Burnt area extraction from high-resolution satellite images based on anomaly detection

TL;DR

This paper presents an unsupervised method for burnt area detection in high-resolution satellite images using anomaly detection with VQ-VAE, enhanced by post-processing with vegetation, water, and brightness indexes, demonstrating promising results.

Contribution

It introduces a novel unsupervised burnt area extraction approach based on VQ-VAE and specialized post-processing, advancing remote sensing techniques for wildfire monitoring.

Findings

High potential of the VQ-VAE based method in burnt area detection

Effective integration of post-processing indexes improves accuracy

Promising results on high-resolution SPOT-6/7 images

Abstract

Wildfire detection using satellite images is a widely studied task in remote sensing with many applications to fire delineation and mapping. Recently, deep learning methods have become a scalable solution to automate this task, especially in the field of unsupervised learning where no training data is available. This is particularly important in the context of emergency risk monitoring where fast and effective detection is needed, generally based on high-resolution satellite data. Among various approaches, Anomaly Detection (AD) appears to be highly potential thanks to its broad applications in computer vision, medical imaging, as well as remote sensing. In this work, we build upon the framework of Vector Quantized Variational Autoencoder (VQ-VAE), a popular reconstruction-based AD method with discrete latent spaces, to perform unsupervised burnt area extraction. We integrate VQ-VAE into an end-to-end framework with an intensive post-processing step using dedicated vegetation, water and brightness indexes. Our experiments conducted on high-resolution SPOT-6/7 images provide promising results of the proposed technique, showing its high potential in future research on unsupervised burnt area extraction.