曹建农,李德仁,关泽群

• 1:西安建筑科技大学建筑学院
• 2:武汉大学遥感信息工程学院
• 3:武汉大学遥感信息工程学院 西安　710055 武汉大学遥感信息工程学院武汉　430079
• 4:武汉　430079

摘要(Abstract)：

研究了马尔科夫网的概念、方法,分析它在空间数据挖掘中的作用与意义,并以视频图像为例广泛研究以多种粒度(节点数)建立视频图像间的马尔科夫网,通过网络结构分析检测视频图像中的目标差异。研究表明马尔科夫网可以很好地揭示数据间的抽象近邻关系,并且这种网络自身就具有表达知识的意义。

关键词(KeyWords)： 马尔科夫网;目标检测;概率距离;数据挖掘

Abstract:

Keywords:

基金项目(Foundation): 地理信息工程国家测绘局重点实验室基金(1469990324233);; 国家自然科学基金资助项目(60175022);; 国家高科技发展计划(863)项目(2001AA135081)

作者(Author): 曹建农,李德仁,关泽群

参考文献(References)：

• [1]NarendraA Mosaicmodelsfortextures[A] In:ImageModeling[C],NewYork:Academic,1980:1 7
• [2]LaveenNKanal Markovmeshmodels[A] In:ImageModeling[C],NewYork:Academic,1980,1981:240 243
• [3]JWModestino,RWFriesandALVickers Stochasticimagemodelsgeneratedrandomtessellationsoftheplane[A]In:ImageModeling[C],NewYork:Academic,1980,1981:302 325
• [4]JohnWoods Two dimensionaldiscreteMarkovfields[J]IEEETransactionsonInformationTheory1972,IT 18(2):232 240
• [5]JAStuller,BKurz Two dimensionalMarkovianrepresentationsofsampledimages[J] IEEETrans Commun 1976,Com 24:1148 1152
• [6]R L Kashyap,R Chellappa Estimationandchoiceofneighborsinspatial interactionmodelsofimages[J] IEEETrans InformationTheory,1983,Vol IT 29,NO 1:6072
• [7]PatrickAKelly,HalukDerin,KeithDHartt Adaptivesegmentationofspeckledimagesusingahierarchicalrandomfieldmodel[J] IEEETransactionsAcoustics,Speech,andSignal,Processing,1988,36(10):1628 1641
• [8]FlorenceTupin,HenriMaitre,Jean FrancoisMangin,JeanMarieNicolas,EugenePechersky DetectionoflinearfeaturesinSARimages:applicationtoroadnetworkextraction[J] IEEETransactionsGeoscienceandRemoteSensing,1998,36(2):434 453
• [9]FlorenceTupin,IsabelleBloch,HenriMaitre AfirststeptowardautomaticinterpretationofSARimagesusingevidentialfusionofseveralstructuredetectors[J] IEEETransactionsonGeoscienceRemoteSensing,1999,37(3):13271343
• [10]FlorenceTupin MarkovianmethodsforSARimageprocessing[J] 0 7803 7031 7/01/＄10 00(C)2001IEEE,1164 1166
• [11]SGeman,DGeman StochasticrelaxationGibbsdistribution,andtheBayesianrestorationofimages[J] IEEETrans PatternAnal MachineIntell 1984,PAMI 16:721 741
• [12]YXiang,SKMWong,NCercone Amicroscopic'studyofminimumentropysearchinlearningdecomposableMarkovnetworks[A] In:MachineLearning[C],Copy1997KluwerAcademicPublishers ManufacturedTheNetherlands:2665 2692
• [13]JiaweiHan,MichelineKamber Dataminingconceptsandtechniques[M] Beijing:ChinaMachinePress,2001
• [14]DavidHeckerman Bayesiannetworkfordataminingdataminingandknowledgediscovery[A] In:Copy1997KluwerAcademicPublishers[C],ManufacturedTheNetherlands:79 119
• [15]MeeraBalakrishnan,AndrewReibman CharacterizingalumpingheuristicforaMarkovnetworkreliabilitymodel[J] Copy1993IEEE:0731 3071
• [16]YXiangandJLee Localscorecomputationinlearningbeliefnetworks[J] ScienceofSurveyingandMapping,2004,(2)
• [17]HepingPan LearningBayesiannetworksII Acomputationalgorithm[J] ScienceofSurveyingandMapping,2002,(4)
• [18]HepingPan LearningBayesiannetworksI AtheorybasedonMAP MDLcriteria[J] ISIFCopy2002,IEEE:769776
• [19]MCGolumbic Algorithmicgraphtheoryandperfectgraphs[M] AcademicPress,1980
• [20]JPearlProbabilisticreasoninginintelligentsystems[M\〗NetworksofPlausibleInference MorganKaufmann,1988
• [21]JamesWModestino,JunZhang AMarkovrandomfieldmodel basedapproachtoimageinterpretation[J] IEEETransactiononPatternAnalysisandMachineIntelligent,1992,14(6):606 615
• [22]郑肇保 图像分析的马尔科夫随机场方法[M] 武汉:武汉测绘科技大学出版社 2000
• [23]阿兰图克尔著,申大维译 数学的原理与实践[M]北京:高等教育出版社 1998
• [24]YXiang,JHu,NCercone,HHamilton Learningpseudoindependentmodels:analyticalandexperimentalresults[J] ScienceofSurveyingandMapping,2004,(3)
• [25]RRBouckaert Propertiesofbayesianbeliefnetworklearningalgorithms[A] In:RLopezdeMantarasandDPooleeditors,Proc Of10thConf OnUncertaintyinArtificialIntelligence,Seattle[C],Washington,1994 MorganKaufmann:102 109