Gender Identification System from Facial Image Using Artificial Neural Network

Abstract

This paper presents the implementation of gender identification system from a front view facial image using artificial neural network. In this system the facial images from different persons were taken as its input. At first, face images were projected onto a feature space that span the significant variations among face images. The features of the images were extracted using a compression technique that used the artificial neural network for compression. The extracted feature was then given to the input of the multi-layer feed forward neural network. Thus the network was trained and created a knowledge base. Finally, when a facial image was given for identification, the recognition part of the identification system identified the gender with the help of previously stored knowledge base.