基于马氏距离局部边界fisher研究降维算法

《基于马氏距离局部边界fisher研究降维算法》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、基于马氏距离局部边界Fisher研究降维算法　　文章编号:10019081(2013)07193005doi:10.11772/j.issn.10019081.2013.07.1930摘要：针对人脸识别应用中的高维数据图像以及欧氏距离不能准确体现样本间的相似度的问题，提出了一种基于马氏距离的局部边界Fisher分析(MLMFA)降维算法。该算法从现有的样本中学习得到一个马氏度量，然后在近邻选择以及新样本降维过程中用马氏距离作为相似性度量。同时，通过马氏度量构造出类内“相似”图和类间“代价”图来描述数据集的类内紧凑性和类间分离性。MLMFA很好地保持了数据

2、集的局部结构。用YALE和FERET人脸库进行实验，MLMFA的最大识别率比传统基于欧氏距离算法的最大识别率平均分别提高了1.03%和6%。实验结果表明，算法MLMFA具有很好的分类和识别性能。关键词：马氏距离；局部边界Fisher分析；降维；人脸识别中图分类号：TP391.413文献标志码:A英文标题4MahalanobisdistancebasedlocalmarginalfisheranalysisdimensionalityreductionalgorithmDimensionalityreductionalgorithmoflocalmarginal

3、FisheranalysisbasedonMahalanobisdistance英文作者名LIFeng1*,WANGZhengqun1,XUChunlin2,ZHOUZhongxia1,XUEWei1英文地址(1.CollegeofInformationEngineering,YangzhouUniversity,YangzhouJiangsu225127,China;2.DepartmentofLaserApplicationTechnology,NorthLaserTechnologyGroupCompanyLimited,Yangzho

4、uJiangsu225009,China英文摘要)Abstract：ConsideringhighdimensionaldataimageinfacerecognitionapplicationandEuclideandistancecannotaccuratelyreflectthesimilaritybetweensamples,aMahalanobisdistancebasedLocal4MarginalFisherAnalysis(MLMFA)dimensionalityreductionalgorithmwasproposed.AMahalanobisdist

5、ancecouldbeascertainedfromtheexistingsamples.Then,theMahalanobisdistancewasusedtochooseneighborsandtoreducethedimensionalityofnewsamples.Meanwhile,todescribetheintraclasscompactnessandtheinterclassseparability,intraclass“similarity”graphandinterclass“penalty”graphwereconstructedbyusin

6、gMahalanobisdistance,andlocalstructureofdatasetwaspreservedwell.WiththeproposedalgorithmbeingconductedonYALEandFERET,MLMFAoutperformsthealgorithmsbasedontraditionalEuclideandistancewithmaximumaveragerecognitionrateby1.03%and6%respectively.Theresultsdemonstratethattheproposedalgorithmhasve

7、rygoodclassificationandrecognitionperformance.ConsideringhighdimensionaldataimageinfacerecognitionapplicationandEuclideandistancecannotaccuratelyreflectthesimilaritybetweensamples,aMahalanobisdistancebasedLocalMarginalFisherAnalysis(MLMFA)dimensionality4reductionalg