Dictionary-based Stochastic Expectation-Maximization for SAR amplitude probability density function estimation

Gabriele Moser (1) , Josiane Zerubia , Sebastiano B. Serpico

1 ARIANA - Inverse problems in earth monitoring

In the context of remotely sensed data analysis, a crucial problem is represented by the need to develop accurate models for the statistics of the pixel intensities. In the current research report, we address the problem of parametric probability density function (PDF) estimation in the context of Synthetic Aperture Radar (SAR) amplitude data analysis. Specifically, several theoretical and heuristic models for the PDFs of SAR data have been proposed in the literature, and have been proved to be effective for different land-cover typologies, thus making the choice of a single optimal SAR parametric PDF a hard task. In thia report, an innovative estimation algorithm is proposed, which addresses this problem by adopting a finite mixture model (FMM) for the amplitude PDF, with mixture components belonging to a given dictionary of SAR-specific PDFs. The proposed method automatically integrates the procedures of selection of the optimal model for each component, of parameter estimation, and of optimization of the number of components, by combining the Stochastic Expectation Maximization (SEM) iterative methodology and the recently proposed «method-of-log-cumulants» (MoLC) for parametric PDF estimation for non-negative random variables. Experimental results on several real SAR images are presented, showing the proposed method is accurately modelling the statistics of SAR amplitude data.

Mots clés: SYNTHETIC APERTURE RADAR (SAR) IMAGES; PROBABILITY DENSITY FUNCTION ESTIMATION; PARAMETRIC ESTIMATION; FINITE MIXTURE MODELS; STOCHASTIC EXPECTATION MAXIMIZATION (SEM)

Domaines: Autre [cs.OH]

Gabriele Moser, Josiane Zerubia, Sebastiano B. Serpico. Dictionary-based Stochastic Expectation-Maximization for SAR amplitude probability density function estimation. RR-5154, INRIA. 2004. ⟨inria-00071429⟩