In the processing of vertical seismic profile (VSP) data, coupled noise is the main interfering signal. Its complex propagation characteristics and frequency overlap make it difficult for traditional denoising methods to achieve precise suppression. Therefore, a hybrid wavelet-Transformer (HWT) framework was proposed for efficient suppression of coupled noise. Firstly, the multi-scale frequency decomposition of the signal was performed by discrete wavelet transform (DWT), and the original signal was decomposed into different frequency bands to reduce frequency aliasing. Then, the two-branch Transformer architecture was used to capture local and global features respectively, where the local branch extracted the high-frequency interference characteristics of the coupled noise, and the global branch models the time-frequency global dependence of the target signal. In addition, a multistage feature aggregation module (MFAM) was designed. By integrating the frequency domain features decomposed by DWT with the local and global features of the two-branch output, the features were deeply integrated. Based on the mask generation strategy in the time-frequency domain, the time-frequency representation was constrained, which further improved the separation quality. Experiments on synthetic data and real VSP data show that the proposed method can suppress coupling noise significantly while maintaining the integrity of target signal, which provides an efficient and robust solution for the pre-processing of VSP data.
WANG Teng-yu, CHEN Hao, ZHANG Zhen, YI Hao-ran, WEI Wei, MAHEMUTI Maierdan, CAI Han-peng.
VSP Data Coupling Noise Suppression Method Based on Hybrid Wavelet-Transformer Framework[J].
Science Technology and Engineering, 2026, 26(12): 4948-4958 DOI:10.12404/j.issn.1671-1815.2503001
局部分支通过窗口化机制对高频分量{${x}_{\mathrm{h}\mathrm{i}\mathrm{g}\mathrm{h}}^{i}$(t)}进行分片处理,提取细粒度的时间-频率特性[28-29]。具体而言,首先将高频分量按照窗口长度 L 切分为多个子序列$\left\{{x}_{\mathrm{h}\mathrm{i}\mathrm{g}\mathrm{h}}^{i}\right({t}_{k}{\left)\right\}}_{k=1}^{K}$,随后输入局部Transformer模块,可表示为
WuJie, HouQinlong, SunJiaqing, et al. Distributed acoustic sensing VSP coupling noise suppression based on local sparse optimization[J]. Journal of Geophysics, 2023, 66(12): 5123-5140.
SunJiaxin, LiJing, LiuHui, et al. Research progress on denoising methods and applications of distributed acoustic sensing data[J]. Advances in Geophysics, 2025, 40(3): 1279-1295.
OuBinglin, ZengTongsheng, LiuTiancheng, et al. Seismic data reconstruction and de-noising based on Huber-U-Net network[J]. Advances in Geophysics, 2023, 38(6): 2540-2552.
Heyue, LiuJianxin, WangXianying, et al. Application of interpolation denoising based on adaptive dictionary learning[J]. Advances in Geophysics, 2021, 36(6): 2454-2461.
[20]
ZhangY, GuB, Sun Z et al. Seismic data denoising with two-step prediction strategy based on neural network[J]. Computers & Geosciences, 2024, 187: 105595.
[21]
SteinbergA, GaeblerP, HartmannG, et al. Deep neural networks based denoising of regional seismic waveforms and impact on analysis of North Korean nuclear tests[J]. Pure and Applied Geophysics, 2024, 182: 4977-4997.
[22]
ZhuD, FuL, KazeiV, et al. Diffusion model for DAS-VSP data denoising[J]. Sensors, 2023, 23(20): 1-15.
[23]
XuY K, ZhuH, CaoS, et al. DAS coupling noise suppression based on MCA-FK[J]. Acta Geophysica, 2023, 72(4): 2465-2474.
GaoYouhu, YueJinghang, KongJun, et al. Random noise removal method for seismic detection data based on convolutional network[J]. Science Technology and Engineering, 2021, 21(1): 103-108.
[30]
MartinC T. Leveraging deep learning for seismic image denoising[J]. First Break, 2020, 38(7): 41-48.
MaJunzhuo, LiGang, SunJiaying, et al. Seismic signal denoising method based on DNDCNN[J]. Science Technology and Engineering, 2024, 24(34): 14571-14580.
[33]
LiY, LiuL. Three-dimensional seismic denoising based on deep convolutional dictionary learning[J]. Results in Applied Mathematics, 2024, 24: 100516.
YeYuhan, HeZongbin. Seismic data denoising based on efficient dictionary learning algorithm[J]. Science Technology and Engineering, 2024, 24(25): 10677-10687.
[36]
LiuQ, FuL, ZhangM. Deep-seismic-prior-based reconstruction of seismic data using convolutional neural networks[J]. Geophysics, 2019, 86(2): 131-142.
LiuXia, PanHongping, GaoXiaochun, et al. Seismic data de-noising based on wavelet thresholding[J]. Science Technology and Engineering, 2010, 10(29): 7251-7254.
[39]
VaswaniA, ShazeerN, ParmarN, et al. Attention is all you need[J]. arXiv Preprint, arXiv. 1706. 03762, 2017.
[40]
ZhuW, ZhaoC, FengS, et al. Multilevel feature alignment based on spatial attention deformable convolution for cross-scene hyperspectral image classification[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 3227251.
[41]
OonincxP J, SleemanR, EckT V. Application of the DWT in seismic data analysis[J]. Springer Netherlands, 2001, 19: 479-500.
[42]
周若水. 三维地震数据断层智能识别中的域适应方法研究[D]. 成都: 电子科技大学, 2023.
[43]
ZhouRuoshui. 3D seismic data fault intelligent identification of domain adaptation method research[D]. Chengdu: University of Electronic Science and Technology of China, 2023.
[44]
周成. 大视野下的三维地震信号断层识别与建模方法研究[D]. 成都: 电子科技大学, 2024.
[45]
ZhouCheng. Under the view of 3 D seismic signal fault identification and modeling method research[D]. Chengdu: University of Electronic Science and Technology of China, 2024.
[46]
ChengJ, ChenK, SacchiM D. Application of robust principal component analysis?[J]. Journal of Association for Computing Machinery, 2015, 58(3): 37.
[47]
HuangG, BaiM, ZhaoQ, et al. Erratic noise suppression using iterative structure-oriented space-varying median filtering with sparsity constraint[J]. Geophysical Prospecting, 2020, 69(1): 101-121.
PDF
(23310KB)
