1887

Abstract

Summary

Seismic interpretation is a complex and tedious task requiring intensive time-consuming manual interpretation. In the last recent years Convolutional Neural Networks (CNNs) have been introduced to solve several interpretation challenges. Promising results have been already presented for salt, horizons or faults interpretation. The main purpose is to transform long manual interpretation tasks to a few hours of machine learning computation. Hence, geoscientists will have more time to focus on high added value tasks that will help to better understand the subsurface. In this paper, we highlight the criticality of data diversity during the training to get a robust and flexible model. Then we investigate the potential gains for the seismic interpreters in term of quality and object extraction.

Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.201901388
2019-06-03
2020-08-11
Loading full text...

Full text loading...

References

  1. VijayBadrinarayanan, AlexKendall and RobertoCipolla
    "SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation."PAMI, 2017.
    [Google Scholar]
  2. OlafRonneberger, PhilippFischer, ThomasBrox
    “U-Net: Convolutional Networks for Biomedical Image Segmentation”, Medical Image Computing and Computer-Assisted Intervention (MICCAI), Springer, LNCS, Vol.9351: 234--241, 2015.
    [Google Scholar]
  3. Milletari, F., Navab, N., Ahmadi, S.A.
    : V-Net: Fully convolutional neural networks for volumetric medical image segmentation. In: 3DV. pp. 565–571. IEEE (2016)
    [Google Scholar]
  4. Wu, X., L. Liang (Uber), Y. Shi and S.Fomel
    , 2019, FaultSeg3D: using synthetic datasets to train an end-to-end CNN for 3D fault segmentation. Geophysics
    [Google Scholar]
  5. HaibinDi
    , “Developing a seismic pattern interpretation network (SpiNet) for automated seismic interpretation”, Submitted on 19Oct2018.
  6. J.Tremblay*, A.Prakash*, D.Acuna*, M.Brophy*, V.Jampani, C.Anil, T.To, E.Cameracci, S.Boochoon and S.Birchfield
    , “Training Deep Networks with Synthetic Data: Bridging the Reality Gap by Domain Randomization”, Workshop on Autonomous Driving, CVPR-Workshops'18
    [Google Scholar]
  7. J.Tobin, R.Fong, A.Ray, J.Schneider, W.Zaremba, and P.Abbeel
    . Domain randomization for transferring deep neural networks from simulation to the real world. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017.
    [Google Scholar]
  8. C.Steger
    . Removing the Bias from Line Detection. In Computer Vision and Pattern Recognition, pages 116.
    [Google Scholar]
http://instance.metastore.ingenta.com/content/papers/10.3997/2214-4609.201901388
Loading
/content/papers/10.3997/2214-4609.201901388
Loading

Data & Media loading...

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error