吉林大学学报(地球科学版) ›› 2021, Vol. 51 ›› Issue (5): 1597-1604.doi: 10.13278/j.cnki.jjuese.20200291

• 绿色岩土工程 • 上一篇    下一篇

基于岩样细观图像深度学习的岩性自动分类方法

熊越晗1, 刘东燕1, 刘东升2, 王艳磊2, 唐小山3   

  1. 1. 重庆大学土木工程学院, 重庆 400045;
    2. 重庆市地质矿产勘查开发局, 重庆 401121;
    3. 重庆科技学院建筑工程学院, 重庆 401331
  • 收稿日期:2020-12-03 出版日期:2021-09-26 发布日期:2021-09-29
  • 通讯作者: 刘东燕(1959-),男,教授,博士生导师,主要从事岩土工程、防灾减灾工程方面的研究,E-mail:liudy@cqu.edu.cn E-mail:liudy@cqu.edu.cn
  • 作者简介:熊越晗(1995-),男,硕士研究生,主要从事岩土材料智能识别方面的研究,E-mail:xyh@cqu.edu.cn
  • 基金资助:
    重庆市自然科学基金项目(博士后基金)(cstc2020jcyj-bsh0137);重庆市地质矿产勘查开发局科研项目(DKJ-2020DZJ-A-015)

Automatic Lithology Classification Method Based on Deep Learning of Rock Sample Meso-Image

Xiong Yuehan1, Liu Dongyan1, Liu Dongsheng2, Wang Yanlei2, Tang Xiaoshan3   

  1. 1. School of Civil Engineering, Chongqing University, Chongqing 400045, China;
    2. Chongqing Bureau of Geology and Minerals Exploration, Chongqing 401121, China;
    3. School of Civil Engineering and Architecture, Chongqing University of Science and Technology, Chongqing 401331, China
  • Received:2020-12-03 Online:2021-09-26 Published:2021-09-29
  • Supported by:
    Supported by the Natural Science Foundation of Chongqing (Postdoctoral Fund) (cstc2020jcyj-bsh0137) and the Scientific Research Project of Chongqing Bureau of Geology and Minerals Exploration (DKJ-2020DZJ-A-015)

摘要: 在现阶段的岩土工程中,通常采用人工识别的方法来判别岩样种类,不仅耗时长、专业性强,还易受主观因素影响,准确率不理想。随着计算机技术的发展,机器学习逐渐被应用于岩性的自动识别,开启了岩样分类的新路径。本文以重庆市主城区4种典型岩样(泥岩、砂质泥岩、泥质砂岩和砂岩)的细观图像为研究对象,基于Inception V3卷积网络模型和迁移学习算法,建立了岩样细观图像深度学习模型,并完成了训练学习。结果显示:模型在训练1 000次后,训练集中的分类准确率达到92.77%,验证集中的分类准确率为76.31%。其中,验证集中的砂岩识别准确率为97.28%,泥岩识别准确率为81.85%,泥质砂岩识别准确率为72.59%,砂质泥岩识别准确率为72.35%。与现有的机器学习方法相比,本识别模型不仅可以自动识别岩性极为相近的岩样,而且具有较好的识别准确率、鲁棒性和泛化能力。

关键词: 岩样细观图像, 深度学习, 岩性识别, 卷积神经网络, 自动分类

Abstract: In the current geotechnical engineering, manual identification methods are usually used to distinguish the types of rock samples, which are highly specialized, time-consuming,and susceptible to subjective factors, resulting in low accuracy. With the development of computer technology, machine learning is gradually applied to the automatic lithology identification, which opens up a new path for rock sample classification. Based on this, the authors took the meso-images of the four typical rock samples (mudstone, sandy mudstone, argillaceous sandstone, and sandstone) in the main urban area of Chongqing as the research object. Based on the Inception V3 convolutional network model and migration learning algorithm, a deep learning model of rock sample mesoscopic images was eatablished, and training and learning were completed. The results show that after 1 000 times training, the classification accuracy rate in the training set reaches 92.77%, and the classification accuracy rate in the verification set is 76.31%. Among them, the sandstone recognition accuracy rate is 97.28%, the mudstone recognition accuracy rate in the verification set is 81.85%, the argillaceous sandstone recognition accuracy rate is 72.59%, and the sandy mudstone recognition accuracy rate is 72.35%. Compared with the existing machine learning methods, this recognition model can automatically recognize the rock samples with very similar lithology, and has higher recognition accuracy, robustness, and generalization ability.

Key words: rock sample meso-image, deep learning, lithology identification, convolutional neural network, automatic classification

中图分类号: 

  • P588
[1] 陈建平, 李婧, 崔宁, 等.大数据背景下地质云的构建与应用[J]. 地质通报, 2015, 34(7):1260-1265. Chen Jianping, Li Jing, Cui Ning, et al. The Construction and Application of Geological Cloud Under the Big Data Background[J]. Geological Bulletin of China, 2015, 34(7):1260-1265.
[2] 肖克炎, 孙莉, 李楠, 等.大数据思维下的矿产资源评价[J]. 地质通报, 2015, 34(7):1266-1272. Xiao Keyan, Sun Li, Li Nan, et al. Mineral Resources Assessment Under the Thought of Big Data[J]. Geological Bulletin of China, 2015, 34(7):1266-1272.
[3] 严光生, 薛群威, 肖克炎, 等.地质调查大数据研究的主要问题分析[J]. 地质通报, 2015, 34(7):1273-1279. Yan Guangsheng, Xue Qunwei, Xiao Keyan, et al. An Analysis of Major Problems in Geological Survey Big Data[J]. Geological Bulletin of China, 2015, 34(7):1273-1279.
[4] 赵鹏大.大数据时代数字找矿与定量评价[J]. 地质通报, 2015, 34(7):1255-1259. Zhao Pengda. Digital Mineral Exploration and Quantitative Evaluation in the Big Data Age[J]. Geological Bulletin of China, 2015, 34(7):1255-1259.
[5] 刘延保, 曹树刚, 刘玉成.基于LS-SVM的岩石细观图像分析方法探讨[J]. 岩石力学与工程学报, 2008, 27(5):1059-1065. Liu Yanbao, Cao Shugang, Liu Yucheng. Discussion on Analytical Method for LS-SVM Based Mesoscopic Rock Images[J]. Chinese Journal of Rock Mechanics and Engineering, 2008, 27(5):1059-1065.
[6] Singh N, Singh T N, Tiwary A, et al. Textural Identification of Basaltic Rock Mass Using Image Processing and Neural Network[J]. Computational Geosciences, 2010, 14(2):301-310.
[7] Młynarczuk M, Górszczyk A, Ślipek B. The Application of Pattern Recognition in the Automatic Classification of Microscopic Rock Images[J]. Computers & Geosciences, 2013, 60(5):126-133.
[8] 郭超, 刘烨.多色彩空间下的岩石图像识别研究[J]. 科学技术与工程, 2014, 14(18):247-251, 255. Guo Chao, Liu Ye. Recognition of Rock Images Based on Multiple Color Spaces[J]. Science Technology and Engineering, 2014, 14(18):247-251, 255.
[9] 程国建, 杨静, 黄全舟, 等.基于概率神经网络的岩石薄片图像分类识别研究[J]. 科学技术与工程, 2013, 13(31):9231-9235. Cheng Guojian, Yang Jing, Huang Quanzhou, et al. Rock Image Classification Recognition Based on Probabilistic Neural Networks[J]. Science Technology and Engineering, 2013, 13(31):9231-9235.
[10] 刘烨, 程国建, 马微, 等.基于铸体薄片图像颜色空间与形态学梯度的岩石分类[J]. 中南大学学报(自然科学版), 2016, 47(7):2375-2382. Liu Ye, Cheng Guojian, Ma Wei, et al. Rock Classification Based on Features Form Color Space and Morphological Gradient of Rock Thin Section Image[J]. Journal of Central South University(Science and Technology), 2016, 47(7):2375-2382.
[11] 程国建, 刘丽婷.深度学习算法应用于岩石图像处理的可行性研究[J]. 软件导刊, 2016, 15(9):163-166. Cheng Guojian, Liu Liting. Feasibility Study of Deep Learning Algorithm Applied to Rock Image Processing[J]. Software Guide, 2016, 15(9):163-166.
[12] 张野, 李明超, 韩帅.基于岩石图像深度学习的岩性自动识别与分类方法[J]. 岩石学报, 2018, 34(2):333-342. Zhang Ye, Li Mingchao, Han Shuai. Automatic Identification and Classification in Lithology Based on Deep Learning in Rock Images[J]. Acta Petrologica Sinica, 2018, 34(2):333-342.
[13] 白林, 姚钰, 李双涛, 等.基于深度学习特征提取的岩石图像矿物成分分析[J]. 中国矿业, 2018, 27(7):178-182. Bai Lin, Yao Yu, Li Shuangtao, et al. Mineral Composition Analysis of Rock Image Based on Deep Learning Feature Extraction[J]. China Mining Magazine, 2018, 27(7):178-182.
[14] 白林, 魏昕, 刘禹, 等.基于VGG模型的岩石薄片图像识别[J]. 地质通报, 2019, 38(12):2053-2058. Bai Lin, Wei Xin, Liu Yu, et al. Rock Thin Section Image Recognition and Classification Based on VGG Model[J]. Geological Bulletin of China, 2019, 38(12):2053-2058.
[15] 王恒, 姜亚楠, 张欣, 等.基于梯度提升算法的岩性识别方法[J]. 吉林大学学报(地球科学版), 2021, 51(3):940-950. Wang Heng, Jiang Yanan, Zhang Xin, et al. Lithology Identification Method Based on Gradient Boosting Algorithm[J]. Journal of Jilin University(Earth Science Edition), 2021, 51(3):940-950.
[16] 刘云鹏, 郭春影, 秦明宽, 等.基于PCA-SVM算法对稀土元素与稀土判别指标耦合数据集的铀矿床分类[J]. 吉林大学学报(地球科学版), 2021, 51(3):723-733. Liu Yunpeng, Guo Chunying, Qin Mingkuan, et al. Classfication of Uranium Deposits Based on PCA-SVM Algorithm for Coupling Data Set of Rare Earth Elements and Rare Earth Discrimination Indexes[J]. Journal of Jilin University(Earth Science Edition), 2021, 51(3):723-733.
[17] Szegedy C, Liu W, Jia Y, et al. Going Deeper with Convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Boston:Hynes Convention Center, 2015:1-9.
[18] Szegedy C, Vanhoucke V, Ioffe S, et al. Rethinking the Inception Architecture for Computer Vision[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas:Caesars Palace, 2016:2818-2826.
[19] Lin M, Chen Q, Yan S C. Network in Network[EB/OL]. ArXiv Preprint, 2013:(2013-12-16)[2020-11-21]. arxiv.org/abs/1312.4400.
[20] 程国建, 范鹏召.基于深度信念网络的岩石粒度分类[J]. 西安石油大学学报(自然科学版), 2018, 33(3):107-112. Cheng Guojian, Fan Pengzhao. Analysis of Rock Granularity by Deep Belief Network[J]. Journal of Xi'an Shiyou University(Natural Science Edition), 2018, 33(3):107-112.
[21] 程国建, 马微, 魏新善, 等.基于图像处理与神经网络的岩石组构识别[J]. 西安石油大学学报(自然科学版), 2013, 28(5):105-110. Cheng Guojian, Ma Wei, Wei Xinshan, et al. Research of Rock Texture Identification Based on Image Processing and Neural Network[J]. Journal of Xi'an Shiyou University(Natural Science Edition), 2013, 28(5):105-110.
[1] 王恒, 姜亚楠, 张欣, 仲鸿儒, 陈庆轩, 高世臣. 基于梯度提升算法的岩性识别方法[J]. 吉林大学学报(地球科学版), 2021, 51(3): 940-950.
[2] 牟丹, 张丽春, 徐长玲. 3种经典机器学习算法在火山岩测井岩性识别中的对比[J]. 吉林大学学报(地球科学版), 2021, 51(3): 951-956.
[3] 王新民, 张超超. 基于深度学习的旧金山湾水质预测[J]. 吉林大学学报(地球科学版), 2021, 51(1): 222-230.
[4] 代丽艳, 董宏丽, 李学贵. 微地震数据去噪方法综述[J]. 吉林大学学报(地球科学版), 2019, 49(4): 1145-1159.
[5] 王英伟, 张建民, 王满, 潘保芝, 邢艳娟, 石丹红. 基于序贯指示模拟方法的火山岩储层岩性及孔隙度模拟[J]. J4, 2010, 40(2): 455-460.
[6] 张治国,杨毅恒,夏立显. RPROP算法在测井岩性识别中的应用[J]. J4, 2005, 35(03): 389-0393.
[7] 周波,李舟波,潘保芝. 火山岩岩性识别方法研究[J]. J4, 2005, 35(03): 394-0397.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 程立人,张予杰,张以春. 西藏申扎地区奥陶纪鹦鹉螺化石[J]. J4, 2005, 35(03): 273 -0282 .
[2] 陈 力,佴 磊,王秀范,李 金. 绥中某电力设备站场区地震危险性分析[J]. J4, 2005, 35(05): 641 -645 .
[3] 李斌,孟自芳,李相博,卢红选,郑民. 泌阳凹陷下第三系构造特征与沉积体系[J]. J4, 2005, 35(03): 332 -0339 .
[4] 赵宏光,孙景贵,陈军强,赵俊康,姚凤良,段 展. 延边小西南岔富金斑岩铜矿床的含矿流体起源与演化——H,O,C,S,Pb同位素示踪[J]. J4, 2005, 35(05): 601 -606 .
[5] 孟元林,高建军,刘德来,牛嘉玉,孙洪斌,周玥,肖丽华,王粤川. 辽河坳陷鸳鸯沟地区成岩相分析与异常高孔带预测[J]. J4, 2006, 36(02): 227 -0233 .
[6] 曾昭发,吴燕冈,郝立波,王者江,黄 航. 基于泊松定理的重磁异常分析方法及应用[J]. J4, 2006, 36(02): 279 -0283 .
[7] 常秋玲,卢欣祥,刘东华,李明立. 东秦岭五朵山花岗岩体及金矿关系探讨[J]. J4, 2006, 36(03): 319 -325 .
[8] 马艳梅,崔启良,周强,黄伟军,刘冶,彭刚,邹广田. 橄榄石原位高温拉曼光谱研究[J]. J4, 2006, 36(03): 342 -345 .
[9] 郝琦,刘震,查明,李春霞. 辽河茨榆坨潜山太古界裂缝型储层特征及其控制因素[J]. J4, 2006, 36(03): 384 -390 .
[10] 曾道明,纪宏金,陈 满,胡大千,朱永正. 胶东山城金矿地质与地球化学变量的关系[J]. J4, 2006, 36(04): 511 -515 .