Digital photos to map remote areas

The idea was inspired by a trip to Nepal

Researchers have come up with a way of turning digital photographs into 3D computer models which could help predict landslides.

Currently, accurate models of remote regions of the world are created using global positioning systems, satellite technology and other surveying techniques to create digital elevation models, known as DEMs.

But these methods are time consuming, expensive and sometimes physically impossible to carry out.

Now, two professors from Dartmouth College in New Hampshire, US, have come up with a way to solve the problem.

Idea on a napkin

We sketched out the idea on a napkin over lunch

Professor Farid

"It started after I got back from one of my trips to Nepal," explained Assistant Professor of Earth Sciences Arjun Heimsath.

"I wasn't able to survey the area I wanted because it was so hard to get to on foot. I'd seen Hany's work and I wondered if he could create the models I needed from photographs," he said.

"We sketched out the idea on a napkin over lunch," explained Professor Hany Farid, Assistant Professor of Computer Science at Dartmouth.

"I asked Arjun to take some photographs on his next trip, and we tested our theory within about three weeks. It didn't really work at first, but it worked well enough to keep going," he said.

The method relies on at least three digital photographs of the region, taken from slightly different vantage points. The photos need to be taken from a hillside adjacent to the area rather than from below.

Breeze in

With our method, you breeze in with a digital camera, and with relative ease, you get the digital elevation model

Professor Heimsath

Once the images are in the computer, a researcher can manually pick spots on each picture that correspond, such as identifying the same shrub, the same boulder, and so on.

"After you pick somewhere between 50 and 100 points, the mathematical algorithm takes over and automatically estimates the elevation map," explained Professor Farid.

With traditional mapping, researchers must carry around cumbersome equipment often in dangerous or rough terrain.

"With our method, you breeze in with a digital camera, and with relative ease, you get the DEM," says Professor Heimsath.

From the original lunch in a cafe to publication of the results in the November issue of the Mathematical Geology Journal took eight months.

The pair admit that the algorithms are not without limitations and caution that the method has yet to be tested in the field.

Close collaboration

Two students will travel to New Zealand in the New Year to compare the new method with conventional mapping techniques.

The research has proved how disparate academic disciplines can collaborate.

"What was nice about the work is that I'm taking tools from the mathematics and computer vision community and applying them to a real world problem," said Professor Farid.

The fact that the two professors lived next door to each other also helped, he added.