Researchers build a cheap camera that sees what we can't

Normally, our eyes can see things better than our cameras, which is why those smartphone shots of a full moon at night are so bad. Hyperspectral imaging is a way to fix that, since it can capture parts of the electromagnetic spectrum, like near-infra-red light, that humans are incapable of seeing. Unfortunately, these cameras are hideously expensive, so you're more likely to find one on a military satellite than in Best Buy. Thankfully, a team made up of members from the University of Washington and Microsoft Research think that they've been able to create a hyperspectral camera smartphone accessory that would cost just $50. It's not just useful for improving your Instagrams, either, since these devices can tell you if fruit has gone bad, peer into your body and even find mineral deposits underground.

Washington and Microsoft's version is called Hypercam, and will capture a series of 17 images at various points of the spectrum. Then, much like a HDR image, software will mash them together in the hope of finding things that wouldn't be visible otherwise. There's a security element to all of this, too, since these cameras can spot the structure of someone's veins and skin pattern -- similar to Fujitsu's palm vein sensing technology. Researchers have also used the gear to scan avocados and were able to tell with a 94 percent success rate if they were ripe enough to eat or not. Admittedly, it's not the first time that we've seen "cheap" hyperspectral cameras: the University of Vienna was able to knock one up with a DSLR and some PVC pipe back in 2012.

Now that the team has shown off its findings, however, it's back to the lab in order to try and refine the technology some more. For instance, the gear doesn't work particularly well in bright light, and there's still the issue of making it small enough to clip onto a smartphone. Whatever happens, the idea of using your phone to check if food is safe to eat is enough to get us excited on this chilly October day.

[Image Credit: University of Washington]