Who We Are
Our Leadership
Our Commitment
Our Campuses
Why LSU?
Ready to Apply?
Learn More
Uncontained wildfires can quickly engulf homes and infrastructure, leading to significant economic losses, displacement of communities, and health risks due to smoke. They also present significant threats to forests, grasslands, and other natural habitats, leading to the death or displacement of countless animals and plants.
Providing firefighting teams with the best technology improves their chances of containing wildfires before they cause widespread damage.
Accuracy using traditional wildfire forecasting methods
Accuracy using LSU's A.I.-based DeepFire wildfire forecasting system
Using large datasets of previous wildfires, real-time weather conditions, predictions of lightning strikes, land cover, vegetation type, etc., the AI engine is able to predict if there will be a wildfire in a designated area in the next few days or even in the medium term.
The DeepFire prediction model produces real-time maps showing the most wildfire-prone areas, where resources can be preemptively deployed. The prediction results serve as input to the detection model, which then pays attention to sensors in the high-risk areas.
AI can be combined with long-range cameras both on the ground and in space to detect wildfires in a nascent stage when they are difficult to detect for the human eye. AI is also able to predict how a wildfire is going to spread. Together, these capabilities enable effectively tackling the fire before it can cause much damage.
Department of Environmental Sciences Professor Supratik Mukhopadhyay
In DeepFire, systems of wildfire prediction and detection work in tandem. The technology predicts locations of potential fires by examining satellite and weather station data, as well as information about previous fire behavior.
“Our system combines a prediction system, a detection system, and a spread modeling system that cooperate with each other,” Mukhopadhyay said. “This enables us to pinpoint our detection system to areas that are predicted to have a high risk of wildfire and deploy resources accordingly.”
Student research has been invaluable to creating and continually improving this technology, Mukhopadhyay said.
“The main work on the prediction model was done by LSU Stamps Scholar Dylan Wichman who graduated in computer science,” he said. “The main work on the detection model was done by Robert DiBiano who graduated with a PhD from LSU.” DiBiano received his PhD in Artificial Intelligence, Machine Learning and Computer Vision.
Growing up in Montana, LSU graduate Dylan Wichman is familiar with wildfires. But his interest in trying to stop them set him on a path of working with artificial intelligence.
Wichman's high school science fair project was based on the premise of using AI for wildfire prediction. When he learned of a similar project being led by Supratik Mukhopadhyay at LSU, his next step was a "no-brainer," he says.
Wichman would become a research assistant under Mukhopadhyay and part of the DeepFire team using AI to predict and detect wildfires.
Dylan Wichman graduated in December 2023 with a degree in computer science and now works as a research engineer.
LSU’s successes are shaping our state and the world in remarkable ways. As we build teams that win for Louisiana, the nation, and the world, we put our state and its citizens on firmer footing for a brighter tomorrow. Your journey starts here.
