欢迎来到天天文库
浏览记录
ID:10022343
大小:2.94 MB
页数:40页
时间:2018-05-21
《人脸识别方法的研究与实现-翻译》由会员上传分享,免费在线阅读,更多相关内容在行业资料-天天文库。
1、苏州大学本科生毕业设计(论文)附件:外文文献资料与中文翻译稿外文文献资料收集:苏州大学应用技术学院11电子班(学号1116405021)靳冉InternationalJournalofArtificialIntelligence&Applications(IJAIA),Vol.2,No.3,July2011DOI:10.5121/ijaia.2011.230545RealtimefacerecognitionusingadaboostimprovedfastpcaalgorithmABSTRACTThispaperpresentsanautomatedsystemforhum
2、anfacerecognitioninarealtimebackgroundworldforalargehomemadedatasetofpersonsface.Thetaskisverydifficultastherealtimebackgroundsubtractioninanimageisstillachallenge.Additiontothisthereisahugevariationinhumanfaceimageintermsofsize,poseandexpression.Thesystemproposedcollapsesmostofthisvariance
3、.TodetectrealtimehumanfaceAdaBoostwithHaarcascadeisusedandasimplefastPCAandLDAisusedtorecognizethefacesdetected.Thematchedfaceisthenusedtomarkattendanceinthelaboratory,inourcase.Thisbiometricsystemisarealtimeattendancesystembasedonthehumanfacerecognitionwithasimpleandfastalgorithmsandgainin
4、gahighaccuracyrate..KEYWORDSFacerecognition,Eigenface,AdaBoost,HaarCascadeClassifier,PrincipalComponentAnalysis(PCA),FastPCA,LinearDiscriminantAnalysis(LDA).1.INTRODUCTIONOverthelasttenyearsorso,facerecognitionhasbecomeapopularareaofresearchincomputervision.Facerecognitionisalsooneofthemost
5、successfulapplicationsofimageanalysisandunderstanding.Becauseofthenatureoftheproblemoffacerecognition,notonlycomputerscienceresearchersareinterestedinit,butneuroscientistsandpsychologistsarealsointerestedforthesame.Itisthegeneralopinionthatadvancesincomputervisionresearchwillprovideuseful40
6、苏州大学本科生毕业设计(论文)附件:外文文献资料与中文翻译稿insightstoneuroscientistsandpsychologistsintohowhumanbrainworks,andviceversa.Thetopicofrealtimefacerecognitionforvideoandcomplexreal-worldenvironmentshasgarneredtremendousattentionforstudenttoattendclassdailymeansonlineattendancesystemaswellassecuritysystembase
7、donfacerecognition.Automatedfacerecognitionsystemisabigchallengingproblemandhasgainedmuchattentionfromlastfewdecades.Therearemanyapproachesinthisfield.Manyproposedalgorithmsaretheretoidentifyandrecognizehumanbeingfaceformgivendataset.Therecentdevelopment
此文档下载收益归作者所有