Face Recognition using the Gabor Wavelet Transform and Feed Forward Neural Network

In recent years, an explosion in research on pattern recognition systems using neural network methods has been observed. Face recognition (FR) is a specialized pattern recognition task for several applications such as security: access to limited areas, banking: identity confirmation and identification of wanted people at airports. Biometric techniques deal with identifying individual with the help of their biological data. Face plays an important role in conveying identity and emotion. People can recognize thousands of faces learned throughout their lifetime and identify familiar faces at a glance even after many years of separation. In this paper, we will explain about the tasks involved in FR and outline a complete face recognition system (FRS) based on the Gabor transform and feed forward neural network (FFNN) system. The FRS includes two stages: training and testing series. We use the Gabor wavelet transform for function extraction and then the FFNN as a classifier. Gabor transform is used to extract features at grid points and graph matching for the proper positioning of the grid. The neural network is used for the purpose of classification because it is a mathematical model which solves the problem of a group of highly connected neurons to realize compositions of non-linear functions. Experimentation is carried out on the FRS by using Yale Face Database and Olivetti Research Laboratory (ORL) database.