THE LARGE-SCALE WILDFIRE SPREAD PREDICTION USING A MULTI-KERNEL CONVOLUTIONAL NEURAL NETWORK

Abstract. In the last twenty years, destructive wildfires have affected environment to tune of billions dollars. An accurate model is crucial for predicting spreading in a variety conditions. this study, multi-kernel convolution neural network (CNN) deep learning was proposed based on elevation, wind direction, and speed, minimum maximum temperatures, humidity, precipitation, drought index, normalized difference vegetation index (NDVI), energy release component predict wildfire spread across United States. Using CNN, it possible whether pixel will be fire at future time. Compared presented by other authors, CNN achieved high accuracy F1 score. comparison with CNNs without mechanism fixed kernel size, predicted more results test data set. The reached an overall 98.6 score 70.97 data.