Neural network-based minutiae extraction for fingerprint verification system

Fingerprint is one of the most important biometrics that has been employed for verification systems. Fingerprint is characterized by two fundamental properties; Easy to acquire, and it is unique for each person. This paper presents minutia extraction method based on Neural Network-based. These features can be used in verification systems. The verification process includes four main phases: image acquisition, preprocessing, feature extraction, and pattern matching. The method is applied on a set of fingerprint images and the results show that the average matching accuracy of fingerprints is 91.6%.