Picture a camera that has no lens, no moving parts, costs fractions of a penny to make, and sees as dimly as a short-sighted worm. Doesn’t exactly sound like a game changer–but it is.
Cornell scientists have achieved the breakthrough by producing what’s called a Planar Fourier Capture Array camera from a super-cheap material, doped silicon, that’s currently used in all sorts of chip technology. It’s just one-hundredth of a millimeter deep and a half a millimeter on each side, which means several of them could fit on the head of a pin. The magical aspect of the cam is that it doesn’t need a lens because it makes use of the wave-like properties of light to work out what it’s looking at, and all the image construction is done by algorithms in a computer later.
The camera can only see about 20 pixels on a side–check out the image to see how the camera views the Mona Lisa. Despite its limitations, its utility is impressive. It starts with the fact it’s so tiny and has no moving parts, which means it could be integrated almost anywhere in a gadget, and will fit into gadgets of any size. It’s super-cheap, which means it could find elegant and surprising uses in situations you may not expect. And its dim vision powers could actually change the capabilities of almost anything you integrate it into.
Cornell’s research was intended to develop an imaging system for implanting near a living brain to detect light from specially engineered neurons–enabling all sorts of extraordinary neural science. And that’s still a use it could be put to. But how about an array of the camera sensors being used in a very small endoscope, letting medics peep inside the body with only the most minimal of invasions? What about a scalpel blade that can image the tissue it’s cutting through? How about a tiny in-body medical robot that needs very basic imaging for navigation?
The camera is so simple it could be built into the fingertips of android robots to help them work out how to best hold an object, or even to aid with object recognition. Sensor arrays on buildings could use the camera to work out the angle of the sun, removing the need for more expensive timing units. Dozens of the devices could be integrated into the screen of a smartphone, to act as a lens-less webcam, or even a more sophisticated touch recognition system. Vacuum-cleaner robots could benefit from multiple camera angles to help them navigate and find every grubby corner of your home to clean.
The potential applications are almost as limitless as the number of ways we use LEDs nowadays–most of which wouldn’t have occurred to the device’s inventors.
[Image: Flickr user robboudon]