[ ] SIENA

Stimulation Initiative for European Neural Applications

Esprit Project 9811

Case Studies of Successful Applications

ZN-Face: Access Control Using Automated Face Recognition

Company background

The following application has been developed in co-operation with a banking-house in Frankfurt, Germany, with a great number of branches all over the world.

The problem

Absolutely safe verification of the true identity of a person is the ultimate goal of all access control systems. However, this goal can not be achieved under all circumstances with conventional automated systems: A stolen key or an illicitly acquired personal identification number are most common safety leakages in high-security areas, e.g. sensible areas in industry, banking, powre plants, computer centers, etc. .

Neural network application

In co-operation with a banking-house in Frankfurt, Germany, an automated face recognition system on the basis of a neural network has been developed which reliably verifies the true identity of a person. A camera -integrated in a verification console- takes a picture of the person and verifies whether the picture matches the stored image features of this person. Even under different facial expression, position or size of the head the neural network-part of the system reliably recognizes the person and prevents admittance of unauthorized persons.

Benefits

The described face recognition system based on a neural network in combination with equipment for image acquisition offers an automated control system which performs an identification of persons from their facial image. It works objective, fast, without fatigue and causes no extra personnel costs even in 24h-operation.

Generalization

In contrast to customary applications of neural networks in which great amounts of data have to be processed, this application opens up possibilities in fields such as image processing where the focus is on data comparison.

Contact person

Dr. Jörg Kopecz - Zentrum für Neuroinformatik GmbH. Universitätsstraße 160, D-44801 Bochum. Phone: +49 234 9787-25, Fax: +49 234 9787-77, E-mail:joerg@berlin.ZN.ruhr-uni-bochum.de

References for further reading

J. Kopecz et al. ZN-Face: A system for Access Control Using Automated Face Recognition. In: Neural Networks: Artificial Intelligence and Industrial Applications. Proceedings of the 3rd Annual SNN Symposium on Neural Networks, Nijmegen, The Netherlands, 14-15 September 1995, Nijmegen. B. Kappen and S. Gielen (eds). Springer-Verlag, London.