Classification of seismic events based on CL-MobileViT network

JiaYing SUN, Gang LI, Ling ZHANG, JunZhuo MA, ChaoFan WEI

Prog Geophy ›› 2024, Vol. 39 ›› Issue (6) : 2153-2164.

PDF(5114 KB)
Home Journals Progress in Geophysics
Progress in Geophysics

Abbreviation (ISO4): Prog Geophy      Editor in chief:

About  /  Aim & scope  /  Editorial board  /  Indexed  /  Contact  / 
PDF(5114 KB)
Prog Geophy ›› 2024, Vol. 39 ›› Issue (6) : 2153-2164. DOI: 10.6038/pg2024HH0592

Classification of seismic events based on CL-MobileViT network

Author information +
History +

Abstract

The accurate classification of seismic signals is the key link in constructing seismic catalog, which is of great significance for seismic catalog cleaning, earthquake monitoring and alerting, and seismological research. Aiming at the existing seismic event classification algorithms with low accuracy and large computational overhead, this paper designs a deep learning network CL-MobileViT for automatic classification of seismic events. CL-MobileViT comprehensively considers the performance and efficiency of the algorithm, selects MobileViT as the main body of the network, adds the attention mechanism to improve the sensitivity of the network to effective features, and uses the idea of large kernel convolution decomposition to reduce the computing overhead of the network. At the same time, the AdamW optimization strategy is adopted to guarantee that the final model can maximize the performance of the network. Specifically, first of all, add Coordinate Attention in the skip connection of MobileViT block, so that the network can pay fine attention to the information of different locations, strengthen the interaction modeling between long-distance seismic phase features, and improve the classification accuracy; Secondly, the traditional convolution used in the local feature extraction part of MobileViT block is replaced by multiple small-size convolution kernels decomposed by a large kernel convolution, which improves the nonlinear fitting ability of the network while reducing the computation and parameter number, thus improving the accuracy of seismic event classification. Finally, AdamW optimizer is used to prevent network from being overfitted and improve the training effect. By comparison with 11 existing mainstream deep learning classification models, it is found that CL-MobileViT can reach 97.3% accuracy in recognizing three seismic events, namely natural earthquake, collapse and blasting, which is superior to the comparison methods. Moreover, the number of parameters of CL-MobileViT is only 1.19 M, which is far lower than the comparison methods. It is proved that the method in this paper has better ability of seismic event classification.

Key words

Seismic classification / Transformer / Coordinate attention / Large kernel convolution decomposition / AdamW / Time-frequency map

Cite this article

Download Citations
JiaYing SUN , Gang LI , Ling ZHANG , et al . Classification of seismic events based on CL-MobileViT network[J]. Progress in Geophysics. 2024, 39(6): 2153-2164 https://doi.org/10.6038/pg2024HH0592

References

Cao Z L , Huang D R . Time-frequency characteristic analyses of measured and artificial seismic waves using the XGBoost algorithm. Journal of Tsinghua University (Science and Technology). 2022, 62(8 1330 1340
Gao J Y , Liu X F , Yan R . Time-frequency analysis of seismic records generated by natural earthquakes, blasts, and collapses near Pingmei mine based on STFT. Seismological and Geomagnetic Observation and Research. 2020, 41(3): 67-74
Guo M H , Lu C Z , Liu Z N , et al. Visual attention network. Computational Visual Media. 2023, 9(4): 733-752
Howard A G , Zhu M L , Chen B , et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv. 2017 1704.04861
Kim G , Ku B , Ko H . Multifeature fusion-based earthquake event classification using transfer learning. IEEE Geoscience and Remote Sensing Letters. 2021, 18(6): 974-978
Krizhevsky A , Sutskever I , Hinton G E . ImageNet classification with deep convolutional neural networks. Communications of the ACM. 2017, 60(6): 84-90
Ku B , Min J , Ahn J K , et al. Earthquake event classification using multitasking deep learning. IEEE Geoscience and Remote Sensing Letters. 2021, 18(7): 1149-1153
Li B J , Huang H M , Wang T T , et al. Research on seismic signal classification and recognition based on STFT and CNN. Progress in Geophysics. 2021, 36(4): 1404-1411
Li X R , Zhang N , Yao L P , et al. Sorting of seismic event triggering and seismic phase picking algorithms in ObsPy package. Earthquake Research in Shanxi. 2020 3): 42-46
Loshchilov I , Hutter F . Decoupled weight decay regularization. arXiv preprint arXiv. 2019 1711.05101
Mehta S , Rastegari M . MobileViT: light-weight, general-purpose, and mobile-friendly vision transformer. arXiv preprint arXiv. 2022 2110.02178
Ren T , Lin M N , Chen H F , et al. Seismic event classification based on bagging ensemble learning algorithm. Chinese Journal of Geophysics. 2019, 62(1): 383-392
Shi J P , Huang H M , Xue S M , et al. Research on seismic signal classification and recognition based on PSO-SVM. Computer & Digital Engineering. 2021, 49(6): 1037-1041 1037-1041, 1046
Wang W H , Yin X X , Yang X L , et al. Research on Huating earthquake classification in Gansu Province based on machine learning method. Progress in Geophysics. 2022, 37(6): 2308-2317
Xu Z S , Wen X T , Tan Z T , et al. A brief introduction to non-natural earthquakes. City and Disaster Reduction. 2021 6): 54-58
Yang X , Li Y H , Ge Z X . Machine learning and its application in seismology. Reviews of Geophysics and Planetary Physics. 2021, 52(1): 76-88
Zhang F , Yang X Z , Wu L F , et al. Classification of seismic events based on short-time Fourier transform and convolutional neural network. Acta Seismologica Sinica. 2021, 43(4): 463-473
Zhang S , Ku B , Ko H . Learnable maximum amplitude structure for earthquake event classification. IEEE Geoscience and Remote Sensing Letters. 2022, 19 3005805
Zhong Z C , Li H Q . Recognition and prediction of ground vibration signal based on machine learning algorithm. Neural Computing and Applications. 2020, 32(7): 1937-1947
Zhou S H , Jiang H K , Li J , et al. Research on identification of seismic events based on deep learning: Taking the records of Shandong seismic network as an example. Seismology and Geology. 2021a, 43(3): 663-676
Zhou S H , Jiang H K , Qu J H , et al. A review on research progress in recognition of blasting and collapse. Earthquake Research in China. 2021b, 37(2): 508-522
子龙 , 杜若 . 基于XGBoost算法的工程场地实测和人工地震波时频特征分析与判别. 清华大学学报(自然科学版). 2022, 62(8): 1330-1340
家乙 , 晓锋 , . 河南平顶山平煤矿区天然地震、爆破、塌陷时频特征分析. 地震地磁观测与研究. 2020, 41(3): 67-74
炳君 , 汉明 , 婷婷 , et al. 基于STFT和CNN的地震信号分类识别研究. 地球物理学进展. 2021, 36(4): 1404-1411
晓锐 , , 林鹏 , et al. ObsPy包中的地震事件触发和震相拾取算法梳理. 山西地震. 2020 3): 42-46
, 梦楠 , 宏峰 , et al. 基于Bagging集成学习算法的地震事件性质识别分类. 地球物理学报. 2019, 62(1): 383-392
佳朋 , 汉明 , 思敏 , et al. 基于PSO-SVM的地震信号分类识别研究. 计算机与数字工程. 2021, 49(6): 1037-1041 1037-1041, 1046
维欢 , 欣欣 , 秀隆 , et al. 基于机器学习方法的甘肃华亭地震分类研究. 地球物理学进展. 2022, 37(6): 2308-2317
志双 , 鑫涛 , 专条 , et al. 浅谈非天然地震. 城市与减灾. 2021 6): 54-58
, 永华 , 增喜 . 机器学习在地震学中的应用进展. 地球与行星物理论评. 2021, 52(1): 76-88
, 晓忠 , 立飞 , et al. 基于短时傅里叶变换和卷积神经网络的地震事件分类. 地震学报. 2021, 43(4): 463-473
少辉 , 海昆 , , et al. 基于深度学习的地震事件分类识别——以山东地震台网记录为例. 地震地质. 2021a, 43(3): 663-676
少辉 , 海昆 , 均浩 , et al. 爆破、塌陷识别研究进展综述. 中国地震. 2021b, 37(2): 508-522

感谢审稿专家提出的修改意见和编辑部的大力支持!

RIGHTS & PERMISSIONS

Copyright ©2024 Progress in Geophysics. All rights reserved.
PDF(5114 KB)

Accesses

Citation

Detail

Sections
Recommended

/