ARCHIVED - A sharper eye in the sky

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

October 03, 2011 — Ottawa, Ontario

A remote sensing technology used for oil and mineral exploration as well as crop surveillance may someday help soldiers spot roadside explosives, police find murder victims, and biologists monitor species at risk. 

Called hyperspectral imaging, the technology gathers information on the intensity of reflected light at wavelengths that are visible to the eye as well as many that aren’t. The result is hundreds of possible images of the same target — instead of the typical red, blue and green provided by standard imaging. 

After extensive testing of this technology, scientists at the NRC Institute for Aerospace Research in Ottawa have made some key enhancements and begun developing new applications. The NRC team designed a polarmetric shortwave infrared airborne spectrographic imager — one of the world’s first — with 160 spectral channels covering the 850 nm to 2500 nm wavelength range. Like a pair of sunglasses, the SASI’s linear polarizer reduces haze and glare, providing a much clearer picture from the air than conventional hyperspectral imagers.

NRC's hyperspectral imager is installed aboard a Twin Otter airplane.

NRC's hyperspectral imager is installed aboard a Twin Otter airplane.

One of the hyperspectral imager’s potential uses is to help spot improvised explosive devices (IEDs) from the air. To test this application, a team led by Dr. George Leblanc implanted 39 dummy roadside IEDs at NRC’s Uplands facility near the Ottawa airport, and then flew over the road to take images. The goal of these flights was to find out whether IEDs have a characteristic “signature” that can be readily identified using an airborne hyperspectral camera. 

The limits of detection

At an altitude of 1 kilometre, the NRC hyperspectral imager can detect objects with a minimum resolution of 70 centimetres — or about three-quarters of the width of an office door.

In other research, NRC is collaborating with various partners to apply hyperspectral imaging toward the identification of clandestine mass graves and unmarked single graves from the air. For example, McGill University’s departments of geography and law want to have imaging data on mass graves accepted by the International Criminal Court. To help McGill, NRC has been acquiring hyperspectral imaging data and analyzing the spectral characteristics of large mammal burial sites that resemble human mass graves.  

Identifying missing persons 

Similarly, NRC is working with McGill University, the Canadian Police Research Centre and various law enforcement agencies to determine whether hyperspectral imaging can locate and identify missing persons buried in single graves. To gather data, the researchers buried several pigs at various depths and under various conditions at the Uplands campus. “Every time we take the instrument up for a flight, we plan to fly over their graves and take an image,” says Dr. Leblanc.

He and his colleagues will monitor the appearance of vegetation and soil at the “analogue” burial site over the next five years, as the pigs decompose. Meanwhile, McGill University researchers will periodically take samples of soil, methane and other trace gases. “We plan to analyze our data as a function of time because the grave of a human who has just gone missing would look substantially different than the grave of someone who has been buried for five or ten years.” 

Aerial image of NRC’s Uplands campus, which contains an “analogue” pig burial site — above and left of the white “H” buildings — that will be monitored for several years.

Aerial image of NRC’s Uplands campus, which contains an “analogue” pig burial site — above and left of the white “H” buildings — that will be monitored for several years.

Besides its forensic potential, hyperspectral imaging could also be used for identifying and counting species at risk — such as polar bear populations — to determine if their numbers are decreasing or stable. In March 2011, the NRC team conducted an exploratory study for the Canadian Wildlife Service, which is interesting in placing an imaging system in an unmanned aerial vehicle for species enumeration. The study involved laying polar bear, caribou, muskox and seal pelts on snow and then taking ground spectrometer readings to see whether different species’ pelts can be distinguished from each other and from the snow. 

“Right now, polar bears are often counted from a helicopter, but if you get too close you might stress them and increase their mortality rate,” says Dr. Leblanc. “It would be better to count wildlife populations from a kilometre up rather than 100 metres up.” This would also reduce the likelihood of counting the same animals twice. 

NRC’s Twin Otter aircraft in flight.

NRC’s Twin Otter aircraft in flight.

Once these research projects are completed, the NRC team hopes to transfer its technology to the private sector to develop commercial cameras for various imaging applications. “For example, we’re talking to a Canadian instrument manufacturer about developing a polarized hyperspectral system based on our mass grave research,” says Dr. Leblanc.

Related information

Enquiries: Media relations
National Research Council of Canada
613-991-1431
media@nrc-cnrc.gc.ca

Stay connected

Flickr Subscribe