1887

Abstract

Summary

Can a data scientist without domain knowledge build a machine learning algorithm to identify salt? Can a deep learning network identify salt by training purely on images without any knowledge in geology? There are the fundamental questions at the core of of the Kaggle-TGS Salt Identification Challenge. These are the fundamental questions at the core of the Kaggle-TGS Salt Identification Challenge. In this study, we present an overview of our approach towards constructing the dataset and some results from the Gulf of Mexico showing the performance of salt segmentation algorithms that help us generate useful prior models for seismic interpreters. We also show the comparison of top results from Kaggle competition for TGS Salt Identification Challenge.

Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.201901271
2019-06-03
2024-05-25
Loading full text...

Full text loading...

References

  1. Berman, M., Triki, A.R. and Blaschko, M.B
    . [2018] The Lovász-Softmax loss: A tractable surrogate for the optimization of the intersection-over-union measure in neural networks.
    [Google Scholar]
  2. Long, J., Shelhamer, E. and Darrell, T
    . [2015] Fully Convolutional Networks for Semantic Segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 3431–3440.
    [Google Scholar]
  3. Mosher, C., Keskula, E., Malloy, J., Keys, R., Zhang, H. and Jin, S
    . [2007] Iterative imaging for subsalt interpretation and model building. The Leading Edge, 26(11), 1424–1428.
    [Google Scholar]
  4. Radosavovic, I., Dollár, P., Girshick, R., Gkioxari, G. and He, K
    . [2018] Data distillation: Towards omnisupervised learning. In: 2018IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE, 4119–4128.
    [Google Scholar]
  5. Reasnor, M.D
    . [2007] Salt interpretation practices for depth imaging in the Gulf of Mexico. The Leading Edge, 26(11), 1438–1441.
    [Google Scholar]
  6. Ronneberger, O., Fischer, P. and Brox, T
    . [2015] U-Net: Convolutional networks for biomedical image segmentation. In: International Conference on Medical image computing and computer-assisted intervention. Springer, 234–241.
    [Google Scholar]
/content/papers/10.3997/2214-4609.201901271
Loading
/content/papers/10.3997/2214-4609.201901271
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