|
|
 AFIS: Automatic Fingerprint Identification System
Fingerprint-based identification is the oldest and most used method
which has been successfully used in numerous applications. The
uniqueness of a fingerprint can be determined by the pattern of ridges
and furrows as well as the minutiae points, endings and bifurcations
of the ridges.
Fingerprint matching techniques can be placed into two categories:
minutiae-based and correlation based. Minutiae-based techniques first
find minutiae points and then map their relative placement on the
finger. Correlation-based techniques require the precise location of a
registration point and are affected by image translation and rotation.
The first approximation is the most robust to image translation and
rotation but it used to be difficult to achieve a good minutiae
extraction from noisy fingerprint image. To accomplish this difficult
task a sequence of preprocessing techniques are applied to the
fingerprint image:
A commented view of the preprocessing of fingerprints.
From each fingerprint the above preprocessing is accomplished
obtaining the corresponding set of minutiaes. Once the minutiae points
are detected it is needed a measure of the similarity of two fingerprint
using these minutiae points as the unique information from the
fingerprint. The invariance to translation, scaling and rotation is
accomplished by performing all these operations in the matching of the
two set of minutiaes. The best match is selected and the similarity of
the two fingerprint is given in terms of the similarity of the
minutiae sets.
An identification of a fingerprint facing it with two reference fingerprints.
|
|