星空无限传媒

When communities face increasingly destructive wildfires, the most powerful tool isn鈥檛 just response, it鈥檚 prevention. At LSU, a multidisciplinary research team is pushing the boundaries of what鈥檚 possible, using artificial intelligence (AI) to help predict wildfires before they ignite, giving fire managers critical time to act.   

鈥淒eepFire allows fire managers to know when and where to deploy their resources ahead of time, so they can prevent disasters like the January 2025 wildfires in Southern California,鈥� said Supratik Mukhopadhyay, professor of environmental science and Center of Computation & Technology (CCT) faculty member.  

Led by Mukhopadhyay, the DeepFire team has emerged as a global leader in wildfire innovation. Alongside Thomas Douthat, associate professor of environmental science, Rubayet Bin Mostaviz, research assistant professor with the LSU AgCenter and LaHouse, and Saiful Sajol, graduate research assistant in the Department of Environmental Sciences and a PhD student in the Department of Computer Science, the team has reached the finals of the $11 million Wildfire XPRIZE, placing among the top teams worldwide in the space-based detection and intelligence track.    

Out of more than 300 international teams, DeepFire stands out for developing an AI-driven engine capable of predicting wildfires weeks in advance with more than 90% accuracy. The system integrates fire prediction, early detection, and fire-spread modeling by factoring in weather conditions, analyzing visual data for smoke, and cross-referencing findings with predictive risk maps.      

The DeepFire team reflects LSU鈥檚 commitment to innovation and teamwork.  

Mukhopadhyay provides the strategic vision and leadership behind the project, while Douthat contributes deep expertise in environmental systems and wildfire behavior. Bin Mostaviz and Rahim strengthen the applied research and outreach components through their work with the AgCenter and LaHouse, ensuring the technology connects to real-world resilience and hazard mitigation efforts.    

Students are at the heart of DeepFire鈥檚 success. 2023 LSU graduate Dylan Wichman, a computer science graduate who grew up in wildfire-prone Montana, led much of the work on the prediction model. His early interest in using AI to prevent wildfires found a natural home 星空无限传媒. The detection model was driven by Robert DiBiano, who earned his PhD 星空无限传媒 in artificial intelligence, machine learning, and computer vision.   

The impact of the team鈥檚 work has reached a national audience, with DeepFire featured in LSU鈥檚 airing during LSU football games and other major sporting events.   

Through innovation, collaboration, and a shared commitment to public good, the DeepFire team is advancing LSU鈥檚 research mission while positioning the university as a leader in solutions that protect lives, communities, and natural resources.     

Next Step

Discover stories showcasing LSU鈥檚 academic excellence, innovation, culture, and impact across Louisiana.

See more LSU News