简介:Dispersedfringesensor(DFS)isanimportantphasingsensorofnext-generationopticalastronomicaltelescopes.ThemeasurementerrorsinducedbythemeasurementnoiseofthreepistonestimationmethodsfortheDFSincludingleastsquaredfitting(LSF)method,frequencypeaklocation(FPL)methodandmainpeakposition(MPP)method,areanalyzedtheoreticallyandvalidatedexperimentallyinthispaper.Theexperimentalresultscoincidewellwiththetheoreticalanalyses.TheMPP,FPL,LSFareusedrespectivelywhentheDFSoperateswithbroadbandlight(centralwavelength:706nm,bandwidth:23nm).Thecorrespondingrootmeansquare(RMS)valueofestimatedpistonerrorcanbeachievedtobe1nm,3nm,26nm,respectively.Additionally,therangeofDFSwiththeFPLcanbemorethan100μmatthesametime.TheFPLmethodcanworkwellbothincoarseandfinephasingstageswithacceptableaccuracy,comparedwithLSFmethodandMPPmethod.
简介:
简介:Nickel-irontypecastironelectrodesarewidelyusedforweldingQT600-3nodulariron.Nevertheless,theresearchworkhasdemonstratedthatthestrengthoftheseheterogeneousweldedjointsislow.Inthispaper,theuni-axialtensionofthejointsisstudiedbytheMoirefringemethod,andthedis-placement,strainandstressinthetotalfieldsareobtaind.Basedontheunevendistributionofstrainandstress,thereasonofcrackinitiationandpropagationisdiscussed.Throughexaminingtheappearanceofthefracturebyscanningelec-tronmicroscope,alotofsphericalsubstancesdistributedonthefracturesurfacearefound,whichmaybeanotherreasonleadingtocrackingoftheweldedjointunderthelowertensilestress.Thesenewfindingswillhelptoimprovethepropertiesofnickel-irontypecastironelectrodesandthestrengthoftheweldedjoint.
简介:Anew3Dsurfacecontouringandrangingsystembasedondigitalfringeprojectionandphaseshiftingtechniqueispresented.Usingthephase-shifttechnique,pointscloudwithhighspatialresolutionandlimitedaccuracycanbegenerated.Stereo-pairimagesobtainedfromtwocamerascanbeusedtocompute3Dworldcoordinatesofapointusingtraditionalactivetriangulationapproach,yetthecameracalibrationiscrucial.Neuralnetworkisawell-knownapproachtoapproximateanonlinearsystemwithoutanexplicitphysicalmodel,inthisworkitisusedtotrainthestereovisionapplicationsystemtocalculating3Dworldcoordinatessuchthatthecameracalibrationcanbebypassed.Thetrainingsetforneuralnetworkconsistsofavarietyofstereo-pairimagesandthecorresponding3Dworldcoordinates.Thepictureelementscorrespondenceproblemissolvedbyusingprojectedcolor-codedfringeswithdifferentorientations.Colorimbalanceiscompletelyeliminatedbythenewcolor-codedmethod.Oncethehighaccuracycorrespondenceof2Dimageswith3Dpointsisacquired,highprecision3Dpointscloudcanberecognizedbythewelltrainednet.Theobviousadvantageofthisapproachisthathighspatialresolutioncanbeobtainedbythephase-shiftingtechniqueandhighaccuracy3Dobjectpointcoordinatesareachievedbythewelltrainednetwhichisindependentofthecameramodelworksforanytypeofcamera.Someexperimentsverifiedtheperformanceofthemethod.
简介:Althoughthestructuredlightsystemthatusesdigitalfringeprojectionhasbeenwidelyimplementedinthree-dimensionalsurfaceprofilemeasurement,themeasurementsystemissusceptibletonon-linearerror.Inthiswork,weproposeaconvenientlook-up-table-based(LUT-based)methodtocompensateforthenon-linearerrorincapturedfringepatterns.Withoutextracalibration,thisLUT-basedmethodcompletelyutilizesthecapturedfringepatternbyrecordingthefull-fielddifferences.Then,aphasecompensationmapisestablishedtorevisethemeasuredphase.Experimentalresultsdemonstratethatthismethodworkseffectively.