The wildfire that ravaged the Yeongnam region in late March is considered the worst in South Korean history. The main fire took over 213 hours to extinguish, and the affected area reached approximately 80% of Seoul’s size (about 48,238.61 hectares). The disaster claimed 75 victims, including 30 fatalities.
As the climate crisis worsens, massive wildfires are becoming more frequent worldwide. In January, a devastating wildfire in Southern California resulted in over 28 fatalities and billions of dollars in economic losses.
On Tuesday, information technology (IT) sources reported that various companies are researching ways to minimize wildfire damage using advanced technologies such as artificial intelligence (AI).
A research team from the Information Science Institute of the University of Southern California’s (USC) Viterbi School of Engineering is developing an AI-powered monitoring system for early wildfire detection.
Current satellite-based wildfire detection systems face challenges due to limited observation frequency and low resolution, making early detection difficult. Moreover, in areas where forests meet urban environments, rooftops and solar panels can further reduce the accuracy of satellite imagery.
USC’s ISI is developing an AI method for analyzing satellite images to address these issues. Their approach uses deep learning algorithms to analyze images captured across various light wavelengths, aiming to identify fire situations accurately.
The research team anticipates this technology will achieve 95% accuracy in identifying fire situations while reducing false alarms to as low as 0.1%. Their ultimate goal is to integrate the AI-based algorithm directly into satellites, enabling real-time fire detection and significantly improving response times.
German company Dryad Networks is developing devices to detect and extinguish wildfires using AI and unmanned drones.
Dryad Networks plans to deploy devices equipped with gas sensors capable of detecting wildfires in forests. Each device can monitor an area the size of a soccer field and identify fires before they become visible to the naked eye.
When sensors detect unusual conditions, unmanned drones automatically deploy to the site. These drones collect optical and infrared images and report back. The system’s key advantage is its automation, significantly reducing the time needed to identify a fire.
Dryad Networks aims to use unmanned drones to extinguish fires in the future. Instead of using chemicals or water, the drones will generate low-frequency sound waves to disrupt combustion. This technology could enable rapid fire suppression in areas difficult for firefighters or helicopters to access.
