EDMONTON—Part of Vancouver’s airport runway is sinking, the Lions Gate bridge in North Vancouver may not be as stable as it seems, and land in the Arctic is thawing.

All of this is known due to satellite images that have been analysed to predict changes on the Earth’s surface over time. Now, thanks to the University of Alberta (U of A), this information can be known even faster.

“The problem is, it takes time,” said Irene Cheng, a lead researcher with the U of A’s computer science department. The process of analysing the images involves many steps, each of them needing a human expert to sign-off on the data before moving on to the next step.

To troubleshoot this, Vancouver-based company 3vGeometrics (3vG), which analyses the satellite image data to help cities and companies predict disasters before they strike, has enlisted the help of the U of A to develop artificial intelligence (AI) technology that can collect the data faster and more accurately.

“The work that we’re doing with the University of Alberta is paving the way for the analysis to become more automated,” said Parwant Ghuman, chief technology officer at 3vG.

Cheng’s team is now working on a prototype that could decrease the time of the data collection process from six months to just one month, all while making the images sharper, and therefore more accurate.

This technology is a breakthrough for 3vG, whose services are being used in more than 40 countries, Ghuman said. Their key clients include companies in the oil and mining industry, both of which require extraction of resources from below the Earth’s surface, which could shift the ground’s movement over long periods of time, causing a risk of landslides.

Ghuman said the consequences of the ground’s movement could be catastrophic if not monitored in advance. With oil extraction, for example, an underlying oil reservoir may explode and cause irreversible damage to the environment.

“In summary, we can tell you when something is moving, where it is moving, and how fast it’s moving,” Ghuman said of 3vG.

The satellite images could examine a large area of land at once, he added. For example, the company can monitor pipelines that run through the entire province of Alberta. The technology is also used to monitor the impact of climate change in the Nunavut and the Northwest Territories.

Monitoring these movements means receiving an early warning before disaster strikes, Ghuman said, and companies, as well as governments, can use this information to change their production methods or infrastructure, ultimately preventing harm.

“(The technology) takes the surprise element out of these geohazards that humans have been affected by for a long time,” Ghuman said.

Cheng added the U of A’s involvement with the project is thanks to Consortium for Aerospace Research and Innovation in Canada, a national not-for-profit organization that funds research and development projects in the aerospace field.

“Initially, they paid more attention to Ontario and the East Coast,” Cheng said, but this is the first time Alberta has chosen to participate in a project that links AI with the aerospace field.

It comes as no surprise, Cheng said, as U of A was one of three universities to receive a $125-million federal grant last year to enhance AI research nationwide, shared collectively between Edmonton, Montreal, and Toronto-Waterloo. Cheng said the funding helped bolster the university as a hub for AI innovation.

The U of A began developing new AI technology with 3vG last year.

Ghuman said he’s excited with the progress made so far, and believes the collaboration with the university will help the company reach its ultimate goal: to eventually monitor the whole Earth for risk of disasters.

“We believe AI and machine-learning has an important role to play in that vision,” Ghuman said.

This article originally appeared in The Star