Abstract
The linear discriminant analysis (LDA) is one of the most traditional linear dimensionality reduction methods. This paper incorporates the inter-class relationships as relevance weights into the estimation of the overall within-class scatter matrix in order to improve the performance of the basic LDA method and some of its improved variants. We demonstrate that in some specific situations the standard multi-class LDA almost totally fails to find a discriminative subspace if the proposed relevance weights are not incorporated. In order to estimate the relevance weights of individual within-class scatter matrices, we propose several methods of which one employs the evolution strategies. © 2004 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.
Original language | English |
---|---|
Pages (from-to) | 485-493 |
Number of pages | 9 |
Journal | Pattern Recognition |
Volume | 38 |
Issue number | 4 |
Early online date | 15 Dec 2004 |
DOIs | |
Publication status | Published - Apr 2005 |
Externally published | Yes |
Keywords
- Approximate pairwise accuracy criterion
- Chernoff criterion
- Evolution strategies
- Feature extraction
- Linear discriminant analysis
- Mahalanobis distance
- Weighted LDA