资源描述:
《A hierarchical statistical framework for the segmentation of deformable objects in image se》由会员上传分享,免费在线阅读,更多相关内容在行业资料-天天文库。
1、AhierarchicalstatisticalframeworkforthesegmentationofdeformableobjectsinimagesequencesCharlesKervrannandFabriceHeitzIRISA/INRIA,CampusUniversitairedeBeaulieu,35042RennesCedex,FranceE-mail:kervrann@irisa.fr,heitz@irisa.frIEEEComputerVisionPatternRecognitionJune1994,Seattle,USAA
2、HierarchicalStatisticalFrameworkfortheSegmentationofDeformableObjectsinImageSequencesCharlesKervrannandFabriceHeitzIRISA/INRIA-RennesCampusUniversitairedeBeaulieuF-35042RennesCedex,FranceAbstractprocess;theycanbeseenasarenementoftheglobaldeformationsappliedtotheoriginalshape
3、.ThejointInthispaper,weproposeanewstatisticalframeworkfordistributionofthedeformabletemplateisderivedandmodelingandextracting2DmovingdeformableobjectsfromaMaximumAPosteriori(map)estimateofthede-imagesequences.Theobjectrepresentationreliesonahie-formationsisobtainedbyminimizing
4、aglobalenergyrarchicaldescriptionofthedeformationsappliedtoatem-(objective)functiondescribingtheinteractionsbet-plate.GlobaldeformationsaremodeledusingaKarhunenweenobservations(spatialortemporalgradientsex-Loeveexpansionofthedistorsionsobservedonarepre-tractedfromtheimage)andt
5、hedeformationprocess.sentativepopulation.LocaldeformationsaremodeledbyThemethodcombinestheadvantagesoffastglobala(rst-order)Markovprocess.Theoptimalbayesianes-optimizationtechniqueswithacompacthierarchicaltimateoftheglobalandlocaldeformationsisobtainedbystatisticaldescription
6、ofdeformations.Thisyieldsfastmaximizinganon-linearjointprobabilitydistributionusingmodeladjustmentandrobustsegmentation.stochasticanddeterministicoptimizationtechniques.TheComputervisionmethodsrelyingondeformableuseofglobaloptimizationtechniquesyieldsrobustandre-templatesareof
7、tenexpressedastheminimizationofliablesegmentationsinadversesituationssuchaslowsignal-(global)energyfunctionsdescribingtheinteractionsto-noiseratio,non-gaussiannoiseorocclusions.Moreo-betweentheobserveddataandthevariablesofthever,nohumaninteractionisrequiredtoinitializethemo-mo
8、del[4,8,9].Inmostmethods[1,5,7,8,9](apartdel.Theapproachisdem