An affordable camera technology being developed by the University of Washington and Microsoft Research might enable consumers of the future to tell which piece of fruit is perfectly ripe or what’s rotting in the fridge.
The team of computer science and electrical engineers developed HyperCam, a lower-cost hyperspectral camera that uses both visible and invisible near-infrared light to “see” beneath surfaces and capture unseen details.
The human eye, though a complex organ, can only see a limited range. Of the entire electromagnetic spectrum, our eyes only perceive three color bands—red, green and blue. So scientists have long used hyperspectral imaging—a technology that carves the electromagnetic spectrum into hundreds of bands to create detailed data images beyond what the eye can see—for a variety of purposes. It’s used in agriculture and mining to look at things like the mineral content and moisture level of soil. In aerial hyperspectral photographs, certain types of soil or minerals will have specific spectral signatures that form patterns. Food safety inspectors can use hyperspectral cameras to assess foods for nutritional content or contamination by a non-food material. This type of camera is typically used other industrial applications and can cost between several thousand to tens of thousands of dollars.
“[HyperCam] automatically tries to define what’s useful in a scene,” explains lead author Mayank Goel, a doctoral student at the University of Washington who works on the HyperCam project. “It exaggerates what the human eye can’t see.”
For instance, HyperCam can see veins beneath human skin. These vein patterns, combined with the camera’s ultra-detailed images of a skin’s surface patterns, can be used for identification purposes. In an experiment involving 25 subjects, HyperCam was able to match photographs of hands and their subjects with more than 99 percent accuracy. This high level of accuracy suggests HyperCam could have potential biometric uses, using skin patterns to unlock smart phones, for example, or even as an ID for online payment purposes.
The ability to make such detailed images of skin patterns could also have a number of medical uses, Goel says. It could, for example, be used to monitor a wound healing over time, capturing fine-grained changes the human eye can’t see.
In a paper presented at the UbiComp 2015 conference, the team detailed a hardware solution that costs roughly $800, or potentially as little as $50 to add to a mobile phone camera. They also developed intelligent software that easily finds “hidden” differences between what the hyperspectral camera captures and what can be seen with the naked eye.
HyperCam’s technology has a few challenges. It can’t be used in bright daylight, as too much light will overwhelm its abilities to carve up the spectrum. Even in a very brightly lit grocery store, a user might have to hold a HyperCam fairly close to the produce—say one foot or so—to get an accurate reading.
While being able to pick out the best peaches in the market would no doubt be useful, HyperCam’s inventors say it has many more potential uses. And while there are no immediate plans to plant HyperCams in cell phones, the researchers hope to work towards this in the near future.
“We want to work with [other] scientists,” says Goel. “The idea is that we educate people on how they make this camera, and then they can generate it for their own applications.”