Graph Convolution Networks for Seismic Events Classification Using Raw Waveform Data from Multiple Stations

Gwantae Kim; Bonhwa Ku; Jae Kwang Ahn; Hanseok Ko

doi:10.1109/LGRS.2021.3127874

Graph Convolution Networks for Seismic Events Classification Using Raw Waveform Data from Multiple Stations

Gwantae Kim
, Bonhwa Ku
, Jae Kwang Ahn
, Hanseok Ko^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

29 Citations (Scopus)

Abstract

This letter proposes a multiple station-based seismic event classification model using a deep convolution neural network (CNN) and graph convolution network (GCN). To classify various seismic events, such as natural earthquakes, artificial earthquakes, and noise, the proposed model consists of weight-shared convolution layers, graph convolution layers, and fully connected layers. We employed graph convolution layers in order to aggregate features from multiple stations. Representative experimental results with the Korean peninsula earthquake datasets from 2016 to 2019 showed that the proposed model is superior to the single-station based state-of the-art methods. Moreover, the proposed model significantly reduced false alarms when using continuous waveforms of long duration. The code is available at.1

Original language	English
Journal	IEEE Geoscience and Remote Sensing Letters
Volume	19
DOIs	https://doi.org/10.1109/LGRS.2021.3127874
Publication status	Published - 2022

Bibliographical note

Funding Information:
This work was supported by the Development of earthquake, tsunami, volcano monitoring and prediction technology under Grant NTIS: 1365003423.

Publisher Copyright:
© 2004-2012 IEEE.

Keywords

Convolution neural network (CNN)
deep learning
graph convolution network (GCN)
multiple station
seismic event classification

ASJC Scopus subject areas

Geotechnical Engineering and Engineering Geology
Electrical and Electronic Engineering

Access to Document

10.1109/LGRS.2021.3127874

Cite this

@article{c0a1c8b034a745a4861d53e959b96bb3,

title = "Graph Convolution Networks for Seismic Events Classification Using Raw Waveform Data from Multiple Stations",

abstract = "This letter proposes a multiple station-based seismic event classification model using a deep convolution neural network (CNN) and graph convolution network (GCN). To classify various seismic events, such as natural earthquakes, artificial earthquakes, and noise, the proposed model consists of weight-shared convolution layers, graph convolution layers, and fully connected layers. We employed graph convolution layers in order to aggregate features from multiple stations. Representative experimental results with the Korean peninsula earthquake datasets from 2016 to 2019 showed that the proposed model is superior to the single-station based state-of the-art methods. Moreover, the proposed model significantly reduced false alarms when using continuous waveforms of long duration. The code is available at.1 ",

keywords = "Convolution neural network (CNN), deep learning, graph convolution network (GCN), multiple station, seismic event classification",

author = "Gwantae Kim and Bonhwa Ku and Ahn, \{Jae Kwang\} and Hanseok Ko",

note = "Funding Information: This work was supported by the Development of earthquake, tsunami, volcano monitoring and prediction technology under Grant NTIS: 1365003423. Publisher Copyright: {\textcopyright} 2004-2012 IEEE.",

year = "2022",

doi = "10.1109/LGRS.2021.3127874",

language = "English",

volume = "19",

journal = "IEEE Geoscience and Remote Sensing Letters",

issn = "1545-598X",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Graph Convolution Networks for Seismic Events Classification Using Raw Waveform Data from Multiple Stations

AU - Kim, Gwantae

AU - Ku, Bonhwa

AU - Ahn, Jae Kwang

AU - Ko, Hanseok

N1 - Funding Information: This work was supported by the Development of earthquake, tsunami, volcano monitoring and prediction technology under Grant NTIS: 1365003423. Publisher Copyright: © 2004-2012 IEEE.

PY - 2022

Y1 - 2022

N2 - This letter proposes a multiple station-based seismic event classification model using a deep convolution neural network (CNN) and graph convolution network (GCN). To classify various seismic events, such as natural earthquakes, artificial earthquakes, and noise, the proposed model consists of weight-shared convolution layers, graph convolution layers, and fully connected layers. We employed graph convolution layers in order to aggregate features from multiple stations. Representative experimental results with the Korean peninsula earthquake datasets from 2016 to 2019 showed that the proposed model is superior to the single-station based state-of the-art methods. Moreover, the proposed model significantly reduced false alarms when using continuous waveforms of long duration. The code is available at.1

AB - This letter proposes a multiple station-based seismic event classification model using a deep convolution neural network (CNN) and graph convolution network (GCN). To classify various seismic events, such as natural earthquakes, artificial earthquakes, and noise, the proposed model consists of weight-shared convolution layers, graph convolution layers, and fully connected layers. We employed graph convolution layers in order to aggregate features from multiple stations. Representative experimental results with the Korean peninsula earthquake datasets from 2016 to 2019 showed that the proposed model is superior to the single-station based state-of the-art methods. Moreover, the proposed model significantly reduced false alarms when using continuous waveforms of long duration. The code is available at.1

KW - Convolution neural network (CNN)

KW - deep learning

KW - graph convolution network (GCN)

KW - multiple station

KW - seismic event classification

UR - https://www.scopus.com/pages/publications/85119448017

U2 - 10.1109/LGRS.2021.3127874

DO - 10.1109/LGRS.2021.3127874

M3 - Article

AN - SCOPUS:85119448017

SN - 1545-598X

VL - 19

JO - IEEE Geoscience and Remote Sensing Letters

JF - IEEE Geoscience and Remote Sensing Letters

ER -

Graph Convolution Networks for Seismic Events Classification Using Raw Waveform Data from Multiple Stations

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this