Since "Minority Report", where Tom Cruise toted a squishy bag full of spare eyeballs around to hold up in front of iris scanners, thus fooling the access systems, biometrics has been a buzzword, if only a minor one, but it has failed to catch on in a meaningful way. A few years back I speculated that this is because every existing biometric method has serious drawbacks. Fingerprints fade as you grow older, and some people don't have any because they are afflicted with a rare disease called "Naegeli syndrome" or dermatopathia pigmentosa reticularis (DPR) that can cause vexing social problems. Recently, two identical twins were indicted for robbing the department store KdW in Berlin, but had to be released when the authorities found that it was impossible to determine which of them had been actually done the heist since they share the same DNA. And many people instinctively refuse to put their eye to an iris scanner because they worry that they may be blinded by a flash of light from a malfunctioning machine.
Now, the weekly newsmagazine The Economist has come up with what may prove to be the perfect biometric identifier: the human knee. According to the story, Lior Shamir, a geneticist at the National Institutes of Health in Maryland, has developed a knee-analysing mathematical algorithm for medical use. Knees, it seems, are unique in each individual human. By exploring X-ray images of the general structure of various knees and then using their brand-new algorithm to look at them in greater detail, for instance by measuring the texture of the bone through monitoring differences in individual pixels, the researchers found that the best identification was possible by concentrating on a smaller image of the centre of the joint rather than the entire knee. The team also points out that the algorithm can correctly identify a given pair of knees and match it to a specific individual in the database even if the original X-ray were taken several years earlier.
According to Mr Shamir, the success rate still needs to be improved. In the International Journal of Biometrics, his team reports it achieved a correct match 34% of the time. It was also able to pick the ten closest matches to a particular knee 56% of the time - still far from the degree of accuracy provided by established biometric systems. But as Shamir remarks, it's early days yet for the science of knee identity management, and given time (and grant money) they hope to get there.
Naturally, this raises the question of how to build a viable world-wide identity infrastructure based on knee ID. Rumors have it that Samsung is secretly developing a "deskbottom" knee scanner (DKS) which fits comfortably under a table. Portable models can't be that far away, and we can easily imagine laptops with built-in knee scanners.
Of course there are still numerous social issues which need attention. Baring one's knees in public is frowned on in some cultures, and it may prove akward in places like airplane seats or boardroom meetings. However, over time we can expect to see a shift in cultural biases, given the obvious advantages of knee-based recognition systems. In the end, the Economist's tongue-in-cheek sum-up may well prove prescient: not the ayes (or eyes), but "the knees have it"...