Face Recognition using PCA-LDA and Wavelet Packet Analysis

초록

In this paper, a novel method for recognition of frontal views of human faces is presented. The proposed method is based on the wavelet packet analysis and PCA-LDA (Principal Component Analysis - Linear Discriminant Analysis). Wavelet packet analysis is adopted to decompose an image to sub-images with low resolution and a particular frequency band is selected for PCA-LDA. In PCA-LDA method, the features are obtained through eigenvector analysis of scatter matrices with the objective of maximizing between-class variations and minimizing within-class variations. The experimental result with the proposed method is more accurate and more computation efficient in comparing with the original Principal Component Analysis (PCA).