QuakeSet is a dataset to analyze different attributes of earthquakes. It contains bi-temporal time series of images and ground truth annotations for magnitudes, hypocenters, and affected areas.
You can find an official implementation on TorchGeo.
The images are taken from the Sentinel-1 mission using the Interferometric Wide swath mode.
The International Seismological Centre provides information about earthquakes.
The dataset is divided into three folds with equal distribution of magnitudes and balanced in positive and negative examples.
Each sample contains:
NOTE: the main file is earthquakes.h5, while p_earthquakes.h5 contains the private set for the SMAC Challenge. This set contains only placeholder ground truth.
BibTeX:
@article{Rege_Cambrin_2024,
title={QuakeSet: A Dataset and Low-Resource Models to Monitor Earthquakes through Sentinel-1},
ISSN={2411-3387},
url={http://dx.doi.org/10.59297/n89yc374},
DOI={10.59297/n89yc374},
journal={Proceedings of the International ISCRAM Conference},
publisher={Information Systems for Crisis Response and Management},
author={Rege Cambrin, Daniele and Garza, Paolo},
year={2024},
month=may
}
23 commits
QuakeSet is a dataset to analyze different attributes of earthquakes. It contains bi-temporal time series of images and ground truth annotations for magnitudes, hypocenters, and affected areas.
You can find an official implementation on TorchGeo.
The images are taken from the Sentinel-1 mission using the Interferometric Wide swath mode.
The International Seismological Centre provides information about earthquakes.
The dataset is divided into three folds with equal distribution of magnitudes and balanced in positive and negative examples.
Each sample contains:
NOTE: the main file is earthquakes.h5, while p_earthquakes.h5 contains the private set for the SMAC Challenge. This set contains only placeholder ground truth.
BibTeX:
@article{Rege_Cambrin_2024,
title={QuakeSet: A Dataset and Low-Resource Models to Monitor Earthquakes through Sentinel-1},
ISSN={2411-3387},
url={http://dx.doi.org/10.59297/n89yc374},
DOI={10.59297/n89yc374},
journal={Proceedings of the International ISCRAM Conference},
publisher={Information Systems for Crisis Response and Management},
author={Rege Cambrin, Daniele and Garza, Paolo},
year={2024},
month=may
}
23 commits