Discriminant analysis of principal components for face recognition

TLDR

Combining PCA and LDA improves LDA’s generalization when few samples per class are available. The paper proposes a face recognition method that integrates PCA and LDA. The method projects images to a PCA‑derived subspace and then applies LDA for classification. On the FERET dataset, the PCA–LDA hybrid yields significant performance gains and offers a framework for other image recognition tasks.

Abstract

In this paper we describe a face recognition method based on PCA (Principal Component Analysis) and LDA (Linear Discriminant Analysis). The method consists of two steps: first we project the face image from the original vector space to a face subspace via PCA, second we use LDA to obtain a best linear classifier. The basic idea of combining PCA and LDA is to improve the generalization capability of LDA when only few samples per class are available. Using PCA, we are able to construct a face subspace in which we apply LDA to perform classification. Using FERET dataset we demonstrate a significant improvement when principal components rather than original images are fed to the LDA classifier. The hybrid classifier using PCA and LDA provides a useful framework for other image recognition tasks as well.