Multichannel SAR Image Classification by Finite Mixtures, Copula Theory and Markov Random Fields

Vladimir Krylov (1, 2) , Gabriele Moser (3) , Sebastiano B. Serpico (3) , Josiane Zerubia (1)

1 ARIANA - Inverse problems in earth monitoring
2 Lomonosov Moscow State University - Faculty of Computational Mathematics and Cybernetics
3 DIBE - Department of Biophysical and Electronic Engineering [Genoa]

In this paper we develop a supervised classification approach for medium and high resolution multichannel synthetic aperture radar (SAR) amplitude images. The proposed technique combines finite mixture modeling for probability density function estimation, copulas for multivariate distribution modeling and a Markov random field (MRF) approach to Bayesian classification. The novelty of this research is in introduction of copulas to classification of D-channel SAR, with D>2, within the mainframe of finite mixtures - MRF approach. This generalization results in a flexible and well performing multichannel SAR classification technique. Its accuracy is validated on several multichannel Quad-pol RADARSAT-2 images and compared to benchmark classification techniques.

Mots clés: multichannel SAR; amplitude; classification; dictionary; probability density function estimation; Markov random field; copula

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Vladimir Krylov, Gabriele Moser, Sebastiano B. Serpico, Josiane Zerubia. Multichannel SAR Image Classification by Finite Mixtures, Copula Theory and Markov Random Fields. International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering (MaxEnt), Jul 2010, Chamonix, France. ⟨inria-00495557⟩