期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2010
卷号：XXXVIII - Part 7A
页码：175-180
出版社：Copernicus Publications
摘要：The objective of this research is to evaluatemulti-temporalRADARSAT-2 polarimetric SAR data for urban land-cover classificationusing a novel classification scheme.Six-date RADARSAT-2 Polarimetric SAR datain both ascending and descending orbitswereacquired during Juneto September2008 in the rural-urban fringe of the Greater Toronto Area. The major land-cover types are built-up areas, roads, golf courses, forest, water and several types of agricultural crops.In this research, the different urban land-covertypes and their corresponding polarimetric behaviorswerestudied. The polarimetric signaturesof the various urban land-cover typeswereextracted fromtheRADARSAT-2 SAR images and analyzed using a new hierarchicalmultitemporalclassification method.Theresults showed that the newclassificationmethodyieldedhigh classificationaccuracy, withoverall accuracyof82.1%andKappacoefficient 0.80forthe major11land-cover classes. Theclassificationscheme can effectively extract the urban structuresbymapping urban related classessuch asstreetsandmajorroadswith the higher user's accuracy, which isdifficult to achieve usingasingle-date data
