| 144 | 0 | 743 |
| 下载次数 | 被引频次 | 阅读次数 |
在深度学习地质建模技术的显式化过程中，传统规则栅格采样方式面临巨大的计算与存储压力，且常出现模型边界的锯齿效应，难以兼顾效率与可视化精度。为此，该文提出一种基于自适应八叉树的深度学习地质建模并行显式化采样方法。首先，基于钻孔数据训练深度学习网络，利用训练好的深度学习网络学习地下岩性的空间分布规律；其次，基于自适应八叉树的数据结构，在识别到岩性界面区域时自动对相应节点进行细化分裂，从而在关键部位采用高分辨率采样、在均质区域采用粗分辨率采样以减少不必要的三角面数；最后，通过多进程并行计算框架提高该方法对大规模数据的处理效率。实验结果表明，相比基于规则栅格的常规显式化方法，该方法在保证建模精度的前提下，模型面数减少60%以上，且模型边界光顺性明显提升，在有效缓解渲染性能压力的同时，能提高模型可视化的精细度。
Abstract：In the explicit modeling phase of deep learning-based geological modeling techniques, traditional regular grid sampling methods often encounter significant computational and storage demands.They frequently produce aliasing effects at model boundaries, making it difficult to balance efficiency with visualization accuracy.To overcome these limitations, this paper proposes a parallel explicit sampling method for deep learning-driven geological modeling based on an adaptive octree structure.The proposed approach involves three key steps: training a deep learning classification network on borehole data to capture the spatial distribution patterns of subsurface lithology; employing an adaptive octree data structure to automatically refine nodes in regions identified as lithological interfaces, enabling high-resolution sampling in critical zones while maintaining coarse resolution in homogeneous areas to minimize redundant triangular facets; and implementing a multi-process parallel computing framework to enhance processing efficiency for large-scale data.Experimental results indicate that, compared to conventional regular grid-based explicit methods, the proposed approach reduces the number of model facets by over 60% without compromising modeling accuracy, while markedly improving boundary smoothness.This method effectively alleviates rendering performance bottlenecks and enhances the precision of model visualization, providing a robust solution for efficient and accurate geological modeling.
[1] 周进威，苏栋，韩文龙，等.基于三维地质模型的城市深层地下空间开发适宜性评估[J/OL].地学前缘，2025:1-12.(2025-02-26)[2025-07-30].https://doi.org/10.13745/j.esf.sf.2025.2.6.
[2] 刘志斌，张健桥，杜晓峰，等.矿区复杂地质构造隐式三维集成建模[J].东北大学学报(自然科学版),2024,45(9):1317-1325.
[3] 尚浩杰，陈叔阳，何云峰，等.断控缝洞型碳酸盐岩储层结构表征与三维地质建模：以顺北4号断裂带为例[J].石油学报，2024,45(11):1662-1679.
[4] 车德福，吴立新，殷作如，等.基于GTP的断层三维交互建模方法[J].东北大学学报(自然科学版),2008,29(3):395-398.
[5] WU L X.Topological relations embodied in a generalized tri-prism (GTP) model for a 3D geoscience modeling system[J].Computers & Geosciences,2004,30(4):405-418.
[6] 余淑娟，郭飞，李想，等.约束Delaunay四面体剖分在三维地质建模中的应用[J].地理与地理信息科学，2013,29(1):41-44.
[7] HILLIER M J,SCHETSELAAR E M,DE KEMP E A,et al.Three-dimensional modelling of geological surfaces using generalized interpolation with radial basis functions[J].Mathematical Geosciences,2014,46(8):931-953.
[8] 郭甲腾，吴立新，周文辉.基于径向基函数曲面的矿体隐式自动三维建模方法[J].煤炭学报，2016,41(8):2130-2135.
[9] GUO J T,WANG X L,WANG J M,et al.Three-dimensional geological modeling and spatial analysis from geotechnical borehole data using an implicit surface and marching tetrahedra algorithm[J].Engineering Geology,2021,284:106047.DOI:10.1016/j.enggeo.2021.106047.
[10] GUO J T,XU X C,WANG L Y,et al.GeoPDNN 1.0:a semi-supervised deep learning neural network using pseudo-labels for three-dimensional shallow strata modelling and uncertainty analysis in urban areas from borehole data[J].Geoscientific Model Development,2024,17(3):957-973.
[11] HILLIER M,WELLMANN F,DE KEMP E A,et al.GeoINR 1.0:an implicit neural network approach to three-dimensional geological modelling[J].Geoscientific Model Development,2023,16(23):6987-7012.
[12] HILLIER M,WELLMANN F,BRODARIC B,et al.Three-dimensional structural geological modeling using graph neural networks[J].Mathematical Geosciences,2021,53(8):1725-1749.
[13] ZHANG B Y,XU K,KHAN U,et al.A lightweight convolutional neural network with end-to-end learning for three-dimensional mineral prospectivity modeling:a case study of the Sanhetun Area,Heilongjiang Province,Northeastern China[J].Ore Geology Reviews,2023,163:105788.DOI:10.1016/j.oregeorev.2023.105788.
[14] YANG Z X,CHEN Q Y,CUI Z S,et al.Automatic reconstruction method of 3D geological models based on deep convolutional generative adversarial networks[J].Computational Geosciences,2022,26(5):1135-1150.
[15] 徐学闯，张恒兵，符韶华，等.基于钻孔伪标签半监督深度学习的沈阳市浅表地层三维建模研究[J].地理与地理信息科学，2023,39(6):9-17.
[16] 郭甲腾，刘寅贺，韩英夫，等.基于机器学习的钻孔数据隐式三维地质建模方法[J].东北大学学报(自然科学版),2019,40(9):1337-1342.
[17] JIA R,LV Y K,WANG G W,et al.A stacking methodology of machine learning for 3D geological modeling with geological-geophysical datasets,Laochang Sn camp,Gejiu (China)[J].Computers & Geosciences,2021,151:104754.DOI:10.1016/j.cageo.2021.104754.
[18] GRACIANO A,RUEDA A J,FEITO F R.A formal framework for the representation of stack-based terrains[J].International Journal of Geographical Information Science,2018,32(10):1999-2022.
[19] LIU L,SUN Y,JI M,et al.Efficient construction of voxel models for ore bodies using an improved winding number algorithm and CUDA parallel computing[J].ISPRS International Journal of Geo-information,2023,12(12):473.DOI:10.3390/ijgi12120473.
[20] ZAMYADI A,POULIOT J,BÉDARD Y.Towards precise metadata-set for discovering 3D geospatial models in geo-portals[J].ISPRS Annals of the Photogrammetry Remote Sensing and Spatial Information Sciences,2013,II-2/W1:299-308.DOI:10.5194/isprsannals-II-2-W1-299-2013.
[21] DUNCAN E E,RAHMAN A A.A unified 3D spatial data model for surface and subsurface spatial objects[J].Ghana Mining Journal,2013,14:7-13.
[22] LI J,LIU P R,WANG X Y,et al.3D geological implicit modeling method of regular voxel splitting based on layered interpolation data[J].Scientific Reports,2022,12(1):13840.DOI:10.1038/s41598-022-17231-x.
[23] CHEN Q Y,LIU G,LI X C,et al.A corner-point-grid-based voxelization method for the complex geological structure model with folds[J].Journal of Visualization,2017,20(4):875-888.
基本信息:
DOI：
中图分类号:TP18;P628
引用信息:
[1]杨长义,尹崧宇,李俊昆,等.基于自适应八叉树的深度学习地质建模并行显式化采样方法[J].地理与地理信息科学,2025,41(06):1-6.
基金信息:
国家自然科学基金项目(42172327)