WHAT YOU NEED TO KNOW:
- A new technology using the vein patterns from the fists of people can soon be used to digitally identify a person to grant access to devices or data.
- Using 3D cameras and artificial intelligence, outcomes showed that vein patterns from photographed fists of 35 participants are 99% accurate in identifying an individual.
- The team of scientists from Australia’s University of New South Wales said they used the vein pattern approach because not only it is hard to circumvent the technique, it also doesn’t leave imprints and cannot be obtained through social media.
Facial recognition technology may become passé when you can soon be identified using the veins of your hands, say researchers from the University of New South Wales in Australia.
From airport security checks to police departments, as well as other sectors, biometric recognition has become widespread over the years.
But a team of researchers, who conducted the study, have pointed out ‘well-known weaknesses’ of some biometric methods.
Such weaknesses include being able to create dummy prints from collected fingerprints from surfaces touched, using contact lenses to confuse iris-based mechanisms, or copy images from social media to bypass facial recognition technology.
A vein-based approach was taken up instead in the study which Syed Shah and his team from the university’s School of Computer Science and Engineering, believed to be a more difficult technique to bypass.
“Vein patterns lie underneath the skin, thus do not leave any imprint, unlike fingerprints, are not available over social media, unlike facial photographs, and cannot be obtained surreptitiously, unlike irises,” Shah told CNN in an email.
Because the study required specialist technology, the researchers used an Intel RealSense D415 Depth Camera, an off-the-shelf 3D camera, to take some 17,500 images of the fists of 35 participants.
With the use of artificial intelligence, distinguishing features were drawn from the vein patterns of the back of the hands, which they say, can be used to identify a person with an accuracy of 99 percent.
“Specially, the requirement of making (a) fist for vein extraction makes it difficult for an adversary to obtain vein patterns furtively,” Shah said, adding the method could also be used to validate individuals on laptops, mobile phones and other personal devices.
The study was published in IET Biometrics.