MuLoG, or How to apply Gaussian denoisers to multi-channel SAR speckle reduction[2017][IEEE Trans Image Process][10.1109tip.2017.2713946]

《MuLoG, or How to apply Gaussian denoisers to multi-channel SAR speckle reduction[2017][IEEE Trans Image Process][10.1109tip.2017.2713946]》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、Thisarticlehasbeenacceptedforpublicationinafutureissueofthisjournal,buthasnotbeenfullyedited.Contentmaychangepriortofinalpublication.Citationinformation:DOI10.1109/TIP.2017.2713946,IEEETransactionsonImageProcessingJOURNALOFLATEXCLASSFILES,VOL.14,NO.8,AUGUST20151MuLoG,orHo

2、wtoapplyGaussiandenoiserstomulti-channelSARspecklereduction?Charles-AlbanDeledalle,Lo¨ıcDenis,SoniaTabti,FlorenceTupin,SeniorMember,IEEEAbstract—Specklereductionisalongstandingtopicinsyntheticwholelineofdenoisingmethodshavebeendesignedspecif-apertureradar(SAR)imaging.Sinc

3、emostcurrentandplannedicallyforspecklereduction.RestorationofintensityimagesSARimagingsatellitesoperateinpolarimetric,interferometricor(i.e.,single-channelSARimages)remainsthefirstandmosttomographicmodes,SARimagesaremulti-channelandspecklestudiedcase.Manydifferentschemesha

4、vebeenproposedreductiontechniquesmustjointlyprocessallchannelstorecoverpolarimetricandinterferometricinformation.Thedistinctivena-inthiscontext,seetherecentreviews[4]and[18].AmongtureofSARsignal(complex-valued,corruptedbymultiplicativethedifferentpossiblestrategies,select

5、ion-basedmethodsusefluctuations)callsforthedevelopmentofspecializedmethodsvariouscriteriatoidentifyhomogeneouscollectionsofneigh-forspecklereduction.Imagedenoisingisaveryactivetopicboringpixels.Pixelsmaybeselectedbylocallychoosingtheinimageprocessingwithawidevarietyofappro

6、achesandbestneighborhoodwithinafixedsetoforientedwindows[44].manydenoisingalgorithmsavailable,almostalwaysdesignedforadditiveGaussiannoisesuppression.ThispaperproposesaAnotherapproachconsistsofselectingconnectedpixelsbygeneralscheme,calledMuLoG(MUlti-channelLOgarithmwithre

7、giongrowing[67].Patch-comparisonhasbeenshowntoGaussiandenoising),toincludesuchGaussiandenoiserswithinprovidearobustwaytogeta(weighted)selectionofsimilaramulti-channelSARspecklereductiontechnique.Anewfamilypixels[22],[18].Variationalmethodsformulatetheestimationofspecklere

8、ductionalgorithmscanthusbeobtained,benefitingproblemasanoptimizationproblemdependingonthewholefro