Intelligent Seismic Stratigraphic Identification Based on BiX-NAS: A Case Study from the F3 Dataset in Netherlands Offshore Area
-
摘要: 近些年来，深度学习方法在地震数据处理和解释领域得到了广泛关注和应用，其中大多数深度学习算法采用了端到端的深度卷积神经网络以实现地质体特征的提取与识别（如地层、断裂以及盐丘等）.然而，这些算法往往含有数十万甚至百万的可训练参数，导致模型存在参数冗余、训练效率低等问题.为了解决上述问题，构建了一个轻量化的双向多尺度网络结构模型用于地震层序智能识别.该模型通过两阶段神经网络体系结构搜索算法（neural architecture search，NAS）剔除了双向多尺度网络结构的冗余连接，使得网络结构大幅简化，从而减少参数冗余，进而提高训练效率.采用荷兰近海地区的F3地震数据集对基于NAS算法简化的双向多尺度网络结构地层识别模型进行训练、验证和预测.结果表明：在实际的地层识别任务中，该轻量化模型的平均识别准确率达到了95.52%，并对远离训练工区的预测集具有良好的泛化性.此外，该模型的参数量仅为U形卷积神经网络（U-Net）模型的4.4%，在训练效率、模型参数量等方面优于前人的相关研究工作；并对地震振幅中的噪声干扰具有鲁棒性.因此，这些结果展现了BiX-NAS网络模型在实际地震地层自动识别中良好的应用前景.
-
关键词:
- 地层自动识别 /
- 深度学习 /
- 神经网络体系结构搜索算法 /
- 双向多尺度网络
Abstract: In recent years, deep learning methods have been widely focused and applied in the field of seismic data processing and interpretation, where most deep learning algorithms employ end-to-end deep convolutional neural networks for the extraction and identification of geological features (e.g., stratum, fault, and salt dome). However, these algorithms often contain hundreds of thousands or even millions of trainable parameters, which lead to the model of parameter redundancy and low training efficiency. Therefore, a lightweight bi-directional multi-scale network is constructed for the intelligent identification of stratum. Specifically, the model eliminates the obvious redundant connections of the bi-directional multi-scale network structure through the two-stage Neural Architecture Search (NAS), which greatly simplifies the network structure, reduces the parameter redundancy, and improves the training efficiency. The Netherlands F3 dataset was used to train, verify and predict the simplified bi-directional multi-scale network by the NAS. The results show that the average recognition accuracy of the lightweight model reaches 95.52% in the actual stratigraphic identification task, and it has well generalization to the prediction work area far from the training work area. In addition, the number of parameters of the proposed model is only 4.4% of the U-shaped convolutional neural network (U-Net), and which outperforms the previous related work in terms of training efficiency and the number of model parameters. It is also robust when processing noisy seismic data. Therefore, the BiX-NAS network model has good prospects for application in practical automatic seismic stratigraphic identification. -
图 4 迭代次数为2、网络层数为4的BiO-Net++网络结构图
Fig. 4. Structure of BiO-Net++ with 4 levels and 2 iterations
图 5 BiO-Net++(N=5, L=4)的选择矩阵的两种情况示意图
Fig. 5. Two cases of selection matrix, BiO-Net++(N=5, L=4)
图 9 基于BiX-NAS的地层识别算法流程示意图
Fig. 9. Stratigraphic identification flow diagram based on BiX-NAS
图 12 BiX-NAS模型与U-Net模型不同轮数下的训练损失
Fig. 12. Training loss of BiX-NAS model and U-Net model at different epochs
图 13 验证集不同模型识别结果对比图
a. 标签数据；b. U-Net模型识别结果；c. BiX-NAS模型识别结果
Fig. 13. Comparison of identification results of different modelson validation set
图 14 预测集不同模型识别结果对比图
a.U-Net模型识别结果；b. BiX-NAS模型识别结果
Fig. 14. Comparison of identification results of different models on prediction set
图 16 三维地层识别剖面示意图
a.地震剖面振幅数据（Time=1 360 ms）；b.BiX-NAS模型识别结果（Time=1 360 ms）；c.地震剖面振幅数据（Time=1 600 ms）；d. BiX-NAS模型识别结果（Time=1 600 ms）
Fig. 16. The profilemap of 3D stratum identification
图 17 BiX-NAS模型与U-Net模型不同高斯噪声下的预测集精度
Fig. 17. Prediction set MIoU of BiX-NAS model and U-Net model in the different Gaussian noise environment
图 18 联络测线切片#1025不同噪声和不同模型下的地层识别结果
a. 标签数据；b.U-Net模型不同高斯噪声（σ=10~50）下的识别结果（i）；c. BiX-NAS模型不同高斯噪声（σ=10~50）下的识别结果（i）；d. U-Net模型不同高斯噪声（σ=10~50）下的识别结果（ii）；e. BiX-NAS模型不同高斯噪声（σ=10~50）下的识别结果（ii）
Fig. 18. Stratum identification results of crossline #1025 in the different Gaussian noise environment and models
算法1：NAS搜索算法： 输入：网络迭代次数T；各相邻提取阶段所有跳跃连接组合的集合C；头部网络结构H；各相邻提取阶段的最优网络保留数S；网络结构评价标准Rank；
#C={C(1), C(2), ..., C(t), ..., C(2T-1)}，C(t)表示第t个提取阶段与第t+1个提取阶段间的所有跳跃连接组合的集合；
#C(t)={C1(t), C2(t), ..., Ci(t), ...}，Ci(t)表示第t个提取阶段与第t+1个提取阶段间的第i种跳跃连接组合，nt为C(t)的基数；
#H(1)→(t)表示第1个提取阶段到第t个提取阶段的头部网络结构；
#S={s1→2, s1→2, ..., st→(t+1), ..., s（2T-1）→2T}，st→（t+1）, 表示第t个提取阶段到第t+1个提取阶段的最优网络结构保留数；
#Rank表示网络评价标准；
输出：第m个提取阶段至第2T个提取阶段间的最简跳跃连接组合的集合E(m)→(2T)，其中m∈[1, 2T-1].
#E(m)→(2T)={E1(m)→(2T), E2(m)→(2T), ...，Ei(m)→(2T), ...，$ {E}_{{s}_{m}\to (m+1)}^{\left(m\right)\to \left(2T\right)} $}，Ei(m)→(2T)表示第m个提取阶段至第2T个提取阶段间的第i种最简跳跃连接组合；
1: for t=2T-1, …, 1 do
2: if t=2T-1 then
3: for j=1, …, nt do
4: Forward head network H(1)→(t)
5: Forward tail network with sampled skips cj(t)
6: Rank (H(1)→(t), cj(t))
7: end for
8: E(2T-1)→(2T)←Optimal solution of Rank (H(1)→(t), cj(t))
9: else
10: for i=1, …, s(t+1)→(t+2) do
11: Forward head network H(1)→(t)
12: for j=1, …, nt do
13: Forward tail network with sampled skips cj(t), Ei(t+1)→(2T)
14: Rank (H(1)→(t), cj(t), Ei(t+1)→(2T))
15: end for
16: end for
17: H(t)→(2T)←Optimal solution of Rank (H(1)→(t), cj(t), Ei(t+1)→(2T))
18: end if
19: end for
下载: 导出CSV
表 1 不同层位的地震相特征解释
Table 1. Interpretation of seismic facies characteristics in different horizons
层位序号 地震相特征解释 1 顶部反射具有分层的地震相，下部为具有均匀的地震相，无明显的反射 2 该层的反射信号主要由低振幅且连续的反射信号组成 3 该层的反射信号主要呈现不连续的丘状特征 4 该层的反射信号主要呈现为亚平行态 5 该层的反射信号主要呈现为中低幅度的S型曲线结构 6 该层的反射信号主要由平行的高振幅反射信号组成 7 该层的反射信号主要由半连续和低振幅的反射信号组成 8 该层的反射信号主要呈现为扭曲状和低振幅相 9 该层内没有明显反射信号，由低振幅组成
下载: 导出CSV
表 2 不同迭代次数的BiO-Net网络模型在不同数据集上的训练结果
Table 2. Training results of BiO-Net with different iterations on different data sets
NO. Dataset Model Epochs Params
(×106)MACs
(×109)Val_MIoU #1 #1 BiO-Net(iter=1) 100 14.964 1 11.300 2 0.967 5 #2 #2 BiO-Net(iter=1) 100 14.964 1 11.300 2 0.957 0 #3 #1 BiO-Net(iter=2) 100 14.980 1 36.427 1 0.962 9 #4 #2 BiO-Net(iter=2) 100 14.980 1 36.427 1 0.953 7 #5 #1 BiO-Net(iter=3) 100 14.996 0 77.277 2 0.942 0 #6 #2 BiO-Net(iter=3) 100 14.996 0 77.277 2 0.947 5 #7 #1 BiO-Net(iter=4) 100 15.011 9 133.850 3 0.794 2 #8 #2 BiO-Net(iter=4) 100 15.011 9 133.850 3 0.934 6 #9 #1 BiO-Net(iter=5) 100 15.027 9 206.146 6 0.324 0 #10 #2 BiO-Net(iter=5) 100 15.027 9 206.146 6 0.795 1
下载: 导出CSV
表 3 不同迭代次数的Phase1-Searched-Net网络模型在不同数据集上的训练结果
Table 3. Training results of Phase1-Searched-Net with different iterations on different data sets
NO. Dataset Model Epochs Params
(×106)MACs
(×109)Val_MIoU #11 #1 Phase1-Searched-Net(iter=1) 100 0.422 1 6.735 0 0.954 7 #12 #2 Phase1-Searched-Net(iter=1) 100 0.422 1 6.735 0 0.946 2 #13 #1 Phase1-Searched-Net(iter=2) 100 0.423 6 50.142 7 0.953 1 #14 #2 Phase1-Searched-Net(iter=2) 100 0.423 6 49.799 5 0.943 3 #15 #1 Phase1-Searched-Net(iter=3) 100 0.425 1 151.297 7 0.951 8 #16 #2 Phase1-Searched-Net(iter=3) 100 0.425 1 124.930 6 0.944 1
下载: 导出CSV
表 4 BiX-NAS网络模型在不同数据集上的训练结果
Table 4. Training results of BiX-NASon different data sets
NO. Dataset Model Epochs Rank Re_Epochs Params
(×106)MACs
(×109)Val_MIoU #17 #1 BiX-NAS 10 #1 100 0.380 5 5.369 6 0.950 5 #18 #1 BiX-NAS 10 #2 100 0.339 1 5.027 2 0.955 2 #19 #2 BiX-NAS 10 #1 100 0.380 5 5.370 9 0.945 9 #20 #2 BiX-NAS 10 #2 100 0.380 5 5.370 9 0.943 7 #21 #1 BiX-NAS 20 #1 100 0.380 5 5.371 6 0.947 6 #22 #1 BiX-NAS 20 #2 100 0.339 1 5.027 2 0.942 6 #23 #2 BiX-NAS 20 #1 100 0.380 5 6.397 7 0.947 7 #24 #2 BiX-NAS 20 #2 100 0.339 1 5.027 3 0.936 8
下载: 导出CSV
表 5 不同模型的性能评估统计
Table 5. Performance evaluation statistics of different models
NO. Dataset Model Epochs Params
(×106)MACs
(×109)Val_MIoU Training time(s) - #1 U-Net 100 7.762 8 13.719 4 0.969 1 1 039.206 9 #11 #1 Phase1-Searched-Net(iter=1) 100 0.422 1 6.735 0 0.954 7 935.148 5 #18 #1 BiX-NAS 100 0.339 1 5.027 2 0.955 2 829.392 7 - #2 U-Net 100 7.762 8 13.719 4 0.958 6 2 000.413 2 #12 #2 Phase1-Searched-Net(iter=1) 100 0.422 1 6.735 0 0.946 2 1 776.755 9 #23 #2 BiX-NAS 100 0.380 5 6.397 7 0.947 7 1 710.181 1
下载: 导出CSV
-
Abdellatif, A., Elsheikh, A.H., Graham, G., et al., 2022. Generating Unrepresented Proportions of Geological Facies using Generative Adversarial Networks. Computers & Geosciences, 162: 105085. https://doi.org/10.1016/j.cageo.2022.105085 Araya-Polo, M., Jennings, J., Adler, A., et al., 2018. Deep-Learning Tomography. The Leading Edge, 37(1): 58-66. https://doi.org/10.1190/tle37010058.1 Badrinarayanan, V., Kendall, A., Cipolla, R., 2017. SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(12): 2481-2495. https://doi.org/10.1109/TPAMI.2016.2644615 Bahorich, M., Farmer, S., 1995. 3-D Seismic Discontinuity for Faults and Stratigraphic Features: The Coherence Cube. The Leading Edge, 14(10): 1053-1058. https://doi.org/10.1190/1.1437077 Cao, S., 2019. Application of Geostatistical Inversion Method in Reservoir Prediction of Coal Measure Strata in Hangjinqi Area. Petroleum Geology and engineering, 33(5): 41-44. (in Chinese with English abstract) Ciresan, D., Giusti, A., Gambardella, L., et al., 2012. Deep Neural Networks Segment Neuronal Membranes in Electron Microscopy Images. Advances in Neural Information Processing Systems, 25: 2843-2851. Di, H., Wang, Z., Alregib, G., 2018. Deep Convolutional Neural Networks for Seismic Salt-Body Delineation. In: 2018 AAPG Annual Convention and Exhibition, Search and Discovery Article, 90323. Donald, A. S., 2021. How Deep Learning Networks could be Designed to Locate Mineral Deposits. Journal of Earth Science, 32(2): 288-292. https://doi.org/10.1007/s12583-020-1399-2 Dorn, G.A., 1998. Modern 3-D Seismic Interpretation. The Leading Edge, 17(9): 1262-1262. https://doi.org/10.1190/1.1438121 Du, G., Cao, X., Liang, J., et al., 2020. Medical Image Segmentation Based on U-Net: A Review. Journal of Imaging Science and Technology, 64(2): 20508-1-20508-12. https://doi.org/10.2352/J.ImagingSci.Technol.2020.64.2.020508 Fan, Y., Huang, L., Dai, S., 1999. Application of Crossplot Technique to the Determination of Lithology Composition and Fracture Identification of Igneous Rock. Well Logging Technology, 23(1): 53-56(in Chinese with English abstract). Gao, K., Huang, L., Zheng, Y., et al., 2022. Automatic Fault Detection on Seismic Images Using a Multiscale Attention Convolutional Neural Network. Geophysics, 87(1): N13-N29. https://doi.org/10.1190/geo2020-0945.1 Guo, Z., Zhang, X., Mu, H., et al., 2020. Single Path One-Shot Neural Architecture Search with Uniform Sampling. In: 2020 European Conference on Computer Vision(ECCV), Springer, Cham, 544-560. https://doi.org/10.1007/978-3-030-58517-4_32 He, K., Zhang, X., Ren, S., et al., 2016. Deep Residual Learning for Image Recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), Las Vegas, NV, USA, 770-778. https://doi.org/10.1109/CVPR.2016.90 Ioffe, S., Szegedy, C., 2015. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. In: 2015 International Conference on Machine Learning(ICML), PMLR, 448-456. Jang, E., Gu, S., Poole, B., 2017. Categorical Reparameterization with Gumbel-Softmax. In: 2017 International Conference on Learning Representations(ICLR). https://doi.org/10.48550/arXiv.1611.01144 Kingma, D.P., Ba, J., 2014. Adam: A Method for Stochastic Optimization. Computer Science, 14(6): 123-126. https://doi.org/10.48550/arXiv.1412.6980 Laloy, E., Hérault, R., Lee, J., et al., 2017. Inversion Using a New Low-Dimensional Representation of Complex Binary Geological Media Based on a Deep Neural Network. Advances in Water Resources, 110: 387-405. https://doi.org/10.1016/j.advwatres.2017.09.029 Laloy, E., Hérault, R., Jacques, D., et al., 2018. Training-Image Based Geostatistical Inversion Using a Spatial Generative Adversarial Neural Network. Water Resources Research, 54(1): 381-406. https://doi.org/10.1002/2017WR022148 Liu, H., Simonyan, K., Yang, Y., 2018. DARTS: Differentiable Architecture Search. In: 2018 International Conference on Learning Representations(ICLR). https://doi.org/10.48550/arXiv.1806.09055 Long, J., Shelhamer, E., Darrell, T., 2015. Fully Convolutional Networks for Semantic Segmentation. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), 3431-3440. https://doi.org/10.1109/CVPR.2015.7298965 Lopez-Alvis, J., Laloy, E., Nguyen, F., et al., 2021. Deep Generative Models in Inversion: a Review and Development of a New Approach Based on a Variational Autoencoder. Computers & Geosciences, 152: 104762. https://doi.org/10.1016/j.cageo.2021.104762 Ma, G., Wu, Q., Xiong, S., et al., 2021. Ratio Method for Calculating the Source Location of Gravity and Magnetic Anomalies Based on Deep Learning. Earth Science, 46(9): 3365-3375. https://doi.org/10.3799/dqkx.2020.350 (in Chinese with English abstract) Marfurt, K.J., Kirlin, R.L., Farmer, S.L., et al., 1998. 3-D Seismic Attributes Using a Semblance-Based Coherency Algorithm. Geophysics, 63(4): 1150-1165. https://doi.org/10.1190/1.1444415 Mosser, L., Dubrule, O., Blunt, M., 2018. Stochastic Seismic Waveform Inversion Using Generative Adversarial Networks As A Geological Prior. Mathematical Geosciences, 52(1), 53-79. https://doi.org/10.3997/2214-4609.201803018 Nair, V., Hinton, G.E., 2010. Rectified Linear Units Improve Restricted Boltzmann Machines. In: 2010 International Conference on Machine Learning(ICML), Haifa, Israel, 807-814. Qi, J., Lyu, B., Wu, X., et al., 2020. Comparing Convolutional Neural Networking and Image Processing Seismic Fault Detection Methods. In: 2020 SEG International Exposition and Annual Meeting. OnePetro, 1111-1115. https://doi.org/10.1190/segam2020-3428171.1 Real, E., Moore, S., Selle, A., et al., 2017. Large-Scale Evolution of Image Classifiers. In: 2017 International Conference on Machine Learning(ICML), PMLR, 2902-2911. Real, E., Aggarwal, A., Huang, Y., et al., 2019. Regularized Evolution for Image Classifier Architecture Search. In: 2019 AAAI Conference on Artificial Intelligence 33: 4780-4789. https://doi.org/10.1609/aaai.v33i01.33014780 Ronneberger, O., Fischer, P., Brox, T., 2015. U-Net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Computing and Computer-Assisted Intervention, 9351: 234-241. https://doi.org/10.1007/978-3-319-24574-4_28 Shi, Y., Wu, X., Fomel, S., 2019. SaltSeg: Automatic 3D Salt Segmentation Using a Deep Convolutional Neural Network. Interpretation, 7(3), SE113-SE122. https://doi.org/10.1190/int-2018-0235.1 Silva, R.M., Baroni, L., Ferreira, R.S., et al., 2019. Netherlands Dataset: A New Public Dataset for Machine Learning in Seismic Interpretation. arXiv preprint, 2019(6): 38-42. https://doi.org/10.48550/arXiv.1904.00770 Tao, H., Cheng, R., Zhao, X., et al., 2011. Well Logging Response to the Volcaniclastic Rocks of Hailar Basin and Application. Chinese Journal of geophysics, 54(2): 534-544. https://doi.org/10.3969/j.issn.0001-5733. 2011. 02.033(in Chinese with English abstract) doi: 10.3969/j.issn.0001-5733.2011.02.033 Wang, D., Chen, G., 2021. Seismic Stratum Segmentation Using an Encoder-Decoder Convolutional Neural Network. Mathematical Geosciences, 53(6): 1355-1374. https://doi.org/10.1007/s11004-020-09916-8 Wang, X., Xiang, T., Zhang, C., et al., 2021. BiX-NAS: Searching Efficient Bi-Directional Architecture for Medical Image Segmentation. Medical Image Computing and Computer-Assisted Intervention, 229-238. https://doi.org/10.1007/978-3-030-87193-2_22 Wang, D., Chen, G., 2022. Seismic Wave Impedance Inversion Based on Temporal Convolutional Network. Earth Science, 47(4): 1492-1506(in Chinese with English abstract). Wu, X., Shi, Y., Fomel, S., et al., 2018. Convolutional Neural Networks for Fault Interpretation in Seismic Images. In: 2018 SEG International Exposition and Annual Meeting. OnePetro: Anaheim, California; 1946-1950. https://doi.org/10.1190/segam2018-2995341.1 Wu, H., Zhang, B., Lin, T., et al., 2019a. Semiautomated Seismic Horizon Interpretation Using the Encoder-Decoder Convolutional Neural Network. Geophysics, 84(6): B403-B417. https://doi.org/10.1190/geo2018-0672.1 Wu, X., Liang, L., Shi, Y., et al., 2019b. FaultSeg3D: Using Synthetic Data Sets to Train an End-to-End Convolutional Neural Network for 3D Seismic Fault Segmentation. Geophysics, 84: IM35-IM45. https://doi.org/10.1190/geo2018-0646.1 Wu, X., Liang, L., Shi, Y., et al., 2019c. Multitask Learning for Local Seismic Image Processing: Fault Detection, Structure-Oriented Smoothing with Edge-Preserving, and Seismic Normal Estimation by Using a Single Convolutional Neural Network. Geophysical Journal International, 219(3): 2097-2109. https://doi.org/10.1093/gji/ggz418 Xiang, T., Zhang, C., Liu, D., et al., 2020. BiO-Net: Learning Recurrent Bi-Directional Connections for Encoder-Decoder Architecture. Medical Image Computing and Computer: Assisted Intervention, 74-84. https://doi.org/10.1007/978-3-030-59710-8_8 Xue, W., Chen, B., Zhang, Z., 2019. Recognition of Stratigraphic Lithology by BP-Neural Network-A Case Study of Yiner Basin. Pretrochemical Technology, 26(11): 103-107. https://doi.org/CNKI:SUN:SHJS.0.2019-11-058 (in Chinese with English abstract)Yang, F., Ma, J., 2019. Deep-Learning Inversion: a Next Generation Seismic Velocity-Model Building Method. Geophysics, 84(4): R583-R599. https://doi.org/10.1190/GEO2018-0249.1 Yang, D., Cai, Y., Hu, G., et al., 2020a. Seismic Fault Detection Based on 3D Unet++ Model. In: 2020 SEG International Exposition and Annual Meeting. OnePetro, 1631-1635. https://doi.org/10.1190/segam2020-3426516.1 Yang, Z., Wang, Y., Chen, X., et al. 2020b. CARS: Continuous Evolution for Efficient Neural Architecture Search. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), 1829-1838. https://doi.org/10.48550/arXiv.1909.04977 Yu, S., Ma, J., 2021. Deep Learning for Geophysics: Current and Future Trends. Reviews of Geophysics, 59(3), e2021RG000742. https://doi.org/10.1029/2021RG000742 Zeng, Y., Jiang, K., Chen, J., 2018. Automatic Seismic Salt Interpretation with Deep Convolutional Neural Networks. In: 2019 International Conference on Information System and Data Mining, 16-20. https://doi.org/10.1145/3325917.3325926 Zhang, H., Liu, Y., Zhang, Y., et al., 2019. Automatic Seismic Facies Interpretation Based on an Enhanced Encoder-Decoder Structure. In: 2019 SEG Technical Program Expanded Abstracts, 2408-2412. https://doi.org/10.1190/segam2019-3215516.1 Zhang, H., Chen, T., Liu, Y., et al., 2021. Automatic Seismic Facies Interpretation Using Supervised Deep Learning. Geophysics, 86(1): IM15-IM33. https://doi.org/10.1190/geo2019-0425.1 Zheng, Y., Li, G., 2009. Application of Support Vector Machine to Stratum Recognition. Journal of Henan Normal University(Natural Science), 37(2): 37-39(in Chinese with English abstract). Zheng, Z.H., Kavousi, P., Di, H.B., 2014. Multi-Attributes and Neural Network-Based Fault Detection in 3D Seismic Interpretation. Advanced Materials Research, 838-841: 1497-1502. https://doi.org/10.4028/www.scientific.net/AMR.838-841.1497 Zhou, Z., Siddiquee, M.M.R., Tajbakhsh, N., et al., 2018. UNet++: A Nested U-Net Architecture for Medical Image Segmentation. In: 2018 Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, 3-11. https://doi.org/10.1007/978-3-030-00889-5_1 曹绍贺, 2019. 地质统计学反演方法在杭锦旗区块煤系地层储层预测中的应用. 石油地质与工程, 33(5): 4. https://www.cnki.com.cn/Article/CJFDTOTAL-SYHN201905010.htm 范宜仁, 黄隆基, 代诗华, 1999. 交会图技术在火山岩岩性与裂缝识别中的应用. 测井技术, 23(1): 53-56. https://www.cnki.com.cn/Article/CJFDTOTAL-CJJS901.012.htm 马国庆, 吴琪, 熊盛青, 等, 2021. 基于重磁数据梯度比值的深度学习技术实现场源位置反演方法. 地球科学, 46(9): 3365-3375. doi: 10.3799/dqkx.2020.350 陶宏根, 程日辉, 赵小青, 等, 2011. 海拉尔盆地火山碎屑岩的测井响应与应用. 地球物理学报, (2): 534-544. https://www.cnki.com.cn/Article/CJFDTOTAL-DQWX201102035.htm 王德涛, 陈国雄, 2022. 基于时间卷积网络的地震波阻抗反演. 地球科学, 47(4): 1492-1506. doi: 10.3799/dqkx.2021.070 薛文卓, 陈彪, 张哲豪, 2019. BP-神经网络识别地层岩性-以银额盆地为例. 石化技术, (11): 103-107. https://www.cnki.com.cn/Article/CJFDTOTAL-SHJS201911058.htm 郑延斌, 李国和, 2009. 支持向量机在地层识别中的应用. 河南师范大学学报(自然科学版), (2): 37-39. https://www.cnki.com.cn/Article/CJFDTOTAL-HNSX200902011.htm -
点击查看大图
图(18) / 表(6)
计量
- 文章访问数: 2216
- HTML全文浏览量: 1220
- PDF下载量: 184
- 被引次数: 0
