Development and Application of a Sentinel-2 Satellite Imagery Dataset for Deep-Learning Driven Forest Wildfire Detection

TL;DR

This paper introduces the CWGID, a large labeled Sentinel-2 satellite imagery dataset for deep learning-based forest wildfire detection, demonstrating high accuracy with pre-trained CNN models.

Contribution

It presents the creation of a new high-resolution wildfire dataset from Sentinel-2 imagery and evaluates its effectiveness using pre-trained CNN architectures.

Findings

EF EfficientNet-B0 achieves over 92% accuracy

CWGID is a valuable resource for wildfire detection research

Deep learning models can effectively utilize the dataset for wildfire detection

Abstract

Forest loss due to natural events, such as wildfires, represents an increasing global challenge that demands advanced analytical methods for effective detection and mitigation. To this end, the integration of satellite imagery with deep learning (DL) methods has become essential. Nevertheless, this approach requires substantial amounts of labeled data to produce accurate results. In this study, we use bi-temporal Sentinel-2 satellite imagery sourced from Google Earth Engine (GEE) to build the California Wildfire GeoImaging Dataset (CWGID), a high-resolution labeled satellite imagery dataset with over 100,000 labeled before and after forest wildfire image pairs for wildfire detection through DL. Our methods include data acquisition from authoritative sources, data processing, and an initial dataset analysis using three pre-trained Convolutional Neural Network (CNN) architectures. Our results show that the EF EfficientNet-B0 model achieves the highest accuracy of over 92% in detecting forest wildfires. The CWGID and the methodology used to build it, prove to be a valuable resource for training and testing DL architectures for forest wildfire detection.