System, method and device for dynamic wildfire risk prediction

A method of predicting a risk of wildfire using artificial intelligence, the method comprising:

• receiving satellite image data covering a target area;
• processing the satellite image data to generate key vegetation indices;
• generating weather maps for the target area based on the key vegetation indices, wherein generating the weather maps includes interpolating weather data;
• generating static maps for the target area indicating proximity to potential ignition sources based on the key vegetation indices;
• analyzing historical wildfire data to identify past fire locations in the target area;
• integrating the received satellite image data, the generated weather maps and the generated static maps, and the analyzed historical wildfire data to obtain a comprehensive dataset for the target area;
• and
• analyzing the comprehensive dataset using a risk prediction model trained to predict the risk of wildfire for the targeted area, the risk prediction model comprising a machine-learning-based pattern detection model configured to receive the comprehensive dataset as an input and generate prediction data describing a predicted risk as an output.

A device for predicting a risk of wildfire using artificial intelligence, the device comprising:

• a network interface;
• a processor;
• and
• a non-transitory computer readable memory having stored thereon instructions which, when executed by the processor, cause the device to:
• receive satellite image data covering a target area;
• process the satellite image data to generate key vegetation indices;
• generate weather maps for the target area based on the key vegetation indices, wherein generating the weather maps includes interpolating weather data;
• generate static maps for the target area indicating proximity to potential ignition sources based on the key vegetation indices;
• analyze historical wildfire data to identify past fire locations in the target area;
• integrate the received satellite image data, the generated weather maps and the generated static maps, and the analyzed historical wildfire data to obtain a comprehensive dataset for the target area;
• and
• analyze the comprehensive dataset using a risk prediction model trained to predict the risk of wildfire for the targeted area, the risk prediction model comprising a machine-learning-based pattern detection model configured to receive the comprehensive dataset as an input and generate prediction data describing a predicted risk as an output.