Parametric and nonparametric tests for speckled im

Authors: Renato J Cintra Alejandro C Frery Abraão D C Nascimento

Publish Date: 2011/10/15

Volume: 16, Issue: 2, Pages: 141-161

PDF Link

Abstract

Synthetic aperture radar SAR has a pivotal role as a remote imaging method Obtained by means of coherent illumination SAR images are contaminated with speckle noise The statistical modeling of such contamination is well described according to the multiplicative model and its implied fancyscriptG0 distribution The understanding of SAR imagery and scene element identification is an important objective in the field In particular reliable image contrast tools are sought Aiming the proposition of new tools for evaluating SAR image contrast we investigated new methods based on stochastic divergence We propose several divergence measures specifically tailored for fancyscriptG0 distributed data We also introduce a nonparametric approach based on the Kolmogorov–Smirnov distance for fancyscriptG0 data We devised and assessed tests based on such measures and their performances were quantified according to their test sizes and powers Using Monte Carlo simulation we present a robustness analysis of test statistics and of maximum likelihood estimators for several degrees of innovative contamination It was identified that the proposed tests based on triangular and arithmeticgeometric measures outperformed the Kolmogorov–Smirnov methodologySimulations of the h ϕdistance based parametric tests were performed in Ox programming environment 10 function QAGI was employed for numerical integration The Kolmogorov–Smirnov test was coded in R and function kstest was called 35 The computation time of the triangular and arithmeticgeometric distances took typically less than one and four millisecond respectively when performed in a Pentium processor at 320 GHz We used the George Marsaglia’s multiplywithcarry with 52 bits pseudorandom number generator which has an approximate period of 28222

Keywords: