The Solution Space for Fisher Discriminant Analysis and the Uniqueness Under Constraints

Weishi Zheng, Jianhuang Lai, Pong Chi Yuen

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

Abstract

This paper studies the solution space of Fisher Criteria. The space is large and it is impossible to find the best solution generally. This paper intends to construct an optimal projection, which solves the Fisher criteria and is the unique solution under nonsingular linear transformation if some constraints are0020given. Therefore a theorem is proposed which shows the feasible for constructing the projection, with a simple way to process the construction from the traditional LDA. Experiment result shows the ability and feasible of the proposed solution.

Original languageEnglish
Title of host publicationAdvances in Biometric Person Authentication
Subtitle of host publication5th Chinese Conference on Biometric Recognition, SINOBIOMETRICS 2004, Guangzhou, China, December 13-14, 2004, Proceedings
EditorsStan Z. Li, Jianhuang Lai, Tieniu Tan, Guocan Feng, Yunhong Wang
Place of PublicationBerlin, Heidelberg
PublisherSpringer
Pages313-319
Number of pages7
Edition1st
ISBN (Electronic)9783540305484
ISBN (Print)9783540240297
DOIs
Publication statusPublished - 29 Nov 2004
Event5th Chinese Conference on Biometric Recognition, SINOBIOMETRICS 2004 - Guangzhou, China
Duration: 13 Dec 200414 Dec 2004
https://link.springer.com/book/10.1007/b104239

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume3338
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349
NameSINOBIOMETRICS: Chinese Conference on Biometric Recognition

Conference

Conference5th Chinese Conference on Biometric Recognition, SINOBIOMETRICS 2004
Country/TerritoryChina
CityGuangzhou
Period13/12/0414/12/04
Internet address

Scopus Subject Areas

  • Theoretical Computer Science
  • General Computer Science

User-Defined Keywords

  • Linear Discriminant Analysis
  • Solution Space
  • Constraint Qualification
  • Scatter Matrix
  • Optimal Projection

Fingerprint

Dive into the research topics of 'The Solution Space for Fisher Discriminant Analysis and the Uniqueness Under Constraints'. Together they form a unique fingerprint.

Cite this