基于TM数据的植被覆盖度反演

31120061ScienceofSurveyingandMappingVol31No1Jan:20050125:(2003DIA6N015)TM,,,,(,100039;,,266510)!∀本文首先对TM影像进行了几何纠正辐射校正大气校正;然后根据混合像元的结构特征,利用TM数据从植被指数(NDVI)中采用#等密度模型∃和#非密度模型∃提取了宜昌南部地区的植被覆盖度在用#非密度模型∃反演植被覆盖度的过程中,叶面积指数(LAI)是一个必要的参数,本文提出了一种改进的借助可见光波段和近红外波段反射值来提取叶面积指数(LAI)的方法通过和MODIS数据反演结果比较表明:#非密度模型∃的估算精度要高于#等密度模型∃;利用#等密度模型∃和#非密度模型∃反演植被覆盖度是可行!∀植被覆盖度;归一化植被指数;叶面积指数!∀TP79!∀A!∀10092307(2006)010043031[1],,[2],,,,,,,:[3],;,Gutman,,#∃,LAI,DN(LAI),,DN,[4],,TM#∃#∃#∃,,,TMMODIS2(NDVI),[5]NDVIfg()LAI()[3],NDVI[6],#∃#∃21#∃,fg1[7]BEER(NDVI)(LAI):NDVI=NDVI%-(NDVI%-NDVI0)exp(-kLAI)(1)NDVI0NDVI%(LAI&0)(LAI&%)NDVI22,:[7]221#∃,(LAI&%,NDVIg&NDVI%)[7]NDVI=fgNDVI%+(1-fg)NDVI0(2)fg=NDVI-NDVI0NDVI%-NDVI0(3)NDVI0NDVI,NDVI%NDVIGutman[8]NDVINDVI0,NDVI%,222#∃,(LAI∋%)[7]NDVI=fgNDVIg+(1-fg)NDVI0fg=NDVI-NDVI0NDVIg-NDVI0(4):NDVIg=NDVI%-(NDVI%-NDVI0)exp(-kLAI)#∃,NDVI0NDVI%kLAINDVI0NDVI%Gutman[7],k=1,LAI:#∃,LAI,3TM,#∃∃∃31:1(5,,,:,,)))[9]:DNL;L:6S[10],NDVI(0606557)(0723077),(0189955)(0340533)(1)16SNDVI32233LAI(LAI),LAI,LAI,,PRICE(LAI)[11]1)LAI:DNsi=DNi(e2cLAIi-r2%i)+DN%i(1-e2cILAI)1-r2%i∗e2ciLAI-DNi∗r2%i(1-e2ciLAI)/DN%i(5)(6):DNS2=a∗DNS1+b(6)LAIDN::R()=r%+Dr%/(1+D)(7)D=rs-r%(1/r%)-rs∗e-2c+LAI(8):r%:;s:;(8)(7):rsi=Ri(r2%i-e2ciLAI)+r%i(e2ciLAI-1)Rir%i(1-e2ciLAI)+r2%ie2ciLAI-1(9):Ri(i=1,2);:rs2=a∗rs1+b(10)(9)(10),LAIR2(r2%2-e2c2LAI)+r%i2(e2c2LAI-1)R2r%2(1-e2c2LAI)+r2%2e2c2LAI-1=a∗R1(r2%1-e2c1LAI)+r%i1(e2c1LAI-1)R1r%1(1-e2c1LAI)+r2%1e2c1LAI-1+b(11)2)(11);a,b;TM4()TM3(),3,ab,:rs2=118876∗rs1+001114767(12)3(,)PRICE[11],c1=06,c2=021,r%1=005,r%2=0073)LAILAI34,#∃#∃,,535MODIS#∃,MODISNDVI250m,44314LAI5#∃#∃LAI250m,#∃,,TM30m,250m;TMMODIS,;(),(6)6#∃#∃MODIS110%(a)004965601313040423500-0216001625860%(b)003819301096480199969-0216001684400%4,#∃#∃,(LAI)MODIS,62586%,,,(LAI),(LAI),(NDVI),(LAI)[1],[J]2003,8(11):13041309[2],[J],2004,26(4):01570164[3],[J],20015(6):416422[4],MODIS[J],2004,(1)[5][D]:,2003[6]TobyN,Carlson,DavidARipleyOntheRelationbetweenNDVIFractionalVegetationCoverandLeafAreaIndex[J]RemoteSensingofEnvironment,1997(62):241252[7]GGUTMANandAIGNATOVThederivationofthegreenvegetationfractionfromNOAA/AVHRRdataforuseinnumericalweatherpredictionmodels[J]RemoteSensing,1998,19(8):15331543[8]TobyN,Carlson,DavidARipleyOntheRelationbetweenNDVIFractionalVegetationCoverandLeafAreaIndex[J]RemoteSensingofEnvironment,1997(62):241252[9],,ETM+[J],2004,(2)[10],,TM)))[J],2004,20(2):3438[11]JohnCPriceEstimatingLeafAreaIndexfromSatelliteData[J]IEEETransactionsonGeoscienceandRemoteSensing,1993,31(3):727734:(1979),,,:,Emailding_0@126com451TMbiningsketchdispartandcollectionwaterlineoftheregionWeidentifyandclassifycharacterpointsusingdispartandcollectionwaterline,finallyextractRidgeandValleylinethemethodcombininggeometryandphysicsofRidgeorValleyline.TheresultsshowcoicidenceofRidgeandValleylinewithterrain.Keywords:douglas-peuker;ridgeline;valleyline;digita;lfeaturepointZHANGWeijun,KONGJinlingWANGWenkeWENGXiaopeng(ShoolofEarthScienceandResourcesManagement,Chang,anUniversity,Xi,an710054;ShoolofEnviromentScienceandEngineering,Chang'anUniversity,Xi'an710054)TheeffectsofICRSnon-conservativeforcesaccelerationbylinearinterpolationCHAMPsatelltedisconnectedattitudedataAbstract:ThispapermainlydiscussthediscussesofthedatagapsproussingattitudedataprovidedbyGFZAstatisticalresultof15daysattitudedatagapispresentedThevaretycharacterofattitudedataisanalysed,andputforwardthatthechangeofattitudequaternionisperiodicWehavecomparedtheattitudedatafromlinearinterpolationwithoriginaldataandgotsomeusefulresultsKeyword:CHAMPsatellite;attitudedata;non-conservativeforcesaccelerationSONGLei,PENGBiBo,ZHOUXuHua(KeyLaboratoryofDynamicalGeodesy,InstituteofGeodesyandGeophysics,ChineseAcademyofSciences,Wuhan430077;ThegraduatestudentdepartmentofChineseAcademyofSciences,Beijing100049)PatternrecognitionandanalysisofdeformationobjectinlandslideofYunyangbaotaAbstract:RecognizingthemovingblocksandmodelingthemovementofdeformationobjectisanimportantcontentindeformationanalysisThetraditionalmethodofthedeformationmodelrecognizingistodrawthemovingvectorsoftheobservedpoints,soastojudgethemovingblocksmanuallyandtoselectsuitablemovementmodelsThispaperintroducesanewideaofrecognizingthedeformationmodelsandcorrespondingalgorithm,whichisbasedonconstraintsoftopologicalrelationshipsandotheraprioriinformationThemodelrecognizingcanbeautomaticallyaccomplishedbythenewmethodFinallyanexampleisgivenKeywords:deformationpattern;similarityofdeformationvectors;topologicalconstrainingmatrix;classificationalgorithmunderconstraintLIUXuchun,,ZHANGZhenglu(WuhanUniversitySchoolofGeodesyandGeomatics,Wuhan430070;BeijingInstituteofCivilEngineeringandArchitectureSurveyingandMappingEngineeringDepartment,Beijing100044)AnalyticgeometryalgorithmofunderwaterGPSpositioningsystemAbstract:UnderwaterGPSpositioningsystemusesdistancedifferencebetweentworangesfromtargettoGPSbuoystoresolvethetargets'positionIts'veryeasytomakepositioningalgorithmdiversebyhyperbolas'intersectionInthispaper,ananalyticgeometryalgorithmisgivenindetails,whichisbasedonspacegeometricrelationshipbetwe