Modeling synthetic aperture radar image data

AUTOR(ES)
DATA DE PUBLICAÇÃO

2008

RESUMO

In this thesis we study maximum likelihood estimation (MLE) of the roughness parameter of the G 0 A distribution for speckled imagery (Frery et al., 1997). We discover that when a certain criteria is satisfied by the sample moments, the likelihood function is monotone and MLE estimates are infinite, implying an extremely homogeneous region. We implement four bias correcting estimators in an attempt to obtain finite MLE estimates. Three of the estimators are taken from the literature on monotone likelihood (Firth, 1993; Jeffreys, 1946) and one, based on resampling, is proposed by the author. We perform Monte Carlo experiments to compare the four estimators and find that there is no clear favorite, except when one of the parameters (which is given before estimation) takes on a specific value. We also apply the estimators to real data obtained from synthetic aperture radar (SAR). It becomes clear from this analysis that the estimators need to be compared based on their ability to classify regions correctly as rough, smooth, or intermediate and not on their biases and mean squared errors

ASSUNTO(S)

speckle matematica da computacao g 0 a distribution maximum likelihood distribuiÃÃo g 0 a coherent imaging radar de abertura sintÃtica (sar) reamostragem resampling speckle imagens coerentes bootstrap synthetic aperture radar (sar) mÃxima verossimilhanÃa bootstrap verossimilhanÃa monÃtona monotone likelihood

ACESSO AO ARTIGO

Documentos Relacionados