You've successfully subscribed to Warp News
Great! Next, complete checkout for full access to Warp News
Welcome back! You've successfully signed in.
Thank you! Check your email inbox to activate your account.
Success! Your billing info is updated.
Billing info update failed.
🔥 AI system spots wildfires immediately

🔥 AI system spots wildfires immediately

This new AI system can spot wildfires before humans, and thus help prevent the great harm a wildfire can cause.

Linn Winge
Linn Winge

Quick detection is vital when it comes to wildfires. Only a few minutes can make a difference. Now, an AI system called ALERTwildfire has been developed by Graham Kent, director of the Nevada Seismological Laboratory at the University of Nevada. The system is currently in training and is installed in Northern California.

ALERTwildfire uses a number of cameras attached to towers standing in fire-prone regions. The images from the cameras are fed through an AI system that immediately detects a developing fire. The 21 initial testing sites for the system are located in Sonoma County where the Tubbs Fire took place in 2017.

Up next: Prediction of new fires

The system also delivers image updates from each tower to the county’s fire emergency center every 10 minutes. This procedure is used in order to add a layer of human confirmation. This far, the notifications from the system have foreshadowed 911 calls from residents with 10 minutes.

The AI will become more and more accurate over time. Right now, it is still learning what a fire looks like and how to tell the difference between smoke from a fire and steam from a geyser field, for example. The researchers believe the system needs to see about 70 fires in real life to fully understand the region’s fire landscape.

In the future, ALERTwildfire’s developers hope to install 850 cameras across six Western states to significantly reduce the harm of wildfires. They even believe that the AI system, with the right amount of training, will be able to analyze fire-prone landscapes and burn scars to predict where it’s most likely that a new fire will start.