Local discriminant nearest feature analysis for image feature extraction

被引：0

作者：

机构：

[1] Huang, Chun-Ta

[2] Yan, Lijun

[3] Pan, Jeng-Shyang

来源：

Huang, C.-T. (huang146@purdue.edu) | 1600年 / ICIC Express Letters Office, Tokai University, Kumamoto Campus, 9-1-1, Toroku, Kumamoto, 862-8652, Japan卷 / 07期

关键词：

Discriminant informations - Finger knuckle prints - Geometric information - Image feature extractions - Nearest feature lines - Nearest feature plane - ORL face database - Subspace learning;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

A novel image feature extraction algorithm, named Local Discriminant Nearest Feature Analysis (LDNFA), is proposed in this paper. LDNFA is a nearest feature (NF) based subspace learning approach. Different from most of current NF based methods, LDNFA uses local features to preserve the geometric information of the samples while extracting discriminant information. The proposed LDNFA is applied to image classification on ORL face database, AR face databases, and PolyU Finger-Knuckle-Print Database. The experimental results demonstrate the efficiency of the proposed LDNFA. © 2013 ICIC International.

引用