Methodology for the construction of a biometric facial recognition system based on a neural classifier

This work focuses on the development of a methodology that can be useful in the construction and selection of classifiers based on artificial neural networks biometric systems. The general hypothesis is that it is possible to find an acceptable neural model as a classifier for face recognition (FR) according to data provided during training, according to automated procedures. This could be very useful to assist in the construction process of FR systems, such as for example, it could facilitate the construction of a FR system optimized for a client database at a bank or a government organization.