Scientists develop inexpensive X-ray camera technology that can identify people
A team of scientists at the University of Washington have developed an inexpensive hyperspectral camera technology that can capture X-ray photos of small items, such as a hand, to identify people, according to a report by International Business Times.
Using hyperspectral imaging, the HyperCam is able to scan 17 wavelengths of visible and invisible near-infrared light from across the electromagnetic spectrum to view details that are otherwise invisible to the human eye.
Hyperspectral imaging is frequently used in a variety of fields. For example, the technology is used to identify new oil fields or mines, to scan ancient manuscripts and artefacts, to diagnose potential medical issues, to spot problems in food processing, and for military surveillance.
Although there are a few hyperspectral cameras currently on the market, they are quite expensive and not designed for consumer use.
The researchers plan to apply the hyperspectral camera technology to develop a pocket-sized camera that costs either $800 (£518) on its own, or $50 as a camera attachment for smartphones.
The hyperspectral camera could be used for biometrics purposes to identify an individual, as well as to recognize a gamer’s gestures when interfacing with a video game.
The HyperCam includes intelligent software that is designed to find the hidden discrepancies between photos taken by a regular camera that simply divides light into red, green and blue, and x-ray photos taken using hyperspectral imaging.
“It’s not there yet, but the way this hardware was built you can probably imagine putting it in a mobile phone,” said Shwetak Patel, Washington Research Foundation Endowed Professor of Computer Science & Engineering and Electrical Engineering at University of Washington.
In one test, researchers captured photos of the hands of 25 participants, photographing the back of the hand.
The HyperCam images unveiled a significant number of details about the veins in each hand, enabling the software to identify the 25 people with 99% accuracy.
The technology does not currently work well in bright light. However, researchers are working to improve this capability as well as developing a smaller prototype of the camera that can be attached to a smartphone.
The scientists detailed their findings in a report entitled, ‘HyperCam: Hyperspectral Imaging For Ubiquitous Computing Applications,’ which was published in the Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing.