遥感影像分类英文文献
7026IEEETRANSACTIONSONGEOSCIENCEANDREMOTESENSING,VOL.52,NO.11,NOVEMBER

2014
Fig.1.Flowchartofthehybridobject-orientedCRFclassi cationframework.
Intheclassi cationfusionstep,thesegmentationis rstper-formedbytheconnected-componentlabelingalgorithm,usingtheCRF-LOGandCRF-QGclassi cationresultsobtainedinthesecondstep.Object-orientedclassi cationmapscanbethenobtainedbyamajorityvotingstrategy,usingthesegmentationmapandtheclassi cationmap.The nalclassi cationresultisachievedbyintegratingtheformerobject-orientedclassi ca-tionsattheobjectlevel.
twoclasses.Finally,intheclassi cation,amajorityvotingstrategyisused.Moredetailscanbefoundin[45].B.CRFClassi cationFramework
Consideranorderedsetofvariablesy={y1,y2,...,yN},whichdenotesthesetofspectralvectorsofanHSRimage,whereNisthetotalnumberofpixelsoftheimage,andyiisthespectralvectoroftheimagepixeli∈V={1,2,...,N}withlengthd(thenumberofspectralbands).Theclassi cationimageisdenotedbyx={x1,x2,...,xn},wherexiinxtakesthevalueofthelabelsetL={1,2,...,K}.Kisthenumberofclasses.
UnderBayes’framework,theimageclassi cationcancor-respondto ndingthemaximumaposteriori(MAP)estimateofthelabelimage,suchthatxMAP=argmaxxP(x|y).IfxconditionedonymeetstheMarkovproperty,CRFdirectlymodelstheposteriorprobabilityasaGibbsdistributionin(3).ThecorrespondingGibbsenergy[40],[41]isde nedas
ψc(xc,y).(4)E(x)= logP(x|y) logZ(y)=
c∈C
A.ProbabilisticPixelwiseClassi cation
The rststepoftheproposedframeworkconsistsofper-formingaprobabilisticpixelwiseclassi cationoftheHSR
imagery.Variousclassi ersthathaveaprobabilisticoutputcanbeused.However,theSVMclassi erissuggestedduetoitshighcapacityandgreateradaptability.Inotherwords,whetherinthecaseofhighdimensionsandasmalltrainingsetornot,SVMalwaysshowsagoodperformance.MoredetailsabouttheSVMclassi ercanbefoundin[45].Inthisstep,theclassi cationmapandthecorrespondingprobabilitymapwiththeposteriorprobabilitiesthatthesamplesbelongingtotheclassesareexpectedtoobtain.However,theoutputofgeneralSVM,viathedecisionfunction,isonlyaclasslabelwithoutprobabilityinformation.Inordertoobtaintheprobabilityes-timates,Platt’sformulationisusedtomaptheoutputintotheprobability[45],[46].Then,theremainingproblemishowtoexpandthealgorithmfrombinaryclassi cationtomulticlassclassi cation.The“one-against-one”approachissuggestedin[45].Assumingthatkisthenumberofclasses,weconstructk(k 1)/2classi erssothateachclassi ertrainsdatafrom
Therefore,theMAPlabelingxMAPoftherandom eldisgivenby
xMAP=argmaxP(x|y)=argminE(x).
x
x
(5)
Equation(5)showsthatthemaximizationoftheposteriorprobabilityP(x|y)isequivalenttotheminimizationoftheenergyfunctionE(x).AcommonlyusedexampleofaCRF
object as multiple instances and allowed the object to have multiple labels ... A Hybrid Conditional R... 11页 免费 Conditional Random Fie... 10页 ...
CVPR, 2007 Adaptive Contour Features in Oriented Granular Space for Human ... A Dynamic Conditional Random Field Model for Joint Labeling of Object and ...
A distributed object-oriented model for_专业资料。A distributed object-oriented model forA distributed object-oriented model for telecommunication services P....
A Haptic Object-Oriented Texture Rendering System Jason P. Fritz University of Delaware/A.I. duPont Institute Applied Science and Engineering Laboratories, ...
Object-Oriented Analysis 240页 免费 Designing test suites fo... 185页 ...We also introduce Frame-based Hidden-state Conditional Random Field model, a...
Even typical fact-oriented applications, such as information extraction, can ...er that combines a Conditional Random Field sequence tagging model (Lafferty ...
Even typical fact-oriented applications, such as information extraction, can ...er that combines a Conditional Random Field sequence tagging model (Lafferty ...
热门文档