탄성파 층서 구분을 위한 합성곱 신경망 기법 비교 연구
Comparison of Convolutional Neural Networks for Dividing Seismic Sequences
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
머신 러닝 기술은 탄성파탐사 분야로 그 적용 범위를 확장하고 있다. 탄성파 해석에서 중요한 탄성파 층서 구분에 머신 러닝의 적용 가능성을 알아보았다. 이미지 분야에 탁월한 결과를 보여온 합성곱 신경망 기법 중 4가지 모델을 네덜란드 F3 block에 적용시켰다. 4가지 모델은 ResNet34 모델, 인코더-디코더 형태를 가지는 U-Net, Residual U-Net, FD U-Net이다. 예측된 이미지의 정성적 분석 수행 후 정량적 분석을 위해 pixel accuracy, mean class accuracy, mean intersection over union, frequency weighted IU의 수식을 활용하였다. 본 연구의 분석 결과 ResNet34의 정확도 결과가 가장 낮았고, 인코더-디코더 형태를 가지는 모델들이 높은 정확도를 보여주었다. 그리고 계산에 필요한 파라미터수와 학습시간을 고려 할 때 U-Net이 가장 효율적임을 확인할 수 있었다.
Abstract
Application of the machine learning technique is expanding to the field of seismic exploration. For the purpose of feasibility assessment, we applied the machine learning technique to seismic sequence identification, which is important in seismic interpretation. From among the convolutional neural network techniques used in image analysis, we applied four models to seismic data obtained in the F3 block, offshore Netherlands, which have yielded remarkable results. One of the four models was ResNet34. The others were encoder– decoder models: U-Net, Residual U-Net, and FD U-Net. We first performed a qualitative analysis of the predicted images and then conducted quantitative analysis using pixel accuracy, mean class accuracy, mean intersection over union, and frequency weighted IU equations. The numerical results showed that ResNet34 had the lowest accuracy and that the encoder–decoder type models had higher accuracy. Considering the number of parameters required for calculation and the learning time, we confirmed that U-Net is the most efficient model.
참고문헌 (39)
- "Facies_classification_benchmark" (2019)[URL]
- Alaudah, Y., "A machine-learning benchmark for facies classification", Interpretation, 7(3), pp.SE175~SE187 (2019)
- Bengio, Y., "Learning long-term dependencies with gradient descent is difficult", IEEE Transactions on Neural Networks, 5(2), pp.157~166 (1994)
- Chaki, S., "Reservoir characterization: A machine learning approach", MS, pp.2015
- Chevitarese, D., "Seismic facies segmentation using deep learning", AAPG Annual Convention and Exhibition (2018)
- Cho, Y., "Semi-auto horizon tracking guided by strata histograms generated with transdimensional Markov-chain Monte Carlo", Geophysical Prospecting, 68(5), pp.1456~1475 (2020)
- 최우창, "기계 학습 기반 탄성파 자료 단층 해석: 연구동향 및 기술", 지구물리와 물리탐사, 23(2), pp.97~114 (2020), 한국지구물리.물리탐사학회
- dGB Earth Sciences, "Seismic interpretation software &services", ~ (2018)
- Dramsch, J. S., "SEG Technical Program Expanded Abstracts 2018", pp.2036~2040 (2018), Society of Exploration Geophysicists
- Glorot, X., "Understanding the difficulty of training deep feedforward neural networks", Proceedings of the thirteenth international conference on artificial intelligence and statistics, pp.249~256 (2010), AISTATS
- Griffith, D. P., "Deep learning applied to seismic attribute computation", Interpretation, 7(3), pp.SE141~SE150 (2019)
- Guillen, P., "SEG Technical Program Expanded Abstracts 2015", pp.1826~1829 (2015), Society of Exploration Geophysicists
- He, K., "Deep residual learning for image recognition", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.770~778 (2016)
- He, K., "Identity mappings in deep residual networks", European conference on computer vision, pp.630~645 (2016), Springer
- Keynejad, S., "Assessment of machine-learning techniques in predicting lithofluid facies logs in hydrocarbon wells", Interpretation, 7(3), pp.SF1~SF13 (2019)
- Krizhevsky, A., "Imagenet classification with deep convolutional neural networks", Advances in neural information processing systems, pp.1097~1105 (2012)
- Kumar, P. C., "Attribute amalgamation-aiding interpretation of faults from seismic data : An example from Waitara 3D prospect in Taranaki basin off New Zealand", Journal of Applied Geophysics, 159, pp.52~68 (2018)
- Noh, H., "Learning deconvolution network for semantic segmentation", Proceedings of the IEEE international conference on computer vision(ICCV), pp.1520~1528 (2015)
- Onajite, E., "Seismic data analysis techniques in hydrocarbon exploration", pp.256~ (2013), Elsevier
- Qian, F., "Unsupervised seismic facies analysis via deep convolutional autoencoders", Geophysics, 83(3), pp.A39~A43 (2018)
- Ronneberger, O., "U-net:Convolutional networks for biomedical image segmentation", International Conference on Medical image computing and computer-assisted intervention, pp.234~241 (2015), Springer
- Sen, S., "Regularization strategies for deep-learning-based salt model building", Interpretation, 7(4), pp.T911~T922 (2019)
- Shi, Y., "SaltSeg : Automatic 3D salt segmentation using a deep convolutional neural network", Interpretation, 7(3), pp.SE113~SE122 (2019)
- Silva, R. M., "Netherlands dataset : A new public dataset for machine learning in seismic interpretation", arXiv preprint arXiv:1904.00770, ~ (2019)
- Simonyan, K., "Very deep convolutional networks for large-scale image recognition", arXiv preprint arXiv : 1409, 1556~ (2014)
- Smith, R., "Correlating geologic and seismic data with unconventional resource production curves using machine learning", Geophysics, 84(2), pp.O39~O47 (2019)
- Szegedy, C., "Going deeper with convolutions", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.1~9 (2015)
- Tian, X., "Machine-learning-based object detection in images for reservoir characterization : A case study of fracture detection in shales", The Leading Edge, 37(6), pp.435~442 (2018)
- Waldeland, A., "3D Attributes and Classification of Salt Bodies on Unlabelled Datasets", 78th EAGE Conference and Exhibition 2016 (2016), EAGE Publications BV
- Wrona, T., "Seismic facies analysis using machine learning", Geophysics, 83(5), pp.O83~O95 (2018)
- Xiong, W., "Seismic fault detection with convolutional neural network", Geophysics, 83(5), pp.O97~O103 (2018)
- 이희욱, "광업 분야의 딥러닝 기술 적용연구 사례분석", 한국자원공학회지, 56(5), pp.435~446 (2019), 한국자원공학회
- Yu, F., "Multi-scale context aggregation by dilated convolutions", arXiv preprint arXiv:1511.07122, ~ (2015)
- Zhang, C., "Machine-learning based automated fault detection in seismic traces", 76th EAGE Conference and Exhibition 2014, European Association of Geoscientists & Engineers, 2014(1), pp.1~5 (2014)
- Zhang, W., "A Robust Iris Segmentation Scheme Based on Improved U-Net", IEEE Access, 7, pp.85082~85089 (2019)
- Zhang, Z., "Road extraction by deep residual u-net", IEEE Geoscience and Remote Sensing Letters, 15(5), pp.749~753 (2018)
- Zhao, T., "SEG Technical Program Expanded Abstracts 2018", pp.2046~2050 (2018), Society of Exploration Geophysicists
- Zhao, T., "Seismic attribute selection for unsupervised seismic facies analysis using user-guided data-adaptive weights", Geophysics, 83(2), pp.O31~O44 (2018)
- Zhou, X. Y., "Normalization in training U-Net for 2-D biomedical semantic segmentation", IEEE Robotics and Automation Letters, 4(2), pp.1792~1799 (2019)