The California Department of Forestry and Fire Protection (Cal Fire) recently announced a new project that uses artificial intelligence (AI) to detect forest fires. The project, developed in partnership with the University of California, San Diego (UCSD), called “Alert California AI,” uses video feeds from 1,032 360-degree rotating cameras to use AI to “identify anomalies in camera footage.” Once a potential fire is detected, emergency services and other relevant authorities are notified to determine whether a response is required.
The project was launched in July, and according to Reuters, at least one fire that could have sparked a fire has been successfully extinguished. In that incident, a camera captured an incipient flame at 3 a.m. in the remote Cleveland National Forest east of San Diego. The AI spotted the fire immediately and notified a fire chief who called about 60 firefighters, including 7 fire trucks, 2 bulldozers, 2 water tankers, and 2 work teams, and the fire was extinguished within 45 minutes. put out.
A fraction of the 1,032 live camera feeds used by Cal Fire, nine of which are seen showing forest and desert scenes in different parts of California
The Alert California AI technology website says they use LiDAR scans from airplanes and drones to generate “3D information about the scanned surface.” Combines physical characteristics of tree species to understand California forest biomass and carbon content. Cal Fire said that the machine learning (ML) model uses the camera’s petabyte-scale (1PB=1000TB) data to distinguish between smoke and other air particles.
The system was developed by UCSD engineers using AI provided by the California company DigitalPath. Cal Fire has invested more than 20 million US dollars on the project in the past four years, and has committed to An additional US$3.516 million will be invested in the near future.
