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
文献类型：期刊论文
| 作者 | Zhang, Baoyi1; Xu, Kun1; Khan, Umair2 ; Li, Xuefeng3; Du, Linze1; Xu, Zhanghao1
|
| 刊名 | ORE GEOLOGY REVIEWS
 |
| 出版日期 | 2023-12-01 |
| 卷号 | 163页码:20 |
| 关键词 | 3D geological models Convolutional neural networks Deconvolution Weight of evidence 3D mineral prospectivity modeling |
| ISSN号 | 0169-1368 |
| DOI | 10.1016/j.oregeorev.2023.105788 |
| 通讯作者 | Khan, Umair(umairkhanbku@mails.ucas.ac.cn) |
| 英文摘要 | With the continuous exploitation of surface and shallow mineral resources, the global demand for concealed ore deposit exploration is increasing. However, concealed mineral prospectivity modeling (MPM) requires significant efforts, particularly in terms of methods and technologies for three-dimensional modeling. In this study, we propose a lightweight three-dimensional convolutional neural network (3D CNN) for MPM, which adopts the inception structure of GoogleNet and combines the idea of end-to-end learning. We replace the fully connected layer in the conventional CNN with deconvolution, which can greatly reduce the required parameters during the training process and accelerate the convergence. The proposed method overcomes the disadvantage that other shallow machine learning methods cannot extract spatial neighborhood information, while it can extract crosscorrelations among geological factors and generates less parameters by a lightweight network when facing massive data. Additionally, compared with the patch-wise method commonly used in previous studies, we use the pixel-wise method for end-to-end learning, which not only overcomes the drawbacks of random sampling but also considers the influence of each voxel when calculating the loss function. The three-dimensional multi-source geoscientific characteristics obtained from the geophysical inversion and 3D geological models are not discretized in order to promote effective CNN training while facilitating the ore-controlling representation. Comparing the predicted results between the 3D weight of evidence (WofE) and our proposed 3D CNN method for MPM, our proposed method and WofE delineated 100% of the known mineralization in high-favorability areas with voxel numbers of 70% and 95%, respectively. A case study of a structure-controlled hydrothermal gold deposit in the Sanhetun area of Heilongjiang Province demonstrates that the proposed 3D CNN method performs better than WofE in terms of prediction effectiveness and efficiency and effectively reveals the correlation between mineralization and adjacent ore-controlling characteristics. Moreover, the proposed 3D CNN method can simulate non-linear metallogenic processes and mine hidden relationships to reveal complex orecontrolling characteristics. In conclusion, the proposed 3D CNN method can reduce the exploration effort in 3D MPM, thus greatly improving the efficiency of discovering concealed ore deposits. |
| WOS关键词 | EPITHERMAL GOLD DEPOSIT ; GREAT XINGAN RANGE ; NE CHINA ; GEOCHEMICAL ANOMALIES ; 3D ; GEOCHRONOLOGY ; PREDICTION ; ROCKS ; FIELD ; ORE |
| 资助项目 | National Natural Science Foundation of China[42072326] ; China Geological Survey Work Project[DD20190156] ; National Engineering Research Center for Geographic Information System of China ; Central South University for providing MapGIS (R) software (Wuhan Zondy Cyber-Tech Co. Ltd., Wuhan, China) |
| WOS研究方向 | Geology ; Mineralogy ; Mining & Mineral Processing |
| 语种 | 英语 |
| WOS记录号 | WOS:001123971100001 |
| 出版者 | ELSEVIER |
| 资助机构 | National Natural Science Foundation of China ; China Geological Survey Work Project ; National Engineering Research Center for Geographic Information System of China ; Central South University for providing MapGIS (R) software (Wuhan Zondy Cyber-Tech Co. Ltd., Wuhan, China) |
| 源URL | [http://ir.idsse.ac.cn/handle/183446/10745]  |
| 专题 | 研究生部 深海科学研究部_深海地球物理与资源研究室 |
| 通讯作者 | Khan, Umair |
| 作者单位 | 1.Cent South Univ, Sch Geosci & Info Phys, Key Lab Metallogen Predict Nonferrous Met & Geol E, Minist Educ, Changsha 410083, Peoples R China 2.Chinese Acad Sci, Inst Deep Sea Sci & Engn, Sanya 572000, Peoples R China 3.Heilongjiang Inst Nat Resources Survey, Harbin 150036, Peoples R China |
| 推荐引用方式 GB/T 7714 | Zhang, Baoyi,Xu, Kun,Khan, Umair,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,2023,163:20. |
| APA | Zhang, Baoyi,Xu, Kun,Khan, Umair,Li, Xuefeng,Du, Linze,&Xu, Zhanghao.(2023).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.ORE GEOLOGY REVIEWS,163,20. |
| MLA | Zhang, Baoyi,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".ORE GEOLOGY REVIEWS 163(2023):20. |
入库方式： OAI收割
来源：深海科学与工程研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。