page 2

Types of biometric technologies

Finger scanning is by far the most widely used biometric technology, and is also the oldest and most well-researched. Extensive use in the private sector would make it the top biometric even if law enforcement agencies were excluded. As a mature technology with decades of development behind it, finger scanning is inexpensive, easy to use, deployable in many different environments, and very reliable.

This technology does have weaknesses, however. Most private-sector applications provide for scanning only one finger, which makes them less reliable than law enforcement applications that rely on full sets of fingerprints. Also, scanners cannot accurately record fingerprints if skin is too dry or too moist or if the scanner itself is not regularly cleaned.

Facial scanning is the most controversial technology. Although facial recognition technologies meet with high levels of acceptance when used overtly for identification and access, acceptance drops significantly when they are used covertly. Because it is the only biometric that can be used at a distance, without the subject’s knowledge or consent, it poses greater privacy problems than any of the other technologies.

There are advantages to this software-based technology, such as ease of use and the low hardware costs. But it also has the lowest accuracy rate of the top five biometrics. Changes of facial expression, hairstyle, makeup, even body weight, can cause errors, as can lighting that is too bright or too dim.

Hand geometry scanning does not use handprints but records distinctive geometric structures, in particular the height and width of the back of the hand and the fingers. This is a well-established technology, but it has limited application because hand measurements are not as distinctive as fingerprints or iris patterns.

The strong points of hand geometry scanning are that the equipment can be deployed in challenging environments, it’s more resistant to fraud than fingerprints, and it has a high level of user acceptance. Its drawbacks are a low level of accuracy — higher than facial and voice recognition, but lower than finger and iris scans — and the high cost of scanning hardware.

Iris scanning technology uses the unique pattern formed by the iris. The iris is an ideal biometric because no two irises are alike, even in the same person, and because iris patterns contain even more unique data than fingerprints. Of all the leading biometrics, iris scanning has the highest level of accuracy. This fact means it can be used not only for verification — confirming that a person has presented biometric data that matches the data on file for that person — but also for the far more difficult job of identification — identifying an individual by searching a large database of biometric data.

The weaknesses of iris scanning are that it requires some training and reliance on proprietary hardware and software, making it relatively expensive. Subjects also report physical and psychological discomfort with the technology.

Voice recognition combines elements of behavioral and physiological biometrics. Text-dependent systems prompt subjects to say something specific while text-independent systems process anything the subject says. The strong points of this technology are that it is easy to use, can make use of existing telephony hardware, is inexpensive, and has a higher level of public acceptance than any of the other leading biometrics.

The major weakness of this technology is that it has a lower level of accuracy than all but facial recognition. Individual variations in voice, such as those caused by infections and allergies, make this biometric too unstable for accurate readings.

< previous page | 1 | 2 | 3 | next page >