MINUTIAE EXTRACTION BASED ON ARTIFICIAL NEURAL NETWORKS FOR AUTOMATIC FINGERPRINT RECOGNITION SYSTEMS

Abstract

Automatic fingerprint recognition systems are utilised for personal identification with the use of comparisons of local ridge characteristics and their relationships. Critical stages in personal identification are to extract features automatically, fast and reliably from the input fingerprint images. In this study, a new approach based on artificial neural networks to extract minutiae from fingerprint images is developed and introduced. The results have shown that artificial neural networks achieve the minutiae extraction from fingerprint images with high accuracy.

Keywords

• Biyometri, Parmakizi, Özellik noktaları, Yapay sinir ağları, Çok katlı perseptronlar.