Minutiae Matching Algorithm Using Artificial Neural Network for Fingerprint Recognition

Abstract

This article present a minutiae matching algorithm using hardware based artificial neural network intended to be implemented in embedded system environment. Hardware based artificial neural network CM1K consist of 1024 neuron in Cognistix device is used in this research. Fingerprint template as a vectors of minutiae points is used as an input to artificial neural network device. Matching process is based on the calculation of the distance from two vectors from two set of minutiae points comming from two fingerprints. Using hardware based of artificial neural network provide faster processing in matching process and the used of Cognistix making faster and easy in development of prototype.