Search
Menu
CASTECH INC - New Building the Bridge of Light

Camera Captures 3-D Images from 1 km Away

Facebook X LinkedIn Email
EDINBURGH, Scotland, April 4, 2013 — A time-of-flight imaging system can now gather high-resolution 3-D information from up to a kilometer away from an object that typically is very difficult to image.

Standard cameras take flat, two-dimensional pictures. To get 3-D information, such as the distance to a far-away object, scientists can use a technique called time-of-flight (ToF) to bounce a laser beam off an object and measure the time it takes the light to travel back to a detector. This method is already used in navigation systems for autonomous vehicles and in machine vision as well as other applications, but many ToF systems have a relatively short range and struggle to image objects that do not reflect laser light well.

Now Heriot-Watt University physicists, led by professor Gerald Buller, have tackled these limitations by creating a system that sweeps a low-power infrared laser beam rapidly over an object, recording, pixel-by-pixel, the round-trip flight time of the photons in the beam as they bounce off the object and arrive back at the source. Their system resolves depth on the millimeter scale over long distances using a detector that “counts” individual photons.

3-D images of a mannequin (top) and person (bottom) from 325 m away using a camera developed at Heriot-Watt University.
3-D images of a mannequin (top) and person (bottom) from 325 m away using a camera developed at Heriot-Watt University. The left-hand panels show close-up photos of the targets taken with a standard camera. In the center are 3-D images of these targets taken by the scanner from 325 m away. On the right is a color-coded map showing the number of photons that bounce off the targets and return to the detector, with black indicating a low number of photons. Notice that human skin does not show up well using the scanner: The mannequin's face includes depth information, but the person's face does not. Images courtesy of Optics Express.

Although other approaches have achieved exceptional depth resolution, the ability of the new system to image objects such as clothing items that do not easily reflect laser pulses makes it useful for a wider range of field situations, said Aongus McCarthy, a research fellow at the university.

“Our approach gives a low-power route to the depth imaging of ordinary, small targets at very long range,” McCarthy said. Although it is possible for other depth-ranging techniques to match or outperform some characteristics of these measurements, “this single-photon counting approach gives a unique trade-off between depth resolution, range, data-acquisition time and laser power levels.”

The primary use of the system will be scanning static man-made targets such as vehicles, the investigators said. With some modifications to the image-processing software, it also could determine a target’s speed and direction.

Deposition Sciences Inc. - Difficult Coatings - MR-8/23

A key characteristic of the system is the 1560-nm wavelength of laser light the researchers chose. This long-wavelength light travels more easily through the atmosphere, is not drowned out by sunlight, and is safe for eyes at low power. Previous ToF systems could not detect this extra-long wavelength.

The scanner is good at identifying objects hidden behind clutter, such as foliage, but it has difficulty rendering human faces, instead drawing them as dark, featureless areas. This is because at the long wavelength used by the system, human skin does not reflect back a large enough number of photons to obtain a depth measurement. However, the reflectivity of skin can change under different circumstances.

“Some reports indicate that humans under duress — for example, with perspiring skin — will have significantly greater return signals,” which should produce better images, McCarthy said.

3-D images of two of the Optics Express authors, taken in daylight from 910 m away.
3-D images of two of the Optics Express authors, taken in daylight from 910 m away. Each standard photograph shows a close-up view of what the scanner sees. The middle-left panels show 3-D images with slightly more depth detail than the right-hand panels; this is because the detector spent more time collecting the returning photons for the images on the left than on the right.

Photon-counting depth imaging could also be used for a number of scientific purposes, including the remote examination of the health and volume of vegetation and the movement of rock faces, to assess potential hazards. Ultimately, McCarthy says, it could scan and image objects located as far as 10 km away.

”It is clear that the system would have to be miniaturized and ruggedized, but we believe that a lightweight, fully portable scanning depth imager is possible and could be a product in less than five years,” he said.

Next, the investigators hope to make the scanner work faster. Although the data for the high-resolution depth images can be acquired in a matter of seconds, currently it takes about five to six minutes from the onset of scanning until a depth image is created by the system. Most of that lag, McCarthy says, is the result of the relatively slow processing time of the team's available computer resources.

“We are working on reducing this time by using a solid-state drive and a higher-specification computer, which could reduce the total time to well under a minute,” he said. “In the longer term, the use of more dedicated processors will further reduce this time.”

The research, funded by the United Kingdom's Engineering and Physical Sciences Research Council, appeared in Optics Express (doi: 10.1364/OE.21.008904).  

For more information, visit: www.hw.ac.uk

Published: April 2013
Glossary
machine vision
Machine vision, also known as computer vision or computer sight, refers to the technology that enables machines, typically computers, to interpret and understand visual information from the world, much like the human visual system. It involves the development and application of algorithms and systems that allow machines to acquire, process, analyze, and make decisions based on visual data. Key aspects of machine vision include: Image acquisition: Machine vision systems use various...
machine visioncameras3-D imagingAongus McCarthyEngineering and Physical Sciences Research CouncilEuropeGerald BullerHeriot-Watt UniversityImagingphoton countingResearch & TechnologyScotlandSensors & Detectorstarget identificationtime-of-flight imagingToF imaging systemvehicle scanning

We use cookies to improve user experience and analyze our website traffic as stated in our Privacy Policy. By using this website, you agree to the use of cookies unless you have disabled them.