Summary: | SeaicetypeclassificationofSentinel-1SARdualpolarizationEWimages. Priortoclassification,Sentinel-1SARimageshavebeenpre-processedtoremovenoise[1].Theprocedureofclassificationiscomprisedofthefollowingsteps:1.Haralicktexturefeaturesarecomputed,2.PrincipalComponentAnalysis(PCA)isapplied,3.K-meansclusteringisusedtogroupthedatainto15clustersusingfactorscoresofthePCAasinput,4.The15clustersareanalysedbyaniceexpertandclassifiedasopenwateranddifferenttypesofseaice. [1]J.-W.Park,A.A.Korosov,M.Babiker,S.Sandven,J.-S.Won,EfficientThermalNoiseRemovalforSentinel-1TOPSARCross-PolarizationChannel,IEEETransactionsonGeoscienceandRemoteSensing,2018;56,3,DOI:10.1109/TGRS.2017.2765248.
|