pdf

Theoretical Analysis of Direct LDA under Small Sample Conditions

  • 2013-09-19
  • 226.44KB
  • Points it Requires : 2

Direct Linear Discriminant Analysis (DLDA) is an extended LDA method proposed to overcome the small sample problem. It is claimed to utilize all information outside the null space containing the intra-class scatter matrix. However, many counterexamples show that this is not the case. In order to gain a deeper understanding of the characteristics of DLDA, this paper analyzes it theoretically and concludes that DLDA based on the traditional Fisher criterion hardly utilizes the null space and will lose some useful discriminant information; while DLDA based on the generalized Fisher criterion is equivalent to null space LDA and orthogonal LDA if it meets certain conditions (generally satisfied in high-dimensional small sample data applications) and the optimal discriminant vector is orthogonal. The comparative experimental results on the face databases ORL and YALE are also consistent with the theoretical analysis.

unfold

You Might Like

Uploader
justyouandmehr
 

Recommended ContentMore

Popular Components

Just Take a LookMore

EEWorld
subscription
account

EEWorld
service
account

Automotive
development
circle

About Us Customer Service Contact Information Datasheet Sitemap LatestNews


Room 1530, 15th Floor, Building B, No.18 Zhongguancun Street, Haidian District, Beijing, Postal Code: 100190 China Telephone: 008610 8235 0740

Copyright © 2005-2024 EEWORLD.com.cn, Inc. All rights reserved 京ICP证060456号 京ICP备10001474号-1 电信业务审批[2006]字第258号函 京公网安备 11010802033920号
×