Noise reduction plays a pivotal role in enhancing the signal-to-noise ratio (SNR) of microseismic data during the preprocessing stage. However, due to the heterogeneous nature and high complexity of noise sources, conventional model-driven denoising methods frequently exhibit suboptimal performance, thereby compromising the fidelity and reliability of downstream microseismic processing tasks. Moreover, while deep learning-based approaches have shown promise for microseismic noise suppression, the resulting models often exhibit limited generalization capability and insufficient preservation of physically meaningful signal features. To address these challenges, this study introduces RD-SAUNet—a novel U-Net–based architecture incorporating hybrid dilation convolutions and a residual spatial attention mechanism. Built upon the foundational U-type convolutional neural network (U-Net) framework, RD-SAUNet integrates three key innovations: (i) residual connections to facilitate gradient flow and accelerate training convergence; (ii) a spatial attention mechanism (SAM) that adaptively emphasizes discriminative signal features while suppressing spurious noise; and (iii) hybrid dilation convolutions that expand the receptive field to capture broader contextual information without significantly increasing computational overhead. Collectively, these design choices enhance both model generalizability and signal fidelity. The proposed method is rigorously evaluated on both synthetic and field-collected microseismic datasets and benchmarked against state-of-the-art denoising techniques. Experimental results confirm that RD-SAUNet achieves robust suppression of random noise, yields substantial SNR improvement, and faithfully preserves the intrinsic temporal-spectral characteristics essential for accurate microseismic event detection and characterization.
- Open Access
- Article
Denoising of Microseismic Data Using U-Net Network with Residual Hybrid Dilated Convolutional Space Attention Mechanism
- Huashuai Cui 1,2,3,4,
- Fei Han 1,2,3,4,*,
- Yuhang Sun 1,2,3,4,
- Huanjun Chen 1,2,3,4
Author Information
Received: 14 Jan 2026 | Revised: 02 Mar 2026 | Accepted: 05 May 2026 | Published: 22 Jun 2026
Abstract
Keywords
micro-seismic | noise suppression | spatial attention mechanism | residual network | dilated convolution | signal-to-noise ratio
References
- 1.
Birnie, C.; Ravasi, M.; Liu, S.; et al. The potential of self-supervised networks for random noise suppression in seismic data. Artif. Intell. Geosci. 2021, 2, 47–59.
- 2.
Gao, H.Y.; Zhang, M.; Hou, N.; et al. Dynamic-transmission-based recursive filtering algorithm for microseismic event detection under sensor saturations. Measurement 2021, 186, 110197.
- 3.
Guan, C.; Zheng, L.; Wang, C. A novel CEEMD-based multichannel denoising autoencoder for noise attenuation of surface microseismic data. Int. J. Netw. Dyn. Intell. 2025, 4, 100026.
- 4.
Fomel, S.; Liu, Y. Seislet transform and seislet frame. Geophysics 2010, 75, V25–V38.
- 5.
Liu, S.; Wang, L.; Zhang, Y.; et al. Recursive filtering of networked systems with communication protocol scheduling: A survey, Int. J. Syst. Sci. 2025, 56, 2499–2516.
- 6.
Zhang, Y.; Liu, G.; Song, X. Unscented recursive three-step filter based unbiased minimum-variance estimation for a class of nonlinear systems. Int. J. Syst. Sci. 2025, 56, 227–236.
- 7.
Zhi, Y.-L.; Wu, Y.-Y.; Zhang, Y.; et al. A new result on reachable set estimation for Markovian jump neural networks with time-varying delays. Int. J. Syst. Sci. 2025, 56, 4131–4143.
- 8.
Meng, C.; Gao, J.; Tian, Y.; et al. Attenuation of seismic random noise with unknown distribution: A gaussianization framework. IEEE Trans. Geosci. Remote. Sens. 2023, 61, 5915915.
- 9.
Mousavi, S.M.; Langston, C.A. Hybrid seismic denoising using higher-order statistics and improved wavelet block thresholding. Bull. Seismol. Soc. Am. 2016, 106, 1380–1393.
- 10.
Labate, D.; Lim, W.-Q.; Kutyniok, G.; et al. Sparse multidimensional representation using shearlets. In Wavelets XI; SPIE: Bellingham, DC, USA. 2005; pp. 254–262.
- 11.
Han, J.; van der Baan, M. Microseismic and seismic denoising via ensemble empirical mode decomposition and adaptive thresholding. Geophysics 2015, 80, KS69–KS80.
- 12.
Zhang, X.; Cao, L.; Chen, Y.; et al. Microseismic signal denoising by combining variational mode decomposition with permutation entropy. Appl. Geophys. 2022, 19, 65–80.
- 13.
Mousavi, S.M.; Langston, C.A.; Horton, S.P. Automatic microseismic denoising and onset detection using the synchrosqueezed continuous wavelet transform. Geophysics 2016, 81, V341–V355.
- 14.
Zhu, W.; Zhang, C.; Qiu, T.; et al. The seismic signal technology with variable velocity FK filtering-the auto-adapted polarization filtering. Prog. Geophys. 2009, 24, 1776–1786.
- 15.
Tselentis, G.-A.; Martakis, N.; Paraskevopoulos, P.; et al. Strategy for automated analysis of passive microseismic data based on S-transform, Otsus thresholding, and higher order statistics. Geophysics 2012, 77, KS43–KS54.
- 16.
Song, B.; Zhao, S.; Dang, L.; et al. A survey on learning from data with label noise via deep neural networks, Syst. Sci. Control. Eng. 2025, 13, 2488120.
- 17.
Zhan, Y.; Yang, R.; You, J.; et al. A systematic literature review on incomplete multimodal learning: techniques and challenges. Syst. Sci. Control. Eng. 2025, 13, 2467083.
- 18.
Lin, J.; Zheng, J.; Li, D.; et al. Research on microseismic denoising method based on CBDNet. Artif. Intell. Geosci. 2023, 4, 28–38.
- 19.
Zhang, Y.; Li, X.; Wang, B.; et al. Robust seismic data denoising based on deep learning. Oil Geophys. Prospect. 2022, 57, 12–25.
- 20.
Wang, Y.; Wen, C.; Wu, X. Fault detection and isolation of floating wind turbine pitch system based on Kalman filter and multi-attention 1DCNN. Syst. Sci. Control. Eng. 2024, 12, 2362169.
- 21.
Liu, L.; Song, W.; Zeng, C.; et al. Microseismic event detection and classification based on convolutional neural network. J. Appl. Geophys. 2021, 192, 104380.
- 22.
Mandelli, S.; Lipari, V.; Bestagini, P.; et al. Interpolation and denoising of seismic data using convolutional neural networks. arXiv 2019, arXiv:1901.07927.
- 23.
Han, W.; Zhou, Y.; Chi, Y. Deep learning convolutional neural networks for random noise attenuation in seismic data. Geophys. Prospect. Pet. 2018, 57, 862–869.
- 24.
Yang, L.; Chen, W.; Liu, W.; et al. Random noise attenuation based on residual convolutional neural network in seismic datasets. IEEE Access 2020, 8, 30271–30286.
- 25.
Zheng, J.; Jiang, T.; Wu, Z.; et al. Application of residual learning to microseismic random noise attenuation. Acta Geophys. 2021, 69, 1151–1161.
- 26.
Wang, Y.; Lu, W.; Liu, J.; et al. Random seismic noise attenuation based on data augmentation and CNN. Chin. J. Geophys. 2019, 62, 421–433.
- 27.
Luo, R.; Li, Y. Random seismic noise attenuation based on RUNet convolutional neural network. Geophys. Prospect. Pet. 2020, 59, 51–59.
- 28.
Zhang, H.; Ma, C.; Pazzi, V.; et al. Microseismic signal denoising and separation based on fully convolutional encoder–decoder network. Appl. Sci. 2020, 10, 6621.
- 29.
Chen, H.; Wu, R.; Tao, C.; et al. Multi-scale class attention network for diabetes retinopathy grading. Int. J. Netw. Dyn. Intell. 2024, 3, 100012.
- 30.
Chen, L.; Wu, P.; Tan, W.; et al. A novel UAV-based road damage detection algorithm with lightweight convolution and attention mechanism. Int. J. Netw. Dyn. Intell. 2025, 4, 100025.
- 31.
Chun, W.; Xu, Y.; Li, Z.; et al. Attention-guided CNN for image denoising. Neural Netw. 2020, 124, 117–129.
- 32.
Yang, C.; Zhou, Y.; He, H.; et al. Global context and attention-based deep convolutional neural network for seismic data denoising. Geophys. Prospect. Pet. 2021, 60, 751–762.
- 33.
Ehab, W.; Huang, L.; Li, Y. UNet and variants for medical image segmentation. Int. J. Netw. Dyn. Intell. 2024, 3, 100009.
- 34.
Chen, H.; Gao, J.; Gao, Z.; et al. A sequential iterative deep learning seismic blind high-resolution inversion. IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. 2021, 14, 7817–7829.
- 35.
Bai, T.; Zhao, H.; Wang, Z. A U-Net based deep learning approach for seismic random noise suppression. In Proceedings of the IGARSS 2022—2022 IEEE International Geoscience and Remote Sensing Symposium, Kuala Lumpur, Malaysia, 17–22 July 2022; pp. 6165–6168.
- 36.
Saad, O.M.; Oboue, Y.A.S.I.; Bai, M.; et al. Self-attention deep image prior network for unsupervised 3-D seismic data enhancement. IEEE Trans. Geosci. Remote. Sens. 2021, 60, 5907014.
- 37.
Yu, F.; Koltun, V. Multi-scale context aggregation by dilated convolutions. arXiv 2015, arXiv:1511.07122.
- 38.
Zhang, Y.; Li, X.; Wang, B.; et al. Random noise suppression of seismic data based on joint deep learning. Oil Geophys. Prospect. 2021, 56, 9–25.
- 39.
Zagoruyko, S.; Komodakis, N. Paying more attention to attention: Improving the performance of convolutional neural networks via attention transfer. arXiv 2016, arXiv:1612.03928.
- 40.
Woo, S.; Park, J.; Lee, J.-Y.; et al. Cbam: Convolutional block attention module. In Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany, 8–14 September 2018; pp. 3–19.
- 41.
Guo, C.; Szemenyei, M.; Yi, Y.; et al. Sa-UNet: Spatial attention U-Net for retinal vessel segmentation. In Proceedings of the 2020 25th International Conference on Pattern Recognition (ICPR), Milan, Italy, 10–15 January 2021; pp. 1236–1242.
- 42.
Shirrmeister, R.T.; Springenberg, J.T.; Fiederer, L.D.J. Deep learning with convolutional neural networks for EEG decoding and visualization. Hum. Brain Mapp. 2017, 38, 5391.
- 43.
Chen, T.; Yi, Y. Random noise suppression of seismic data based on deep convolution neural network. Acta Seismol. Sin. 2021, 43, 474–482.
- 44.
Lopez, V.; Fernandez, A.; Herrera, F. On the importance of the validation technique for classification with imbalanced datasets: Addressing covariate shift when data is skewed. Inf. Sci. 2014, 257, 1–13.
- 45.
Chen, J.; Chen, G.; Li, J.; et al. Efficient seismic data denoising via deep learning with improved mca-scunet. IEEE Trans. Geosci. Remote. Sens. 2024, 62, 5903614.
- 46.
Wang, Z.; Bovik, A.C.; Sheikh, H.R.; et al. Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 2004, 13, 600–612.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Suite 4002 Level 4, 447 Collins Street, Melbourne, Victoria 3000, Australia
General Inquiries: info@sciltp.com