Title :
Supervised High-Resolution Dual-Polarization SAR Image Classification by Finite Mixtures and Copulas
Author :
Krylov, Vladimir A. ; Moser, Gabriele ; Serpico, Sebastiano B. ; Zerubia, Josiane
Author_Institution :
Fac. of Comput. Math. & Cybern., Lomonosov Moscow State Univ., Moscow, Russia
fDate :
6/1/2011 12:00:00 AM
Abstract :
In this paper, a novel supervised classification approach is proposed for high-resolution dual-polarization (dual-pol) amplitude satellite synthetic aperture radar (SAR) images. A novel probability density function (pdf) model of the dual-pol SAR data is developed that combines finite mixture modeling for marginal probability density functions estimation and copulas for multivariate distribution modeling. The finite mixture modeling is performed via a recently proposed SAR-specific dictionary-based stochastic expectation maximization approach to SAR amplitude pdf estimation. For modeling the joint distribution of dual-pol data the statistical concept of copulas is employed, and a novel dictionary-based copula-selection method method is proposed. In order to take into account the contextual information, the developed joint pdf model is combined with a Markov random field approach for Bayesian image classification. The accuracy of the developed dual-pol supervised classification approach is validated and compared with benchmark approaches on two high-resolution dual-pol TerraSAR-X scenes, acquired during an epidemiological study. A corresponding single-channel version of the classification algorithm is also developed and validated on a single polarization COSMO-SkyMed scene.
Keywords :
Bayes methods; Markov processes; geophysical image processing; image classification; radar imaging; spaceborne radar; synthetic aperture radar; Bayesian image classification; Markov random field approach; PDF model; contextual information; dictionary-based copula-selection method; dual-polarization amplitude satellite synthetic aperture radar images; epidemiological study; finite mixture modeling; high-resolution dual-pol TerraSAR-X scenes; multivariate distribution modeling; probability density function model; single polarization COSMO-SkyMed scene; stochastic expectation maximization approach; supervised high-resolution dual-polarization SAR image classification; Data models; Dictionaries; Estimation; Image resolution; Joints; Mathematical model; Probability density function; Copula; Markov random field; dictionary-based pdf estimation; polarimetric synthetic aperture radar; probability density function (pdf); supervised classification;
Journal_Title :
Selected Topics in Signal Processing, IEEE Journal of
DOI :
10.1109/JSTSP.2010.2103925