A machine learning-based lithologic mapping method
-
摘要: 通过野外地质调查与机器学习方法的有机融合,提出了一种基于梯度提升决策树算法的岩性单元填图方法。研究以多龙矿集区为模型试验区,选择1∶5万勘查地球化学数据为基础预测数据,以1∶5万区域地质图为参考,进行基于梯度提升决策树算法的岩性预测填图模型试验。首先选择研究区内小范围空白区开展野外填图,建立原始数据集并初步构建岩性单元与预测数据对应关系;其次利用机器学习方法对预测数据进行多分类任务,进而开展目标填图区预测填图工作;最后通过概率选区选定概率较低目标区,开展进一步的小范围野外地质调查填图,对原始数据和知识库进行补充,迭代循环以上流程,直至预测填图达到要求。试验显示,随着迭代次数的增加,模型精度不断提高,并在7次迭代后模型准确率达到87%。该方法强调在实际应用中野外地质调查与基于机器学习预测填图的深度融合,以及野外实地工作在整个流程中的重要性和不可或缺性;同时能够充分挖掘已有数据资料的有用信息,用于辅助修正已有岩性填图内容,或根据已勘探区资料对邻近的未勘探区进行岩性分类,有效减少野外填图工作量,是对岩性填图方法、地质单元定量预测识别的有益探索,为区域地质填图工作提供了新的参考思路和辅助手段。Abstract: In this study, a gradient boosting decision tree (GBDT)-based lithologic mapping method constituted by field survey and machine learning is introduced. The Duolong mineral district, Tibet, China is currently chosen for model test. During the practical application, geochemical data at a 1:50000 scale is analyzed to identify lithologic units, while a geological map at the same scale currently provides lithologic units identified by field survey. Lithologic units within a small area are firstly collected from the geological map. Correspondence between geochemical data and lithologic units within the small area can consequently be marked, by which the GBDT method is applied to reclassify the geochemical data and further predict lithologic units in the Duolong district. Transforming the result to a probability distribution, areas with low probability can be identified, and further investigation will be implemented to update geological knowledge and correspondence between geochemical and lithologic units. Iteration of the process will lead a reasonable lithologic mapping result. It is shown that the model accuracy increases with iteration growing, and reaches 87% after 7 iterations. The currently proposed method highlights deep integration of field survey and machine learning algorithm, and emphasizes importance of field work in the whole modeling process. Useful geo-information can be deeply mined from existing data and further updates former geological understandings. Meanwhile, lithologic units within un-explored areas can be identified based on the knowledge in explored areas. The GBDT-based method which effectively reduces field work is a meaningful exploration in lithologic mapping and will provide a new reference and supplementary way to geological mapping.
-
Key words:
- data mining /
- information fusion /
- geologic unit /
- decision tree /
- geological mapping
-
表 1 模型迭代性能统计表
Table 1. Performance of model iteration
迭代次数 宏平均精确率 宏平均召回率 宏平均F1分数 准确率 1 0.348 0.193 0.185 0.473 2 0.472 0.362 0.361 0.571 3 0.633 0.472 0.507 0.635 4 0.737 0.600 0.638 0.707 5 0.761 0.683 0.701 0.768 6 0.797 0.715 0.746 0.821 7 0.827 0.765 0.789 0.870 表 2 迭代分析结果信息统计表
Table 2. Statistics table of iteration results
迭代次数 面积占比 岩性种类数 1 0.088 13 2 0.171 17 3 0.262 18 4 0.357 19 5 0.445 19 6 0.531 19 7 0.622 19 表 3 模型分类精度表
Table 3. Table of classification accuracy of the current model
岩性单元 岩性符号 宏平均F1分数 流纹岩 λ53 0.880 第四系残坡积物 Q4 0.926 上第三系康托组棕红色粘土及砂砾石层 N1k 0.794 下白垩统美日切组上段火山角砾岩 K1m3 0.834 下白垩统美日切组中段火山碎屑岩 K1m2 0.810 下白垩统美日切组下段安山玢岩、安山质玄武岩 K1m1 0.843 中侏罗统色哇组二段:变石英砂岩、变长石石英砂岩夹深灰色粉砂质板岩 J2s2 0.909 中侏罗统色哇组一段:变长石石英砂岩砂、砾岩夹深灰色至深黑色变石英粉砂岩 J2s1 0.933 中侏罗统曲色组二段:变长石石英砂岩、粉砂岩、粉砂质板岩、夹硅质岩、灰绿色玄武岩、基性火山熔岩 J2q2 0.870 中侏罗统曲色组一段:深灰色粉砂质板岩夹变长石石英砂岩、灰岩条带及透镜体 J2q1 0.849 不明火山角砾岩铁帽 B53|Fe 0.839 褐红色、褐灰色安山岩 α53 0.739 浅绿灰色辉长岩 ν53 0.841 灰绿色辉绿岩 βμ53 0.683 灰绿色闪长岩、石英闪长岩 δ53 0.935 英安岩 ξ53 0.906 花岗闪长斑岩 γδπ53 0.756 绿帘石化玄武质安山岩 β53 0.882 蚀变体 SB 0.834 -
BIE X J, ZHANG T B, SUN C M, et al., 2013. Extraction of remote sensing anomaly and metallogenic prediction in Duolong ore-concentrated area of Tibet[J]. Journal of Guilin University of Technology, 33(2): 252-258. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTOTAL-GLGX201302009.htm CHEN H Q, QU X M, FAN S F, 2015. Geological characteristics and metallogenic-prospecting model of Duolong porphyry copper-gold ore concentration area in Gerze County, Tibet[J]. Mineral Deposits, 34(2): 321-332. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTOTAL-KCDZ201502008.htm CHEN S, CHEN C J, WU J, et al., 2017. Application and exploration of geophysical methods in geological mapping in strongly weathered area[J]. Journal of Geomechanics, 23(2): 206-213. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTOTAL-DZLX201702004.htm CRACKNELL M J, READING A M, 2014. Geological mapping using remote sensing data: A comparison of five machine learning algorithms, their response to variations in the spatial distribution of training data and the use of explicit spatial information[J]. Computers & Geosciences, 63: 22-33, doi:10.1016/j.cageo. 2013.10.008. DAI J J, WANG R J, QU X M, et al., 2013. Application of remote sensing alteration information in prospecting of Duolong ore concentration area in Tibet[J]. Acta Mineralogica Sinica, 33(S2): 753-754. (in Chinese) DUAN Y X, ZHAO Y S, MA C F, et al., 2020. Lithology identification method based on multi-layer ensemble learning[J]. Journal of Data Acquisition and Processing, 35(3): 572-581. (in Chinese with English abstract) FRIEDMAN J H, 2001. Greedy function approximation: a gradient boosting machine[J]. The Annals of Statistics, 29(5): 1189-1536. doi: 10.1214/aos/1013203450 FU J J, ZHAO Y Y, GUO S, 2014. Geochemical characteristics and significance of granodiorite porphyry in the Duolong ore concentration area, Tibet[J]. Acta Petrologica et Mineralogica, 33(6): 1039-1051. (in Chinese with English abstract) http://www.researchgate.net/publication/302925032_Geochemical_characteristics_and_significance_of_granodiorite_porphyry_in_the_Duolong_ore_concentration_area_Tibet GUO N, SHI W X, HUANG Y R, et al., 2018. Alteration mapping and prospecting model construction in the Tiegelongnan ore deposit of the Duolong ore concentration area, northern Tibet, based on shortwave infrared technique[J]. Geological Bulletin of China, 37(2-3): 446-457. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTotal-ZQYD2018Z1023.htm HARRIS J R, GRUNSKY E C, 2015. Predictive lithological mapping of Canada's North using Random Forest classification applied to geophysical and geochemical data[J]. Computers & Geosciences, 80: 9-25, doi: 10.1016/j.cageo.2015.03.013. HU J M, CHEN H, 2019. Innovation and exploration of the guiding ideology and method system of geological mapping in the coverage area-an overview of the results of the pilot project of mapping special geological and geomorphic areas[J]. Journal of Geomechanics, 25(5): 1001-1002. (in Chinese) JIANG S Q, SUN X G, YANG T Z, et al., 2014. Integrated anomaly model and metallogenic prediction of the Duolong porphyry copper-gold ore concentration area in northern Tibet[J]. Geology in China, 41(2): 497-509. (in Chinese with English abstract) http://www.cnki.com.cn/Article/CJFDTotal-DIZI201402014.htm KUHN S, CRACKNELL M J, Reading A M, 2018. Lithologic mapping using Random Forests applied to geophysical and remote-sensing data: A demonstration study from the Eastern Goldfields of Australia[J]. Geophysics, 83(4): B183-B193, doi: 10.1190/geo2017-0590.1. LI H M, 2017. Hydrothermal alteration mineral (groups) mapping and distribution characterization analysis based on multispectral remote sensing in Duolong ore concentration area, Tibetan[D]. Chengdu: Chengdu University of Technology. (in Chinese with English abstract) LI X K, LI C, WANG M, et al., 2018. Nature and evolution of crustal basement beneath the Duolong ore concentration area, northern Tibet, and their constraints on the metallogenesis: Insights from U-Pb ages of inherited zircons from the Bolong volcanic-intrusive rocks[J]. Geological Bulletin of China, 37(8): 1439-1449. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTotal-ZQYD201808009.htm LI Y Q, FEI G C, WEN C Q, et al., 2020. Characteristics of Ce4+/Ce3+ ratio and oxygen fugacity of Zircon in porphyry from Bolong and Duobuza porphyry Cu-Au deposits in Duolong ore district, Tibet[J]. Mineralogy and Petrology, 40(2): 59-70. (in Chinese with English abstract) LIU Z B, WANG W L, SONG Y, et al., 2017. Geo-information extraction and integration of ore-controlling structure in the Duolong ore concentration area of Tibet[J]. Acta Geoscientica Sinica, 38(5): 803-812. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTOTAL-DQXB201705019.htm OTHMAN A A, GLOAGUEN R, 2017. Integration of spectral, spatial and morphometric data into lithological mapping: A comparison of different Machine Learning Algorithms in the Kurdistan Region, NE Iraq[J]. Journal of Asian Earth Sciences, 146: 90-102, doi: 10.1016/j.jseaes.2017.05.005. QUINLAN J R, 1986. Induction of decision trees[J]. Machine Learning, 1(1): 81-106. http://onlinelibrary.wiley.com/resolve/reference/XREF?id=10.1007/BF00116251 REN J S, NIU B G, ZHAO L, et al, 2019. Basic ideas of the multisphere tectonics of earth system[J]. Journal of Geomechanics, 25(5): 607-612. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTotal-DZLX201905003.htm SHI H Z, LI Y C, HUANG H X, et al., 2019. Genesis of early Cretaceous Meiriqiecuo Formation volcanic rocks in the Duolong ore concentration area, southern margin of Qiangtang, Tibet, China[J]. Journal of Chengdu University of Technology (Science & Technology Edition), 46(4): 421-434. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTotal-CDLG201904003.htm SUN J, MAO J W, LIN B, et al., 2019. Comparison of ore geology and ore-forming processes of ore deposits (ore spots) in Duolong area, Tibet[J]. Mineral Deposits, 38(5): 1159-1184. (in Chinese with English abstract) SUN J, MAO J W, WANG J X, et al., 2020. Timing of Cu-Au mineralization in Nadun Cu-Au deposit of Duolong district, Tibet, and its implication for mineral exploration[J]. Mineral Deposits, 39(6): 1091-1102. (in Chinese with English abstract) WANG J B, 2018. Mineral assemblages mapping of porphyry copper deposits based on normalized multispectral remote sensing data in the Dulong ore concentrating area[D]. Chengdu: Chengdu University of Technology. (in Chinese with English abstract) WANG Q, LIN B, TANG J X, et al., 2018. Diagenesis, lithogenesis and geodynamic setting of intrusions in Senadong Area, Duolong district, Tibet[J]. Earth Science-Journal of China University of Geosciences, 43(4): 1125-1141. (in Chinese with English abstract) doi: 10.3799/dqkx.2017.613 WANG Q, TANG J X, CHEN Y C, et al., 2019. The metallogenic model and prospecting direction for the Duolong super large copper (gold) district, Tibet[J]. Acta Petrologica Sinica, 35(3): 879-896. (in Chinese with English abstract) doi: 10.18654/1000-0569/2019.03.16 WANG Q, TANG J X, FANG X, et al., 2015. Petrogenetic setting of andsites in Rongna ore block, Tiegelong Cu (Au-Ag) deposit, Duolong ore concentration area, Tibet: Evidence from zircon U-Pb LA-ICP-MS dating and petrogeochemistry of andsites[J]. Geology in China, 42(5): 1324-1336. (in Chinese with English abstract) http://www.researchgate.net/publication/288005950_Petrogenetic_setting_of_andsites_in_Rongna_ore_block_Tiegelong_Cu_Au-Ag_deposit_Duolong_ore_concentration_area_Tibet_Evidence_from_zircon_U-Pb_LA-ICP-MS_dating_and_petrogeochemistry_of_andsites WANG Z Y, ZUO R G, DONG Y N, 2020a. Mapping Himalayan leucogranites using a hybrid method of metric learning and support vector machine[J]. Computers & Geosciences, 138: 104455, doi: 10.1016/j.cageo.2020.104455. WANG Z Y, ZUO R G, JING L H, 2020b. Fusion of geochemical and remote-sensing data for lithological mapping using random forest metric learning[J]. Mathematical Geosciences, doi: 10.1007/s11004-020-09897-8. WEI S G, SONG Y, TANG J X, et al., 2019. Geochemistry, Si-O isotopic compositions and its tectonic significance of the siliceous rocks in the Duolong deposit, Tibet[J]. Acta Geologica Sinica, 93(2): 428-439. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTotal-DZXE201902010.htm WEI S G, TANG J X, SONG Y, et al., 2017. Zircons LA-MC-ICP-MS U-Pb Ages, Petrochemical, petrological and its significance of the potassic monzonitic granite porphyry from the Duolong Ore-concentrated district, Gaize County, Xizang (Tibet)[J]. Geological Review, 63(1): 189-206. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTOTAL-DZLP201701021.htm WU G P, CHEN G X, CHENG Q M, et al., 2021. Unsupervised machine learning for lithological mapping using geochemical data in covered areas of Jining, China[J]. Natural Resources Research, 30(2): 1053-1068, doi: 10.1007/s11053-020-09788-z. WU J, BU J J, XIE G G, et al., 2016. Application of regional geochemical data in geological mapping in strongly weathered area in southern China[J]. Journal of Geomechanics, 22(4): 955-966. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTOTAL-DZLX201604013.htm YAN H W, LIU J, TIAN Y, 2017. Magnetic characteristics and airborne radioactive anomaly characteristics of the shallow coverage in Baiyintuga area in Inner Mongolia[J]. Modern Mining, 33(3): 35-39, 45. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTOTAL-KYKB201703006.htm YANG X C, YE M N, YE P S, et al, 2020. Information construction method of geological survey projects based on digital mapping technology[J]. Journal of Geomechanics, 26(2): 263-270. (in Chinese with English abstract) YANG H H, SONG Y, DILLES J H, et al., 2019. The thermal-tectonic history of the Duolong ore district: evidence from apatite (U-Th)[J]. Acta Petrologica Sinica, 35(3): 867-878. (in Chinese with English abstract) http://www.researchgate.net/publication/333191639_The_thermal-tectonic_history_of_the_Duolong_ore_district_Evidence_from_apatite_U-ThHe_dating ZHANG X G, LI Y C, SUN R B, 2020. State, features, trends and enlightenment of geological mapping in major countries of the world[J]. Mineral Exploration, 11(2): 301-310. (in Chinese with English abstract) ZHANG Y, SUN J, YU C C, et al., 2019. Classification of Quaternary Coverings in desert grassland shallow cover area based on multi-source remote sensing data: a case of 1: 50000 pilot geological mapping in Qigandianzi, Inner Mongolia[J]. Geological Science and Technology Information, 38(2): 281-290. (in Chinese with English abstract) http://en.cnki.com.cn/Article_en/CJFDTotal-DZKQ201902034.htm ZHAO Z O, QIAO D H, ZHAO Y Y, 2020. Alteration mineralogical and geochemical features of the Rongna deposit in Duolong mining district of Tibet and their deep prospecting significances[J]. Acta Petrologica Sinica, 36(9): 2785-2798. (in Chinese with English abstract) doi: 10.18654/1000-0569/2020.09.11 ZHENG Y, 2017. Research on lithology recognition based on deep learning[D]. Beijing: China University of Petroleum (Beijing). (in Chinese with English abstract) ZHU M Y, LI B Q, FU H Z, et al., 2020. SVM lithological classification based on multi-source data collaboration: a case study in Jianggalesayi area[J]. Uranium Geology, 36(4): 288-292, 317. (in Chinese with English abstract) 别小娟, 张廷斌, 孙传敏, 等, 2013. 西藏多龙矿集区遥感异常提取与成矿预测[J]. 桂林理工大学学报, 33(2): 252-258. doi: 10.3969/j.issn.1674-9057.2013.02.009 陈红旗, 曲晓明, 范淑芳, 2015. 西藏改则县多龙矿集区斑岩型铜金矿床的地质特征与成矿-找矿模型[J]. 矿床地质, 34(2): 321-332. https://www.cnki.com.cn/Article/CJFDTOTAL-KCDZ201502008.htm 陈松, 陈长敬, 吴俊, 等, 2017. 物探方法在强风化区填图中的应用探索[J]. 地质力学学报, 23(2): 206-213. doi: 10.3969/j.issn.1006-6616.2017.02.003 代晶晶, 王瑞江, 曲晓明, 等, 2013. 遥感蚀变信息在西藏多龙矿集区找矿中的应用研究[J]. 矿物学报, 33(S2): 753-754. https://www.cnki.com.cn/Article/CJFDTOTAL-KWXB2013S2421.htm 段友祥, 赵云山, 马存飞, 等, 2020. 基于多层集成学习的岩性识别方法[J]. 数据采集与处理, 35(3): 572-581. https://www.cnki.com.cn/Article/CJFDTOTAL-SJCJ202003020.htm 符家骏, 赵元艺, 郭硕, 2014. 西藏多龙矿集区花岗闪长斑岩地球化学特征及其意义[J]. 岩石矿物学杂志, 33(6): 1039-1051. doi: 10.3969/j.issn.1000-6524.2014.06.004 郭娜, 史维鑫, 黄一入, 等, 2018. 基于短波红外技术的西藏多龙矿集区铁格隆南矿床荣那矿段及其外围蚀变填图-勘查模型构建[J]. 地质通报, 37(2-3): 446-457. https://www.cnki.com.cn/Article/CJFDTOTAL-ZQYD2018Z1023.htm 胡健民, 陈虹, 2019. 覆盖区区域地质填图指导思想与方法体系的创新与探索: 特殊地质地貌区填图试点项目成果概述[J]. 地质力学学报, 25(5): 1001-1002. https://www.cnki.com.cn/Article/CJFDTOTAL-DZLX201905027.htm 江少卿, 孙兴国, 杨铁铮, 等, 2014. 藏北多龙斑岩铜金矿集区综合信息找矿模型研究[J]. 中国地质, 41(2): 497-509. doi: 10.3969/j.issn.1000-3657.2014.02.014 李红梅, 2017. 多龙矿集区遥感蚀变矿物(组合)提取及蚀变分带特征研究[D]. 成都: 成都理工大学. 李兴奎, 李才, 王明, 等, 2018. 藏北多龙矿集区地壳基底性质、演化及其对成矿的制约: 来自波龙火山-侵入岩中继承锆石U-Pb年龄的信息[J]. 地质通报, 37(8): 1439-1449. https://www.cnki.com.cn/Article/CJFDTOTAL-ZQYD201808009.htm 李云强, 费光春, 温春齐, 等, 2020. 西藏多龙矿集区波龙、多不杂斑岩铜金矿床岩体锆石Ce4+/Ce3+比值及氧逸度特征[J]. 矿物岩石, 40(2): 59-70. https://www.cnki.com.cn/Article/CJFDTOTAL-KWYS202002006.htm 刘治博, 王文磊, 宋扬, 等, 2017. 多龙矿集区控矿构造信息提取、识别与融合[J]. 地球学报, 38(5): 803-812. https://www.cnki.com.cn/Article/CJFDTOTAL-DQXB201705019.htm 任纪舜, 牛宝贵, 赵磊, 等, 2019. 地球系统多圈层构造观的基本内涵[J]. 地质力学学报, 25(5): 607-612. https://www.cnki.com.cn/Article/CJFDTOTAL-DZLX201905003.htm 石洪召, 李玉昌, 黄瀚霄, 等, 2019. 西藏多龙矿集区早白垩世美日切错组火山岩成因[J]. 成都理工大学学报(自然科学版), 46(4): 421-434. doi: 10.3969/j.issn.1671-9727.2019.04.03 孙嘉, 毛景文, 林彬, 等, 2019. 西藏多龙矿集区典型矿床(点)矿化特征与成矿作用对比研究[J]. 矿床地质, 38(5): 1159-1184. https://www.cnki.com.cn/Article/CJFDTOTAL-KCDZ201905014.htm 孙嘉, 毛景文, 王佳新, 等, 2020. 西藏多龙矿集区拿顿铜金矿床成矿时代的厘定及其找矿指示意义[J]. 矿床地质, 39(6): 1091-1102. 王继斌, 2018. 基于归一化多光谱遥感数据的多龙矿集区斑岩铜矿蚀变矿物组合提取[D]. 成都: 成都理工大学. 王勤, 唐菊兴, 方向, 等, 2015. 西藏多龙矿集区铁格隆南铜(金银)矿床荣那矿段安山岩成岩背景: 来自锆石U-Pb年代学、岩石地球化学的证据[J]. 中国地质, 42(5): 1324-1336. doi: 10.3969/j.issn.1000-3657.2015.05.011 王勤, 林彬, 唐菊兴, 等, 2018. 多龙矿集区色那东岩体年龄、成因与动力学背景[J]. 地球科学-中国地质大学学报, 43(4): 1125-1141. https://www.cnki.com.cn/Article/CJFDTOTAL-DQKX201804013.htm 王勤, 唐菊兴, 陈毓川, 等, 2019. 西藏多龙超大型铜(金)矿集区成矿模式与找矿方向[J]. 岩石学报, 35(3): 879-896. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXB201903016.htm 韦少港, 唐菊兴, 宋扬, 等, 2017. 西藏改则多龙矿集区地堡那木岗矿床钾玄质二长花岗斑岩锆石LA-MC-ICP-MS U-Pb年龄、地球化学特征及其地质意义[J]. 地质论评, 63(1): 189-206. https://www.cnki.com.cn/Article/CJFDTOTAL-DZLP201701021.htm 韦少港, 宋扬, 唐菊兴, 等, 2019. 西藏多龙矿集区硅质岩岩石地球化学、Si-O同位素特征及其构造意义[J]. 地质学报, 93(2): 428-439. doi: 10.3969/j.issn.0001-5717.2019.02.011 吴俊, 卜建军, 谢国刚, 等, 2016. 区域化探数据在华南强烈风化区地质填图中的应用[J]. 地质力学学报, 22(4): 955-966. doi: 10.3969/j.issn.1006-6616.2016.04.013 严昊伟, 刘君, 田野, 2017. 内蒙古白音图嘎浅覆盖区地层磁性及航空放射性异常特征[J]. 现代矿业, 33(3): 35-39, 45. https://www.cnki.com.cn/Article/CJFDTOTAL-KYKB201703006.htm 杨星辰, 叶梦旎, 叶培盛, 等, 2020. 地质调查成果信息化建设方法探索: 基于数字填图技术[J]. 地质力学学报, 26(2): 263-270. https://www.cnki.com.cn/Article/CJFDTOTAL-DZLX202002011.htm 杨欢欢, 宋扬, DILLES J H, et al., 2019. 西藏多龙矿集区热构造演化历史: 来自磷灰石(U-Th)/He的证据[J]. 岩石学报, 35(3): 867-878. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXB201903015.htm 张鑫刚, 李仰春, 孙仁斌, 2020. 世界主要国家地质填图现状、特点、趋势及启示[J]. 矿产勘查, 11(2): 301-310. https://www.cnki.com.cn/Article/CJFDTOTAL-YSJS202002016.htm 张艳, 孙杰, 于长春, 等, 2019. 基于多源遥感数据的第四系覆盖物分类方法研究: 以内蒙古旗杆甸子幅1: 5万填图试点为例[J]. 地质科技情报, 38(2): 281-290. https://www.cnki.com.cn/Article/CJFDTOTAL-DZKQ201902034.htm 赵子欧, 乔东海, 赵元艺, 2020. 西藏多龙矿集区荣那铜金矿床蚀变矿物学和地球化学及找矿意义[J]. 岩石学报, 36(9): 2785-2798. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXB202009012.htm 郑阳, 2017. 基于深度学习的岩性识别研究[D]. 北京: 中国石油大学(北京). 朱明永, 李炳谦, 付翰泽, 等, 2020. 基于多源数据协同的SVM岩性分类研究: 以江尕勒萨依地区为例[J]. 铀矿地质, 36(4): 288-292, 317. doi: 10.3969/j.issn.1000-0658.2020.04.007