Artificial neural networks for face recognition using PCA and BPNN

In today´s age of automation, face recognition is a vital component for authorization and security. It has received substantial attention from researchers in various fields of science such as biometrics and computer vision. In this paper, a face recognition system using Principal Component Analysis (PCA) with Back Propagation Neural Networks (BPNN) is analysed. A neural based algorithm is presented to recognize the frontal views of faces. The multi-variate data set of face image is reduced using the PCA technique. BPNN is used for training and learning, leading to efficient and robust face recognition. Experiments and testing were conducted over Olivetti Research Laboratory (ORL) Face database. Results indicate that PCA based execution is faster while the recognition accuracy suffers and vice versa for BPNN, thus suggesting a system incorporating both techniques is preferred.