Automatic Detection of Channels in Seismic Images via Deep Convolutional Neural Networks Learning

Fedor Krasnov, Alexander Butorin, Alexander Sitnikov

Abstract


Collecting relevant evidence from seismic data is one of the standard milestones in the timeline of geological modeling.

Automation of this process is difficult due to the lack of labeled dataset and computing resources.

The authors of this paper used computer vision to identify how the seismic data can be classified. In previous studies, the authors developed a method of spectral decomposition of seismic signals in order to obtain false-color images.

Spectral decomposition of seismic signals was used in this study to generate a dataset. The deep neural network was applied to solve problems of image classification. The obtained results allow to determine geological units with a test’s accuracy of 90% rendering to F1 score measure.

Based on the results obtained, improvements can be made in the organizational processes of parties engaged in the processing of seismic data.

Automated processes for identifying geological units will speed up the acquisition and at the same time increase the accuracy of information on the geological environments of the deposit for constructing structural oilfield models

Full Text:

PDF (Russian)

References


Chollet F. Xception: Deep Learning with Depthwise Separable Convolutions //arXiv preprint arXiv:1610.02357. – 2016.

Abadi M. et al. Tensorflow: Large-scale machine learning on heterogeneous distributed systems //arXiv preprint arXiv:1603.04467. – 2016.

Butorin A. V. et al. Approaches to the Analysis of Spectral Decomposition for the Purpose of Detailed Geological Interpretation //SPE Russian Petroleum Technology Conference and Exhibition. – Society of Petroleum Engineers, 2016.

LI Y. et al. Seismic multi-attribute analysis based on RGB color blending technology [J] //Journal of China Coal Society. – 2009. – Т. 11. – С. 018.

Simonyan K., Zisserman A. Very deep convolutional networks for large-scale image recognition //arXiv preprint arXiv:1409.1556. – 2014.

Szegedy C. et al. Rethinking the inception architecture for computer vision //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. – 2016. – С. 2818-2826.

He K. et al. Deep residual learning for image recognition //Proceedings of the IEEE conference on computer vision and pattern recognition. – 2016. – С. 770-778.

Krizhevsky A., Sutskever I., Hinton G. E. Imagenet classification with deep convolutional neural networks //Advances in neural information processing systems. – 2012. – С. 1097-1105.

Kingma D., Ba J. Adam: A method for stochastic optimization //arXiv preprint arXiv:1412.6980. – 2014.


Refbacks

  • There are currently no refbacks.


Abava   MSU conference 2018

ISSN: 2307-8162