Future Eye Scanners Must Combat Aging Eyes (Live Science)
The iris — the colored part of the eye that eye-scanners analyze — changes as people age, making the scanners more likely to wrongly lock out people with every passing year, according to a new study.
The finding goes against the established, yet never-proven notion that eye scanners can accurately identify people throughout their lives, said Kevin Bowyer, a computer scientist at the University of Notre Dame who performed the study.
Read the whole thing. It’s an article that gets at an interesting aspect of the algorithm end of the biometric ID management problem. It also has input from two of the speakers at the recent TechConnectWV event: Marios Savvides (Carnegie Mellon) and Bojan Cukic (W. Va. Univ.).
A good biometric modality must be: unique, durable, and easily measurable. If any of these are missing, widespread use for ID management isn’t in the cards. If something is unique and durable but isn’t easily measurable, it can still be useful but it isn’t going to become ubiquitous in automated (or semi-automated) technology. Teeth and DNA fit this model. Teeth have been used to determine the identity of dead bodies with a high degree of certainty for a long time, but we aren’t going to be biting any sensors to get into our computers any time soon — or ever. Likewise with DNA.
There is also the challenge of proving that a modality is in fact unique, durable and easily measurable which requires a whole lot of experimental data, and especially regarding uniqueness, a healthy dose of statistical analysis. I’m no statistician, and from what I understand, the statistical rules for proving biometric uniqueness aren’t fully developed yet anyway, so let’s just leave things in layman’s terms and say that if you’re wanting to invent a new biometric modality and someone asks you how big a data set of samples of the relevant body part you need, your best answer is “how much can you get me?”
In order to ascertain uniqueness you need samples from as many different people as you can get. For durability you biometric samples for the same person taken over a period of time and multiplied by a lot of people.
Ease of measure is more experiential and will be discovered during the experimentation process. The scientists charged with collecting the samples from real people will quickly get a feel for the likelihood that people would adapt to a given ID protocol.
For two of the “big three” biometric modalities, face and fingerprint, huge data repositories have existed since well before there was any such thing as a biometric algorithm. Jails (among others) had been collecting this information for a hundred years and the nature of the jail business means you’ll get several samples from the same subject often enough to test durability, too, over their criminal life. These data could be selected such that they were as good as they could be to assess both uniqueness and durability. For face, other records such as school year books exist and were readily available to researchers who sought to measure uniqueness and durability.
Which brings us to iris.
Where do you look to find a database of several million high-resolution images of human irises collected by professionals who took good notes? Well there’s your problem.
The solution is to go about building such a data set yourself and several organizations have been doing just that. One can make considerable progress on in the question of uniqueness with a big push, collecting more data quickly. Assessing durability, however, takes time no matter how much money and effort can be applied. Some processes can be sped up with more resources; some can’t (nine women can’t make a baby in a month) and the real bummer with determining biometric durability is that you can’t really know in advance how much time it’s going to take to prove it to a satisfactory degree.
So it’s not a surprise that the uniqueness of the human iris was determined before its durability, and it may come about that the iris is, like the face, “durable enough.” We are all too aware that the face changes, but certain aspects of it don’t change so much that facial recognition is pointless. The same may be true of the iris. It, too, may be durable enough.
It may also turn out to be the case that irises change in a predictable way and that those changes can be accounted for on the software side, so all this isn’t to say that iris isn’t among, or won’t solidify its position among the “big three”; it’s just had a harder road to get there.
