Recognition of Face Using Neural Network

Advancement in Artificial Intelligence has lead to the developments of various “smart” devices. The task of face Recognition has been actively researched in recent years. Wide usage of biometric information for person identity verification purposes, terrorist acts prevention measures and authentication process simplification in computer systems has raised significant attention to reliability and efficiency of biometric systems. Modern biometric systems still face much reliability and efficiency related issues such as database search speed, errors while recognizing of biometric information or automating biometric feature extraction. In face recognition, many methods are used but due to advancement there are some new methods and algorithm used for recognition of face i.e. line edges map, support vector machine etc. for face recognition. A number of current face recognition algorithms use face representations found by supervised and unsupervised statistical methods. In this paper we use a neural network approach for face recognition system. We are creating a supervised multilayer feed forward network model. We will design a neural network for face recognition system which first detect the face and then recognize the face. Specifically some network models use a set of desired outputs to compare with the output and compute an error to make use of in adjusting their weights. Such learning rules are termed as Supervised Learning One such network with supervised learning rule is the Multi-Layer Perceptron (MLP) model. Back propagation algorithm is used to train the network, calculate error and modify weights. Keywords—Artificial Neural Network, Back Propagation Algorithm, Face Recognition, Multi-Layer Perceptron, Supervised Learning