Fingerprint Identification Is Quick and Well-Established
Law enforcement and security have long used fingerprints as a means of identifying individuals. This is done using the whorls, lines, ridges and bifurcations associated with a specific finger.

It is not the image of the finger that is saved to the user’s access data folder, but rather a mathematical equivalent. Some systems also use the blood vessel patterns inherent in a finger to achieve the same goal.

Fingerprint identification systems have been on the market for some time and the technology is relatively inexpensive compared to some of the other biometric developments. The technology is also relatively straight- forward in operation and easy to use.

Some of the problems operators encounter include the need for regular maintenance. In this regard, routine cleaning of the finger scanner is necessary in order to maintain a relatively low false reject rate. According to one source, an ultrasonic-based fingerprint scanning system is under development that would eliminate this problem. Eye-Based Biometric Systems Gaining in Popularity
There are two flavors of eye identification employed in the biometrics field to date: retina and iris.

Retina identification systems effectively record the blood vessel patterns contained on the retina at the backside of the eyeball using a high-quality camera.

Enrollment is not as easy and quick as it is with other biometric technologies. Using a typical retina identification system, enrollment can take several minutes as the user must position their eye an inch or two from the imaging lens while a low-intensity light reads more than 300 points within a 450° circle. Any movement can render an image unusable, which means the process must begin again.

The typical user access file size for retina authentication is 96 bytes. Once enrolled, a user can expect verification in 1.5 seconds (see sidebar “Two Flavors of Authentication” on page 54 of November issue). When used in identification mode, response time is typically 5 to 8 seconds depending on the number of users in the database.

The iris identification system relies on the colored portion that surrounds the pupil of the eye. Contained in the iris are rifts, fibers, filaments, rings, pits and other elements that distinguish one person from another. In this case, a CCD camera operating at 30 images per second acquires high-resolution images of each user’s iris.

Enrollment in an iris-based system typically takes between 1 and 2 minutes and is distinctly different than that of the retina system. The user is made to look into a LCD rendering of their eye. The system creates a grid that captures the unique characteristics contained within each.

The result is a mathematical equivalent that the system stores in the user’s access data folder. The file size is typically 250 to 500 bytes, depending on the system. Response time during use is between 2 and 5 minutes but could be more when users are inexperienced and unaware of the various issues associated with field use.

Hand Geometry Offers Excellent Security and Quick Access
Like fingerprint identification, the security market has had a long history with hand geometry as a biometric identifier — especially when it comes to government customers. Just as each individual has uniquely different fingerprints, the same applies to the shape of the hand. For example, hand geometry was utilized on Scott Air Force Base (AFB) in 2003 to secure its Shiloh-Scott MetroLink rail system because it provides direct access to the base. In this case, the Department of Defense (DOD) elected to use verification instead of identification for user authentication by employing a PIN in conjunction with hand geometry.

In some systems, a flat pad called a platen is used to electronically map each user’s hand. More recent systems, however, employ a high-resolution camera to do this in conjunction with a pad with pins that fit between each finger set.

Enrollment usually takes only a minute or two, although after enrollment the authentication process only takes 3 to 5 seconds to perform.

Trends in Access Control Include More Than Just Access at the Door
There are three significant, identifiable trends in the electronic access control market that systems integrators should be aware of.

The first involves the placement of biometric identifiers within chip- and optical-based cards when it comes to user authentication. Although the U.S. Government has fueled the most recent move to a biometric attribute in access control, there are plenty of other opportunities to apply this new science in both commercial and retail applications. The second trend involves the rapid convergence of physical and logical access control using biometric identifiers.

“Convergence, which is what we call physical and logical access, is becoming more and more prevalent,” says Matt Halbgewachs, program manager for Axalto’s partners channels and alliances. “As it happens, we want to see a single identifier coupled with a smart card that contains a biometric identifier before you can enter through the door or logon to your computer, access your E-mail, or read data from your computer when you sit down at your desk.”

No longer is management merely concerned with access at motorized and hydraulic drive-up gates and doors. The same biometric access control that provides automatic access to employees, vendors, and visitors is also being put to work in protecting computer data.

The third trend involves the integration of access control with myriad nonsecurity applications. Nowhere is this more obvious than in the protection of benefits.

Halbgewachs says there are 33,000 Medicaid recipients that use Axalto’s smart card equipped with a fingerprint template in Tarrent County, Texas.

“The system uses the security inside the card to protect the information,” he says. “An algorithm inside the card also does the comparison. Texas likes this solution because it does not require them to maintain a huge database.”

Halbgewachs calls this access scheme match-on-card. Although its use will limit the amount of data that must be housed on the network, there may be some concern whether a high-tech criminal or terrorist can in some way create their own card using the same algorithm and data scheme.

Axalto’s Pattinson says that smart card security is such that any attempt to circumvent an authorized card will more than likely negate its use. In addition, the level of expertise associated with such an act of compromise is so high that it’s doubtful anyone would be successful. He maintains that no one has successfully compromised a smart card in 25 years.

For the complete story, see the November print issue of Security Sales & Integration magazine.

If you enjoyed this article and want to receive more valuable industry content like this, click here to sign up for our FREE digital newsletters!

Related Content