A biometric access control system is developed using automated iris recognition. Based on the fact that a human iris contains patterns that are best suited for identification, the use of low cost equipment can help iris recognition to become a standard in security systems. In this paper the database containing images from a low cost camera is assessed, and the overall recognition performance was measured. The pattern matching is performed using statistical metrics. The real-time experimental results on a group of 50 persons are presented. The results showed that a system using low cost equipment can be constructed with a promising 95% accuracy rate.