Advances in Fingerprint Technology.pdf

Fingerprint Enhancement
The performance of a fingerprint feature extraction and image matching
algorithm relies critically on the quality of the input fingerprint images. The
ridge structures in poor-quality fingerprint images are not always well
defined, and hence cannot be correctly detected. This leads to the following
problems: (1) a significant number of spurious minutiae may be created,
(2) a large percentage of genuine minutiae may be ignored, and (3) large
errors in minutiae localization (position and orientation) may be introduced.
To ensure that the performance of the minutiae extraction algorithm will be
robust with respect to the quality of fingerprint images, an enhancement
algorithm that can improve the clarity of the ridge structures is necessary.
Traditionally, forensic applications have been the biggest end users of fingerprint
enhancement algorithms because the important ridge details are frequently
obliterated in the latent fingerprints lifted from a crime scene. Overinking,
under-inking, imperfect friction skin contact, fingerprint smudges
left from previous live-scan acquisitions, adverse imaging conditions, and
improper imaging geometry/optics are some of the systematic reasons for
poor-quality fingerprint images. It is widely acknowledged that at least 2 to
5% of the target population have poor-quality fingerprints: fingerprints that
cannot be reliably processed using automatic image processing methods. We
suspect this fraction is even higher in reality when the target population
consists of (1) older people, (2) people who suffer routine finger injuries in
their occupation, (3) people living in dry weather conditions or having skin
problems, and (4) people who have poor fingerprints due to genetic
attributes. With the increasing demand for cheaper and more compact fingerprint
scanners, fingerprint verification software cannot afford the luxury
of assuming good-quality fingerprints obtained from the optical scanner. The
cheaper and more compact semiconductor sensors not only offer smaller
scan area but also typically poor-quality fingerprints.
Fingerprint enhancement approaches 55-58 often employ frequency
domain techniques 56,58,59 and are computationally demanding. In a small local
neighborhood, the ridges and furrows approximately form a two-dimensional
sinusoidal wave along the direction orthogonal to local ridge orientation.
Thus, the ridges and furrows in a small local neighborhood have welldefined
local frequency and orientation properties. The common approaches
employ bandpass filters that model the frequency domain characteristics of
a good-quality fingerprint image. The poor-quality fingerprint image is processed
using the filter to block the extraneous noise and pass the fingerprint
signal. Some methods can estimate the orientation and/or frequency of ridges
in each block in the fingerprint image and adaptively tune the filter characteristics
to match the ridge characteristics.