線性子空間法人臉識(shí)別的研究

1、山東大學(xué)碩士學(xué)位論文線性子空間法人臉識(shí)別的研究姓名：曲玉強(qiáng)申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：信號(hào)與信息處理指導(dǎo)教師：吳曉娟20081023山東大學(xué)碩士學(xué)位論文ABSTRACTFacerecognitionhasbecomeahotresearchedfieldinbiometricrecognitionfieldComparing、航tIlothertechnologiesofbiometricrecognitionthetechnologyof

2、facerecognitionhasSOmanyad餾Ita舀；ssuchasnonintrusiveeasycollectionandhighaccuraccythatitispopularTherearebroadapplicationsinthefieldsofcommercial，lawenforcementpublicsecurityidentityclassificationentrancecontrol，videosurv

3、eillanceandSOOILFacerecognitionincludesthreeparts：facedetectionandlocalizationfeatureextractionandclassificationThisthesispresentsthestudiesoffeatureextractioninfacerecognitionFeatureextractionisaprocesswhichtransfersthe

4、datafromprimaryspacesintofeaturespace，representingtheminalowerdimensionalspacewithlesseffectivecharactersuptonow，manymethodSoffeatureextractionhavebeenproposedAmongthem，thesubspaceanalysishasreceivedextensiveattentionowi

5、ngtoitsappealingpropertiesNow，thesubspaceanalysismethodhasbeenthemostpopulartechnologyforfeatureextractionandfacerecognitionFeatureextractionbased011linearsubspaceanalysishasadvantages，suchashi曲computingefficencyandstron

6、ggeometryfeaturedescriptionabilityEigenfacemethodbasedonprincipalcomponentanalysisandthemethodbasedonlineardiscriminantanalysisaretwosuceuessfulonesBecameofthedisturbancecomingfromenvironmentoneorderfeatureextractionbase

7、donPCAorLDACanhardlygetthefeaturesexactlystandingforidentificationBycomparingtheselinearsubspaceanalysismethodsintheoryandexperimentresults，Thisthesiswillpointoutthatlineardiscriminateanalysisismoresuitableforfacerecogni

8、tionthanothersubspacemethodsFisherlineardiscriminateanalysishasobtainedsuccessfulapplicationsinfacerecognitionAnewLDA—basedfacerecognitionalgorithmisproposedinthethesiswhichiscalled“weightednullspaceapproach’Duetothefact