Ultra tiny camera has no lens – uses algorithm to develop pictures

Mar 28, 2014 by Bob Yirka report
Ultra tiny camera has no lens – uses algorithm to develop pictures
Image sensing and computational reconstruction of Leonardo’s Mona Lisa from a lensless phase anti-symmetric spiral phase grating sensor. (Left) The input image. (Middle) The simulated response on the photodetectors due to the six-fold grating, and (right) the reconstruction by. This image estimate is of higher fidelity than the estimate based on traditional square-wave amplitude gratings and photodetector arrays of comparable number of pixels and overall noise level described in earlier work.

(Phys.org) —An extremely tiny lensless camera, developed by Rambus, has been slowly making waves over the past year. Researchers for the company, David Stork and Patrick Gill won a Best Paper award at last year's Sencomm 2013 for describing what the company has created. They spoke again at last month's Mobile World Congress, describing their new type of camera—one that might someday soon be used to give virtually any digital device, some degree of vision.

The camera is both simple and complex, it's really just a very (CMOS imager) embedded in a piece of glass. Instead of a lens, a pattern is etched into the glass above the chip—the imager reads the light that is received, processes it using an algorithm developed by Rambus and converts it into a recognizable image. What's amazing is that the etched pattern on the glass and the chip are both roughly the size of a period at the end of a sentence.

Particular etched patterns allow for light to be intentionally refracted in different ways as it passes through the glass—images made from them would appear unrecognizable to the human eye, but the algorithm makes use of refraction properties to reconstruct the light received into a recognizable image.

The whole point of the camera is to show that cameras can be made smaller than has been envisioned by engineers of late. Trying to grind ever smaller lenses has reached its limits, thus something new had be developed. The camera by Rambus is one such possibility. Its images are not sharp—in fact at a resolution of just 128x128, its images are downright blurry—but at this point, that doesn't matter, because images taken by the camera are recognizable, and that's all digital devices of the near future likely need. Perhaps just as remarkable is that the can be used to capture real-time video too, which makes it a likely candidate for future motion sensing devices.

Making a so tiny opens the door for its use in a whole host of new applications, allowing them to become aware of their physical surroundings, all at a very low cost—perhaps just pennies per chip—that means they could be embedded in clothes, toys, mirrors, security systems, etc., bounded only by the imagination of device makers. On the other hand, such tiny cameras could also open a Pandora's box if they are used to invade privacy or for control purposes.

Explore further: Toshiba's dual-camera system enables second-chance manipulations

More information: Lensless Ultra-Miniature CMOS Computational Imagers and Sensors, PDF paper, PDF presentation

add to favorites email to friend print save as pdf

Related Stories

Bending the light with a tiny chip

Mar 10, 2014

(Phys.org) —Imagine that you are in a meeting with coworkers or at a gathering of friends. You pull out your cell phone to show a presentation or a video on YouTube. But you don't use the tiny screen; your ...

3-D, after-the-fact focus image sensors invented

Apr 03, 2012

(PhysOrg.com) -- At the heart of digital photography is a chip called an image sensor that captures a map of the intensity of the light as it comes through the lens and converts it to an electronic signal.

Recommended for you

DIY glove-based tutor indicates muscle-memory potential

20 hours ago

A senior editor at IEEE Spectrum worked on a DIY project that enabled his 11-year-old son to improve his touch typing by use of a vibrating glove. His son was already "pretty quick on the keyboard," said ...

US cyber-warriors battling Islamic State on Twitter

21 hours ago

The United States has launched a social media offensive against the Islamic State and Al-Qaeda, setting out to win the war of ideas by ridiculing the militants with a mixture of blunt language and sarcasm.

User comments : 10

Adjust slider to filter visible comments by rank

Display comments: newest first

antialias_physorg
5 / 5 (4) Mar 28, 2014
Prett cool.

Completely off topic:
that means they could be embedded in clothes, toys, mirrors, security systems, etc.,

Does anyone else start to get a slightly creepy feeling whenever the term "security systems" is mentioned in these articles?
what_the_hell
5 / 5 (4) Mar 28, 2014
Prett cool.

Completely off topic:
that means they could be embedded in clothes, toys, mirrors, security systems, etc.,

Does anyone else start to get a slightly creepy feeling whenever the term "security systems" is mentioned in these articles?
I watched you post this.
antialias_physorg
5 / 5 (1) Mar 28, 2014
I watched you post this.

Does that mean I'm safe? (I notice that the meanings of the words "secure" and "safe" seem to diverge of late)
Noumenon
not rated yet Mar 28, 2014
It's all a scam to sell you cloaking pants.
jimbo92107
not rated yet Mar 29, 2014
Hey, wanna see my new binoculars?

..
alfie_null
not rated yet Mar 29, 2014
How does this compare with a pinhole camera?
antialias_physorg
not rated yet Mar 29, 2014
How does this compare with a pinhole camera?

It's quite a bit smaller.
JimD
not rated yet Mar 30, 2014
The military has probably already latch onto this technology. Everything from cloaking devices to remote monitoring.
Jizby
Mar 30, 2014
This comment has been removed by a moderator.
a_n_k_u_r
not rated yet Apr 04, 2014
Can these or such algorithms be used to improve the images from imperfect lenses?

I am thinking about using these algorithm for big space telescopes like Hubble and James Webb and especially their future successors. These telescopes require extremely high accuracy in mirror construction that dramatically raises their costs. But if we can derive images from inaccurate mirrors then the costs can be reduced and we can go for even larger mirrors.
antialias_physorg
not rated yet Apr 04, 2014
But if we can derive images from inaccurate mirrors then the costs can be reduced and we can go for even larger mirrors.

There's already a lot of image processing going on with these telescopes (and yes: they do have software correction for distortion)

But a software correction is always worse than the real thing (a perfect mirror/lens), because it constitutes a filter. Any filter operation always decreases the information content of the image. So you have to be very careful about adding artificial (hardware or software) filters so as not to add artefacts into your image that would skew the data.