Hephaestus: A large scale multitask dataset towards InSAR understanding

TL;DR

This paper introduces Hephaestus, the first large-scale, manually annotated InSAR dataset with nearly 20,000 interferograms, enabling diverse deep learning applications in Earth observation and geophysical analysis.

Contribution

The creation and release of a comprehensive, annotated InSAR dataset that supports multiple computer vision tasks, filling a critical gap in Earth observation data resources.

Findings

Dataset includes 19,919 interferograms from 44 volcanoes.

Contains 216,106 InSAR patches with annotations.

Supports various computer vision tasks like classification, segmentation, and captioning.

Abstract

Synthetic Aperture Radar (SAR) data and Interferometric SAR (InSAR) products in particular, are one of the largest sources of Earth Observation data. InSAR provides unique information on diverse geophysical processes and geology, and on the geotechnical properties of man-made structures. However, there are only a limited number of applications that exploit the abundance of InSAR data and deep learning methods to extract such knowledge. The main barrier has been the lack of a large curated and annotated InSAR dataset, which would be costly to create and would require an interdisciplinary team of experts experienced on InSAR data interpretation. In this work, we put the effort to create and make available the first of its kind, manually annotated dataset that consists of 19,919 individual Sentinel-1 interferograms acquired over 44 different volcanoes globally, which are split into 216,106 InSAR patches. The annotated dataset is designed to address different computer vision problems, including volcano state classification, semantic segmentation of ground deformation, detection and classification of atmospheric signals in InSAR imagery, interferogram captioning, text to InSAR generation, and InSAR image quality assessment.