Fingerprint registration using genetic algorithms

Abstract

In automated fingerprint identification systems, an efficient and accurate alignment algorithm in the preprocessing stage plays a crucial role in the performance of the whole system. We explore the use of genetic algorithms for optimizing the alignment of a pair of fingerprint images. To test its performance, we compare the implemented genetic algorithm with two other algorithms, namely, 2D and 3D algorithms. Based upon our experiment on 250 pairs of fingerprint images, we find that: genetic algorithms run ten times faster than a 3D algorithm with similar alignment accuracy; and genetic algorithms are 13% more accurate than a 2D algorithm, with the same running time. The conclusion drawn from this study is that a genetic algorithm approach is an efficient and effective approach for fingerprint image registration.