Global Nuclear Explosion Discrimination Using a Convolutional Neural Network

Barama, Louisa; Williams, Jesse; Newman, Andrew V.; Peng, Zhigang

doi:10.1029/2022gl101528

Cited by 4 publications

(2 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We compare the proposed method to several state-of-theart networks, including CapsNet [1], CNN [10], VGG [11], ResNet [11], GoogleNet [11], CNN [12], and CNN [13] with the same training and testing set. Table 1 shows the results of the GA-Net comparison with the benchmark networks.…”

Section: Testing Resultsmentioning

confidence: 99%

See 1 more Smart Citation

A Ghost-Attention Network for Discriminating Tectonic and Non-Tectonic Events on a Small and Imbalanced Dataset

Liu,

Ren,

Chen

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Discrimination between tectonic and non-tectonic events is crucial to assess seismic hazards and manage associated risks. However, the discrimination process is challenging due to the imbalanced distribution of tectonic and non-tectonic events. In this paper, we propose a ghost-attention network (GA-Net) consisting of multiple ghost modules and convolutional block attention modules (CBAMs) to solve this problem. Ghost module allows the network to extract feature maps using cost effective operations, which are suitable for small and unbalanced training sets. CBAM emphasizes meaningful features along channel and spatial axes, effectively learning which information to emphasize or suppress. We train the proposed GA-Net using seismic data from Shaanxi Province in China. The evaluation shows that on the test set, the proposed GA-Net achieves sensitivity, specificity, and accuracy of 96.40%, 93.24%, and 95.75%, respectively. In addition, a comparative analysis with several state-of-the-art networks establishes its superiority. Our GA-Net exhibits robust generalized performance on several training sets with different imbalanced levels and an independent dataset.

show abstract

Section: Testing Resultsmentioning

confidence: 99%

“…Sensitivity Specificity ACC Paras CapsNet [1] 90.31% 80.43% 88.28% 40.32K CNN [10] 93.08% 90.04% 92.45% 1.36M VGG [11] 93.08% 66.19% 87.55% 11.37K ResNet [11] 98.61% 90.34% 96.85% 7.13M GoogleNet [11] 99.08% 91.10% 97.44% 5.12M CNN [12] 94.74% 89.68% 93.70% 5.20M CNN [13] 91…”

Section: Methodsmentioning

confidence: 99%

A Ghost-Attention Network for Discriminating Tectonic and Non-Tectonic Events on a Small and Imbalanced Dataset

Liu,

Ren,

Chen

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

Enhancing the classification of seismic events with supervised machine learning and feature importance

Habbak,

Abdalzaher,

Othman

et al. 2024

Sci Rep

View full text Add to dashboard Cite

The accurate classification of seismic events into natural earthquakes (EQ) and quarry blasts (QB) is crucial for geological understanding, seismic hazard mitigation, and public safety. This paper proposes a machine-learning approach to discriminate seismic events, particularly differentiating between natural EQs and man-made QBs. The core of this study is to integrate different features into a unified dataset to train some linear and nonlinear supervised machine learning (ML) models. The proposed approach considers a collection of 837 events (EQs and QBs) with local magnitudes of $$1.5 \le M_{L} \le 3.3$$ 1.5 ≤ M L ≤ 3.3 from the Egyptian National Seismic Network (ENSN) seismic event catalog between 2009 and 2015. This paper’s principal contribution is applying feature selection techniques and feature importance analysis to identify the best features leading to the best events’ discrimination. In other words, the proposed approach enhances classification accuracy and provides insights into which features are most crucial for distinguishing between EQ and QB events. The results show that with only three features, corner frequency, power of event, and spectral ratio, the best-developed ML model accomplishes a discrimination accuracy of 100% among several benchmarks of linear and non-linear models.

show abstract

Classifying small earthquakes, explosions and collapses in the western United States using physics-based features and machine learning

Koper,

Burlacu,

Armstrong

et al. 2024

Geophysical Journal International

View full text Add to dashboard Cite

SUMMARY Classifying the source type of small seismic events is a key task in seismology. A common goal is distinguishing tectonic earthquakes from explosions and human induced seismicity. To this end, we applied a spectral modelling workflow to Pg and Sg waves from ∼10 000 seismic events that occurred in or near Utah and were recorded by broadband seismometers in the western United States at distances of 10–300 km. The events were a mixture of tectonic earthquakes (EQ), industrial explosions (EX) and mining-induced seismicity (MIS, primarily collapses) and were mostly small (median magnitude of 1.34 MC). Our spectral modelling was successful for 54 per cent of the events, resulting in a new catalogue of M0 and fc values. We evaluated 13 physics-based features—including differential magnitudes, Pg/Sg spectral amplitude ratios, long-period/short-period spectral amplitude ratios and spectral misfit—as source classifiers. We found that Φ ≡ log10(M0) + 3log10(fc) was the most effective individual feature for distinguishing EQ from EX and MIS sources because EQ spectra are relatively enriched in high frequencies. We selected five less correlated features that spanned the feature space and used a naïve Bayes approach to create a three-way classification model. The model had 97.5 per cent accuracy when applied to an independent test data set. Model performance deteriorated when more than six features were combined. We conclude that models developed with a few physics-based waveform features can classify small seismic events with performance comparable to high-dimensional deep-learning models. Simple models that rely on physics-based features require less training data and make more interpretable decisions than deep-learning models, though they may require higher signal-to-noise ratios.

show abstract

Global Nuclear Explosion Discrimination Using a Convolutional Neural Network

Cited by 4 publications

References 52 publications

A Ghost-Attention Network for Discriminating Tectonic and Non-Tectonic Events on a Small and Imbalanced Dataset

A Ghost-Attention Network for Discriminating Tectonic and Non-Tectonic Events on a Small and Imbalanced Dataset

Enhancing the classification of seismic events with supervised machine learning and feature importance

Classifying small earthquakes, explosions and collapses in the western United States using physics-based features and machine learning

Contact Info

Product

Resources

About