An innovative weighted 2DLDA approach to face recognition

Two Dimensional Linear Discrimination Analysis (2DLDA) is an effective feature extraction approach for face recognition, which manipulates on the two dimensional image matrices directly. However, some between-class distances in the projected space are too small and this may produce a large erroneous...

Full description

Bibliographic Details
Main Authors: Lu, C., An, Senjian, Liu, Wan-Quan, Liu, X.
Format: Journal Article
Published: Springer 2011
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/49372
_version_ 1848758226704662528
author Lu, C.
An, Senjian
Liu, Wan-Quan
Liu, X.
author_facet Lu, C.
An, Senjian
Liu, Wan-Quan
Liu, X.
author_sort Lu, C.
building Curtin Institutional Repository
collection Online Access
description Two Dimensional Linear Discrimination Analysis (2DLDA) is an effective feature extraction approach for face recognition, which manipulates on the two dimensional image matrices directly. However, some between-class distances in the projected space are too small and this may produce a large erroneous classification rate. In this paper we propose a new 2DLDA-based approach that can overcome such drawback for the existing 2DLDA. The proposed approach redefines the between-class scatter matrix by putting a weighting function based on the between-class distances, and this will balance the between-class distances in the projected space iteratively. In order to design an effective weighting function, the between-class distances are calculated and then used to iteratively change the between-class scatter matrix, which eventually leads to an optimal projection matrix. Experimental results show that the proposed approach can improve the recognition rates on benchmark data-bases such as the ORL database, the Yale database, the YaleB database and the Feret database in comparison with other 2DLDA variants.
first_indexed 2025-11-14T09:40:37Z
format Journal Article
id curtin-20.500.11937-49372
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T09:40:37Z
publishDate 2011
publisher Springer
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-493722017-09-15T23:25:24Z An innovative weighted 2DLDA approach to face recognition Lu, C. An, Senjian Liu, Wan-Quan Liu, X. Face recognition Two Dimensional Linear Discriminant Analysis Weighted linear discriminant analysis Two Dimensional Linear Discrimination Analysis (2DLDA) is an effective feature extraction approach for face recognition, which manipulates on the two dimensional image matrices directly. However, some between-class distances in the projected space are too small and this may produce a large erroneous classification rate. In this paper we propose a new 2DLDA-based approach that can overcome such drawback for the existing 2DLDA. The proposed approach redefines the between-class scatter matrix by putting a weighting function based on the between-class distances, and this will balance the between-class distances in the projected space iteratively. In order to design an effective weighting function, the between-class distances are calculated and then used to iteratively change the between-class scatter matrix, which eventually leads to an optimal projection matrix. Experimental results show that the proposed approach can improve the recognition rates on benchmark data-bases such as the ORL database, the Yale database, the YaleB database and the Feret database in comparison with other 2DLDA variants. 2011 Journal Article http://hdl.handle.net/20.500.11937/49372 10.1007/s11265-010-0541-2 Springer restricted
spellingShingle Face recognition
Two Dimensional Linear Discriminant Analysis
Weighted linear discriminant analysis
Lu, C.
An, Senjian
Liu, Wan-Quan
Liu, X.
An innovative weighted 2DLDA approach to face recognition
title An innovative weighted 2DLDA approach to face recognition
title_full An innovative weighted 2DLDA approach to face recognition
title_fullStr An innovative weighted 2DLDA approach to face recognition
title_full_unstemmed An innovative weighted 2DLDA approach to face recognition
title_short An innovative weighted 2DLDA approach to face recognition
title_sort innovative weighted 2dlda approach to face recognition
topic Face recognition
Two Dimensional Linear Discriminant Analysis
Weighted linear discriminant analysis
url http://hdl.handle.net/20.500.11937/49372