Scientists are using deep learning data to map structure damage from wildfires

A new computer program powered by artificial intelligence takes mere minutes to determine whether homes and buildings have been destroyed by wildfires once the smoke has cleared.

Developed by scientists at Cal Poly in San Luis Obispo and Stanford University, DamageMap evaluates the destruction by scrutinizing post-wildfire aerial and satellite images. — Santa Cruz Sentinel

The team worked to fix a major flaw in existing modeling systems that required exact lighting and photographic angles in order to produce an accurate survey of structural damage in a specific area. 50,000 images from various fires across California were used as a baseline for the software, which came back with a 92% accuracy rate when tested against another set of 18,000 burn scar images, according to the scientists.

Meanwhile, the rash of recent fires has led many to speculate about a statewide exodus that will continue as a product of future cataclysms which the team says it is hastily preparing for.

“The code works, and we have a lot of data,” co-developer Andrew Fricker told the Santa Cruz Sentinel. “If people were motivated to get this out to the public for the next fire season, it could be done.”

 

Original post: https://archinect.com/news/article/150292971/scientists-are-using-deep-learning-data-to-map-structure-damage-from-wildfires

Leave a Reply

Your email address will not be published. Required fields are marked *