Computer based face recognition using neural networks: A biometric access control system based on the human face.

摘要

Fraud is faced on a daily basis in today's society. An impostor could slip through and access highly sensitive locations, a child could be released to a stranger from a daycare center, a welfare recipient could sign up for benefits under six identities, a voter may vote under two different names using two different social security numbers, and a counterfeiter could make copies of and charges to credit cards. Authentic means of identity verification are urgently in high demand. Biometrics, which is the field that incorporates the measurement of one or more distinctive biological trait(s) in order to be studied, examined, or used to uniquely identify its owner, is the most promising solution.; A biometric system that employs the human face for identity verification and access control was successfully developed and tested. In the course of achieving this goal, two different models were developed and tested. One employs a neural network and the other uses feature extraction by implementing a novel elastic template matching approach.; The neural network based model was considered for further development and modification to build the aimed system. The developed face based biometric access control system incorporates more than one access control technique in order to provide a higher level of security. The system employs the user's knowledge of a password or personal identification number (PIN), possession of a magnetic strip card providing the user's name, and the user's face as a physiological characteristic in order to grant or deny access. The results obtained after rigorous testing of the system are very encouraging.