Improved PCA based face recognition using directional filter bank

This study addresses new face recognition method based on principal component analysis (PCA) and directional filter bank (DFB) responses. Our method consists of two parts. One is the creation of directional images using DFB from the original face image. The other is transforming the directional images into eigenspace by PCA, which is able to optimally classify individual facial representations. PCA analysis is primarily used as a dimensionality reduction technique with least consideration to the recognition aspect. The basic idea of combining PCA and DFB is to provide PCA with some recognition ability. In our system recognition ability of the PCA is enhanced by providing directional images as inputs. The experiment results showed the remarkable improvement of recognition rate of 21. 25% in Olivetti data set.