Multifeature-Based High-Resolution

Multifeature-Based High-Resolution

ID:40721739

大小:2.13 MB

页数:13页

时间:2019-08-06

Multifeature-Based High-Resolution_第1页
Multifeature-Based High-Resolution_第2页
Multifeature-Based High-Resolution_第3页
Multifeature-Based High-Resolution_第4页
Multifeature-Based High-Resolution_第5页
资源描述:

《Multifeature-Based High-Resolution》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、IEEETRANSACTIONSONPATTERNANALYSISANDMACHINEINTELLIGENCE,VOL.33,NO.5,MAY2011945Multifeature-BasedHigh-ResolutionPalmprintRecognitionJifengDaiandJieZhou,SeniorMember,IEEEAbstract—Palmprintisapromisingbiometricfeatureforuseinaccesscontrolandforensicapplications.Previousresearchonpalmprintrecogn

2、itionmainlyconcentratesonlow-resolution(about100ppi)palmprints.Butforhigh-securityapplications(e.g.,forensicusage),high-resolutionpalmprints(500ppiorhigher)arerequiredfromwhichmoreusefulinformationcanbeextracted.Inthispaper,weproposeanovelrecognitionalgorithmforhigh-resolutionpalmprint.Thema

3、incontributionsoftheproposedalgorithmincludethefollowing:1)useofmultiplefeatures,namely,minutiae,density,orientation,andprincipallines,forpalmprintrecognitiontosignificantlyimprovethematchingperformanceoftheconventionalalgorithm.2)Designofaquality-basedandadaptiveorientationfieldestimational

4、gorithmwhichperformsbetterthantheexistingalgorithmincaseofregionswithalargenumberofcreases.3)Useofanovelfusionschemeforanidentificationapplicationwhichperformsbetterthanconventionalfusionmethods,e.g.,weightedsumrule,SVMs,orNeyman-Pearsonrule.Besides,weanalyzethediscriminativepowerofdifferent

5、featurecombinationsandfindthatdensityisveryusefulforpalmprintrecognition.Experimentalresultsonthedatabasecontaining14,576fullpalmprintsshowthattheproposedalgorithmhasachievedagoodperformance.Inthecaseofverification,therecognitionsystem’sFalseRejectionRate(FRR)is16percent,whichis17percentlowe

6、rthanthebestexistingalgorithmataFalseAcceptanceRate(FAR)of105,whileintheidentificationexperiment,therank-1live-scanpartialpalmprintrecognitionrateisimprovedfrom82.0to91.7percent.IndexTerms—Palmprint,orientationfield,thecompositealgorithm,densitymap,datafusion.Ç1INTRODUCTIONPALMPRINTrecognit

7、ionhasconsiderablepotentialasathreeregions:thenar,hypothenar,andinterdigital.Mutablepersonalidentificationtechnique.Palmprintssharecreasesmainlycomefromdryingcracks,whichcomeintomostofthediscriminativefeatureswithfingerprintsand,beinginSpringandWin

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。