基于深度堆栈自编码器的地震相识别
Deep Stacked Auto-Encoder for Seismic Facies Classification
DOI: 10.12677/CSA.2022.1210241, PDF,    科研立项经费支持
作者: 赵云鹤, 硕良勋, 柴变芳, 李增浩：河北地质大学信息工程学院，河北 石家庄
关键词: 地震相识别；自编码器；深度学习；无监督聚类；Seismic Facies Classification； Auto Encoder； Deep Learning； Unsupervised Clustering
摘要: 地震相是沉积环境分析和储层预测的基础。然而在实际地震勘探中，标注的地震数据难以获取。针对这些问题，本文分别借鉴了深度学习与无监督学习的优点，提出了基于深度堆栈自编码器的地震相识别方法，该模型利用深度学习更能捕捉地震数据语义信息的优点，用深度堆栈自编码器对地震数据进行特征提取，然后用提取的特征进行无监督聚类算法对地震相进行识别。在实际工区的实验结果表明，该模型相较于直接使用聚类算法识别地震相，准确率更高。
Abstract: Seismic facies play a key role in depicting subsurface geology horizons and predicting. However, large amounts of annotated seismic data are often unavailable. To overcome these challenges, we proposed a deep stacked auto-encoder network for seismic facies classification, which draws on the advantages of deep learning and unsupervised learning. Deep stacked auto-encoder performs bet-ter to exact deep features from seismic data. Then we cluster exacted features to classify seismic fa-cies without the labels for unsupervised learning. The proposed model has been tested successfully on real seismic data fields, which indicates that the model is more effective than clustering directly.
文章引用：赵云鹤, 硕良勋, 柴变芳, 李增浩. 基于深度堆栈自编码器的地震相识别[J]. 计算机科学与应用, 2022, 12(10): 2357-2361. https://doi.org/10.12677/CSA.2022.1210241

参考文献

[1] 吴宝年. 油藏模型含油饱和度变化的地震振幅响应特征[J]. 物探与化探, 2015, 39(6): 1271-1277.
[2] Dramsch, J.S. and Lüthje, M. (2018) Deep-Learning Seismic Facies on State-of-the-Art CNN Architectures. SEG Technical Pro-gram Expanded Abstracts, Anaheim, 17 October, 2018, 2036-2040. [Google Scholar] [CrossRef] 
[3] Zhao, T. (2018) Seismic Facies Classification Using Different Deep Convolutional Neural Networks: 88th Annual International Meeting. SEG Technical Program Expanded Abstracts, An-aheim, 17 October, 2018, 2046-2050. [Google Scholar] [CrossRef] 
[4] Liu, M., Jervis, M., Li, W., et al. (2020) Seismic Facies Classifica-tion Using Supervised Convolutional Neural Network and Semi-Supervised Generative Adversarial Networks. Geo-physics, 85, 1-57. [Google Scholar] [CrossRef] 
[5] 闫星宇, 顾汉明, 罗红梅, 等. 基于改进深度学习方法的地震相智能识别[J]. 石油地球物理勘探, 2020, 55(6): 1169-1177.
[6] Li, F.Y., Zhou, H.L., Wang, Z.Y. and Wu, X.M. (2021) ADDCNN: An Attention-Based Deep Dilated Convolutional Neural Network for Seismic Facies Analysis with Inter-pretable Spatial-Spectral Maps. IEEE Transactions on Geoscience and Remote Sensing, 59, 1733-1744. [Google Scholar] [CrossRef] 
[7] Chen, X.Y., Zou, Q., Xu, X.X. and Wang, N. (2022) A Stronger Baseline for Seismic Facies Classification with Less Data. IEEE Transactions on Geoscience and Remote Sensing, 60, Article No. 5914910. [Google Scholar] [CrossRef] 
[8] MacQueen, J.B. (1967) Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability. University of California Press. Berkeley, 281-297.
[9] Silva, R.M., Baroni, L., Ferreira, R.S., et al. (2019) Netherlands Dataset: A New Public Dataset for Machine Learning in Seismic Interpretation. https://arxiv.org/abs/1904.00770v1