backtop

Print 14 comment(s) - last by 91TTZ.. on Jul 16 at 2:28 PM
Studies find iris changes with age

There are numerous different security technologies on the market today to limit access to digital information and for physical security on doors or gates. One common method of limiting access to specific areas in a building includes iris recognition systems. These are the systems that scan a person's eyes to determine their identity and allow them access to restricted areas.
 
Since the early days of iris recognition technology, users and developers have assumed that the iris doesn't change over time. If the iris doesn't change over time that means a person can be enrolled once and the system should recognize them indefinitely. However, recent studies have shown that the iris of a person's eye can in fact change over time leading to the possibility of false negatives and false positives.
 
"The biometric community has long accepted that there is no 'template aging effect' for iris recognition, meaning that once you are enrolled in an iris recognition system, your chances of experiencing a false non-match error remain constant over time," stated Kevin Bowyer, Notre Dame's Schubmel-Prein Family Chair in Computer Science and Engineering.
 
"This was sometimes expressed as 'a single enrollment for life.' Our experimental results show that, in fact, the false non-match rate increases over time, which means that the single enrollment for life idea is wrong."
 
Bowyer worked with Sam Fenker, an undergraduate from Notre Dame, to analyze a very large data set containing iris images acquired over a longer period of time. The researchers were able to analyze iris images year to year for three successive years to determine if changes occurred. The research found that the iris could in fact change over time requiring reenrollment.
 
Bowyer added, "I do not see this as a major problem for security systems going forward… In the long run, researchers may develop new approaches that are 'aging-resistant.' The iris template aging effect will only be a problem for those who for some reason refuse to believe that it exists."

Source: Eurek Alert


Comments     Threshold


Simple solution
By Belard on 7/13/2012 10:48:06 AM , Rating: 2
Every other year, re-scan the eye... takes about a minute or two. The person enters the room, confirm with scan to be true. hits the scanner. leaves.

The changes would be very very minor, but the system will be up to date.

Besides, people in the field in which such tech is used maybe only need it for 4~20 years.




RE: Simple solution
By 91TTZ on 7/13/2012 11:00:21 AM , Rating: 2
Or, they can allow for a certain percentage of change per unit of time. For instance scientists might find that the shape of the iris changes 2% every year, so the programming can allow for a 2% deviation within one year of the initial scan and it would progress to 4% deviation allowed 2 years after the scan. But if someone else tries to fraudulently use the system and get in, it would detect a drastic change overnight.


RE: Simple solution
By bupkus on 7/13/2012 12:02:04 PM , Rating: 3
Certainly the change delta will vary between individuals and most certainly those with a degenerative and/or disease condition.
This offers a huge opportunity in ophthalmology to track and study eye change perhaps offering an early warning system for some eye diseases.

As for this security problem, the stored profile can be changed each time a person tests for entry. If the test scan is defined to be of poor quality (by some metric) then the test scan would not replace the stored identifier and either trigger a fail or report to admin that the scanner is out of adjustment.
In either case, a history of scans can be kept on each member and certainly of each fail. This tracking could provide a learning module within the testing system to adjust to progressive alterations.


RE: Simple solution
By bupkus on 7/13/2012 12:09:42 PM , Rating: 3
In addition, it is possible that this technology can be used to screen for substance abuse and either trigger a urine test request by the HR dept or termination having detected a progressive disease that the HMO would flag as a high risk factor.
Yes, abuses are a very possible result.


RE: Simple solution
By Bad-Karma on 7/13/2012 5:55:44 PM , Rating: 2
If a company doesn't want drug users in their employ then that's their right. If you go to work for a company that has a strict no-drug-use policy, and you decide to do so anyways, then you deserve to be let go.


RE: Simple solution
By bupkus on 7/13/2012 8:18:47 PM , Rating: 2
quote:
or termination having detected a progressive disease that the HMO would flag as a high risk factor. Yes, abuses are a very possible result.
This is the part of the sentence that you seemed to have ignored. Detecting drug use through covert means is another issue I had not intended to address but as long as you've raised that point I am willing to combine both as abuse of power.


RE: Simple solution
By bupkus on 7/13/2012 8:21:44 PM , Rating: 2
Edit:
quote:
I am willing to combine both as abuse of privacy.


RE: Simple solution
By Bad-Karma on 7/14/2012 2:25:27 AM , Rating: 3
Termination for a a genetic defect or disease has already been ruled against by the U.S courts as long as it does not affect your ability to perform the job.

How it is an abuse of privacy if you understood and signed the terms of employment?

And again, it doesn't matter if it is a covert sampling. If you willingly violate the terms of your employment then the onus is on you and you should be removed.

You want to keep your job,then simply follow the agreed to rules.


RE: Simple solution
By 91TTZ on 7/16/2012 2:28:48 PM , Rating: 2
Actually it does matter if it was a covert sampling. How information is obtained matters greatly when it comes to law. Usually a person needs to consent to a drug test.


RE: Simple solution
By bebimbap on 7/13/2012 12:02:40 PM , Rating: 2
eventually you would have to rescan the eye. After a while you'll hit the upper limit and even a completely different eye might eventually become a false positive.

Honestly no bio-metric will work as a once in a lifetime. Even DNA changes over time, for instance when you get infected by a DNA virus like Heb B.


RE: Simple solution
By MozeeToby on 7/13/2012 1:08:23 PM , Rating: 2
So, lets say the cutoff is 2 years. After 2 years, your iris will have changed enough so that, oh I don't know... .01% of people would match if you allowed the necessary fudge factor. So long as the person uses the system at least once every 2 years, there's no problem. User wants to unlock the door, system scans the eye, verifies you are who you say you are, updates the records with the new scan, opens the door. Now your record is never out of date.

As for DNA, identification only looks at a tiny subsection of your DNA. The odds of a virus infecting a significant number of cells in your body and making edits to the important sections and those cells surviving the infection long term are near enough to zero to make no difference.


RE: Simple solution
By MrBlastman on 7/13/2012 12:09:27 PM , Rating: 4
Or better yet, have a living database of scans for each subject. Each time a participant has their eyes scanned for entry, since it was already analyzed for variance to obtain entry, as long as it was within reasonable parameters have the eye updated into the system each and every time with a more current representation.

I find it amusing that someone would actually think a part of the human body doesn't change over time. We might have DNA but even that denatures. We're organic, electro-chemical machines--of course we will never remain 100% the same forever.


work at home
By PittmanJack on 7/15/12, Rating: 0
work at home
By PittmanJack on 7/15/12, Rating: 0
"Nowadays, security guys break the Mac every single day. Every single day, they come out with a total exploit, your machine can be taken over totally. I dare anybody to do that once a month on the Windows machine." -- Bill Gates









botimage
Copyright 2012 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki