基于信息融合的人臉識別研究.pdf

1、聲明本學(xué)位論文是我在導(dǎo)師的指導(dǎo)下取得的研究成果，盡我所知，在本學(xué)位論文中，除了加以標(biāo)注和致謝的部分外，不包含其他人已經(jīng)發(fā)表或公布過的研究成果，也不包含我為獲得任何教育機(jī)構(gòu)的學(xué)位或?qū)W歷而使用過的材料。與我一同工作的同事對本學(xué)位論文做出的貢獻(xiàn)均已在論文中作了明確的說明。砧年；月鴣日學(xué)位論文使用授權(quán)聲明南京理工大學(xué)有權(quán)保存本學(xué)位論文的電子和紙質(zhì)文檔，可以借閱或上網(wǎng)公布本學(xué)位論文的部分或全部內(nèi)容，可以向有關(guān)部門或機(jī)構(gòu)送交并授權(quán)其保存、借閱或上網(wǎng)

2、公布本學(xué)位論文的部分或全部內(nèi)容。對于保密論文，按保密的有關(guān)規(guī)定和程序處理。砂侈年弓月夥日Abstmct碩士學(xué)位論文AbstractBasedon也ehighidentiflcationof2Dfacialtexturefeaturesandthegoodrobustnessagainstvariationofposition，illumination，aJldexpressionof3Dfacedata，wefocusonthegeom

3、e仃icfeatureex仃actionof3Dfacepoilltcloud，featureex仃actionof3Ddepthimageand2Dfaceimageand如sionmemodsoffacerecognitionOurmainworkiIlcludes(1)Whileconsid砸ngt11epanicul撕t)，ofthepo協(xié)tcloudofnoseregion，wepresentsame也odforregistr

4、ationusingthelocalpoiIltc10udoffacetocorrectfacepostureFirstly’meregistrationpe南manceallde伍ciencyoft11eICPalgor砒11：nareweightedtodetemlinetheareasizeWimthesymme仃yofface，t11eholesgenerateddudngdataacquisitionprocesscanbef

5、illedbymirrorExperimentresultsshowtllatit’se丘’ectiVetocorrectposturebynleaureaaroundnlenosetipwhenweweight11eregis仃ationerroragaillstthecostsoftime(2)Abomt11epointcloudgeometricfeatures，t11emost似or印resentatiVecontourline

6、soffaceareex仃actedConside血gnlerobus廿lessagaillstpositionandeXpression，noseaureacontoursfeatIlreswi也outmou廿1a11dfaceedgeaureex仃actedThenidenti毋t11elocalcontourlinest11atex仃acted，aJldanalyzenle證1pactofdi丘’erentextemalfacto

7、rsnshowsthehi曲discrimillationofea’ective10calcontouTfeaturesFurthermore，recognitionratehasbeenimproVedondi疏renttestsetwhen如sillg鉚ocontourfeatureswithdi骶rentweightsThetwocontoullshaddi虢rentinnuenceonrecognitionrate(3)Mult

8、ipleinf0肋ation如sionisstudiedi11detailOntheonehand，theweightedLBPfeatureof2Dfaceimagesand3DdepmimagesareextractedTheyareweightedto如sei11matchinglayertorecognitionOnt11eotllerhaIld，山efacecontourfeatures，2Dfaceimagef已a(bǔ)tures

9、and3Dd印t11imagefbaturesareVotedto如sei11decision1ayerResultsshow也att11erecognitionpe面n】他ncehasbeen鏟eatlyimproVedby如sionThemoreinfomlation，mehi曲erreco弘itionped’0nnanceKeyword：3Dfacereco鰣tion，Pr印rocessing，E能ctiVefacecontouI