基于增强残差特征蒸馏的地震图像超分辨率重建

doi:10.3799/dqkx.2026.076

基于增强残差特征蒸馏的地震图像超分辨率重建

doi: 10.3799/dqkx.2026.076

吴迪^1,2,3,4,5,
刘刚^{1,2,3,4,5,, ,},
陈麒玉^1,2,3,4,5,
崔哲思^1,2,3,4,5,
方洪峰^1,2,3,4,5,
张策^1,2,3,4,5

1. 国家地理信息系统工程技术研究中心, 中国地质大学(武汉)计算机学院, 湖北武汉 430074;
2. 智能地学信息处理湖北省重点实验室, 湖北武汉 430074;
3. 自然资源信息管理与数字孪生工程软件教育部工程研究中心, 湖北武汉 430074;
4. 贵州省战略矿产智慧勘查重点实验室, 贵州贵阳, 550081;
5. 自然资源部资源定量评价与信息工程重点实验室, 湖北武汉 430074

基金项目:

国家自然科学基金项目(Nos.42372345，42172333)；湖北省自然科学基金（No.2025AFB179）；贵州省科技重大专项（No.黔科合重大[2025]016）。

详细信息

作者简介:
吴迪(1999-)，男，硕士研究生，研究方向为智能地学信息处理、三维地质建模，ORCID：0009-0006-5062-0461，E-mail： wu.di666@qq.com

通讯作者:
刘刚，ORCID： 0000-0002-9651-4473，E-mail： liugang@cug.edu.cn

中图分类号: P618
计量
- 文章访问数: 25
- HTML全文浏览量: 0
- PDF下载量: 1
- 被引次数: 0
出版历程
- 收稿日期: 2025-12-23
- 网络出版日期: 2026-05-13

Seismic Image Super-Resolution Reconstruction Based on Enhanced Residual Feature Distillation

Wu Di^1,2,3,4,5,
Liu Gang^{1,2,3,4,5,
, ,},
Chen Qiyu^1,2,3,4,5,
Cui Zhesi^1,2,3,4,5,
Fang Hongfeng^1,2,3,4,5,
Zhang Ce^1,2,3,4,5

1. National Engineering Research Center for Geographic Information System, School of Computer Science, China University of Geosciences, Wuhan 430074, China;
2. Hubei Key Laboratory of Intelligent Geo-Information Processing, Wuhan 430074, China;
3. Engineering Research Center of Natural Resources Information Management and Digital Twin Engineering Software, Ministry of Education, Wuhan 430074, China;
4. Guizhou Key Laboratory of Intelligent Exploration of Strategic Mineral Resources, Guiyang 550081, China;
5. Key Laboratory of Resource Quantitative Assessment and Information Engineering, Ministry of Natural Resources, Wuhan 430074, China

摘要

摘要: 地震图像超分辨率重建对提升地震资料解释与地质建模精度具有重要作用，但现有方法模型规模较大、计算开销高，在强噪声和复杂地质结构条件下细节恢复能力有限。针对上述问题，本文提出了基于增强型残差特征蒸馏生成对抗网络（ERFDN-GAN）的地震图像超分辨率重建方法。该方法以增强型残差特征蒸馏模块（ERFDB）为基本单元，通过特征蒸馏机制降低模型复杂度，并引入空间注意力和通道注意力以增强特征表达能力。实验表明，该方法在PSNR和SSIM等指标上整体优于目前代表性方法，并能在复杂场景下保持较低复杂度和稳定的重建性能。所提出的ERFDN-GAN方法兼顾效率与精度，为地震资料解释、地质建模以及多源地学数据采集和挖掘的智能化处理提供了技术支撑。
- 地震图像超分辨率重建 /
- 生成对抗网络 /
- 残差特征蒸馏 /
- 注意力机制
Abstract: Seismic image super-resolution reconstruction is of great significance for improving the accuracy of seismic interpretation and geological modeling. However, existing methods often suffer from large model sizes and high computational costs, and exhibit limited detail recovery capability under strong noise and complex geological structures. To address these issues, this paper proposes an enhanced residual feature distillation generative adversarial network (ERFDN-GAN) for seismic image super-resolution reconstruction. The network is built upon enhanced residual feature distillation blocks (ERFDBs), where a feature distillation mechanism is employed to reduce model complexity. Spatial attention and channel attention mechanisms are further introduced to enhance feature representation capability. Experimental results demonstrate that the proposed method achieves overall superior performance compared with representative methods in terms of PSNR and SSIM, while maintaining low model complexity and stable reconstruction performance under complex scenarios. The proposed ERFDN-GAN method effectively balances computational efficiency and reconstruction quality, providing technical support for seismic interpretation, geological modeling, and the intelligent acquisition and mining of multi-source geoscience data.
- seismic image super-resolution reconstruction /
- generative adversarial network /
- residual feature distillation /
- attention mechanism

HTML全文

参考文献(41)

Bashir, S.M.A., Wang, Y., Khan, M., Niu, Y., 2021. A Comprehensive Review of Deep Learning-Based Single Image Super-Resolution. PeerJ Computer Science, 7: e621. doi: 10.7717/peerj-cs.621
Cheng, Z.H., Yuan, Y.J., Liu, S., Fan, C.X., Li, Z.H., Li, Y., 2023. Application of Broadband Wavelet Deconvolution in Seismic Data Processing of Tight Reservoirs. Oil Geophysical Prospecting, 62(1): 119-129 doi: 10.3969/j.issn.1000-1441.2023.01.010 (in Chinese with English abstract).
Dong, X., Lu, S., Lin, J., Zhang, S., Ren, K., Cheng, M., 2024a. Can Deep Learning Compensate for Sparse Shots in the Imaging Domain? A Potential Alternative for Reducing the Acquisition Cost of Seismic Data. Geophysics, 89(2): V119-V137. doi: 10.1190/geo2022-0711.1
Dong, X., Wei, C., Zhong, T., Cheng, M., Dong, S., Li, F., 2024b. Seismic Data Reconstruction Based on a Multicascade Self-Guided Network. Geophysics, 89(3): V179-V195. doi: 10.1190/geo2022-0712.1
Duan, J., Zhang, G., Li, X., Zhang, Y., Li, L., Liu, W., Ran, S., Cao, C., Tang, W., Yang, F., Luo, Y., 2024. Seismic Facies-Guided Trace-by-Trace High-Precision Strong and Weak Reflection Separation. IEEE Transactions on Geoscience and Remote Sensing, 62: 1-17. doi: 10.1109/TGRS.2024.3386749
Elgendy, N.H., Reda, M., Elmashaly, M.M., Raef, A., Al-Hashim, M.H., Barakat, M.Kh., 2025. Three-Dimensional Reservoir Modeling of the Pliocene Reservoir Based on Seismic Data Advances for New Prospect Assessment. Marine Georesources & Geotechnology, 43(7): 1281-1298. doi: 10.1080/1064119X.2024.2400689
Hamida, A., Alfarraj, M., Al-Shuhail, A.A., Zummo, S.A., 2023. Facies-Guided Seismic Image Super-Resolution. IEEE Transactions on Geoscience and Remote Sensing, 61: 1-13. doi: 10.1109/TGRS.2023.3289151
Jiang, C., Lü, Z. Y., Fang, L. H., 2024. Earthquake Detection Model Trained on Velocity and Acceleration Records and Its Application in Xinfengjiang Reservoir. Earth Science, 49(2): 469-479. doi: 10.3799/dqkx.2023.186 (in Chinese with English abstract).
Kong, L., Wang, Y., Chang, D., Zhao, Y., 2022. Contour Enhanced Image Super-Resolution. Journal of Visual Communication and Image Representation, 89: 103659. doi: 10.1016/j.jvcir.2022.103659
Li, F., Sun, W.Y., Wang, R., Li, J.P., 2025. Seismic Rock Physics Modeling and Key Reservoir Parameter Prediction of Ultra-Shallow Loose Sandstone Reservoirs in Ultra-Deepwater Areas. Natural Gas Industry, 45(6): 112-120. (in Chinese with English abstract)
Li, J., Fang, F., Li, J., Mei, K., Zhang, G., 2021. MDCN: Multi-Scale Dense Cross Network for Image Super-Resolution. IEEE Transactions on Circuits and Systems for Video Technology, 31(7): 2547-2561. doi: 10.1109/TCSVT.2020.3027732
Li, J., Wu, X., Hu, Z., 2022a. Deep Learning for Simultaneous Seismic Image Super-Resolution and Denoising. IEEE Transactions on Geoscience and Remote Sensing, 60: 1-11. doi: 10.1109/TGRS.2021.3057857
Li, X.G., Wu, X., 2020. New Advances of Artificial Intelligence in Seismic Data Processing and Interpretation: Insights from SEG Annual Meeting. World Petroleum Industry, 27(4): 27-35. (in Chinese with English abstract).
Li, Y., Zhang, G., Du, H., Li, L., Qian, F., Duan, J., Zhan, Y., 2022b. Multiscale Adaptive Side Window Filtering and Its Application on Seismic Data. IEEE Geoscience and Remote Sensing Letters, 19: 1-5. doi: 10.1109/LGRS.2022.3147837
Lin, L., Zhong, Z., Cai, C., Li, C., Zhang, H., 2024. SeisGAN: Improving Seismic Image Resolution and Reducing Random Noise Using a Generative Adversarial Network. Mathematical Geosciences, 56(4): 723-749. doi: 10.1007/s11004-023-10103-8
Liu, J., Cao, J., Zhao, L., You, J., Li, H., 2024. Super-Resolution Reconstruction of Seismic Images Based on Deep Residual Channel Attention Mechanism. IEEE Access, 12: 149032-149044. doi: 10.1109/ACCESS.2024.3477984
Liu, Y.M., Wei, H.J., Yuan, S., An, Z.W., 2022. Seismic Data Denoising Method Based on Convolutional Neural Networks. Journal of Jilin University (Information Science Edition), 40(2): 231-239. (in Chinese with English abstract).
Min, F., Wang, L., Pan, S., Song, G., 2023. D2UNet: Dual Decoder U-Net for Seismic Image Super-Resolution Reconstruction. IEEE Transactions on Geoscience and Remote Sensing, 61: 1-13. doi: 10.1109/TGRS.2023.3264459
Mousavi, S.M., Beroza, G.C., 2022. Deep-Learning Seismology. Science, 377(6607): eabm4470. doi: 10.1126/science.abm4470
Mousavi, S.M., Beroza, G.C., 2023. Machine Learning in Earthquake Seismology. Annual Review of Earth and Planetary Sciences, 51(1): 105-129. doi: 10.1146/annurev-earth-071822-100323
Ni, W.J., Liu, S.Y., Wang, L.P., Han, B.K., Sheng, S., 2023. Wavelet Shaping Deconvolution Method Based on Deep Learning. Oil Geophysical Prospecting, 58(6): 1313-1321. doi: 10.13810/j.cnki.issn.1000-7210.2023.06.002 (in Chinese with English abstract)
Piao, J., Xiong, J., Zheng, D. C., Wang, L. T., Yun, N. D., Pei, W. L., Zhou, Y. J., Zhou, S. Y., 2025. Advances in Intelligent Processing of Seismic Data: Focal Mechanisms, Fault Zone Head Waves, and Dynamic Triggering Detection and Analysis. Earth Science, 51(1): 56-73. doi: 10.3799/dqkx.2025.252 (in Chinese with English abstract)
21-01447-0
Wang, F., Yin, D., Song, R., 2023. Image Super-Resolution Using Only Low-Resolution Images. The Visual Computer, 39(10): 5069-5084. doi: 10.1007/s00371-022-02646-4
Wu, H.B., Zhang, P.S., Xu, S.A., Huang, Y.P., Dong, S.H., Cheng, Y., Li, D.H., Zhang, S., Zhang, M., 2025. Development Status and Trends of Seismic Prediction and Evaluation Techniques for Coalbed Methane Reservoirs. Coal Geology & Exploration, 53(6): 233-245. doi: 10.12363/issn.1001-1986.25.04.0301 (in Chinese with English abstract).
Wu, X., Geng, Z., Shi, Y., Pham, N., Fomel, S., Caumon, G., 2020. Building Realistic Structure Models to Train Convolutional Neural Networks for Seismic Structural Interpretation. Geophysics, 85(4): WA27-WA39. doi: 10.1190/geo2019-0375.1
Wu, X., Liang, L., Shi, Y., Fomel, S., 2019. FaultSeg3D: Using Synthetic Data Sets to Train an End-to-End Convolutional Neural Network for 3D Seismic Fault Segmentation. Geophysics, 84(3): IM35-IM45. doi: 10.1190/geo2018-0646.1
Yuan, Y., Li, Y., Zhou, S., 2020. Multichannel Statistical Broadband Wavelet Deconvolution for Improving Resolution of Seismic Signals. IEEE Transactions on Geoscience and Remote Sensing, 59(2): 1772-1783. doi: 10.1109/TGRS.2020.2997977
Zeng, D., Xu, Q., Pan, S., Song, G., Min, F., 2023. Seismic Image Super-Resolution Reconstruction Through Deep Feature Mining Network. Applied Intelligence, 53(19): 21875-21890. doi: 10.1007/s10489-023-04660-y
Zhang, J., Yin, X., Gu, Y., Zhou, Y., 2024. Sandstone Reservoir Rock Physics Modeling and Time-Lapse Seismic Analysis. Journal of Applied Geophysics, 222: 105318. doi: 10.1016/j.jappgeo.2024.105318
Zhao, Y.S., Yang, P., Xu, H.Q., Huang, X.P., Nie, R., Yang, M.Q., 2023. Seismic High-Resolution Processing Method Based on Adjustable Gabor Wavelet Transform. Oil Geophysical Prospecting, 58(2): 345-350. doi: 10.13810/j.cnki.issn.1000-7210.2023.02.011 (in Chinese with English abstract)
Zhong, T., Yang, F., Dong, X., Dong, S., Luo, Y., 2024. SHBGAN: Hybrid Bilateral Attention GAN for Seismic Image Super-Resolution Reconstruction. IEEE Transactions on Geoscience and Remote Sensing, 62: 1-12. doi: 10.1109/TGRS.2024.3492142
程朝辉, 苑益军, 刘晟, 范承祥, 李子卉, 李影. 2023. 宽带子波反褶积方法在致密储层地震资料处理中的应用. 石油物探, 62(1): 119-129
蒋策, 吕作勇, 房立华, 2024. 融合处理速度和加速度记录的地震检测模型及其在新丰江水库的应用. 地球科学, 49(2): 469-479.
李芳, 孙万元, 汪锐, 李进鹏. 2025. 超深水超浅层疏松砂岩储层地震岩石物理建模及储层关键参数预测. 天然气工业, 45(06): 112-120
李晓光, 吴潇. 2020. 从SEG年会看人工智能在地震数据处理与解释中的新进展. 世界石油工业, 27(4): 27-35
刘玉敏, 魏海军, 袁硕, 安志伟. 2022. 基于卷积神经网络的地震数据去噪方法. 吉林大学学报(信息科学版), 40(2): 231-239
倪文军, 刘少勇, 王丽萍, 韩冰凯, 盛燊. 2023. 基于深度学习的子波整形反褶积方法. 石油地球物理勘探, 58(6): 1313-1321
朴健, 熊健, 郑定昌, 汪龙潭, 运乃丹, 裴玮来, 周一剑, 周仕勇, 2025. 地震资料智能化处理进展：震源机制、断层首波，以及动态触发的识别与分析. 地球科学，51(1): 56-73.
吴海波, 张平松, 许时昂, 黄亚平, 董守华, 程彦, 李东会, 张生, 张敏. 2025. 煤层气储层地震预测与评价技术发展现状及趋势. 煤田地质与勘探, 53(06): 233-245
赵桠松, 杨平, 许辉群, 黄鑫鹏, 聂荣, 杨梦琼. 2023. 基于可调因子Gabor小波变换的地震高分辨处理方法. 石油地球物理勘探, 58(2): 345-350

施引文献

资源附件(0)

访问统计

点击查看大图

计量

文章访问数: 25
HTML全文浏览量: 0
PDF下载量: 1
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

留言板

基于增强残差特征蒸馏的地震图像超分辨率重建

doi: 10.3799/dqkx.2026.076

作者简介:
吴迪(1999-)，男，硕士研究生，研究方向为智能地学信息处理、三维地质建模，ORCID：0009-0006-5062-0461，E-mail： wu.di666@qq.com

通讯作者:
刘刚，ORCID： 0000-0002-9651-4473，E-mail： liugang@cug.edu.cn

计量

Seismic Image Super-Resolution Reconstruction Based on Enhanced Residual Feature Distillation

计量

目录

留言板

基于增强残差特征蒸馏的地震图像超分辨率重建

doi: 10.3799/dqkx.2026.076

作者简介: 吴迪(1999-)，男，硕士研究生，研究方向为智能地学信息处理、三维地质建模，ORCID：0009-0006-5062-0461，E-mail： wu.di666@qq.com

通讯作者: 刘刚，ORCID： 0000-0002-9651-4473，E-mail： liugang@cug.edu.cn

计量

出版历程

Seismic Image Super-Resolution Reconstruction Based on Enhanced Residual Feature Distillation

计量

出版历程

目录

作者简介:
吴迪(1999-)，男，硕士研究生，研究方向为智能地学信息处理、三维地质建模，ORCID：0009-0006-5062-0461，E-mail： wu.di666@qq.com

通讯作者:
刘刚，ORCID： 0000-0002-9651-4473，E-mail： liugang@cug.edu.cn