Deep Learning Approach for Earthquake Detection and Classification using MobileNet V2 Transfer Learning

Giel Utami Putri Siregar; Agus Subekti; Hasri Akbar Awal Rozaq; Oktay Yildiz

doi:10.59395/ijadis.v7i1.1400

Authors

Giel Utami Putri Siregar Faculty of Information Technology, Universitas Nusa Mandiri, Jakarta, Indonesia
Agus Subekti Faculty of Information Technology, Universitas Nusa Mandiri, Jakarta, Indonesia
Hasri Akbar Awal Rozaq Graduate School of Informatics, Department of Computer Science, Gazi University, Ankara, Türkiye
Oktay Yildiz Faculty of Engineering, Department of Computer Engineering, Gazi University, Ankara, Türkiye

DOI:

https://doi.org/10.59395/ijadis.v7i1.1400

Keywords:

Earthquake Detection, MobileNetV2, Spectogram (STFT), Seismic Time Series

Abstract

Earthquake detection is a critical component of disaster management, as early identification of seismic events can help mitigate potential damage and support timely response efforts. This study evaluates the application of deep learning for binary earthquake detection using a lightweight Convolutional Neural Network (CNN) based on the MobileNetV2 architecture. The experiments were conducted using seismic waveform data from the Stanford Earthquake Dataset (STEAD), which were transformed into time–frequency representations through the Short-Time Fourier Transform (STFT). Spectrogram images derived from the seismic signals were used as input to the CNN models. Transfer learning was applied to MobileNetV2 to adapt the pretrained architecture to the earthquake detection task. The proposed approach achieved an accuracy of 99%, precision of 100%, recall of 97.96%, and an F1-score of 98.97% on the test dataset. In terms of model complexity, MobileNetV2 has 7,176,600 total parameters and 1,639,538 trainable parameters, indicating a favorable balance between performance and computational efficiency. For comparative evaluation, MobileNetV2 was benchmarked against several commonly used CNN architectures, including CNN Vanilla, MobileNetV1, VGG16, and ResNet, under the same experimental conditions. The results indicate that MobileNetV2 provides competitive detection performance while maintaining a significantly smaller model size. Although real-time deployment on mobile devices was not implemented in this study, the findings suggest that lightweight CNN architectures, such as MobileNetV2, hold promise for future earthquake detection systems operating in resource-constrained environments.

1065 477

Downloads

Download data is not yet available.

References

[1] T. Perol, M. Gharbi, and M. Denolle, Convolutional neural network for earthquake detection and location, Sci. Adv., vol. 4, no. 2, Feb. 2018, doi: 10.1126/sciadv.1700578. DOI: https://doi.org/10.1126/sciadv.1700578

[2] Z. E. Ross, M. Meier, E. Hauksson, and T. H. Heaton, Generalized Seismic Phase Detection with Deep Learning, Bull. Seismol. Soc. Am., vol. 108, no. 5A, pp. 28942901, Oct. 2018, doi: 10.1785/0120180080. DOI: https://doi.org/10.1785/0120180080

[3] D. Zhexebay et al., Deep Learning for Early Earthquake Detection: Application of Convolutional Neural Networks for P-Wave Detection, Appl. Sci., vol. 15, no. 7, p. 3864, Apr. 2025, doi: 10.3390/app15073864. DOI: https://doi.org/10.3390/app15073864

[4] S. M. Mousavi, Y. Sheng, W. Zhu, and G. C. Beroza, STanford EArthquake Dataset (STEAD): A Global Data Set of Seismic Signals for AI, IEEE Access, vol. 7, pp. 179464179476, 2019, doi: 10.1109/ACCESS.2019.2947848. DOI: https://doi.org/10.1109/ACCESS.2019.2947848

[5] S. M. Mousavi and G. C. Beroza, Deep-learning seismology, Science (80-. )., vol. 377, no. 6607, Aug. 2022, doi: 10.1126/science.abm4470. DOI: https://doi.org/10.1126/science.abm4470

[6] Z. E. Ross, M. Meier, and E. Hauksson, P Wave Arrival Picking and FirstMotion Polarity Determination With Deep Learning, J. Geophys. Res. Solid Earth, vol. 123, no. 6, pp. 51205129, Jun. 2018, doi: 10.1029/2017JB015251. DOI: https://doi.org/10.1029/2017JB015251

[7] S. M. Mousavi, W. Zhu, Y. Sheng, and G. C. Beroza, CRED: A Deep Residual Network of Convolutional and Recurrent Units for Earthquake Signal Detection, Sci. Rep., vol. 9, no. 1, p. 10267, Jul. 2019, doi: 10.1038/s41598-019-45748-1. DOI: https://doi.org/10.1038/s41598-019-45748-1

[8] W. Zhu and G. C. Beroza, PhaseNet: A Deep-Neural-Network-Based Seismic Arrival Time Picking Method, Geophys. J. Int., Oct. 2018, doi: 10.1093/gji/ggy423. DOI: https://doi.org/10.1093/gji/ggy423

[9] S. M. Mousavi, W. L. Ellsworth, W. Zhu, L. Y. Chuang, and G. C. Beroza, Earthquake transformeran attentive deep-learning model for simultaneous earthquake detection and phase picking, Nat. Commun., vol. 11, no. 1, p. 3952, Aug. 2020, doi: 10.1038/s41467-020-17591-w. DOI: https://doi.org/10.1038/s41467-020-17591-w

[10] T. Lu et al., EQTypeNet: deep learning tri-branch earthquake automatic classification model and its application in China, Geophys. J. Int., vol. 242, no. 3, Jul. 2025, doi: 10.1093/gji/ggaf263. DOI: https://doi.org/10.1093/gji/ggaf263

[11] A. Navarro-Rodrguez et al., Recent Advances in Early Earthquake Magnitude Estimation by Using Machine Learning Algorithms: A Systematic Review, Appl. Sci., vol. 15, no. 7, p. 3492, Mar. 2025, doi: 10.3390/app15073492. DOI: https://doi.org/10.3390/app15073492

[12] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, MobileNetV2: Inverted Residuals and Linear Bottlenecks, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, Jun. 2018, pp. 45104520. doi: 10.1109/CVPR.2018.00474. DOI: https://doi.org/10.1109/CVPR.2018.00474

[13] T. M. Pham et al., A CNN-LSTM Framework for Time-Windowed Detection of Local Earthquakes from Ocean Bottom Seismometer Data, J. Robot. Control, vol. 6, no. 6, p. 2025, 2025, doi: 10.18196/jrc.v6i6.28381.

[14] D. Baar and R. N. elik, A Hybrid CNN-LSTM Architecture for Seismic Event Detection Using High-Rate GNSS Velocity Time Series, Sensors, vol. 26, no. 2, p. 519, Jan. 2026, doi: 10.3390/s26020519. DOI: https://doi.org/10.3390/s26020519

[15] I. Kaftan, Machine Learning Applications for Earthquake Magnitude Prediction in Western Turkey, Appl. Sci., vol. 15, no. 20, p. 10909, Oct. 2025, doi: 10.3390/app152010909. DOI: https://doi.org/10.3390/app152010909

[16] P. Welch, The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms, IEEE Trans. Audio Electroacoust., vol. 15, no. 2, pp. 7073, Jun. 1967, doi: 10.1109/TAU.1967.1161901. DOI: https://doi.org/10.1109/TAU.1967.1161901

[17] Y. Shi, H. Tang, M. J. Baine, M. A. Hollingsworth, H. Du, and ..., Adversarial Networks with a 3D U-Net Based Generator to Achieve the Accurate and Effective Synthesis of Clinical Tumor Image Data for Pancreatic Cancer, Cancers (Basel)., 2023, [Online]. Available: https://www.mdpi.com/2072-6694/15/23/5496 DOI: https://doi.org/10.3390/cancers15235496

[18] B. A. Omodunbi et al., Stacked Ensemble Learning for Classification of Parkinsons Disease Using Telemonitoring Vocal Features, Diagnostics, vol. 15, no. 12, p. 1467, Jun. 2025, doi: 10.3390/diagnostics15121467. DOI: https://doi.org/10.3390/diagnostics15121467

[19] E. Patimah, V. B. Haekal, and D. Sandya Prasvita, Klasifikasi Penyakit Liver dengan Menggunakan Metode Decision Tree, Semin. Nas. Mhs. Ilmu Komput. dan Apl. Jakarta-Indonesia, vol. 2, no. 1, pp. 655659, 2021.

[20] M. R. Canal, Comparison of Wavelet and Short Time Fourier Transform Methods in the Analysis of EMG Signals, J. Med. Syst., vol. 34, no. 1, pp. 9194, Feb. 2010, doi: 10.1007/s10916-008-9219-8. DOI: https://doi.org/10.1007/s10916-008-9219-8

[21] D. Griffin and Jae Lim, Signal estimation from modified short-time Fourier transform, IEEE Trans. Acoust., vol. 32, no. 2, pp. 236243, Apr. 1984, doi: 10.1109/TASSP.1984.1164317. DOI: https://doi.org/10.1109/TASSP.1984.1164317

[22] P. Ewert, B. Wicher, and T. Pajchrowski, Application of the STFT for Detection of the Rotor Unbalance of a Servo-Drive System with an Elastic Interconnection, Electronics, vol. 13, no. 2, p. 441, Jan. 2024, doi: 10.3390/electronics13020441. DOI: https://doi.org/10.3390/electronics13020441

[23] M. A. Latumahina, H. Santoso, I. Wasito, and H. Haryono, Emotion Recognition System Using Autoencoder + CNN + Attention, vol. 10, no. 4, pp. 150161, 2023, [Online]. Available: http://jurnal.mdp.ac.id

[24] D. Baby, S. J. Devaraj, J. Hemanth, and M. M. Anishin Raj, Leukocyte classification based on feature selection using extra trees classifier: A transfer learning approach, Turkish J. Electr. Eng. Comput. Sci., vol. 29, no. 8, pp. 27422757, 2021, doi: 10.3906/elk-2104-183. DOI: https://doi.org/10.3906/elk-2104-183

[25] R. Balestriero, L. Bottou, and Y. LeCun, The Effects of Regularization and Data Augmentation are Class Dependent, Adv. Neural Inf. Process. Syst., vol. 35, 2022, [Online]. Available: https://proceedings.neurips.cc/paper_files/paper/2022/hash/f73c04538a5e1cad40ba5586b4b517d3-Abstract-Conference.html

[26] M. Zakariah, S. A. AlQahtani, and M. S. Al-Rakhami, Machine Learning-Based Adaptive Synthetic Sampling Technique for Intrusion Detection, Appl. Sci., vol. 13, no. 11, p. 6504, May 2023, doi: 10.3390/app13116504. DOI: https://doi.org/10.3390/app13116504

[27] A. Krizhevsky, I. Sutskever, and G. E. Hinton, ImageNet classification with deep convolutional neural networks, Commun. ACM, vol. 60, no. 6, pp. 8490, May 2017, doi: 10.1145/3065386. DOI: https://doi.org/10.1145/3065386

[28] G. W. Intyanto, Klasifikasi Citra Bunga dengan Menggunakan Deep Learning: CNN (Convolution Neural Network), J. Arus Elektro Indones., vol. 7, no. 3, p. 80, Dec. 2021, doi: 10.19184/jaei.v7i3.28141. DOI: https://doi.org/10.19184/jaei.v7i3.28141

[29] H. A. A. Rozaq, E. M. zen, M. K. Altparmak, and O. Yldz, A Hybrid Approach Combining Neural Networks and Signal Processing for Indoor Temperature Detection Using Plants Bioelectric Potential, in 2025 9th International Symposium on Innovative Approaches in Smart Technologies (ISAS), IEEE, Jun. 2025, pp. 18. doi: 10.1109/ISAS66241.2025.11101899. DOI: https://doi.org/10.1109/ISAS66241.2025.11101899

[30] I. M. Putra, I. Tahyudin, H. A. A. Rozaq, A. Y. SyafaAt, R. Wahyudi, and E. Winarto, Classification analysis of COVID19 patient data at government hospital of banyumas using machine learning, in 2021 2nd International Conference on Smart Computing and Electronic Enterprise: Ubiquitous, Adaptive, and Sustainable Computing Solutions for New Normal, ICSCEE 2021, Malaysia: IEEE, Jun. 2021, pp. 271274. doi: 10.1109/ICSCEE50312.2021.9498020. DOI: https://doi.org/10.1109/ICSCEE50312.2021.9498020