Fingerprint verification is an important biometric technique for perso
nal identification, In this paper, we describe the design and implemen
tation of a prototype automatic identity-authentication system that us
es fingerprints to authenticate the identity of an individual. We have
developed an improved minutiae-extraction algorithm that is faster an
d more accurate than our earlier algorithm [58]. An alignment-based mi
nutiae-matching algorithm has been proposed. This algorithm is capable
of finding the correspondences between input minutiae and the stored
template without resorting to exhaustive search and has the ability to
compensate adaptively for the nonlinear deformations and inexact tran
sformations between an input and a template. To establish an objective
assessment of our system, both the Michigan State University and the
National Institute of Standards and Technology NIST 9 fingerprint data
bases have been used to estimate the performance numbers. The experim
ental results reveal that our system can achieve a good performance on
these data bases. We also have demonstrated that our system satisfies
the response-time requirement, A complete authentication procedure, o
n average, takes about 1.4 seconds on a Slut ULTRA 1 workstation (it i
s expected to run as fast or faster on a 200 HMz Pentium [7]).