169 Download

Share

Gmail Facebook Twitter LinkedIn More