简介:Distinguishingwithtraditionaltoothprofileofspiralbevelandhypoidgear,itproposedanewtoothprofilenamelythesphericalinvolute.Firstly,anewtheoryofformingthesphericalinvolutetoothprofilewasproposed.Then,thistheorywasappliedtocompleteparametricderivationofeachpartofitstoothprofile.Forenhancingtheprecision,theSWEEPmethodusedforformationofeachpartoftoothsurfaceandG1stitchingschemaforobtainingaunifiedtoothsurfaceareputforwardandmadetheapplicationintheaccuratemodeling.Lastly,owingtothehigheraccuracyoftoothsurfaceofoutputtedmodel,itgavesomeoptimizationapproaches.Givennumericalexampleaboutthemodelcanshowthatthisdesignedgearwithsphericalinvolutetoothprofilecanachievefastandaccurateparametricmodelingandprovideafoundationfortoothcontactanalysis(TCA)indigitizeddesignandmanufacture.更多还原
简介:
简介:Anewmethodforpoint-polygonclassificationispresentedinthepaper.Thebasicideaofthemethodisfirstadvancedandthentwopropertiesofpolygonareintroduced.Thecriterionsforpoint-polygonclassificationaregivenexplicitly.Theanalysisshowsthatthepossessesperfectfunction,extensivesuitabilityandhighefficiency.
简介:Anew3Dsurfacecontouringandrangingsystembasedondigitalfringeprojectionandphaseshiftingtechniqueispresented.Usingthephase-shifttechnique,pointscloudwithhighspatialresolutionandlimitedaccuracycanbegenerated.Stereo-pairimagesobtainedfromtwocamerascanbeusedtocompute3Dworldcoordinatesofapointusingtraditionalactivetriangulationapproach,yetthecameracalibrationiscrucial.Neuralnetworkisawell-knownapproachtoapproximateanonlinearsystemwithoutanexplicitphysicalmodel,inthisworkitisusedtotrainthestereovisionapplicationsystemtocalculating3Dworldcoordinatessuchthatthecameracalibrationcanbebypassed.Thetrainingsetforneuralnetworkconsistsofavarietyofstereo-pairimagesandthecorresponding3Dworldcoordinates.Thepictureelementscorrespondenceproblemissolvedbyusingprojectedcolor-codedfringeswithdifferentorientations.Colorimbalanceiscompletelyeliminatedbythenewcolor-codedmethod.Oncethehighaccuracycorrespondenceof2Dimageswith3Dpointsisacquired,highprecision3Dpointscloudcanberecognizedbythewelltrainednet.Theobviousadvantageofthisapproachisthathighspatialresolutioncanbeobtainedbythephase-shiftingtechniqueandhighaccuracy3Dobjectpointcoordinatesareachievedbythewelltrainednetwhichisindependentofthecameramodelworksforanytypeofcamera.Someexperimentsverifiedtheperformanceofthemethod.