Multi-Instance Face Recognition System Using Pca And Ann

Abstract

Multi-instance Biometrics is a biometric system configuration in which biometric information is taken in different
conditions, coming from the same biometric type. This study uses biometric from two-dimensional image of face that is taken
from two sides of the face itself and compares the effect of Principal Component Analysis (PCA) on face recognition where the
system is run by NN Backpropagation artificial neural network. This study finds that the recognition performance and learning
speed of network systems are better when PCA is used and have an accuracy of up to 97%