Facial image processing in conjunction with password for intelligent access control

Abstract

In most of the access control systems employed in any organizations, one or more level authentication steps are used to protect their assets. These authentication systems in general involve one or more level knowledge based or possession based authentication steps or sometimes the combination of both, in order to have a high security. However, with the advance of technology these methods are more vulnerable to soft attacks, i.e. more prone to tampering problems. Hence there is a need to have an authentication factor which is highly impossible to synthesize it in any manner. We present a frame work of secured intelligent access control system using two authentication factors namely, inherence factor and knowledge factor. Facial image and a personalized unique password are used as inherence and knowledge factors respectively. Gabor filter along with a subspace technique is employed for feature extraction and matching test and training images. Proposed two factor authentication system reported far better results than any other access control systems existing in literature.