预览加载中，请您耐心等待几秒...

基于距离-夹角形状分布的三维CAD模型检索算法.docx

预览
1/3
2/3
3/3

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

基于距离-夹角形状分布的三维CAD模型检索算法 【Abstract】Withtherapiddevelopmentof3Dmodelingtechnology,thenumberof3DCADmodelsisincreasingrapidly,whichposeschallengestotheretrievalofrelevantmodels.Inthispaper,weproposeamodelretrievalalgorithmbasedonthedistance-angleshapedistributionin3Dspace.First,weconvertthe3Dmodeltoasetofpointcloudsandcalculatetheshapedistributionofeachpointcloud.Then,weusetheEuclideandistanceandtheanglebetweentheshapedistributionsoftwomodelstoestablishadistance-anglesimilaritymeasurebetweenthem.Finally,weconductexperimentsonalarge-scaleCADmodelrepositoryandcompareouralgorithmtostate-of-the-artmethods.Resultsshowthatouralgorithmachievesbetterretrievalaccuracyandefficiencythanexistingmethods. 【Keywords】3Dmodeling,CADmodel,shapedistribution,distance-anglesimilarity,retrievalalgorithm Introduction As3Dmodelingtechnologycontinuestomature,moreandmore3DCADmodelsarecreatedandusedinvariousfieldssuchasarchitecture,engineering,andmanufacturing.Withtherapidgrowthofthenumberofmodels,ithasbecomeincreasinglydifficulttoretrievethedesiredmodelfromalargerepository.Therefore,efficientandaccurateretrievalalgorithmsarenecessaryforeffectiveuseof3Dmodels. A3DCADmodelistypicallyrepresentedasacollectionofpolygons,vertices,andedges.Theserawdataaredifficulttousedirectlyforsimilaritymeasurement,somanyfeature-basedretrievalmethodshavebeenproposed.Feature-basedmethodsextractfeaturessuchasshapedescriptors,keypoints,andhistogramsfromthe3Dmodelandusethemforsimilaritymeasurement.Amongthem,shapedescriptorsareoneofthemostwidelyusedfeatures. Oneofthemostpopularshapedescriptorsistheshapedistribution,whichencodesthelocalgeometricinformationofa3Dmodel.Theshapedistributionmeasuresthedistributionofsurfacenormalsateachpointofthemodel,whichprovidesusefulinformationforshapematchingandretrieval.Severalmethodshavebeenproposedbasedonshapedistributions,suchasthesphericalharmonicrepresentationandthespinimagerepresentation.However,thesemethodshavelimitationsindescribingtheshapedistributionofamodelaccurately. Toaddresst