Aging Irises a Complication for Biometric Security Systems

Aging irises could cause problems for biometric identity checks based on iris recognition. This significant obstacle affecting biometric identification was discussed in a recent article published at the Homeland Security Newswire. We haven’t seen any real discussion of this in other articles, and it’s interesting to see some of the problems that can and will arise once new biometric technologies become more prevalent in our daily lives.

Confirming someone’s identity through iris recognition is a pretty high tech process and involves matching a scan of their iris to a template in a library. Systems are designed to cope with the fact that scanning processes are variable, meaning that two separate scans of the same eye could be slightly different. But that actual degree of difference (called the Hamming distance) tends to increase over time for scans of the same eye.

“Biometric traits are not only unique but are continuous and consistent for life.” Well, that’s what the marketing tells us. But researchers have found that’s not the case; it seems that when it comes to iris recognition, it’s not so. Iris scans tend to produce subtly different patterns over time, and the older the image of a person’s iris, the more likely the system will fail to make a concrete match from a new scan.

Iris recognition systems work by scanning a person’s eye to create a digital representation, a template of the iris’ texture. To confirm someone’s identity, you have to match a new scan of his or her iris against templates in a library. If two scans of the same eye are made, the Hamming distance should be small, but Kevin Boyer, a researcher at the University of Notre Dame is saying they’ve seen evidence that Hamming distance increases dramatically over time.

In a study, Bowyer and his team compared 26 irises taken two months apart with scans made four years apart. The team then measured an increase of 0.018 percent over the period – which they describe as an increase in the false rejection rate, or the rate at which an iris scan is not matched with its template. A system with a false rejection rate of 1 in 100 using scans two months apart would produce 1.75 false rejections in 100 scans after four years. This could be a serious problem in places such as India, where officials are considering storing iris scans in chips used in national identity cards.

Aging irises would hobble biometric identity checks based on iris recognition

Related articles:

1. Biometric Devices: Ear Scanning, An Alternative in Airport Security
2. Taking Biometric Security to the Next Level With Ear Scanning
3. Iris Recognition: Biometric Security in Mexico
4. Can Biometric Security Balance Between Safety and Efficiency?
5. New Biometric Devices in Hospitals
6. The All-In-One AOptix Face and Iris Recognition System
7. Homeland Security to Start Testing Iris Recognition
8. Minority Report Biometric Devices: Reality or Fiction?
9. Eye Tracking: Iris Scanning Is Coming, Like It Or Not
10. Biometric Security: Crime Fighting Goes Minority Report