ABSTRACT
The detection of earthquakes in seismological time series is central to observational seismology. Generally, seismic sensors passively record data and transmit it to the cloud or edge for integration, storage, and processing. However, transmitting raw data through the network is not an option for sensors deployed in harsh environments like underwater, underground, or in rural areas with limited connectivity. This paper introduces an efficient data processing pipeline and a set of lightweight deep-learning models for seismic event detection deployable on tiny devices such as microcontrollers. We conduct an extensive hyperparameter search and devise three lightweight models. We evaluate our models using the Stanford Earthquake Dataset and compare them with a basic STA/LTA detection algorithm and the state-of-the-art machine learning models, i.e., CRED, EQtransformer, and LCANet. For example, our smallest model consumes 193 kB of RAM and has an F1 score of 0.99 with just 29k parameters. Compared to CRED, which has an F1 score of 0.98 and 293k parameters, we reduce the number of parameters by a factor of 10. Deployed on Cortex M4 microcontrollers, the smallest version of LightEQ-NN has an inference time of 932 ms for 1 minute of raw data, an energy consumption of 5.86 mJ, and a flash requirement of 593 kB. Our results show that resource-efficient, on-device machine learning for seismological time series data is feasible and enables new approaches to seismic monitoring and early warning applications.
- James Bergstra, Rémi Bardenet, Yoshua Bengio, and Balázs Kégl. 2011. Algorithms for hyper-parameter optimization. Advances in neural information processing systems 24 (2011).Google Scholar
- Jörg Bialas and Ernst R Flueh. 1999. Ocean bottom seismometers. Sea Technol 40, 4 (1999), 41–46.Google Scholar
- Ronald Newbold Bracewell and Ronald N Bracewell. 1986. The Fourier transform and its applications. Vol. 31999. McGraw-Hill New York.Google Scholar
- Tai-Lin Chin, Chin-Ya Huang, Shan-Hsiang Shen, You-Cheng Tsai, Yu Hen Hu, and Yih-Min Wu. 2019. Learn to detect: Improving the accuracy of earthquake detection. IEEE Transactions on Geoscience and Remote Sensing 57, 11 (2019), 8867–8878.Google ScholarCross Ref
- Yoni Choukroun, Eli Kravchik, Fan Yang, and Pavel Kisilev. 2019. Low-bit quantization of neural networks for efficient inference. In 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW). IEEE, 3009–3018.Google ScholarCross Ref
- Robert David, Jared Duke, Advait Jain, Vijay Janapa Reddi, Nat Jeffries, Jian Li, Nick Kreeger, Ian Nappier, Meghna Natraj, Tiezhen Wang, 2021. Tensorflow lite micro: Embedded machine learning for tinyml systems. Proceedings of Machine Learning and Systems 3 (2021), 800–811.Google Scholar
- Gregory Giuliani, Hy Dao, Andrea De Bono, Bruno Chatenoux, Karin Allenbach, Pierric De Laborie, Denisa Rodila, Nikos Alexandris, and Pascal Peduzzi. 2017. Live Monitoring of Earth Surface (LiMES): A framework for monitoring environmental changes from Earth Observations. Remote Sensing of Environment 202 (2017), 222–233.Google ScholarCross Ref
- Daniel Griffin and Jae Lim. 1984. Signal estimation from modified short-time Fourier transform. IEEE Transactions on acoustics, speech, and signal processing 32, 2 (1984), 236–243.Google ScholarCross Ref
- Andrew Howard, Mark Sandler, Grace Chu, Liang-Chieh Chen, Bo Chen, Mingxing Tan, Weijun Wang, Yukun Zhu, Ruoming Pang, Vijay Vasudevan, 2019. Searching for mobilenetv3. In Proceedings of the IEEE/CVF international conference on computer vision. 1314–1324.Google ScholarCross Ref
- Forrest N Iandola, Song Han, Matthew W Moskewicz, Khalid Ashraf, William J Dally, and Kurt Keutzer. 2016. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and < 0.5 MB model size. arXiv preprint arXiv:1602.07360 (2016).Google Scholar
- Benoit Jacob, Skirmantas Kligys, Bo Chen, Menglong Zhu, Matthew Tang, Andrew Howard, Hartwig Adam, and Dmitry Kalenichenko. 2018. Quantization and training of neural networks for efficient integer-arithmetic-only inference. In Proceedings of the IEEE conference on computer vision and pattern recognition. 2704–2713.Google ScholarCross Ref
- Benoit Jacob, Skirmantas Kligys, Bo Chen, Menglong Zhu, Matthew Tang, Andrew Howard, Hartwig Adam, and Dmitry Kalenichenko. 2018. Quantization and training of neural networks for efficient integer-arithmetic-only inference. In Proceedings of the IEEE conference on computer vision and pattern recognition. 2704–2713.Google ScholarCross Ref
- Shengming Jiang. 2019. Marine internet for internetworking in oceans: A tutorial. Future Internet 11, 7 (2019), 146.Google ScholarCross Ref
- Supriya Kapur, Asit Mishra, and Debbie Marr. 2017. Low precision rnns: Quantizing rnns without losing accuracy. arXiv preprint arXiv:1710.07706 (2017).Google Scholar
- Fazle Karim, Somshubra Majumdar, Houshang Darabi, and Shun Chen. 2017. LSTM fully convolutional networks for time series classification. IEEE access 6 (2017), 1662–1669.Google Scholar
- Qingkai Kong, Young-Woo Kwon, Louis Schreierz, Steven Allen, Richard Allen, and Jennifer Strauss. 2015. Smartphone-based networks for earthquake detection. In 2015 15th international conference on innovations for community services (i4cs). IEEE, 1–8.Google ScholarCross Ref
- Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. 2012. 2012 AlexNet. Adv. Neural Inf. Process. Syst (2012), 1–9.Google Scholar
- Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. 2015. Deep learning. nature 521, 7553 (2015), 436–444.Google Scholar
- Jongseong Lim, Sunghun Jung, Chan JeGal, Gwanghoon Jung, Jung Ho Yoo, Jin Kyu Gahm, and Giltae Song. 2022. LEQNet: Light Earthquake Deep Neural Network for Earthquake Detection and Phase Picking. Frontiers in Earth Science 10 (2022), 848237.Google ScholarCross Ref
- Han Liu and Jian-zhong ZHANG. 2014. STA/LTA algorithm analysis and improvement of Microseismic signal automatic detection. Progress in Geophysics 29, 4 (2014), 1708–1714.Google Scholar
- JH Luetgert, DH Oppenheimer, and J Hamilton. 2010. The netquakes project-research-quality seismic data transmitted via the internet from citizen-hosted instruments. In AGU Fall Meeting Abstracts, Vol. 2010. S51E–03.Google Scholar
- S Mostafa Mousavi, William L Ellsworth, Weiqiang Zhu, Lindsay Y Chuang, and Gregory C Beroza. 2020. Earthquake transformer—an attentive deep-learning model for simultaneous earthquake detection and phase picking. Nature communications 11, 1 (2020), 1–12.Google Scholar
- S Mostafa Mousavi, Stephen P Horton, Charles A Langston, and Borhan Samei. 2016. Seismic features and automatic discrimination of deep and shallow induced-microearthquakes using neural network and logistic regression. Geophysical Journal International 207, 1 (2016), 29–46.Google ScholarCross Ref
- Seyed Mostafa Mousavi and Charles Langston. 2016. Fast and novel microseismic detection using time-frequency analysis. In SEG Technical Program Expanded Abstracts 2016. Society of Exploration Geophysicists, 2632–2636.Google Scholar
- S Mostafa Mousavi, Yixiao Sheng, Weiqiang Zhu, and Gregory C Beroza. 2019. STanford EArthquake Dataset (STEAD): A global data set of seismic signals for AI. IEEE Access 7 (2019), 179464–179476.Google ScholarCross Ref
- S Mostafa Mousavi, Weiqiang Zhu, Yixiao Sheng, and Gregory C Beroza. 2019. CRED: A deep residual network of convolutional and recurrent units for earthquake signal detection. Scientific reports 9, 1 (2019), 1–14.Google Scholar
- Keiron O’Shea and Ryan Nash. 2015. An introduction to convolutional neural networks. arXiv preprint arXiv:1511.08458 (2015).Google Scholar
- Thibaut Perol, Michaël Gharbi, and Marine Denolle. 2018. Convolutional neural network for earthquake detection and location. Science Advances 4, 2 (2018), e1700578.Google ScholarCross Ref
- Omar M Saad and Yangkang Chen. 2020. Earthquake detection and P-wave arrival time picking using capsule neural network. IEEE Transactions on Geoscience and Remote Sensing 59, 7 (2020), 6234–6243.Google ScholarCross Ref
- Sandra Sendra, Jaime Lloret, Jose Miguel Jimenez, and Lorena Parra. 2015. Underwater acoustic modems. IEEE Sensors Journal 16, 11 (2015), 4063–4071.Google ScholarCross Ref
- BK Sharma, Amod Kumar, and VM Murthy. 2010. Evaluation of seismic events detection algorithms. Journal of the Geological Society of India 75, 3 (2010), 533–538.Google ScholarCross Ref
- Alex Sherstinsky. 2020. Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network. Physica D: Nonlinear Phenomena 404 (2020), 132306.Google ScholarCross Ref
- Amadej Trnkoczy. 2009. Understanding and parameter setting of STA/LTA trigger algorithm. In New Manual of Seismological Observatory Practice (NMSOP). Deutsches GeoForschungsZentrum GFZ, 1–20.Google Scholar
- David A Van Dyk and Xiao-Li Meng. 2001. The art of data augmentation. Journal of Computational and Graphical Statistics 10, 1 (2001), 1–50.Google ScholarCross Ref
- Yue Wu, Youzuo Lin, Zheng Zhou, David Chas Bolton, Ji Liu, and Paul Johnson. 2018. DeepDetect: A cascaded region-based densely connected network for seismic event detection. IEEE Transactions on Geoscience and Remote Sensing 57, 1 (2018), 62–75.Google ScholarCross Ref
- Clara E Yoon, Ossian O’Reilly, Karianne J Bergen, and Gregory C Beroza. 2015. Earthquake detection through computationally efficient similarity search. Science advances 1, 11 (2015), e1501057.Google Scholar
- Yu Zhao, Pan Deng, Junting Liu, Mulan Wang, and Jiafu Wan. 2022. LCANet: Lightweight Context-Aware Attention Networks for Earthquake Detection and Phase-Picking on IoT Edge Devices. IEEE Systems Journal 16, 3 (2022), 4024–4035. https://doi.org/10.1109/JSYST.2021.3114689Google ScholarCross Ref
Index Terms
- LightEQ: On-Device Earthquake Detection with Embedded Machine Learning
Recommendations
Fusion-based volcanic earthquake detection and timing in wireless sensor networks
Volcano monitoring is of great interest to public safety and scientific explorations. However, traditional volcanic instrumentation such as broadband seismometers are expensive, power hungry, bulky, and difficult to install. Wireless sensor networks (...
Small-layered Feed-Forward and Convolutional neural networks for efficient P wave earthquake detection
AbstractThe number and efficiency of seismic networks have steadily increase over time delivering large datasets to be analyzed for earthquake occurrence. Automatic tools for accurate earthquake detection are under emerging and intense ...
Highlights- Small-layered neural networks with high detection efficiency of seismic events.
An automated earthquake classification model based on a new butterfly pattern using seismic signals
Abstract BackgroundSeismic signals are useful for earthquake detection and classification. Therefore, various artificial intelligence (AI) models have been used with seismic signals to develop automated earthquake detection systems. Our primary goal is ...
Comments