Random noise in seismic data impairs the quality of the data,thus affecting the accuracy of subsequent processing and interpretation.Conventional denoising methods,constrained by prior conditions,exhibit low efficiency.Neural networks possess a strong feature extraction ability,which can make up for these shortcomings.However,the limitations of convolution kernels in conventional neural networks may lead to the loss of global information.Hence,this study introduced a retractable attention mechanism to the convolutional neural network (CNN).This mechanism presents both dense and sparse self-attention modules in the CNN.The alternate use of the two self-attention modules can significantly enhance the performance of the CNN and expand the receptive field.The shallow and deep features of seismic data were extracted using the convolutional layer and self-attention modules.Combined with CNN's local modeling ability and Transformer's global modeling ability,they contributed to enhancing CNN's global interaction and ability to reduce noise and deal with details.As indicated by the experimental results of synthetic and field data,the method used in this study can more effectively suppress noise and retain effective information of seismic data compared to Unet and DnCNN,significantly improving the signal-to-noise ratio and thus assisting in the processing and interpretation of seismic data.
张敏, 许一卓, 易继东. 基于可伸缩型注意力机制的神经网络地震数据去噪方法[J]. 物探与化探, 2024, 48(4): 1065-1075.
ZHANG Min, XU Yi-Zhuo, YI Ji-Dong. A method for seismic data denoising based on the neural network with a retractable attention mechanism. Geophysical and Geochemical Exploration, 2024, 48(4): 1065-1075.
Li J, Wu X, Hu Z. Deep learning for simultaneous seismic image super-resolution and denoising[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60(1):1-11.
Cao J J, Yang Z Q, Yang Y, et al. An adaptive seismic random noise elimination method based on Curvelet transform[J]. Geophysical Prospecting for Petroleum, 2018, 57(1):72-78.
doi: 10.3969/j.issn.1000-1441.2018.01.010
Sun M M, Li Z C, Qu Y M, et al. A seismic denoising method based on Curvelet transform with constraint in OVT domain[J]. Geophysical Prospecting for Petroleum, 2019, 58(2):208-218.
Guo L Y, Liu C, Liu Y, et al. Seismic random noise attenuation based on streaming prediction filter in the f-x domain[J]. Chinese Journal of Geophysics, 2020, 63(1):329-338.
[9]
Naghizadeh M, Sacchi M. Multicomponent f-x seismic random noise attenuation via vector autoregressive operators[J]. Geophysics, 2012, 77(2):V91-V99.
[10]
Witten B, Shragge J. Extended wave-equation imaging conditions for passive seismic data[J]. Geophysics, 2015, 80(6):WC61-WC72.
Zhang L, Han L G, Fang J W, et al. Seismic data denoising via double sparsity dictionary and fast iterative shrinkage-thresholding algorithm[J]. Chinese Journal of Geophysics, 2019, 62(7):2071-2683.
Tang J, Meng T, Zhang W Z, et al. Suppression of seismic random noise with overcomplete dictionary signal sparse representation algorithm based on deep learning[J]. Oil Geophysical Prospecting, 2020, 55(6):1202-1209,1160.
Song H, Gao Y, Chen W, et al. Seismic noise suppression based on convolutional denoising autoencoder[J]. Oil Geophysical Prospecting, 2020, 55(6):1210-1219,1160-1161.
[14]
Wang Y Q, Lu W K, Liu J L, et al. Random seismic noise attenuation bas ed on data augmentation and CNN[J]. Chinese Journal of Geophysics, 2019, 62(1):421-433.
Gao H T, Sun N N, Sun K Y, et al. Comparative analysis of DnCNN and U-Net on suppression of seismic random doise[J]. The Journal of China Universities of Posts and Telecommunications, 2021, 36(6):2441-2453.
[16]
Wang F, Chen S. Residual learning of deep convolutional neural network for seismic random noise attenuation[J]. IEEE Geoscience and Remote Sensing Letters, 2019, 16(8):1314-1318.
Chen T, Yi Y Y. Random noise suppression of seismic data based on deep convolutional neural network[J]. Acta Seismologica Sinica, 2021, 43(4):474-482,533.
[18]
Si X, Yuan Y, Si T, et al. Attenuation of Random noise using denoising convolutional neural networks[J]. Interpretation, 2019, 7(3):SE269-SE280.
Wang Y Q, Lu W K, Liu J L, et al. Random seismic noise attention based on data augmentation and CNN[J]. Chinese Journal of Geophysics, 2019, 62(1):421-433.
Luo R Z, Li Y Y. Random seismic noise attenuation based on RUnet convolutional neural network[J]. Geophysical Prospecting for Petroleum, 2020, 59(1):51-59.
doi: 10.3969/j.issn.1000-1441.2020.01.006
[21]
Yang L, Chen W, Wang H, et al. Deep learning seismic random noise attenuation via improved residual convolutional neural network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(9):7968-7981.
Yang C Q, Zhou Y T, He H, et al. Global context and attention-based deep convolutional neural network for seismic data denoising[J]. Geophysical Prospecting for Petroleum, 2021, 60(5):751-762,855.
doi: 10.3969/j.issn.1000-1441.2021.05.006
Zhou W H, Shi M, Zhu D M, et al. Seismic data super-resolution method based on residual attention network[J]. Computer Science, 2021, 48(8):24-31.
[24]
Liu Z, Lin Y T, Cao Y, et al. Swin transformer:Hierarchical vision transformer using shifted windows[C]// Montreal:2021 IEEE/CVF International Conference on Computer Vision(ICCV), 2021:9992-10002.
[25]
Cao H, Wang Y Y, Chen J, et al. Swin-Unet:Unet-like pure transformer for medical image segmentation[J]. Image and Video Processing, 2021.
[26]
Fan C M, Liu T J, Liu K H. SUNet:Swin transformer UNet for image denoising[C]// Austin:2022 IEEE International Symposium on Circuits and Systems(ISCAS), 2022:2333-2337.
[27]
Zhang K, Li Y, Liang J, et al. Practical blind image denoising via swin-conv-UNet and data synthesis[J]. Machine Inteligence Research, 2023, 20(6):822-836.
