Multimodal Wildland Fire Smoke Detection

Research has shown that climate change creates warmer temperatures and drier conditions, leading to longer wildfire seasons increased risks in the United States. These factors have, turn, led increases frequency, extent, severity of wildfires recent years. Given danger posed by wildland fires people, property, wildlife, environment, there is an urgent need provide tools for effective management. Early detection essential minimizing potentially catastrophic destruction. To end, this paper, we present our work on integrating multiple data sources into SmokeyNet, a deep learning model using spatiotemporal information detect smoke from fires. We Multimodal SmokeyNet Ensemble multimodal fire satellite-based detections, weather sensor measurements, optical camera images. An analysis provided compare these approaches baseline terms accuracy metrics, as well time-to-detect, which important early wildfires. Our results show incorporating improves performance numerically both F1 time-to-detect over with single source. With only few minutes, can be used automated notification wildfires, providing useful tool fight against destructive