Trending: Hands-on with the $350 Surface Headphones: Why is Microsoft making them, and are they any good?

The HyperCam prototype. (Photo via University of Washington.)

They may not be X-ray specs, but the University of Washington and Microsoft Research are working together to let you see under your skin — and much more. The researchers think their HyperCam can be used for a range of applications, from analysing blood vessels to making sure that avocado is ripe before cutting into it.

A photo taken with a normal camera on the left, and the HyperCam image on the right. Photo via UW Today
A photo taken with a normal camera on the left, and the HyperCam image on the right. (Photo via UW.)

The heart of the HyperCam is its multi-spectral camera. While normal cameras capture just red, green and blue wavelengths, the HyperCam also takes in near-infrared wavelengths, which reflect back off blood vessels under the skin differently than normal light. The camera flashes with 17 different wavelengths of light in sequence, capturing a different image for each wavelength.

After what looks like a mini-disco party for your subject, you’re left with a handful of pictures that range from pretty normal looking to what-even-is-that territory. Importantly, the researchers’ camera also sifts through the large number of pictures produced to find the one that is most different from a normal image. If a human did that, not only would it be time consuming, but they might get too accustomed to the stranger images and miss the most different and revealing image.

 

“After building the camera we just started pointing it at everyday objects — really anything we could find in our homes and offices — and we were amazed at all the hidden information it revealed,” Neel Joshi, a Microsoft researcher who worked on the project, told UW Today.

The project is described in a paper that was presented by the researchers at the UbiComp 2015 conference. The technology relies on something called hyperspectral imaging, which is already in use today with satellite imagery and energy monitoring system, but at a prohibitively high cost—sometimes reaching tens of thousands of dollars. The University of Washington’s hardware solution costs about $800, but could be added to a standard smartphone camera for just $50.

The tool has particular use in biometric security. By revealing the detailed vein and skin texture patterns on someone’s hand, which are as unique as a fingerprint, the sensor could be used to unlock your front door, safe or even your Xbox if incorporated into the Kinect 3.

But it could also be used for more mundane activities.

HyperCam had 94 percent accuracy when predicting the ripeness of fruits, compared to just 62 percent for standard cameras. Photo via UW
HyperCam had 94 percent accuracy when predicting the ripeness of fruits, compared to just 62 percent for standard cameras. Photo via UW

“With this kind of camera, you could go to the grocery store and know what produce to pick by looking underneath the skin and seeing if there’s anything wrong inside. It’s like having a food safety app in your pocket,” UW computer science and engineer professor Shwetak Patel said.

However, you might need to ask the grocery store manager to turn off the lights while you shop: right now, the tech doesn’t work well in bright lights as the wavelengths may confuse the camera’s sensor. The prototype is also too bulky for practical use, but could be refined for portable applications.

Like what you're reading? Subscribe to GeekWire's free newsletters to catch every headline

Comments

Job Listings on GeekWork

Find more jobs on GeekWork. Employers, post a job here.