基于高分辨率衛(wèi)星影像的城市道路陰影區(qū)域車(chē)輛信息提取研究.pdf

1、北京交通大學(xué)碩士學(xué)位論文基于高分辨率衛(wèi)星影像的城市道路陰影區(qū)域車(chē)輛信息提取研究姓名：郭杜杜申請(qǐng)學(xué)位級(jí)別：碩士專(zhuān)業(yè)：城市交通工程指導(dǎo)教師：梁艷平201106ABSTRACTABsTl認(rèn)CT：Hi曲resolutionsatellite僦lotesensing嘶ngsnewideast0m鋤傾伍ciIlfon】1ationC011ectionWimtheimprov鋤entofⅡlespatialreSolutionofremotesens

2、ingimage，tlleimageofVehiclesontheroadh嬲becomemorealldmorecle強(qiáng)Itc觚proVidemore仃a伍ci11fb鋤atioIl，whiChisagoodsourceof觸伍cflowdatamatisrequiredfortraIlsportpl甜lIlingandmanagementButinlli曲densityurbanenvm)mnent，buildingshadowsb

3、lockedalotofimportantinfomlatioIl，whichresultininaccurateresultsOreVener】r01葛ofimageinterpretationThesenegatiVee虢ctswillreducemeaCquisitionaccuracyofVehicleinfo冊(cè)ation，oreVeIlblocktheprocessofhigh—resolutionsatellite撒note

4、sensingintheapplicationofurb嬲t啪sponinfo肌ationcollectionT1lisp印eraimstostudythetechn0109)rofmeVehicleinfomationex仃actionoftheshadedarea丘omhi曲一resolutionremoteseIlsingimagesThemainresearchcontentsa1easf0nows：(1)CombiIlir培o

5、fshadowimagecharacteristicsofremotesensingarldt11ene。dsofmisanicle，t量lispapers煳撕zesmethodsarldproceduresofimageshadowdetectionbaSedonbri曲t11essaIldshadoweliminationoflinearm印pinggrayscale；鋤dteststhes鋤pledataofshadowsarea

6、oftllehi曲spatialresolution懈notesensingimagetov硎匆meen’ectivellessofthemetllod；a11dprovidesdataSupponforthelaterc|咖actionofmevellicles(2)AgainsttomeincompleteVehicleinfo蛐ationofshadowarea，“spaperproposed也ememodtllatcombine

7、dclassificationofVdliclemorpholo百calmleandHSVcolorspaceaIlalysisarIdtheideaofthemethodaIldimpl鋤entationstepsareconstmctedFirst，itusedthemethodbasedonVehicleclassificationtoextracttheimagedataofcomprehensiVeVehicleaIldthe

8、nusedmethodofHSVc010rspaceanalysistoextractmeremainingponionofmeVehiclebemissedThecombinationofthesetwomethodstoeX劬ctthevehiclesofshadowre西oncallgetbetterresults(3)AccordingtoVehicleinfomationextractionmethodsandst印swhiC

9、hhasbeellbuiltinthechapt％tllispaperselectstheW6rldView一2andGeoeyelimagesaRershadowdetectionaIldelimination硒meexp如饑taldatat0extractVehiclesandveri母mee仃ectiVenessofthemethod，theIlusesthe研oindicatorsofVehideex仃actioncorrect