case_studies
Smart Wildfire Sensor - fighting fires with TensorFlow
"Existing wildfire prevention tools can assess an area’s risk of wildfire by measuring things like wind speed, wind direction, humidity and temperature. But biomass, which consists of fire-fueling materials like dead trees and debris, is hard to measure. Two teenagers created categories for the different types of biomass fuel and then trained a machine learning model to recognize relevant data about the fuels. This technology could help warn emergency responders about high-risk levels of biomass, allowing them to prevent fires before they even start."