There are unresolved issue inherent to this standard. Currently
other countries around the world are willing to adopt the expected
standard. The FBI has been studying the electronic transfer of
fingerprint image data for over four years. This research has
given them a head start over the technical committees of standard
bodies. However many of the specifications that are being adopted,
are tailored for the existing FBI infrastructure. For example,
the standard relies upon FBI certified encryption for the security
of data transmission across networks. It is unclear if international
bodies are willing or able to use American encryption techniques
for the transmission of their own data.
One of the most significant issues regarding this standard is
examining the need for the standard. The push for a standard
in biometric identification has come as countries are grappling
with the need to identify individuals. The need for standardization
has evolved because there is an urgency for countries to be able
to exchange information regarding individuals. However if each
country adopts a different form of biometry, we are no better
off then before the standardization process. What good is a perfect
transmission of a terrorist's subcutaneous vascular network from
Israel's database if U.S. airports are only able to verify people
by their fingerprint? There is the potential for hundreds of
virtual "stand alone" databases. Obviously the standard
does have potential uses as internal security for organizations,
fraud cases with credit cards or EBT's and other homogeneous operations.
However before the international committees take up serious research
on biometric identification they must consider what measurements
can be internationally accepted. In addition to technical specifications,
this standardization process must include an awareness of social
and political implications.
Also, a number of vendors have made claims that their own systems
are nearly perfect and will provide test results as proof. However
one must be careful to understand the nature of the tests. There
are two measurements of success, the False Acceptance Rate (FAR)
or type 2 error and the False Rejection Rate (FRR) also called
a type 1 error. These rates are set according to the comparison
algorithm and may vary by intended application. For example,
a Type 1 error, falsely rejecting an authorized user would be
very bad if its application was an EBT. In this case an elderly
citizen would be denied their social security check because the
system failed to identify them correctly. However the FBI would
be concerned about type 2 errors, an unauthorized person being
identified as valid. This would allow terrorists to walk through
airports successfully passing security. Thus far, systems have
far better false acceptance rates than false rejection rates.
Although science and technology is capable of capturing and transmitting digital fingerprint images, there remains a great deal of logistics to be worked out in order for this standard to become successful. The fingerprinting quality is dependent on the operator's ability to record a good image. It is futile to send perfectly transmitted images, if the original is poor. Therefore it must be understood that image quality is first dependent on the operator's ability to capture high quality images.