ica算法在人臉表情識別中的應(yīng)用研究

1、大連理工大學(xué)碩士學(xué)位論文ICA算法在人臉表情識別中的應(yīng)用研究姓名：呂惠清申請學(xué)位級別：碩士專業(yè)：軟件工程指導(dǎo)教師：周勇20071215ICA算法的人臉表情識別中的應(yīng)用研究StudyoflCA’applicationinfacialexpressionrecognitionAbstracthrecentyears，伯estudyoffacereengmtionhasmadeverygreatprogressSomesystemhasrea

2、lizedcommercialization。Butitdoesnotmeanthatfacerecognitiontechnologyhasbeenalreadyperfect。Atfact，There勰stillalotofproblemsinthefieldoffacerecognition。Amongthose，AutomaticisoneofthedifficultAutomaticfacialexpressionrecogn

3、ition(AFER)whichisdevelopedtecognizeoneof0111“humanspecialemotionalrepresentalionfacialexpressionhasbeenamactedmoreandmoleattentionItis矗攫ofchallenginganddifficultbecauseofthecomplicatedandsubtleproper蛀esoffacialexpressio

4、nAspecificdifficultiesliesinA阿tisitsdataexistsamongthehighenddataIndependentcomponentanalysisalgorithminhighenddataprocessinghasitsuniqueadvamagesIndependentComponentAnalysisisanewlydevelopedandpowerful1echniqueforrecove

5、ringlatentindependent∞蝴sgivenonlytheirmixtures翦磚basicICAmodelassumesthatsourcesarelinearlymixedandmutuallyindependent11“dataanalyzedbyICAcouldoriginate矗ommanydifferentkindsofapplicationfield，includedigitalimagesanddocume

6、ntdatabaseasweH囂economicindicatorsandpsychometricmoasurgmerlisgtc。Independentcomponentanalysisinitsownhighenddataprocessingcapabilityatthesametimehasbroughtinevitableincreaseincomplexityofcomputingthefacerecognitionsyste

7、minordertomeettherequirementsofrealtimethepaperwilltakeafastICAalgorithmffastlCA)。andOnthisbasistonlake內(nèi)婦improvementstoguaranteeSystemforrealtimerequirements。KeyWords：FaceexpressionrecognitionFastlCA：LFastlCA：Featurea：bs