A method to incorporate uncertainty in the classification of remote

A method to incorporate uncertainty in the classification of remote

ID:39711679

大小:100.73 KB

页数:10页

时间:2019-07-09

A method to incorporate uncertainty in the classification of remote_第1页
A method to incorporate uncertainty in the classification of remote_第2页
A method to incorporate uncertainty in the classification of remote_第3页
A method to incorporate uncertainty in the classification of remote_第4页
A method to incorporate uncertainty in the classification of remote_第5页
资源描述:

《A method to incorporate uncertainty in the classification of remote》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、ISBN1-8462-xxxx-xProceedingofthe8thinternationalsymposiumonspatialaccuracyassessmentinnaturalresourcesandenvironmentalsciencesShanghai,June25-27,2008,pp.xxx-xxx(Willbesetbythepublisher)Amethodtoincorporateuncertaintyintheclassificationofremotesensingimages(1)(2,3)(4)(5,6)LuísaMS

2、Gonçalves,CidáliaFonte,EduardoNBSJúlio,MarioCaetano1PolytechnicInstituteofLeiria,SchoolofTechnologyandManagement,Portugal2InstituteforSystemsandComputersEngineeringatCoimbra,Portugal3DepartmentofMathematics,UniversityofCoimbra,P–3001454Coimbra,Portugal4ISISE,CivilEngineeringDepa

3、rtment,UniversityofCoimbra5PortugueseGeographicInstitute(IGP),RemoteSensingUnit(RSU),Lisboa,Portugal6CEGI,InstitutoSuperiordeEstatísticaeGestãodeInformação,ISEGI,UniversidadeNovadeLisboa,1070-312Lisboa,PortugalAbstract.Theauthorsanalyzeinthispaperwhethertheintroductionoftheuncer

4、taintyassociatedtotheclassificationofsurfaceelementsintheclassificationoflandscapeunitscanimprovetheresultsaccuracy.Tothisend,ahybridclassificationmethodisdeveloped,incorporatinguncertaintyinformationintheautomaticclassificationofveryhighspatialresolutionmultispectralsatelliteim

5、agestoobtainamapoflandscapeunits.Thedevelopedclassificationmethodologyincludesthefollowingsteps:1)asoftpixel-basedclassification;2)computationoftheclassificationuncertainty;3)imagesegmentation;and4)objectclassificationbasedondecisionrules.Thefirststepoftheproposedmethodologyisth

6、esoftpixel-basedclassificationperformedwiththemaximumlikelihoodclassifier,aimingtoidentifythesurfaceelements(e.g.,treecrown,shade,baresoil,buildings).Subsequentlytheposteriorprobabilitiesarecomputedtoallpixelsoftheimage.Thisinformationenablesthecomputationoftheclassificationunce

7、rtainty.Animagesegmentationisthenmadetoobtainimage-objects.Theclassificationoftheresultingobjectsintolandscapeunitsisperformedconsideringasetofdecisionrulesthatincorporatetheprobabilitiesassignedtotheseveralclassesateachpixelandthedegreeofuncertaintyassociatedtotheseassignments.

8、Theproposedmethodologywastestedontheclassificat

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。