Removal of Seabed Multiples in Seismic Reflection Data using Machine Learning

머신러닝을 이용한 탄성파 반사법 자료의 해저면 겹반사 제거

  • Nam, Ho-Soo (KT Powertel, Strategic Product Planning Team) ;
  • Lim, Bo-Sung (Korea National Oil Corporation, Domestic Business Dept., Domestic Exploration Team) ;
  • Kweon, Il-Ryong (PODO Inc.) ;
  • Kim, Ji-Soo (Chungbuk National University, Dept. of Earth and Environment Sciences)
  • 남호수 (케이티파워텔 전략상품팀) ;
  • 임보성 (한국석유공사 국내사업처 국내탐사팀) ;
  • 권일룡 (주식회사 포도) ;
  • 김지수 (충북대학교 지구환경과학과)
  • Received : 2020.07.07
  • Accepted : 2020.08.25
  • Published : 2020.08.31


Seabed multiple reflections (seabed multiples) are the main cause of misinterpretations of primary reflections in both shot gathers and stack sections. Accordingly, seabed multiples need to be suppressed throughout data processing. Conventional model-driven methods, such as prediction-error deconvolution, Radon filtering, and data-driven methods, such as the surface-related multiple elimination technique, have been used to attenuate multiple reflections. However, the vast majority of processing workflows require time-consuming steps when testing and selecting the processing parameters in addition to computational power and skilled data-processing techniques. To attenuate seabed multiples in seismic reflection data, input gathers with seabed multiples and label gathers without seabed multiples were generated via numerical modeling using the Marmousi2 velocity structure. The training data consisted of normal-moveout-corrected common midpoint gathers fed into a U-Net neural network. The well-trained model was found to effectively attenuate the seabed multiples according to the image similarity between the prediction result and the target data, and demonstrated good applicability to field data.


  1. Choi, W., Lee, G., Cho, S., Choi, B., and Pyun, S., 2020, Fault Detection for Seismic Data Interpretation Based on Machine Learning: Research Trends and Technological Introduction, Geophys. and Geophys. Explor., 23(2), 97-114 (in Korean with English abstract), doi: 10.7582/GGE.2020.23.2.097.
  2. Chollet, F., 2018, Deep Learning mit Python und Keras, Das Praxis-Handbuch vom Entwickler der Keras-Bibliothek, MITP-Verlags GmbH & Co. KG.
  3. Dahl-Jensen, T., 1989, Reflection Seismic Studies in the Baltic Shield: Special Processing Techniques and Results, Uppsala University, 125p.
  4. Deng, L., and Yu, D., 2014, Deep Learning: Methods and Applications, Found. Trends Signal Process., 7(3-4), doi:10.1561/2000000039.
  5. Hahnloser, R., Sarpeshkar, R., Mahowald, M. A., Douglas, R. J., and Seung, H. S., 2000, Digital Selection and Analogue Amplification Coexist in a Cortex-inspired Silicon Circuit, Nature, 405(6789), 947-951, doi: 10.1038/35016072.
  6. Hampson, D., 1986, Inverse Velocity Stacking for Multiple Elimination, 56th Ann. Internat. Mtg., Soc. Expl. Geophys., Expanded Abstracts, 422-424, doi: 10.1190/1.1893060.
  7. Hatton, L., Worthington, M. H., and Makin, J., 1986, Seismic Data Processing: Theory and Practice, Oxford, Blackwell Scientific Publications.
  8. Jo, J. H., and Ha, W., 2020, Case Analysis of Applications of Seismic Data Denoising Methods using Deep-Learning Techniques, Geophys. and Geophys. Explor., 23(2), 72-88 (in Korean with English abstract), doi: 10.7582/GGE.2020.23.2.072.
  9. Lee, W., 2017, A Deep Learning Analysis of the KOSPI's Directions, Journal of the Korean Data & Information Science Society, 28(2), 287-295 (in Korean with English abstract), doi: 10.7465/jkdi.2017.28.2.287.
  10. Lokshtanov, D., 1995, Multiple Suppression by Single Channel and Multichannel Deconvolution in the Tau-P Domain, 65th Ann. Internat. Mtg., Soc. Expl. Geophys., Expanded Abstracts, 1482-1485, doi: 10.1190/1.1887243.
  11. Luporini, F., Louboutin, M., Lange, M., Kukreja, N., Witte, P., Huckelheim, J., Yount, C., Kelly, P. H. J., Herrmann, F. J., and Gorman, G. J., 2020, Architecture and performance of Devito, a system for automated stencil computation, ACM Trans. Math. Softw., 46(1), 6, doi: 10.1145/3374916.
  12. Martin, G. S., 2004, The Marmousi2 model, elastic synthetic data, and an analysis of imaging and AVO in a structurally complex environment, Master's dissertation, University of Houston.
  13. Naidu, P., Santosh, Chand, S., and Saxena, U. C., 2013, Surface Related Multiple Elimination: A Case study from East Coast India, 10th Biennial International Conference & Exposition., 217p
  14. Nam, H. S., 2020, Attenuation of the Multiples in Seabed Seismic Reflection Data using Machine Learning System, Master's dissertation, Chungbuk National University, 49p.
  15. Peacock, K. L., and Treitel, S., 1969, Predictive Deconvolution: Theory and Practice, Geophysics, 34(2), 155-169, doi:10.1190/1.1440003.
  16. Ronneberger, O., Fischer, P., and Brox, T., 2015, U-Net: Convolutional Networks for Biomedical Image Segmentation, Med. Image Comput. Comput. Assist. Interv., 234-241, doi:10.1007/978-3-319-24574-4_28.
  17. Sheriff, R. E., and Geldart, L.P., 1995, Exploration Seismology, 2nd Ed., Cambridge University Press, doi: 10.1017/CBO9781139168359.
  18. Siahkoohi, A., Verschuur, D. J., and Herrmann, F. J., 2019, Surface-related multiple elimination with deep learning, 89th Ann. Internat. Mtg. Soc. Expl. Geophys., Expanded Abstracts, 4629-4634, doi: 10.1190/segam2019-3216723.1.
  19. Stewart, P. G., Jones, I. F., and Hardy, P. B., 2007, Solutions for Deep Water Imaging, GeoHorizons, 8-22.
  20. Wang, T., Wang, D., and Sun, J., 2017, Closed-loop SRME based on 3D L1-norm sparse inversion. Acta Geophys., 65, 1145-1152, doi: 10.1007/s11600-017-0098-6.