Speckle reduction in PolSAR by multi-channel variance stabilization and Gaussian denoising: MuLoG
TUPIN, Florence
Image, Modélisation, Analyse, GEométrie, Synthèse [IMAGES]
Département Images, Données, Signal [IDS]
Image, Modélisation, Analyse, GEométrie, Synthèse [IMAGES]
Département Images, Données, Signal [IDS]
TUPIN, Florence
Image, Modélisation, Analyse, GEométrie, Synthèse [IMAGES]
Département Images, Données, Signal [IDS]
< Reduce
Image, Modélisation, Analyse, GEométrie, Synthèse [IMAGES]
Département Images, Données, Signal [IDS]
Language
en
Communication dans un congrès
This item was published in
EUSAR, EUSAR, EUSAR, 2018-06, Aachen. 2018-06
English Abstract
Due to speckle phenomenon, some form of filtering must be applied to SAR data prior to performing any polarimet-ric analysis. Beyond the simple multilooking operation (i.e., moving average), several methods have been ...Read more >
Due to speckle phenomenon, some form of filtering must be applied to SAR data prior to performing any polarimet-ric analysis. Beyond the simple multilooking operation (i.e., moving average), several methods have been designedspecifically for PolSAR filtering. The specifics of speckle noise and the correlations between polarimetric channelsmake PolSAR filtering more challenging than usual image restoration problems. Despite their striking performance,existing image denoising algorithms, mostly designed for additive white Gaussian noise, cannot be directly applied toPolSAR data. We bridge this gap with MuLoG by providing a general scheme that stabilizes the variance of the po-larimetric channels and that can embed almost any Gaussian denoiser. We describe MuLoG approach and illustrate itsperformance on airborne PolSAR data using a very recent Gaussian denoiser based on a convolutional neural network.Read less <
English Keywords
Speckle filtering
multiplicative noise
variance stabilization
polarimetry
Origin
Hal imported