基于形状上下文的成矿构造形态控矿特征深度学习及三维成矿预测——以大尹格庄金矿床为例
Study on Deep Learning of Ore-Controlling Characteristics of Geological Morphology and 3D Metallogenic Prediction Based on Shape Context—A Case Study of Dayingezhuang Gold Deposit
摘要: 随着地质找矿工作逐步向深部空间的发展，深部矿、隐伏矿已成为找矿的主要对象。在对深部矿、隐伏矿的预测中，有效的成矿信息提取能够保证预测结果的准确性和可靠性。本文提出一种基于形状上下文的成矿构造形态控矿特征深度学习方法，以大尹格庄金矿床为研究对象，利用胶西北招平断裂带大尹格庄断离面趋势-起伏因素(waF、wbF)、断离面坡度因素(gF)、断离面陡缓转换部位综合场因素(fV)封装的三维形状上下文特征，利用深度学习中的图卷积神经网络，学习获得深层次的地质体形态控矿特征。研究表明，图卷积网络可提取更完备的地质体形态控矿特征，建立具有较好预测准确性的三维成矿预测模型。
Abstract: With the gradual development of geological prospecting to deep space, deep ore and concealed ore have become the main objects of ore prospecting. In the prediction of deep and concealed deposits, the effective extraction of metallogenic information can ensure the accuracy and reliability of the prediction results. This paper proposes a deep learning method based on the shape context of geological morphology based on shape context. Using the 3D shape context features encapsulated by the trend-fluctuation factors (waF, wbF), slope factor (gF), and comprehensive field factors (fV) of the Dayingezhuang gold deposit, the graph convolution neural network in deep learning is used. Learn more complete morphological ore-controlling characteristics of geological bodies. Research shows that the graph convolutional network can extract more complete geological shape ore-controlling characteristics, and establish a three-dimensional ore-forming prediction model with certain prediction accuracy.
文章引用：聂凡, 魏运凤, 郑扬, 邓浩. 基于形状上下文的成矿构造形态控矿特征深度学习及三维成矿预测——以大尹格庄金矿床为例[J]. 地球科学前沿, 2021, 11(2): 137-146. https://doi.org/10.12677/AG.2021.112011

参考文献

[1] 赵鹏大. 成矿定量预测与深部找矿[J]. 地学前缘, 2007, 14(5): 1-10.
[2] 陈建平, 吕鹏, 吴文, 赵洁, 胡青. 基于三维可视化技术的隐伏矿体预测[J]. 地学前缘, 2007, 14(5): 54-61.
[3] 王功文, 郭远生, 杜杨松, 范玉华, 郭欣, 庞振山, 陈建平. 基于GIS的云南普朗斑岩铜矿床三维成矿预测[J]. 矿床地质, 2007, 26(6): 651-658.
[4] 肖克炎, 李楠, 孙莉, 邹伟, 李莹. 基于三维信息技术大比例尺三维立体矿产预测方法及途径[J]. 地质学刊, 2012, 36(3): 229-236.
[5] 赵鹏大. 成矿多样性与矿床谱系[M]. 北京: 中国地质大学, 2009.
[6] 於崇文. 成矿作用动力学[M]. 北京: 地质出版社, 1997.
[7] 汤国安, 龚健雅, 陈正江, 成燕辉, 王占宏. 数字高程模型地形描述精度量化模拟研究[J]. 测绘学报, 2001, 30(4): 361-365.
[8] Mejía-Herrera, P., Royer, J.J., Caumon, G., et al. (2015) Curvature Attribute from Surface-Restoration as Predictor Variable in Kupferschiefer Copper Potentials. Natural Resources Research, 24, 275-290. [Google Scholar] [CrossRef] 
[9] 毛先成, 赵莹, 唐艳华, 彭正林, 陈进, 邓浩. 基于TIN的地质界面三维形态分析方法与应用[J]. 中南大学学报(自然科学版), 2013, 44(4): 1493-1499.
[10] 邓浩. 面向隐伏矿体预测的三维地质建模与空间分析若干技术研究[D]: [硕士学位论文]. 长沙: 中南大学, 2008.
[11] 毛先成, 唐艳华, 邓浩. 地质体的三维形态分析方法与应用[J]. 中南大学学报(自然科学版), 2012, 43(2): 588-595.
[12] 徐述腾, 周永章. 基于深度学习的镜下矿石矿物的智能识别实验研究[J]. 岩石学报, 2018, 34(11): 3244-3252.
[13] 刘艳鹏, 朱立新, 周永章. 卷积神经网络在矿床找矿预测中的应用——以安徽招吉口铅锌矿床为例. 岩石学报, 2018, 34(11): 3217-3224.
[14] 宋明春, 伊丕厚, 徐军祥, 崔书学, 沈昆, 姜洪利, 袁文花, 王化江. 胶西北金矿阶梯式成矿模式[J]. 中国科学: 地球科学, 2012, 42(7): 992-1000.
[15] 宋明春. 胶东三山岛北部海域超大型金矿床的发现及其构造——岩浆背景[J]. 地质学报, 2015, 89(2): 365-383.
[16] 戴雪灵, 邓湘伟, 彭省临, 杨斌. 玲珑花岗岩在大尹格庄金矿成矿系统中的作用[J]. 地质与勘探, 2011, 47(3): 370-379.
[17] 陈进, 毛先成, 邓浩. 山东大尹格庄金矿床深部三维定量成矿预测[J]. 地球学报, 2020, 41(2): 179-191.
[18] Mao, X., Ren, J., Liu, Z., et al. (2019) Three-Dimensional Prospectivity Modeling of the Jiaojia-Type Gold Deposit, Jiaodong Peninsula, Eastern China: A Case Study of the Dayingezhuang Deposit. Journal of Geochemical Exploration, 203, 27-44. [Google Scholar] [CrossRef] 
[19] Körtgen, M., Park, G.J., Novotni, M., et al. (2003) 3D Shape Matching with 3D Shape Contexts. Central European Seminar on Computer Graphics, Bonn, 164.
[20] Wu, Z., Pan, S., Chen, F., et al. (2019) A Comprehensive Survey on Graph Neural Networks. IEEE Transactions on Neural Networks and Learning Systems, 32, 4-24.
[21] Goyal, P. and Ferrara, E. (2018) Graph Embedding Techniques, Applications, and Performance: A Survey. Knowledge-Based Systems, 151, 78-94. [Google Scholar] [CrossRef] 
[22] Kipf, T.N. and Welling, M. (2016) Semi-Supervised Classification with Graph Convolutional Networks.
[23] Hamilton, W.L., Ying, R. and Leskovec, J. (2017) Inductive Representation Learning on Large Graphs. Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, 4-9 December 2017, 1025-1035.
[24] Hamilton, W.L., Ying, R. and Leskovec, J. (2017) Representation Learning on Graphs: Methods and Applications. IEEE Data Engineering Bulletin, 40, 52-74.
[25] Aranha, P.S., Monteiro, F.P., Pontes, P.A.I., et al. (2015) Supervised Classification of Satellite Images to Analyze Multi-Temporal Land Use and Coverage: A Case Study for the Town of MARABA, State of PARA, Brazil. Second International Conference on Computer Science and Information Technology, Marabá City, 9-19. [Google Scholar] [CrossRef] 
[26] Hu, P., Niu, Z., He, T., et al. (2018) Learning Deep Representations in Large Integrated Network for Graph Clustering. 2018 IEEE First International Conference on Artificial Intelligence and Knowledge Engineering (AIKE), Laguna Hills, 26-28 September 2018, 101-105. [Google Scholar] [CrossRef] 
[27] Ye, J., Zhao, J., Ye, K., et al. (2020) How to Build a Graph-Based Deep Learning Architecture in Traffic Domain: A Survey. [Google Scholar] [CrossRef] 
[28] Lv, J., Zhong, J., Liang, J., et al. (2019) ACE: Ant Colony Based Multi-Level Network Embedding for Hierarchical Graph Representation Learning. IEEE Access, 7, 73970-73982. [Google Scholar] [CrossRef]