Knee X-ray biometrics plan to fight spoofing
Accuracy is 'much better than random', apparently
US federal eggheads have proposed a novel method of preventing biometric ID systems being spoofed by the use of such things as contact lenses, fake fingerprints etc. Instead of easily-fooled systems of this sort, people should instead be identified using X-ray photographs of their knees.
Lior Shamir of the US National Institutes of Health and computer engineer Salim Rahimi of the State University of New York propose that knee X-rays be analysed automatically by the publicly-available "wnd-charm" algorithm, normally used for diagnosing medical problems in the knee joints.
According to the two researchers:
The advantage of using a biometric identification process based on this kind of imaging is that it would be so much more difficult for a fraudster to spoof the knees or other internal body part in the way that they might with artificial fingerprints or contact lenses ... 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. Identifiable features correspond to specific persons, rather than the present clinical condition of the joint.
Shamir and Rahimi tried out their theory on a dataset of 1700 kneebone X-rays, scanning them as 8 megapixel images and using the algorithm on a central area of 700x500 pixels.
Apparently the below-the-belt bone probe gear isn't actually very reliable: it isn't as accurate as iris or fingerprint scans. But it is "much better than random results", according to the two men.
They suggest that accuracy could be improved by refining the algorithm. The problem of possible knee cancer in frequent flyers caused by incessant kilt-level X raying at passport control might be dealt with using "an alternative imaging process such as terahertz".
Shamir and Rahimi's paper Biometric identification using knee X-rays is to be published in the International Journal of Biometrics. ®
Sponsored: Global DDoS threat landscape report