Many people have lost their lives, and many more have been displaced as a result of the devastating wildfires that climate change has recently ignited. Amid all of these, the era of artificial intelligence (AI) is here.
Fire departments and new businesses are using cameras equipped with AI to look for smoke. In order to monitor fires from space, a German firm is developing a constellation of satellites. In addition, Microsoft is making use of AI algorithms to foresee potential ignition points for future fires.
There is no denying that firefighters, utilities, and governments are using the newest AI tech to stay ahead of the growing and more severe wildfires caused by global warming.
AI That Detects Smoke From 1,000+ Mountaintop Cameras
This summer, California's primary firefighting agency began pilot-testing an AI system that monitors smoke from over 1,000 mountaintop video feeds. They are now rolling out the software statewide, according to the California Department of Forestry and Fire Protection.
The device is programmed to detect "abnormalities" and send an alarm to emergency command centers, where trained personnel may determine if the smell in the air is indeed smoke.
The cameras, part of a network that employees had to monitor before, feed billions of bytes of data into the AI system. Humans must still verify any smoke detections, but this method will reportedly help minimize worker exhaustion by notifying them to check only when fire or smoke is possible.
It has been useful thus far. According to ABC News, a battalion chief received a smoke alarm at night, verified it on his smartphone, and phoned a San Diego command center to send first responders to the isolated region. If dispatchers had not been warned, the fire would have been considerably bigger since it would not have been spotted until the following morning.
AI Usage via Computer Vision Machine Learning
Pano AI, based in San Francisco, takes a similar tack by installing cameras on cell towers to detect smoke and notify its clients, such as fire agencies, power providers, and ski resorts. Cameras employ AI in the form of computer vision and machine learning.
CEO Sonia Kastner told ABC News, "They're trained very specifically to detect smoke or not, and we train them with images of smoke and images of not smoke."
Images are coupled with data from other sources, such as social media postings and government weather satellites, to locate and analyze hotspots.
Pano AI's technologies still need human validation; for example, managers may play a time-lapse of the camera feed to make sure the haze they see is indeed smoke.
PGE, which serves 51 communities throughout Oregon with electricity, has installed 26 Pano AI cameras. The response time and cooperation with emergency services have both been improved thanks to these devices.
In the past, fire crews wasted time and energy searching in the wrong places. The cameras may reduce response times by up to two hours by helping to spot fires earlier and sending crews to the scene sooner.
