1(630)802-8605 Ravi.das@bn-inc.net
In today’s blog, we continue with Facial Recognition, whether you love it or hate it!

The Application Environments for Facial Recognition System Applications

Unlike the other biometric systems we have reviewed in previous articles (iris recognition, fingerprint recognition, and hand geometry recognition) implementing a facial recognition system can be much more complicated.

For example,  the other biometric systems can work in different kinds of application environments, and to a certain degree,  are not impacted as much by the external variables which are present. This is not the case with facial recognition, as the performance of it can be greatly influenced by the type of application setting it is used for, and the external variables which are present.

The application environments for facial recognition systems can be further subdivided as “controlled” (verification) and “random” (identification).  In a controlled environment, there is not much variation.  The user will look normally into the camera, and good quality enrollment and verification templates will be produced.

A typical example of a controlled environment is that of physical access entry at a particular location or site. However, in a random setting, there is a lot of variation.  A typical example of a random setting is that of surveillance.  Facial recognition systems have been used at airports for such purposes, and the results have been very mixed.

This is because the facial recognition system has to identify and filter faces from different scenes which contain a lot of extraneous background noise (the “external variables”).  The common belief is that the facial recognition system will be as good as the human brain in identifying individuals.  However, reality dictates the opposite.

The human brain has evolved over thousands of years.  Facial recognition technology, although improving steadily, has a very long way to go before it reaches the sophistication of the human brain.  Also, unlike finger scan and iris recognition where a positive match can be achieved, the facial recognition system will return a predetermined number of potential matches in a random environment.  It is then up to the system administrator to determine the positive match.

In tomorrow’s blog, we look at some of the various technologies of Facial Recognition.