Bayesian Approach to Image Interpretation.pdf

《Bayesian Approach to Image Interpretation.pdf》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、BayesianApproachtoImageInterpretationTHEKLUWERINTERNATIONALSERIESINENGINEERINGANDCOMPUTERSCIENCEBAYESIANAPPROACHTOIMAGEINTERPRETATIONSUNILK.KOPPARAPUResearchandDevelopmentGroupAquilaTechnologiesPrivateLimitedBangalore-560068.Indiakopparapu@bigfoot.comUDAYB.DESAISPANNLab.Departme

2、ntofElectricalEngineerinIndianInstituteofTechnology-BombayPowai,Mumbai400076.Indiaubdesai@ee.iitb.ernet.inKluwerAcademicPublishersNewYork/Boston/Dordrecht/London/MoscoweBookISBN:0-306-46996-0PrintISBN:0-792-37372-3©2002KluwerAcademicPublishersNewYork,Boston,Dordrecht,London,Mosc

3、owPrint©2000KluwerAcademic/PlenumPublishersNewYorkAllrightsreservedNopartofthiseBookmaybereproducedortransmittedinanyformorbyanymeans,electronic,mechanical,recording,orotherwise,withoutwrittenconsentfromthePublisherCreatedintheUnitedStatesofAmericaVisitKluwerOnlineat:http://kluw

4、eronline.comandKluwer'seBookstoreat:http://ebooks.kluweronline.comContentsListofFiguresixListofTablesxiPrefacexiiiAcknowledgmentsxv1.OVERVIEW11.Introduction11.1.ScopeofComputerVision22.ImageInterpretation23.LiteratureReview54.Approaches75.LayoutoftheMonograph92.BACKGROUND111.Int

5、roduction112.MarkovRandomFieldModels132.1.MRFModelandGibbsDistribution132.2.MaximumEntropyInterpretation172.3.Application:EstimatingImageAttribute182.4.PosteriorDistribution192.5.DetectingDiscontinuities:LineFieldModel222.6.Minimizingposteriorienergy233.Multiresolution243.1.Gaus

6、sianandLaplacianPyramid263.2.WaveletPyramid29viBAYESIANAPPROACHTOIMAGEINTERPRETATION3.MRFFRAMEWORKFORIMAGEINTERPRETATION351.MRFonaGraph351.1.MRFbasedFormulationforInterpretation371.2.SelectionofCliquesFunctions394.BAYESIANNETAPPROACHTOINTERPRETATION431.Introduction432.MRFmodelle

7、adingtoBayesianNetworkFormulation443.BayesianNetworksandProbabilisticInference453.1.Introduction463.2.DefinitionofaBayesianNetwork473.3.JointProbabilityDistribution473.4.PropertiesofBayesianNetworks484.ProbabilityUpdatinginBayesianNetworks494.1.TheDependenceSemanticsofBayesianNe

8、tworks494.2.DifferentSchemesforUpdatingProbabil