摘要
AimingtosolvethemisclassificationproblemsofunsupervisedpolarimetricWishartclassificationalgorithmbasedonFreemandecomposition,anunsupervisedPolarimetricSyntheticApertureRadar(SAR)Interferomery(PolInSAR)classificationalgorithmbasedonoptimalcoherencesetparametersisstudiedandproposed.ThisalgorithmusestheresultofFreemandecompositiontodividetheimageintothreebasiccategoriesincludingsurfacescattering,volumescattering,anddouble-bounce.Then,thePolInSARoptimalcoherencesetparametersareusedtofinelydivideeachofthethreebasiccategoriesinto9categories,andthewholeimageisdividedinto27categories.BecauseboththeFreemandecompositionresultandoptimalcoherencesetparametersindicatespecificscatteringcharacteristics,thewholeimageismergedinto16categoriesbasedonphysicalmeaning.Atlast,theWishartclusterisemployedtoobtainthefinalclassificationresult.Topreservethepurityofscatteringcharacteristics,pixelswithsimilarscatteringcharacteristicsarerestrictedtobeclassifiedwithotherpixels.Thefinalclassificationresultseffectivelyresolvethemisclassificationproblem,notonlythebuildingscanbeeffectivelydistinguishedfromvegetationinurbanareas,butalsotheroadiswelldistinguishedfromgrass.Inthispaper,theE-SARPolInSARdataofGermanAerospaceCenter(DLR),areusedtoverifytheeffectivenessofthealgorithm.
出版日期
2013年04月14日(中国期刊网平台首次上网日期,不代表论文的发表时间)