融合图像处理与深度学习的亮晶颗粒灰岩岩相学分析应用

余晓露; 李龙龙; 蒋宏; 卢龙飞; 杜崇娇

doi:10.11781/sysydz2023051026

融合图像处理与深度学习的亮晶颗粒灰岩岩相学分析应用

doi: 10.11781/sysydz2023051026

余晓露^{1, 2,},
李龙龙^{1, 2, ,},
蒋宏^{1, 2},
卢龙飞^{1, 2},
杜崇娇^{1, 2}

1.
中国石化油气成藏重点实验室，江苏无锡 214126
2.
中国石化石油勘探开发研究院无锡石油地质研究所，江苏无锡 214126

基金项目:

中国石化优秀青年科技创新项目“岩石(矿物)自动化鉴定分析仪” P19028

详细信息

作者简介:
余晓露(1983—)，女，硕士，高级工程师，从事岩石薄片智能化自动鉴定研究。E-mail: yuxl.syky@sinopec.com

通讯作者:
李龙龙(1994—)，男，硕士，助理研究员，从事油气地质研究。E-mail: lilongl.syky@sinopec.com

中图分类号: TE135
计量
- 文章访问数: 339
- HTML全文浏览量: 107
- PDF下载量: 69
- 被引次数: 0
出版历程
- 收稿日期: 2023-05-25
- 修回日期: 2023-08-11
- 刊出日期: 2023-09-28

Application of sparry grain limestone petrographic analysis combining image processing and deep learning

YU Xiaolu^{1, 2
,},
LI Longlong^{1, 2
, ,},
JIANG Hong^{1, 2},
LU Longfei^{1, 2},
DU Chongjiao^{1, 2}

1.
SINOPEC Key Laboratory of Petroleum Accumulation Mechanisms, Wuxi, Jiangsu 214126, China
2.
Wuxi Research Institute of Petroleum Geology, Wuxi, Jiangsu 214126, China

摘要

摘要: 针对传统碳酸盐岩薄片鉴定基于肉眼观察和描述，存在主观性强、定性评价为主、定量困难等问题，以亮晶颗粒灰岩为对象，设计了涵盖流程与技术的智能化岩石薄片图像信息挖掘模型。通过岩相学分析框架构建了岩相特征与薄片图像之间的映射关系。融合图像处理和深度学习设计了全流程的特征提取算法。通过卷积神经网络获得结构组分特征中对颗粒类型的定性识别，即基于改进的ResNet50模型划分颗粒所属类别(内碎屑、生物碎屑、包粒、球粒和团块)。通过数字图像处理技术获得结构组分特征中对颗粒含量、粒径、形状、接触方式的定量识别，即基于阈值分割计算颗粒含量，基于最小外接圆/最小外接矩形，结合面积比/长宽比、交并比(IoU)等算法计算颗粒形态学参数，并通过对染色图像的HSV色彩空间处理获得矿物组分特征中对方解石和其他矿物组分的定性和定量识别。以顺X井亮晶颗粒灰岩薄片样品为例，通过完整的图像识别过程验证了各个特征点提取算法的有效性，并与人工鉴定报告进行对比。岩相学分析框架能够有效地表征亮晶颗粒灰岩中的有意义信息。通过岩相学分析框架结合图像分析算法的模式，实现了对这一类碳酸盐岩的规范化流程化智能鉴定，为岩石薄片图像智能识别研究提供有效的方法支撑。
- 碳酸盐岩 /
- 岩相学 /
- 卷积神经网络 /
- 深度学习 /
- 人工智能
Abstract: Traditional carbonate rock slice identification is based on manual observation and description, which is highly subjective, mainly qualitative and difficult to quantify. In this paper, an intelligent rock thin-section image information mining model covering the process and technology is designed. The mapping relationship between lithofacies characteristics and thin slice images is constructed through a petrographic analysis framework. A full-flow feature extraction algorithm is also designed by combing image processing and deep learning. Qualitative recognition of grain type in structural component features is obtained by convolutional neural network, that is, the classification of grains (intraclasts, bioclasts, envelopes, spherulites and agglomerates) based on the improved ResNet50 model. Quantitative recognition of grain content, size, shape and contact mode in structural component features are obtained by digital image processing, that is, grain content is calculated based on threshold segmentation, grain morphology parameters are calculated based on minimum peripheral circle/minimum peripheral rectangle and by combining with area ratio/aspect ratio, intersection ratio (IoU) and other algorithms. And the qualitative and quantitative identification of calcite and other minerals in mineral component features are obtained by HSV colour space processing of the stained images. A thin section example from Shun X well is given as an example, and the validity of each feature extraction algorithm is verified through the complete image recognition process and comparison with the manual identification report. The results show that the petrographic analysis framework is effective in representing meaningful information in sparry grain limestone. Through the model of combing petrographic analysis framework with image analysis algorithm, a standard and intelligent identification of this type of carbonate rocks has been achieved, which can provide effective method support for the study of intelligent identification of rock slice images.
- carbonate rock /
- petrography /
- convolutional neural network /
- deep learning /
- artificial intelligence
注释:

利益冲突声明/Conflict of Interests

所有作者声明不存在利益冲突。

All authors disclose no relevant conflict of interests.

作者贡献/Authors’Contributions

余晓露参与实验设计；杜崇娇、蒋宏完成实验操作；余晓露、李龙龙、卢龙飞参与论文写作和修改。所有作者均阅读并同意最终稿件的提交。

The study was designed by YU Xiaolu. The experimental operation was completed by DU Chongjiao and JIANG Hong. The manuscript was drafted and revised by LI Longlong and LU Longfei. All the authors have read the last version of paper and consented for submission.

HTML全文

图 1 基于图像的亮晶颗粒灰岩岩相学分析框架

Figure 1. Petrographic analysis framework of sparry grain limestone based on image

下载: 全尺寸图片幻灯片

图 2 碳酸盐岩常见颗粒类型示例及描述

Figure 2. Examples and descriptions of common grain types in carbonate rocks

下载: 全尺寸图片幻灯片

图 3 改进的ResNet50卷积神经网络结构

Figure 3. Improved structure diagram of ResNet50 convolutional neural network

下载: 全尺寸图片幻灯片

图 4 颗粒形状确定方法示意

Figure 4. Schematic diagram of grain shape determination method

下载: 全尺寸图片幻灯片

图 5 亮晶颗粒灰岩图像结构组分特征提取过程示例

Figure 5. Example of structural component features extraction process of sparry grain limestone images

下载: 全尺寸图片幻灯片

图 6 亮晶颗粒灰岩图像矿物组分特征提取过程示例

Figure 6. Example of mineral component features extraction process of sparry grain limestone images

下载: 全尺寸图片幻灯片

图 7 岩矿鉴定人脑活动层次图示例

Figure 7. Example of human brain activity for rock and mineral identification

下载: 全尺寸图片幻灯片

表 1 卷积神经网络各模型特征对比

Table 1. Comparison of features of convolutional neural network models

模型名称	特点
ResNet	(1)网络深度更深，不会出现梯度消失现象，解决了深层次的网络退化问题； (2)由于使用更深的网络，分类准确率提升。
DenseNet	(1)采用密集连接方式，提升了梯度的反向传播，使得网络更容易训练； (2)参数更小且计算更高效； (3)由于特征复用，最后的分类器使用了低级特征。
InceptionNet	(1)在同一层网络中使用多个尺寸的卷积核来提升模型感知力； (2)使用批标准化来缓解梯度消失现象。
MobileNet	将传统结构改造成两层卷积结构，轻量级网络，计算量更少，精度更高。

下载: 导出CSV

表 2 亮晶颗粒灰岩人工鉴定与图像分析结果对比

Table 2. Comparison of manual identification and image analysis results of sparry grain limestone

项目	结构组分特征					矿物组分特征
项目	颗粒类型	颗粒含量/%	颗粒粒径/mm	颗粒形状	颗粒接触方式	方解石含量/%
人工鉴定	砂屑、鲕粒	80	0.1~0.35	圆、椭圆及不规则状	无接触—点接触	97
图像分析	内碎屑、包粒	71.1	0.14~0.37	圆、次圆及椭圆状	点接触	95.4

下载: 导出CSV

参考文献(25)

[1]	刘宝珺. 沉积岩石学[M]. 北京: 地质出版社, 1980. LIU Baojun. Sedimentary rocks[M]. Beijing: Geology Press, 1980.
[2]	MOORE C H. Carbonate reservoirs: porosity, evolution and diagenesis in a sequence stratigraphic framework[M]. Amsterdam: Elsevier, 2001.
[3]	FLÜGELE. Microfacies of carbonate rocks: analysis, interpretation and application[M]. Berlin: Springer, 2010.
[4]	LIU Jingtuo, DENG Yafeng, BAI Tao, et al. Targeting ultimate accuracy: face recognition via deep embedding[EB/OL]. 2015[2023-06-28]. https://arxiv.org/abs/1506.07310.
[5]	陈利. 基于深度学习的车牌识别系统设计[J]. 计算机技术与发展, 2018, 28(6): 85-89. doi: 10.3969/j.issn.1673-629X.2018.06.019 CHEN Li. Design of license plate recognition system based on deep learning[J]. Computer Technology and Development, 2018, 28(6): 85-89. doi: 10.3969/j.issn.1673-629X.2018.06.019
[6]	ESTEVA A, KUPREL B, NOVOA R A, et al. Dermatologist-level classification of skin cancer with deep neural networks[J]. Nature, 2017, 542(7639): 115-118. doi: 10.1038/nature21056
[7]	张新钰, 高洪波, 赵建辉, 等. 基于深度学习的自动驾驶技术综述[J]. 清华大学学报(自然科学版), 2018, 58(4): 438-444. https://www.cnki.com.cn/Article/CJFDTOTAL-QHXB201804017.htm ZHANG Xinyu, GAO Hongbo, ZHAO Jianhui, et al. Overview of deep learning intelligent driving methods[J]. Journal of Tsinghua University (Science and Technology), 2018, 58(4): 438-444. https://www.cnki.com.cn/Article/CJFDTOTAL-QHXB201804017.htm
[8]	PERRING C S, BARNES S J, VERRALL M, et al. Using automated digital image analysis to provide quantitative petrographic data on olivine-phyric basalts[J]. Computers & Geosciences, 2004, 30(2): 183-195.
[9]	叶润青, 牛瑞卿, 张良培. 基于多尺度分割的岩石图像矿物特征提取及分析[J]. 吉林大学学报(地球科学版), 2011, 41(4): 1253-1261. https://www.cnki.com.cn/Article/CJFDTOTAL-CCDZ201104043.htm YE Runqing, NIU Ruiqing, ZHANG Liangpei. Mineral features extraction and analysis based on multiresolution segmentation of petrographic image[J]. Journal of Jilin University (Earth Science Edition), 2011, 41(4): 1253-1261. https://www.cnki.com.cn/Article/CJFDTOTAL-CCDZ201104043.htm
[10]	刘春, 许强, 施斌, 等. 岩石颗粒与孔隙系统数字图像识别方法及应用[J]. 岩土工程学报, 2018, 40(5): 925-931. https://www.cnki.com.cn/Article/CJFDTOTAL-YTGC201805022.htm LIU Chun, XU Qiang, SHI Bin, et al. Digital image recognition method of rock particle and pore system and its application[J]. Chinese Journal of Geotechnical Engineering, 2018, 40(5): 925-931. https://www.cnki.com.cn/Article/CJFDTOTAL-YTGC201805022.htm
[11]	张吉群, 胡长军, 和冬梅, 等. 孔隙结构图像分析方法及其在岩石图像中的应用[J]. 测井技术, 2015, 39(5): 550-554. https://www.cnki.com.cn/Article/CJFDTOTAL-CJJS201505003.htm ZHANG Jiqun, HU Changjun, HE Dongmei, et al. Image analysis method of pore structure and its application in rock image[J]. Well Logging Technology, 2015, 39(5): 550-554. https://www.cnki.com.cn/Article/CJFDTOTAL-CJJS201505003.htm
[12]	胡祺. 融合多维信息的岩石薄片图像深度学习分类方法[D]. 杭州: 浙江大学, 2019. HU Qi. Thinsection image classification method using deep learning of integrated multi-dimensional information[D]. Hangzhou: Zhejiang University, 2019.
[13]	郝慧珍, 顾庆, 胡修棉. 基于机器学习的矿物智能识别方法研究进展与展望[J]. 地球科学, 2021, 46(9): 3091-3106. https://www.cnki.com.cn/Article/CJFDTOTAL-DQKX202109005.htm HAO Huizhen, GU Qing, HU Xiumian. Research advances and prospective in mineral intelligent identification based on machine learning[J]. Earth Science, 2021, 46(9): 3091-3106. https://www.cnki.com.cn/Article/CJFDTOTAL-DQKX202109005.htm
[14]	余晓露, 叶恺, 杜崇娇, 等. 基于卷积神经网络的碳酸盐岩生物化石显微图像识别[J]. 石油实验地质, 2021, 43(5): 880-885. doi: 10.11781/sysydz202105880 YU Xiaolu, YE Kai, DU Chongjiao, et al. Microscopic recognition of micro fossils in carbonate rocks based on convolutional neural network[J]. Petroleum Geology & Experiment, 2021, 43(5): 880-885. doi: 10.11781/sysydz202105880
[15]	张涛, 雷丹博, 王宾, 等. 陕南寒武系底部宽川铺组微体化石人工智能识别[J]. 古生物学报, 2019, 58(2): 141-151. https://www.cnki.com.cn/Article/CJFDTOTAL-GSWX201902001.htm ZHANG Tao, LEI Danbo, WANG Bin, et al. Artificial intelligence identification of microfossils from the Lower Cambrian Kuchuanpu Formation in southern Shaanxi, China[J]. Acta Palaeontologica Sinica, 2019, 58(2): 141-151. https://www.cnki.com.cn/Article/CJFDTOTAL-GSWX201902001.htm
[16]	姜枫, 顾庆, 郝慧珍, 等. 基于语义特征提取的砂岩薄片图像颗粒分割方法[J]. 中国科学: 信息科学, 2020, 50(1): 109-127. https://www.cnki.com.cn/Article/CJFDTOTAL-PZKX202001005.htm JIANG Feng, GU Qing, HAO Huizhen, et al. Grain segmentation of sandstone thin section images based on semantic feature extraction[J]. Science China Information Sciences, 2020, 50(1): 109-127. https://www.cnki.com.cn/Article/CJFDTOTAL-PZKX202001005.htm
[17]	RUBO R A, DE CARVALHO CARNEIRO C, MICHELON M F, et al. Digital petrography: mineralogy and porosity identification using machine learning algorithms in petrographic thin section images[J]. Journal of Petroleum Science and Engineering, 2019, 183: 106382.
[18]	KOESHIDAYATULLAH A, MORSILLI M, LEHRMANN D J, et al. Fully automated carbonate petrography using deep convolutional neural networks[J]. Marine and Petroleum Geology, 2020, 122: 104687.
[19]	IZADI H, SADRI J, MEHRAN N A. A new intelligent method for minerals segmentation in thin sections based on a novel incremental color clustering[J]. Computers & Geosciences, 2015, 81: 38-52.
[20]	国家能源局. 岩石薄片鉴定: SY/T 5368-2016[S]. 北京: 石油工业出版社, 2017. National Energy Administration. Identification for thin section of rocks: SY/T 5368-2016[S]. Beijing: Petroleum Industry Press, 2017.
[21]	GONZALEZ R C, WOODS R E. Digital image processing[M]. 3rd ed. A Wiley-Interscience Publication, 2010.
[22]	张顺, 龚怡宏, 王进军. 深度卷积神经网络的发展及其在计算机视觉领域的应用[J]. 计算机学报, 2019, 42(3): 453-482. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJX201903001.htm ZHANG Shun, GONG Yihong, WANG Jinjun. The development of deep convolution neural network and its applications on computer vision[J]. Chinese Journal of Computers, 2019, 42(3): 453-482. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJX201903001.htm
[23]	隋微波, 程思. 基于卷积神经网络的砂岩数字岩心绝对渗透率计算方法[J]. 油气地质与采收率, 2022, 29(1): 128-136. https://www.cnki.com.cn/Article/CJFDTOTAL-YQCS202201016.htm SUI Weibo, CHENG Si. Calculation methods for absolute permeability of sandstone digital cores based on convolutional neural networks[J]. Petroleum Geology and Recovery Efficiency, 2022, 29(1): 128-136. https://www.cnki.com.cn/Article/CJFDTOTAL-YQCS202201016.htm
[24]	KRUMBEIN W C, SLOSS L L. Stratigraphy and sedimentation[M]. San Francisco: Freeman, 1963: 676.
[25]	孙世强, 左海维, 赵露婷. 联合特征相似性度量和交并比的检测框优选研究[J]. 电脑知识与技术, 2019, 15(29): 190-193. https://www.cnki.com.cn/Article/CJFDTOTAL-DNZS201929085.htm SUN Shiqiang, ZUO Haiwei, ZHAO Luting. Research on detection box optimization of joint feature similarity measurement and intersection over union[J]. Computer Knowledge and Technology, 2019, 15(29): 190-193. https://www.cnki.com.cn/Article/CJFDTOTAL-DNZS201929085.htm