Performance Analysis of Face Recognition using Feed Forward Neural Network and PCA

Abstract

In this paper, a face recognition system for personal identification and verification using Principal Component Analysis (PCA) with Neural Networks (NN) is proposed. The dimensionality of face images is reduced by the PCA and the recognition is done by the NN for efficient and robust face recognition. In this also focuses on the face database with different sources of variations, especially pose, expression. In this method of face identification covariance matrix of training and testing samples is prepared, which is further utilized for finding the eigenvalues and eigenvectors. These components are utilized for training of the face identification model. The algorithm has been tested on 165 grayscale images (15*11 classes). Face will be categorized as known or unknown face after matching with the present YALE face database. Experimental results in this paper showed that an accuracy of 96.4% was achieved.