The amazing camera that can see around corners (w/ video)

December 8, 2015 by Claire Vallance, The Conversation
It’s just your average corner, but a far from average camera. Credit: Gariepy et al./Heriott-Watt

How can a person see around a blind corner? One answer is to develop X-ray vision. A more mundane approach is to use a mirror. But if neither are an option, a group of scientists led by Genevieve Gariepy have developed a state-of-the-art detector which, with some clever data processing techniques, can turn walls and floors into a "virtual mirror", giving the power to locate and track moving objects out of direct line of sight.

The shiny surface of a mirror works by reflecting from an at a well-defined angle towards your eye. Because light scattered from different points on the object is reflected at the same angle, your eye sees a clear image of the object. In contrast, a non-reflective surface scatters light randomly in all directions, and creates no clear image.

However, as the researchers at Heriot-Watt University and the University of Edinburgh recognised, there is a way to tease out information on the object even from apparently random scattered light. Their method, published in Nature Photonics, relies on laser range-finding technology, which measures the distance to an object based on the time it takes a pulse of light to travel to the object, scatter, and travel back to a detector.

In principle, the measurement is quite simple. A laser pulse is bounced off the floor and scatters in all directions. A small fraction of the laser light strikes the object, and the back-scattered light is recorded on a patch of floor – the "virtual mirror" – next to the spot the laser strikes. Because the speed of light is known and constant, by measuring the time interval between the start of the laser pulse and the scattered light reaching the patch of floor, the position of the object can be triangulated.

The video will load shortly

However, the devil is in the detail. The timing measurement needs to be accurate to within around 500 billionths of a second (5x10-7, or 500 nanoseconds), and the light levels that must be detected are extremely low. Overcoming both of these obstacles requires some serious laser and detector technology. The used for the timing measurement are just ten femtoseconds (100,000 billionths of a second, or 10-15) long, and each pixel in the ultra-sensitive "camera" (known as a single-pixel avalanche diode array, or SPAD) used to image the patch of floor is essentially an ultrafast stopwatch that records the arrival time of the scattered light pulse to within a few hundred billionths of a second.

The complications do not end there. Light scattered from the object of interest reaches the virtual mirror of the floor, but so does light scattered from every other object in the vicinity. The success of this technique requires that the two be separated, the "signal" of the hidden object from the background noise of everything else.

This is achieved by using the fact that the hidden object the device is trying to detect is moving, while other nearby objects are not. Because the moving object generates a signal in the virtual mirror that changes with time, it can be filtered from the constant background signal produced by the stationary objects of the surroundings.

The video will load shortly
Supplementary video. Credit: Nature Photonics/ Gariepy et al.

The final complication is that the timing measurement for scattered light arriving at a single point on the virtual mirror and recorded by a single pixel in the detector unfortunately doesn't locate the object to a single unique position. A similar time delay could result from objects located at any number of different positions located an appropriate distance from the virtual mirror.

While the timing data from a single pixel only locates the object to a range of positions, the range is different for each pixel. However, it turns out that there is only a single position at which the timing condition is satisfied simultaneously for all pixels, and this allows the object to be unambiguously identified from the background signals.

Laser light fired at the floor is scattered in a spherical wave. It reflects from the hidden object and makes its way back to the SPAD detector, where the noisy data is processed to reveal the hidden object. Credit: Gariepy et al./Nature Photonics

The prototype camera system allows the object's position behind the wall to be localised to within a centimetre or two, and by making measurements every few seconds the camera can also detect the speed of a moving object. In contrast to previous methods, which required long times, the new method can track moving objects in real time. At present it's limited to locating objects up to 60cm away from the virtual mirror on the floor, but this should improve to around ten metres, as well as to more closely detect the shapes of hidden objects as well as their positions.

So while it's not quite as promising, or as convenient, as the science-fiction powers of X-ray vision, the study's authors note that the technology has interesting future applications in areas such as surveillance – to detect a moving person behind a wall, for example – or in car safety systems to detect incoming vehicles approaching around corners.

Explore further: Laser-based camera can see around corners

More information: Genevieve Gariepy et al. Detection and tracking of moving objects hidden from view, Nature Photonics (2015). DOI: 10.1038/nphoton.2015.234

Related Stories

Laser-based camera can see around corners

November 17, 2010

(PhysOrg.com) -- Researchers from MIT have developed a camera that can capture images of a scene that is not in its direct line of sight. The camera is equipped with a femtosecond laser, which fires extremely short bursts ...

Creative Cameras exhibit explores light-in-flight imaging

July 6, 2014

The Creative Cameras exhibition team from Heriot-Watt University and the University of Glasgow are asking some interesting questions. How do you take images so fast that you can see light travelling through air? How do you ...

Camera chip provides superfine 3-D resolution

April 3, 2015

Imagine you need to have an almost exact copy of an object. Now imagine that you can just pull your smartphone out of your pocket, take a snapshot with its integrated 3-D imager, send it to your 3-D printer, and within minutes ...

New X-ray method uses scattering to visualize nanostructures

November 19, 2015

Both in materials science and in biomedical research it is important to be able to view minute nanostructures, for example in carbon-fiber materials and bones. A team from the Technical University of Munich (TUM), the University ...

Recommended for you

Electrons at the speed limit

August 26, 2016

Electronic components have become faster and faster over the years, thus making powerful computers and other technologies possible. Researchers at ETH Zurich have now investigated how fast electrons can ultimately be controlled ...

Understanding nature's patterns with plasmas

August 23, 2016

Patterns abound in nature, from zebra stripes and leopard spots to honeycombs and bands of clouds. Somehow, these patterns form and organize all by themselves. To better understand how, researchers have now created a new ...

Light and matter merge in quantum coupling

August 22, 2016

Where light and matter intersect, the world illuminates. Where light and matter interact so strongly that they become one, they illuminate a world of new physics, according to Rice University scientists.

4 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

tpb
not rated yet Dec 08, 2015
"The timing measurement needs to be accurate to within around 500 billionths of a second (5x10-7, or 500 nanoseconds)"

This must be incorrect since this would mean a distance of 150 meters for resolution.

This article is poorly worded. Depending on which sentence you interpret in the article, they probably meant 2 to 5 picoseconds or within a few tenths of a millimeter.
baudrunner
5 / 5 (1) Dec 08, 2015
This is not new. This ability to "see around corners" has been postulated since the seventies, as I recall, and in fact, a much better video of the procedure was posted on Youtube on March 20, 2012 by Nature Video, originating from the MIT Media Lab. The result is a coherent image of the hidden object. https://www.youtu...ocXPy-iQ
gwrede
not rated yet Dec 09, 2015
This is not new. This ability to "see around corners" has been postulated since the seventies, as I recall, and in fact, a much better video of the procedure was posted on Youtube on March 20, 2012 by Nature Video, originating from the MIT Media Lab. The result is a coherent image of the hidden object. https://www.youtu...ocXPy-iQ
That technique looks similar, but under the hood they are very different. The MIT method can see stationary objects, too.
baudrunner
not rated yet Dec 09, 2015
That technique looks similar, but under the hood they are very different.
I don't see how "under the hood" they are actually any different. The video explains pretty much what the experimenters at Heriot-Watt University and the University of Edinburgh are doing, from reading this article, and in the same way, too. How are they different?

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.