PCA and back-propagation neural network based face recognition system

It´s very difficult for an intelligent control system to recognize a person accurately as like humans. Though face recognition has some grand difficulties because of facial expression, illumination, lighting condition, pose etc., but the recognition of faces is very important for many applications such as: video surveillance, retrieval of an identity from a database for criminal investigations, home and office security, facebook and forensic applications. Within the last decade, the technological revolution has taken place all over the world based on computer vision. Face recognition has proved to be a challenging problem in computer vision. So, researches of computer science, biometrics, pattern recognition and neuroscientists have shown their deep interest in developing a consistent and proficient face recognition system. In this paper, we have proposed a method for face recognition by combining PCA and Back-propagation neural network which efficiently works in a constrained environment. Experiments demonstrate good results for face recognition on AT&T(ORL) database.