Object-oriented Landcover Classification of Multi-source Remote Sensing Data in International Trans-boundary River
-
摘要: 作为横跨3个国家(尼泊尔、印度、中国)的国际跨界河流———柯西河流域,地形高差巨大,土地覆被结构组成复杂,进行土地覆被的自动分类研究具有典型意义。基于面向对象方法多源遥感数据、训练规则、丰富的细节信息为复杂土地覆被自动分类研究提供了可能。选择合适的影像分割特征和最优分割尺度,按照数据挖掘中的规则顺序逐步进行各个土地覆被的提取。总体精度说明分类结果与野外点相一致的概率能达到90.05%,说明国际跨界河流土地覆被分类方法是可行的,分类结果是准确、可信的。Abstract: Landcoverclassificationisnoteasyforitsinnercomplexityresultingfromhugeterrainele vationinKosiRiverasinternationaltrans boundaryriver,ithaspossibletoclassificationdifficultiesflowingthroughthreecountries.Theemergencyofmulti sourceremotesensingimagesandtrainingalgorithmmakeitpossibletoclassifythelandcoverforitsampledetailsbasedonobject orientedmethod.Thepaperisaboutlandcoverclassificationmethodselectingfeatureandoptimalscaleofseg mentation,theinnovationofthismethodliesinselectionofproperscaleparameterresultingfromproperimagedataandcertainclassificationorder.Thetotalaccuracyhashighly90.05%comparedwithobjectorientedclassificationresultandactualsamplingpoint,whichisafeasiblemethod,andtheclassificationresultsaremoreaccurate.
-
[1] LiZhifei.Trans boundaryRiverIssuesandtheChi nesePeripherySecurity[J].犃犮犪犱犲犿犻犮犈狓狆犾狅狉犪 狋犻狅狀，２０１１（１）：１１３（李志斐．跨国界河流问题与中国周边关系［Ｊ］．学术探索，２０１１（１）：１１３）[2] ChenYunhao,FengTong,ShiPeijun,etal.Clas sificationofRemotSensingImageBasedonObjectOrientedandClassRules[J].犌犲狅犿犪狋犻犮狊犪狀犱犐狀犳狅狉 犿犪狋犻狅狀犛犮犻犲狀犮犲狅犳犠狌犺犪狀犝狀犻狏犲狉狊犻狋狔，２００６，３１（４）：３１６３２０（陈云浩，冯通，史培军，等．基于面向对象和规则的?杏跋穹掷嘌芯浚郏剩荩浜捍笱аПāば畔⒖蒲О?2006,31(4):316 320)[3] ChenQihao.ResearchandRealizationofMulti SourcesRemoteSensingDataonObjectOriented[D].Wuhan:ChinaUniversityofGeoseiences,2007(陈启浩.面向对象的多源遥感数据分类技术研究与实现[D].武汉:中国地质大学,2007)[4] SuWei,LiJing,ChenYunhao,etal.Object orien tedUrbanLand coverClassificationofMulti scaleImageSegmentationMethod———ACaseStudyinKualaLumpurCityCenter,Malaysia[J].犑狅狌狉狀犪犾狅犳狉犲犿犪狋犲狊犲狀狊犻狀犵，２００７，１１（４）：５２１５２９（苏伟，李京，陈云浩，等．基于多尺度影像分割的面向对象城市土地覆被分类研究———以马来西亚吉隆坡市城市中心区为例［Ｊ］．?醒Пǎ玻埃埃罚保保ǎ矗海担玻豹?29)[5] SunDanfeng,YangJihong,LiuShunxi.Applica tionofHigh SpatialIKNOSRemoteSensingImagesinLandUseClassficationandChangeMonitoring [J].犜狉犪狀狊犪犮狋犻狅狀狊狅犳狋犺犲犆犛犃犈，２００２，１８（２）：１６０ 164(孙丹峰,杨冀红,刘顺喜.高分辨率遥感卫星影像在土地利用分类及其变化监测的应用研究[J].农业工程学报,2002,18(2):160 164)[6] BlaschkET,LangS,LorupE,etal.Object orien tedImageProcessinginanIntegratedGIS/RemoteSensingEnvironmentandPerspectivesforEnviron mentalApplications[C].EnvironmentalInformationforPanning,PoliticsandthePublicMetropolisVer lag,Marburg,2000[7] MauroC,EufemiaT.AccuracyAssessmentofPer fieldClassificationIntegratingveryFineSpatialRes olutionSatelliteImagerywithTopographicData[J].犑狅狌狉狀犪犾狅犳犌犲狅狊狆犪狋犻犪犾犈狀犵犻狀犲犲狉犻狀犵，２００１，３（２）：127 134[8] BenzUC,HofmannP,IllhauckWG,etal.Multi resolution,Object orientedFuzzyAnalysisIsofRe moteSensingDataforGISReadyInformation[J].犐犛犘犚犛犑狅狌狉狀犪犾狅犳犘犺狅狋狅犵狉犪犿犿犲狋狉狔犪狀犱犚犲犿狅狋犲犛犲狀狊犻狀犵，２００４，５８：２３９２５８[9] LoboA,ChicO,CasteradA.ClassificationofMediterraneanCropswithMulti sensorData:Per pixelVersusPer objectStatisticsandImageSeg mentation[J].犐狀狋犲狉狀犪狋犻狅狀犪犾犑狅狌狉狀犪犾狅犳犚犲犿狅狋犲犛犲狀狊犻狀犵，１９９６，１７：２３５８２４００[10] BaatzM,SchapeA.Object OrientedandMulti ScaleImageAnalysisinSemanticNetworks[C].The2ndInternationalSymposiumonOperationali zationofRemoteSensing,Ensehede,1999[11] BaatzM,SchapeA.Multi resolutionSegmenta tion:AnOptimizationApproachforHighQualityMulti scaleImageSegmentation[C].ProceedingsofAGIT’00,Heidelberg,Germany,2000[12] BhaskaranS,ParamanandaS,RamnarayanM.Per pixelandObject OrientedClassificationMethodsforMappingUrbanFeaturesUsingIKONOSSatelliteData[J].犃狆狆犾犻犲犱犌犲狅犵狉犪狆犺狔，２０１０，３０：６５０６６５[13] YuHuan,ZhangShuqing,KongBo,etal.OptimalSegmentationScaleSelectionforObject orientedRe moteSensingImageClassification[J].犑狅狌狉狀犪犾狅犳犐犿犪犵犲犪狀犱犌狉犪狆犺犻犮狊，２０１０，１５（２）：３５２３６０（于欢，张树清，孔博，等．面向对象?杏跋穹掷嗟淖钣欧指畛叨妊≡裱芯浚郏剩荩泄枷裢夹窝Пǎ玻埃保埃?5(2):352 360)[14] LiHui,HuXiaomei.AnalysisandComparisonBe tweenID3AlgorithmandC4.5AlgorithminDeci sionTree[J].犠犪狋犲狉犚犲狊狅狌狉犮犲狊犪狀犱犘狅狑犲狉，２００８，２６（２）：１２９１３２（李会，胡笑梅．决策树中ＩＤ３算法与Ｃ４．５算法分析与比较［Ｊ］．水电能源科学，2008,26(2):129 132) -
点击查看大图
计量
- 文章访问数: 900
- HTML全文浏览量: 24
- PDF下载量: 441
- 被引次数: 0
下载:
