Where Biometric readers are used we utilise a toolkit (SDK) provided by Digital Persona for the technology. Below is the details from their developer documentation covering how it works. Up to GPoS 158, version 1 of the SDK was used and we used the FID method. From GPoS 159 onward you can use Version 2 of the SDK and we then use the FMD method.
Standards Compliance
DigitalPersona U.are.U SDK version 2 Fingerprint Image Data (FID) format is compliant with
ANSI INSITS 381-2004
ISO/IEC 19794-4:2005
DigitalPersona U.are.U SDK Version 2 Fingerprint Minutiae Data (FMD) format is compliant with
ANSI INSITS 378-2004
ISO/IEC 19794-2:2005
DigitalPersona U.are.U SDK version 2 can process FIDs and raw images generated by other systems as long as
they are compliant with the aforementioned standards and:
Are Uncompressed
Have horizontal and vertical resolutions that are the same (square pixels)
Have 8 bit per pixel
For multiview image records, we also require:
No more than 16 views per finger
All views in a multiview image must have the resolution and same pixel dimensions (Width x Height)
We do not support unknown finger positions (finger position = 0).
DigitalPersona U.are.U SDK version 2 can process minutiae data generated by other systems as long as they are
compliant with the aforementioned standards and meet this criterion:
Horizontal and Vertical resolutions are the same (square pixels)
Biometric solutions offer many advantages that other technologies cannot provide.
Uniqueness - Fingerprints from each one of our ten fingers is distinctive, different from one another and
from those of other persons. Even identical twins have different fingerprints.
Convenience - Users no longer have to remember multiple, long and complex, frequently changing
passwords or carry multiple keys.
Non-repudication - Ensures the user is present at the point and time of recognition and later cannot deny
having accessed the system.
Non-transferable - Cannot be shared, lost, stolen, copied, distributed or forgotten unlike passwords, PINs,
and smart cards.
Proven - Long history of successful use in identification tasks - the U.S. and other countries have extensive
real-world experience with fingerprint recognition. Fingerprints have been used in forensics for well over
a century and there is a substantial body of scientific studies and real world data supporting the
distinctiveness and permanence of fingerprints.
The Basics of Fingerprint Identification
The skin on the inside surfaces of our hands, fingers, feet, and toes is “ridged” or covered with concentric raised
patterns. These ridges are called friction ridges and they provide friction making it easier for us to grasp and
hold onto objects and surfaces without slippage. The many differences in the way friction ridges are patterned,
broken, and forked make ridged skin areas, including fingerprints, distinctive.
The distinctiveness of fingerprints is well established. The underlying biological persistence of fingerprint
characteristics is also a well established fact reported in various fingerprint studies conducted in different
scientific fields over the past century.
Fingerprint Characteristics
Fingerprint ridges are not continuous, straight ridges. Instead, they are broken, forked, interrupted or changed
directionally. The points at which ridges end, fork, and change are called minutiae points which provide
distinctive, identifying information.
The most common properties of fingerprint minutiae points are:
1. Type
There are several types of minutiae points, some of which are listed below. The most common are ridge
endings and ridge bifurcations.
· Ridge ending – the abrupt end of a ridge
· Ridge bifurcation – a single ridge that divides into two ridges
· Short ridge, or independent ridge – a ridge that commences, travels a short distance and then ends
· Island – a single small ridge inside a short ridge or ridge ending that is not connected to all other ridges
· Ridge enclosure – a single ridge that bifurcates and reunites shortly afterward to continue as a single ridge
· Spur – a bifurcation with a short ridge branching off a longer ridge
· Crossover or bridge – a short ridge that runs between two parallel ridges
· Delta – a Y-shaped ridge meeting
· Core – a U-turn in the ridge pattern.
2. Direction
3. Position
Issues in Fingerprint Recognition Technology
In a perfect world, it would be a simple matter to determine whether two fingerprints were from the same
finger-- the images would be identical or they would not. However, Even though our fingerprints do not change over time, the fingerprint images can vary a lot, especially for
some people. For example, certain skin conditions and wear due to manual labour can affect fingerprint
images. This makes fingerprint recognition a very challenging problem that does not have a perfect
solution. As the result, captured fingerprint images are often not a perfect match to the stored image
from the same finger.
Fingerprint images from two different fingers of two different people can look similar, especially when,
because of worn fingerprints or temporary creases, there is very little information left about the actual
fingerprint. The larger the population your are working with, the more likelihood of similar fingerprint
images.