A direct locality preserving projections (DLPP) algorithm for image recognition

被引：27

作者：

Feng, Guiyu ^{[1
]}

Hu, Dewen ^{[1
]}

Zhou, Zongtan ^{[1
]}

机构：

[1] Natl Univ Def Technol, Coll Mechatron & Automat, Dept Automat Control, Changsha 410073, Hunan, Peoples R China

来源：

NEURAL PROCESSING LETTERS | 2008年 / 27卷 / 03期

基金：

中国国家自然科学基金;

关键词：

principal component analysis (PCA); linear discriminant analysis (LDA); locality preserving projections (LPP); direct locality preserving projections (DLPP); image recognition;

D O I：

10.1007/s11063-008-9073-1

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper proposes a novel locality preserving projections (LPP) algorithm for image recognition, namely, the direct locality preserving projections (DLPP), which directly optimizes locality preserving criterion on high-dimensional raw images data via simultaneous diagonalization, without any dimensionality reduction preprocessing. Our algorithm is a direct and complete implementation of LPP. Experimental results on the PolyU palmprint database and ORL face database show the effectiveness of the proposed algorithm.

引用

页码：247 / 255

页数：9

共 17 条

[1] Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection [J].