Study of Small Dataset of Images for Seismic Fault Imaging with a Supervised Convolutional Neural Network

Authors

  • Munezero Ntibahanana Centre for Research in Geophysics (C.R.G.), Kinshasa, RD Congo
  • Tondozi Keto Centre for Research in Geophysics (C.R.G.), Kinshasa, RD Congo
  • Moise Luemba School of Geosciences, China University of Petroleum (East China), Qingdao, China
  • Raïs Seki Lenzo Centre for Research in Geophysics (C.R.G.), Kinshasa, RD Congo
  • Yannick Mananga Thamba Centre for Research in Geophysics (C.R.G.), Kinshasa, RD Congo
  • Kevin Lumpungu Lutumba Centre for Research in Geophysics (C.R.G.), Kinshasa, RD Congo
  • Yang’tshi Ndong Olola Faculty of Oil, Gas and Renewable Energies, University of Kinshasa, DR Congo
  • Emmanuel Lokilo Lofiko Faculty of Sciences, University of Gbadolite, Nord-Ubangi, DR Congo
  • Alidor Kazadi Mutambayi Centre for Research in Geophysics (C.R.G.), Kinshasa, RD Congo

DOI:

https://doi.org/10.31695/IJASRE.2022.8.7.2

Keywords:

Binary Image Segmentation, Convolutional Neural Networks, Deep Learning, Seismic Fault Interpretation

Abstract

Recognizing faults in seismic images is crucial for structural modeling, prospect delineation, reservoir characterization, and well placement. Basically, faults have the appearance of lateral reflection discontinuities in seismic images and are interpreted using seismic attributes that measure those discontinuities such as coherence, and curvature. However, methods based on seismic attributes are often more challenging, time-consuming, and may suffer from noises and sensitivity of stratigraphic features, which also tie in reflection discontinuities. Therefore, we propose a solution for delineating faults from 3D seismic images using a supervised fully convolutional neural network (CNN). This approach uses a pixel-by-pixel prediction in 3D seismic images to classify whether a given pixel is a fault or a non-fault. The trained model learned to bank on rich and proper features that are important for the recognition of faults and achieved 97% of accuracy. To test the effectiveness of our model, we used new 3D seismic images, and the results displayed clean and accurate recognition of faults within only milliseconds, saving time and optimizing the accuracy. In this paper, we showed that by using only a few 3D seismic images from a given seismic volume to train the model, not only do we handle one of the difficulties encountered by researchers to obtain a sufficient amount of data needed to train common CNN models but also, interpreters can successfully predict faults in any other seismic image from the same volume.

References

W. Xinming, L. Liang, Y. Shi and S. Fomel, “FaultSeg3D: using synthetic datasets to train an end-to-end convolutional neural network for 3D seismic fault segmentation,” Geophysics, vol. 84, no. 3, pp. IM35-IM45, 2019.

S. Li, Y. Changchun, S. Hui and Z. Hao, “Seismic fault detection using an encoder-decoder convolutional neural network with a small training set,” Journal of Geophysics and Engineering, pp. 1-15, 2019.

Bahorich and Farmer, “3-D seismic discontinuity for faults and stratigraphic features: The coherence cube,” The Leading Edge, vol. 14, pp. 1053-1058, 1995.

A. Roberts, “Curvature attributes and their application to 3D interpreted horizons,” First Break, vol. 19, pp. 85-100, 2001.

T. Boe and A. a. Daber, “Seismic features and the human eye: RGB blending of azimuthal curvatures for enhancement of fault and fracture interpretation,” in SEG Technical Program, 2010.

D. Gao, “Integrating 3D seismic curvature and curvature gradient attributes for fracture detection: Methodologies and Interpretational implications,” Geophysics, vol. 78, no. 2, pp. 021-038, 2013.

K. J. Marfurt, K. R. L., F. S. L. and B. and M. S., “3-D seismic attributes using a semblance-based coherency algorithm,” Geophysics, vol. 63, pp. 1150-1165, 1998.

V. P. Bemmel and R. Pepper, “Seismic signal processing method and apparatus for generating a cube of variance values”. Washington DC. Patent 6,151,555, 2000.

T. Randen, S. Pedersen and L. Sonneland, “Automatic extraction of fault surfaces from three-dimensional seismic data,” in 71st Annual International Meeting, Expanded Abstracts, 2001.

A. Aqrawi and T. Boe, “Improved fault segmentation using a dip guided and modified 3D Sobel filter,” in 81st Annual International Meeting, 2011.

D. Hale, “Methods to compute fault images, extract fault surfaces, and estimate fault throws from 3D seismic images,” Geophysics, vol. 78, pp. 33-43, 2013.

S. I. Pedersen, T. Randen, L. Sonneland and Steen, “Automatic fault extraction using artificial ants,” in 72nd Annual International Meeting, 2002.

N. Al Bin Hassan and K. Marfurt, “Fault detection using Hough transform,” in 73rd Annual International Meeting, 2003.

A. Barnes, “A filter to improve seismic discontinuity data for fault interpretation,” Geophysics, vol. 71, no. 3, pp. 1-4, 2006.

Z. Wang, Z. Long, G. AlRegib, A. Asjad, and M. Deriche, “Automatic fault tracking across seismic volumes via tracking vectors,” in IEEE International Conference on Image Processing, 2014.

D. Haibin, W. Zhen and A. Ghassan, “Real-time seismic image interpretation via deconvolutional neural network,” in SEG International Exposition and 88th Annual Meeting., 2018.

Z. Zheng, P. Kavousi and H. Di, “Multi-attributes and neural network-based fault detection in 3D seismic interpretation,” Advanced Materials Research, pp. 1497-1502, 2014.

G. Huang, Z. Liu, L. VanDerMaaten and K. Weinberger, “Densely connected convolutional networks,” in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2017.

H. Di, M. Shafiq, and G. AlRegib, “Seismic fault detection based on multi-attribute support vector machine analysis,” in 87th Annual International Meeting, 2017.

Y. Alaudah and G. AlRegib, “A weakly-supervised approach to seismic structure labeling,” in 87th Annual International Meeting, 2017.

Z. Tao, Q. Jie, L. Tengfei, L. Fangyu and M. Kurt, “Semisupervised multiattribute seismic facies analysis,” Interpretation, vol. 4, no. 1, pp. SB91-SB106, 2016.

R. Rocky and C. ChingWen, “Interpretation of DHI characteristics with machine learning,” Geophysical Insights, vol. 35, no. 5, pp. 55-63, 2017.

B. Jason, Deep Learning with Python, Develop Deep Learning Models on Theano and TensorFlow using Keras, v1.18 ed., Melbourne: Machine Learning Mastery, 2019.

W. Xinming, Y. Shi, F. Sergey, L. Luming and Uber, “Convolutional neural networks for fault interpretation in seismic images,” in SEG International Exposition and 88th Annual Meeting, 2018.

S. Hyunseok, K. V. V. Masoud Badiei, H. Charles, R. Honyi, X. Ruoxiu, J. Xiao and X. Lei, “Machine Learning Techniques for Biomedical Image Segmentation: An Overview of Technical Aspects and Introduction to State‐of‐Art Applications,” Medical Physics, 2020.

S. Himanshu, Practical Machine Learning and Image Processing for Facial recognition, Object Detection and Pattern Recognition Using Python, Allahabad, Uttar Pradesh, India: Apress, 2019.

A. Antoly, D. Andrew and F. Di, “Application of deep learning for subsurface faults detection with seismic data,” CS230, Stansford University, CA, 2019.

V. Nair and G. Hinton, “Rectified linear units improve restricted Boltzmann machines,” in Proceedings of the 27th International Conference on International Conference on Machine Learning, Haifa, Israel, 2010.

O. Ronneberger, P. Fischer and T. Brox, “U-Net: convolutional networks for biomedical image segmentation,” Medical Image Computing and Computer-Assisted Intervention-MICCAI, pp. 234-241, 2015.

D. Vincent and V. Francesco, “A guide to convolution arithmetic for deep learning,” 12 January 2018. [Online]. Available: https://arxiv.org/abs/1603.07285. [Accessed 19 September 2020].

Wikipedia, “Logistic function,” 2020. [Online]. Available: https://en.wikipedia.org/w/index.php?title=Logistic_function&oldid=977583860. [Accessed 9 September 2020].

R. Adrian, Keras ImageDataGenerator and Data Augmentation, pyimagesearch, 2019.

Keras.a, “Accuracy metrics,” 2020. [Online]. Available: https://keras.io/api/metrics/accuracy_metrics/#accuracy-class. [Accessed 13 October 2020].

Keras.b, “Image segmentation metrics,” 2020. [Online]. Available: https://keras.io/api/metrics/. [Accessed 16 September 2020].

Keras.c, “Classification metrics,” 2020. [Online]. Available: https://keras.io/api/metrics/classification_metrics/classification-metrics-based-on-truefalse-positives-amp-negatives. [Accessed 16 September 2020].

S. Manomar, Mastering Machine Learning with Python in Six Steps: a practical implementation guide to predictive data analytics using python, Bangalore, Karnataka, India: Apress, 2017.

R. Sebastian and M. Vahid, Python Machine Learning, Second Edition ed., MUMBAI, BIRMINGHAM: Packt Publishing, 2017.

Downloads

How to Cite

Munezero Ntibahanana, Tondozi Keto, Moise Luemba, Raïs Seki Lenzo, Yannick Mananga Thamba, Kevin Lumpungu Lutumba, Yang’tshi Ndong Olola, Emmanuel Lokilo Lofiko, & Alidor Kazadi Mutambayi. (2022). Study of Small Dataset of Images for Seismic Fault Imaging with a Supervised Convolutional Neural Network . International Journal of Advances in Scientific Research and Engineering (IJASRE), ISSN:2454-8006, DOI: 10.31695/IJASRE, 8(7), 11–23. https://doi.org/10.31695/IJASRE.2022.8.7.2

Issue

Vol. 8 No. 7: IJASRE July-2022

Section

Articles

License

Copyright (c) 2022 Munezero Ntibahanana, Tondozi Keto, Moise Luemba, Raïs Seki Lenzo, Yannick Mananga Thamba, Kevin Lumpungu Lutumba, Yang’tshi Ndong Olola, Emmanuel Lokilo Lofiko, Alidor Kazadi Mutambayi

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.