New computer program quickly detects homes and buildings damaged by wildfires – Monterey Herald

A new computer program powered by artificial intelligence takes just minutes to determine if homes and buildings have been destroyed by wildfires once the smoke clears.

Developed by scientists at Cal Poly in San Luis Obispo and Stanford University, DamageMap assesses destruction by examining post-fire aerial and satellite imagery.

Residents of the Santa Cruz Mountains who were evacuated during the CZU Lightning Complex fire in the summer of 2020 say having such a program would have relieved them of a great deal of stress and anguish.

“I spent days thinking, ‘My house might be burning down right now,'” said Lisa Smith Beasley, a Boulder Creek resident who was ordered out of her home during the devastating event.

A satellite image shows the aftermath of the 2018 Camp Fire in Butte County. A new computer program called DamageMap identified buildings as damaged (red) or undamaged (green). (M. Galanis et al)

Beasley’s house survived the flames. But she and other Northern California residents often waited weeks to hear from authorities whether their homes survived the wildfires.

Andrew Fricker, a space ecologist at Cal Poly who co-developed DamageMap, said the program should significantly reduce that waiting period. Once fully developed, the program would be freely available to the public and emergency responders.

“There are so many people in California who are affected by this fire season,” Fricker said. “And it will only get worse.”

Fricker and his colleagues at Cal Poly and Stanford published their peer-reviewed work on the program in the November issue of the International Journal of Disaster Risk Reduction.

Computer programs that detect damage caused by natural disasters from aerial and satellite photos have been in development for a quarter of a century. But most of them require before and after photos to be taken with similar angles, lighting and photo quality, an imperfect system that requires an expensive and continually updated image catalog.

To determine which structures have burned, DamageMap relies solely on post-fire imagery and a digital database showing the location of houses and buildings.

Over the past four decades, the number of acres and homes burned in the West has increased dramatically, fueled in part by climate change.

California’s deadliest and most destructive wildfire – the 2018 Butte County Campfire – inspired the creation of DamageMap.

Hell badly damaged Fricker’s childhood home in Chico, where his parents still lived, but luckily the wildfire didn’t burn the house down.

During the evacuation, Fricker struggled to find out if the house was still standing. “I was frantically trying to download all the satellite images I could get, trying to get information for myself and our neighbors,” he recalls.

Wanting to prevent others from experiencing the same distress, Fricker gathered aerial footage of the campfire’s destruction and Cal Fire door-to-door structural damage assessments. With this data, he and a team of Cal Poly students created a rudimentary prototype of DamageMap.

He took the prototype to Google’s Geo for Good Summit 2019, where he met Stanford graduate student Krishna Rao. At the event, the two built an improved version of the program. And in the years that followed, they continued to collaborate and recruited more scientists to work on the project.

Last year’s CZU Lightning Complex fire in Santa Cruz and San Mateo counties was California’s ninth-most destructive wildfire. The event burned more than 80,000 acres and destroyed nearly 1,500 structures, including 911 homes in Santa Cruz County.

Despite evacuation orders, many of Beasley’s neighbors remained behind. If a program like DamageMap had been available to continuously update evacuees on the condition of their homes, she said, more people would likely have fled to safety.

“It was the not knowing that made it so bad because you couldn’t look ahead at all,” said M’Liss Jarvis Bounds, another Boulder Creek evacuee. She waited three weeks to hear that her house had survived the flames.

DamageMap works by first creating a database of pre-fire house and building locations using satellite imagery or aerial photos. Then he reviews post-fire photos and decides which structures are damaged based on features such as collapsed or blackened roofs.

The app uses “machine learning,” a form of artificial intelligence, or AI, to identify burned-out buildings.

Typically, computer programmers feed tens of thousands of images into a program so that it learns to identify specific patterns. Facebook, for example, uses machine learning to recognize faces and suggest people “tag” in photos.

In developing DamageMap, the researchers incorporated nearly 50,000 images of burned and intact structures into the program, including photos of the 2017 Tubbs Fire in Santa Rosa, the 2017 SoCal Fire in Los Angeles, and the 2018 Woolsey Fire in Los Angeles and Ventura counties. Subsequently, the programmers tested how well DamageMap had learned what a fire-damaged structure looked like by showing the app an additional 18,000 images of the Camp Fire and the 2018 Carr Fire in the Shasta and Trinity counties.

The program correctly identified charred structures in the second set of photos at least 92 percent of the time in about 18 minutes, according to the published paper. But he made mistakes when trees or other objects blocked the view of buildings and when rooftops blended in with the surroundings.

While not intended to replace post-fire assessments performed by people, technology that can quickly and accurately assess damage appeals to emergency responders.

“As technology and machine learning technology develops, we will definitely use it in the unfortunate event of another camp fire or a Tubbs fire, where it would mow down many structures at time,” said Will Brewer, geographic information system analyst and developer at Cal Fire.

For now, Fricker and his team are improving the program by providing DamageMap with more data to learn from. The more post-fire images it sees, the better it identifies the damage.

The developers say a lack of funding is preventing the program from being available for wider use. So far, an $18,000 grant from Cal Poly has been the main source of funding, but Fricker estimates an additional $80,000 will be needed to get the app up and running for the public.

Fricker said he had to pay dedicated computer programmers to continue training him, and he had to find a suitable online platform to host the program, which could be expensive.

“The code works and we have a lot of data,” Fricker said. “If people were motivated to let the public know for the next fire season, it could be done.”

Gordon K. Morehouse