Traditional camera lenses can be bulky and unwieldy, but an innovation coming out of the technology licensing company Rambus could make them miniscule. Senior research scientist Patrick Gill developed a glass chip that's only 200 microns in diameter that acts like a camera, capturing images with a tiny sensor using light and algorithms.

It looks like an extremely small and thin piece of glass, but it has a sensor inside of it that's etched with a spiral pattern allowing light to pass through. When light reflects off the object being "photographed," the sensors catches the light, passing it through the pattern, so it reaches a CMOS imager. At this point, the "image" is just spherical light, but software can then translate that light into a replica of the actual object. Gill has used the Mona Lisa and a photo of John Lennon in his examples: the sensor captures the scene, the data is shown as spherical light, and then the final, computerized image is generated. Gill also explains the device could be better as a primitive video camera because being a lensless sensor can capture moving light with higher quality than it can with a single image since it doesn't "lose any pixels to focusing or over-saturation."

Patrick_gill_ex

"Our aim is to add eyes to any digital device, no matter how small."

But using a "lens" that's smaller than a pencil point will not yield high-quality results. The images produced by Gill's invention currently have a maximum resolution of 128 by 128 pixels, however that is an improvement from the original prototype which only shot images with a resolution of 20 pixels across.

Gill doesn't want this to replace larger cameras. Rather than focusing on making high-resolution, microscopic cameras, he's attempting to change the focus of basic image creation. "We shift the burden of image formation from a purely optical task to a joint optical and computational task," Gill told The Verge. This could lead to minuscule, cheap, and easily constructed sensors that could be placed in other digital devices, letting them capture what is going on in the environment and relay that information just well enough for users to understand it. Since there hasn't been much innovation in camera lens technology since shrinking pixels, an MIT expert notes that this project shows promise, however it's still unclear how well the technology will work in practice.