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

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

Research output: Contribution to journalArticlepeer-review

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 languageEnglish
JournalIEEE Geoscience and Remote Sensing Letters
Volume19
DOIs
Publication statusPublished - 2022

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

Fingerprint

Dive into the research topics of 'Graph Convolution Networks for Seismic Events Classification Using Raw Waveform Data from Multiple Stations'. Together they form a unique fingerprint.

Cite this