CatBoost machine learning improves wildfire detection from satellite images
Comparing a gradient boosting algorithm to the GOES FDC for wildfire detection
Computer Vision and Pattern RecognitionMachine Learning
Summary
Detecting wildfires quickly and accurately is important to protect people and nature. This study tested a machine learning tool called CatBoost using satellite images and found it was better at spotting fires than the current GOES FDC system. CatBoost detected more fires, made fewer mistakes, and caught fires earlier, even at night when GOES FDC struggled. This suggests machine learning can help improve wildfire warnings from satellite data.
What this means in practice
- •For wildfire monitoring teams: Improve wildfire alerts by integrating CatBoost-based fire detection from geostationary satellite images to increase early warning accuracy.
- •For emergency response coordinators: Use enhanced fire detection data from CatBoost models to plan faster responses to wildfires detected earlier than current systems allow.
Authors
Asaf Vanunu, Boaz Nadler, Arnon Karnieli
Abstract
Wildfires pose severe risks to human life, ecosystems, and property. This study presents a machine learning approach for wildfire detection from GOES ABI imagery. A CatBoost model was trained on a large dataset with thousands of ABI images and over 300,000 matching VIIRS fire detections. An evaluation on a separate dataset across five regions showed that the learned CatBoost model outperformed the operational GOES Fire Detection and Characterization (FDC) product. It achieved higher precision, recall, and F1 scores both within and outside the training area. The CatBoost model achieved F1 scores that were 0.16 to 0.38 higher than the GOES FDC in all regions. In addition, out of 51 historical fire events, the CatBoost detected 26 fires before both VIIRS and GOES FDC, compared to only six earlier detections by the GOES FDC. Importantly, the CatBoost model achieved accurate wildfire detection also during nighttime, whereas the GOES FDC obtained very low recall values, around 0.03. This study demonstrates that machine learning models may offer significant improvements over existing geostationary fire products, including higher accuracy, fewer false alarms, and earlier detection.