Home / Blog / Eye Spy

Eye Spy

Even if we keep it relatively simple by using the same password – or variants of it – for multiple purposes, or we use some sort of single-sign on application, it’s still hard and irritating to keep track of all the pin-codes and passwords we need to keep everyday life running smoothly. Given how often we all use our phones in the course of a day, one of the most irritating is the phone password. And yet it’s one of the most important ones, given how relatively easy it is to lose a smartphone – as opposed to an IoT device attached to the wall in our home – and how much info so many of us keep on those smartphones.

Biometrics has been coming to the rescue for years to replace the old-school password approach. Most common to date are the fingerprint identification methods that some smartphones offer. Fingerprints, as it turned out, can be stolen. I read somewhere that PlayDo can even be used to do the job. Then there’s the fear that your finger could get hacked off. (Fortunately, it turns out that the technology can detect whether the finger is attached to a live body or not.) Then there’s the more reasonable fear that the fingerprint database in the sky will be hacked. And, unlike with a password, you can’t just go and change your fingerprint.

Now all eyes seem to be turning to iris-scans and/or facial recognition. And one of the technology leaders here is FotoNation, the company that – bless them – came up with the algorithms that cameras and smartphones have been using for years to eliminate “red eye” photos.

“Other algorithms exist for robust iris recognition, but according to FotoNation it has the only IP that does not need connection to cloud computing resources. Running in standalone mode [Sumat] Mehta claims it has a one in 10 million false-acceptance rate, compare to 1-in-10,000 for its closest competitor. Mehta attributes its accuracy to its use of facial tracking of both irises simultaneously.” (Source: EE Times)

Well, you can’t believe everything you read, but this does sound interesting. They don’t need the cloud because they’ve developed hardware accelerator IP that enables extremely fast locally processing.

“[FotoNation’s] algorithm, as almost all artificial intelligence (AI) algorithms today, is based on a multi-layered neural network performing deep learning on user data. And because it tracks your facial features too, you don’t have to stare at the phone to get it to work. It also will cannot be “spoofed” according to Mehta, by taking a photo of the user.”

FotoNation’s not the only player here, of course.  The Samsung Galaxy Note7 has an iris scanner. (Apple is not expected to incorporate it in the iPhone for another couple of years. But with all the problems Galaxy has been having with exploding devices, I’m not sure I’d feel all that confident holding one up to my eye for scan every time I wanted to use it!)

It will be interesting to see whether iris scanning (and facial recognition) will be incorporated in consumer applications, as they have been in security systems and other commercial/industrial settings.