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

被引：19

作者：

Kim, Gwantae ^{[1
]}

Ku, Bonhwa ^{[1
]}

Ahn, Jae-Kwang ^{[2
]}

Ko, Hanseok ^{[1
]}

机构：

[1] Korea Univ, Sch Elect Engn, Seoul 02841, South Korea

[2] Korea Meteorol Adm, Seoul 07062, South Korea

来源：

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS | 2022年 / 19卷

关键词：

Earthquakes; Convolution; Feature extraction; Data models; Convolutional neural networks; Adaptation models; Neural networks; Convolution neural network (CNN); deep learning; graph convolution network (GCN); multiple station; seismic event classification;

D O I：

10.1109/LGRS.2021.3127874

中图分类号：

P3 [地球物理学]; P59 [地球化学];

学科分类号：

0708 ; 070902 ;

摘要：

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)

引用

页数：5

共 24 条

[21]

Trani L., 2020, EARTH SPACE SCI OPEN, P1

[22] Earthquake detection through computationally efficient similarity search [J].