基于卷积神经网络的碳酸盐岩生物化石显微图像识别

余晓露; 叶恺; 杜崇娇; 宫晗凝; 马中良

doi:10.11781/sysydz202105880

基于卷积神经网络的碳酸盐岩生物化石显微图像识别

doi: 10.11781/sysydz202105880

余晓露^{1, 2,},
叶恺^2, ,,
杜崇娇²,
宫晗凝^{1, 2},
马中良^{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

国家自然科学基金 42072156

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化鉴定研究。E-mail: yuxl.syky@sinopec.com

通讯作者:
叶恺(1984-), 男, 工程师, 从事信息化管理、平台搭建及软件开发研究。E-mail: yekai.syky@sinopec.com

中图分类号: TE135
计量
- 文章访问数: 947
- HTML全文浏览量: 266
- PDF下载量: 125
- 被引次数: 0
出版历程
- 收稿日期: 2021-06-06
- 修回日期: 2021-08-31
- 刊出日期: 2021-09-28

Microscopic recognition of micro fossils in carbonate rocks based on convolutional neural network

YU Xiaolu^{1, 2
,},
YE Kai^{2
, ,},
DU Chongjiao²,
GONG Hanning^{1, 2},
MA Zhongliang^{1, 2}

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

摘要

摘要: 碳酸盐岩薄片中的生物化石识别对判断沉积环境研究具有重要的意义，但传统的人工鉴定方法对经验要求高，受主观影响较大。该文提出一种基于ResNet卷积神经网络的碳酸盐岩生物化石显微图像识别方法，通过图像预处理、设计模型、训练模型等步骤，实现了薄片图像中生物化石的智能识别，识别准确率为86%；并同时提出进阶YOLO（You Only Look Once）目标检测模型，可实现薄片图像中生物化石所在区域的检测和识别，识别准确率为85%。该方法验证了使用数字图像处理和深度学习方法对碳酸盐岩生物化石显微图像进行智能识别的可行性，可作为传统人工鉴定方法的有益补充，具有一定的实际应用价值。
- 卷积神经网络 /
- ResNet /
- YOLO /
- 显微图像识别 /
- 生物化石 /
- 碳酸盐岩
Abstract: The identification of microfossils in carbonate rocks with thin-section observation is of great significance for the study of sedimentary environment, but the traditional method by manual identification is highly experience required and is greatly affected by subjective factors.In this paper, a method for microscopic recognition of carbonate rocks based on ResNet convolutional neural network was introduced. Through image preprocessing, model design, model training etc., the intelligent recognition of fossils of organisms within thin section images were realized, and the recognition accuracy showed to be 86%.Meanwhile, an advanced YOLO(You Look Only Once) object detection model was proposed, which could realize the detection and recognition of the area where the organism locates in thin section image, and the recognition accuracy appeared to be 85%.This method verified the feasibility of using digital image processing algorithm and deep learning method to intelligently identify biological microscopic images of carbonate rocks.It can be regarded as a useful supplement to traditional manual identification methods and has certain practical application value.
- Convolutional Neural Network /
- ResNet /
- YOLO /
- biological recognition /
- fossils /
- carbonate rock

HTML全文

图 1 图像预处理对比

Figure 1. Comparison of image preprocessing

下载: 全尺寸图片幻灯片

图 2 碳酸盐岩生物化石图像数据集增强过程示例

Figure 2. Example of enhancement process for micro fossils image data set of carbonate rocks

下载: 全尺寸图片幻灯片

图 3 ResNet模型中残差块结构示意^[17]

Figure 3. Residual block structure in ResNet model^[17]

下载: 全尺寸图片幻灯片

图 4 ResNet卷积神经网络生物化石分类过程

Figure 4. Fossil classification process by ResNet Convolutional Neural Network

下载: 全尺寸图片幻灯片

图 5 ResNet卷积神经网络碳酸盐岩生物化石的识别过程

Figure 5. Fossil dentification of carbonate rocks by ResNet Convolutional Neural Network

下载: 全尺寸图片幻灯片

图 6 YOLO目标检测模型碳酸盐岩生物化石的检测与识别过程

Figure 6. Fossil detection and identification of carbonate rocks by YOLO object detection model

下载: 全尺寸图片幻灯片

表 1 ResNet模型中网格结构示意^[17]

Table 1. Network structure in ResNet model^[17]

卷积层名	输出尺寸	第18层	第34层	第50层	第101层	第152层
卷积层1	112×112	7×7, 64, 步长2
卷积层2_x	56×56	3×3最大池化层, 步长2
卷积层2_x	56×56	$ \left[\begin{array}{l} 3 \times 3, 64 \\ 3 \times 3, 64 \end{array}\right] \times 2$	$ \left[\begin{array}{l} 3 \times 3, 64 \\ 3 \times 3, 64 \end{array}\right] \times 3$	$ \left[\begin{array}{l} 1 \times 1, 64 \\ 3 \times 3, 64 \\ 1 \times 1, 256 \end{array}\right] \times 3$	$ \left[\begin{array}{l} 1 \times 1, 64 \\ 3 \times 3, 64 \\ 1 \times 1, 256 \end{array}\right] \times 3$	$ \left[\begin{array}{l} 1 \times 1, 64 \\ 3 \times 3, 64 \\ 1 \times 1, 256 \end{array}\right] \times 3$
卷积层3_x	28×28	$ \left[\begin{array}{l} 3 \times 3, 128 \\ 3 \times 3, 128 \end{array}\right] \times 2$	$ \left[\begin{array}{l} 3 \times 3, 128 \\ 3 \times 3, 128 \end{array}\right] \times 4$	$ \left[\begin{array}{l} 1 \times 1, 128 \\ 3 \times 3, 128 \\ 1 \times 1, 512 \end{array}\right] \times 4$	$ \left[\begin{array}{l} 1 \times 1, 128 \\ 3 \times 3, 128 \\ 1 \times 1, 512 \end{array}\right] \times 4$	$ \left[\begin{array}{l} 1 \times 1, 128 \\ 3 \times 3, 128 \\ 1 \times 1, 512 \end{array}\right] \times 8$
卷积层4_x	14×14	$ \left[\begin{array}{l} 3 \times 3, 256 \\ 3 \times 3, 256 \end{array}\right] \times 2$	$ \left[\begin{array}{l} 3 \times 3, 256 \\ 3 \times 3, 256 \end{array}\right] \times 6$	$ \left[\begin{array}{l} 1 \times 1, 256 \\ 3 \times 3, 256 \\ 1 \times 1, 1024 \end{array}\right] \times 6$	$ \left[\begin{array}{l} 1 \times 1, 256 \\ 3 \times 3, 256 \\ 1 \times 1, 1~ 024 \end{array}\right] \times 23$	$ \left[\begin{array}{l} 1 \times 1, 256 \\ 3 \times 3, 256 \\ 1 \times 1, 1~024 \end{array}\right] \times 36$
卷积层5_x	7×7	$ \left[\begin{array}{l} 3 \times 3, 512 \\ 3 \times 3, 512 \end{array}\right] \times 2$	$ \left[\begin{array}{l} 3 \times 3, 512 \\ 3 \times 3, 512 \end{array}\right] \times 3$	$ \left[\begin{array}{l} 1 \times 1, 512 \\ 3 \times 3, 512 \\ 1 \times 1, 2~048 \end{array}\right] \times 3$	$ \left[\begin{array}{l} 1 \times 1, 512 \\ 3 \times 3, 512 \\ 1 \times 1, 2~048 \end{array}\right] \times 3$	$ \left[\begin{array}{l} 1 \times 1, 512 \\ 3 \times 3, 512 \\ 1 \times 1, 2~048 \end{array}\right] \times 3$
	1×1	平均池化层, 1 000-d fc, softmax函数
浮点运算每秒		1.8×10⁹	3.6×10⁹	3.8×10⁹	7.6×10⁹	11.3×10⁹

下载: 导出CSV

参考文献(17)

[1]	余素玉. 化石碳酸盐岩微相[M]. 北京: 地质出版社, 1988. YU Suyu. Microfacies of fossil carbonate[M]. Beijing: Geological Publishing House, 1988.
[2]	刘宝郡. 沉积岩石学[M]. 北京: 地质出版社, 1980. LIU Baojun. Sedimentary petrology[M]. Beijing: Geological Publishing House, 1980.
[3]	WILSON J L. Carbonate facies in geologic history[M]. Berlin: Springer-Verlag, 1975.
[4]	赵敬松, 唐洪明, 雷卞军. 矿物岩石薄片研究基础[M]. 北京: 石油工业出版社, 2003. ZHAO Jingsong, TANG Hongming, LEI Bianjun. Research basis of mineralogical thin section[M]. Beijing: Petroleum Industry Press, 2003.
[5]	THOMPSON S, FUETEN F, BOCKUS D. Mineral identification using artificial neural networks and the rotating polarizer stage[J]. Computers and Geosciences, 2001, 27(9): 1081-1089. doi: 10.1016/S0098-3004(00)00153-9
[6]	ROSS B J, FUETEN F, YASHKIR D Y. Automatic mineral identification using genetic programming[J]. Machine Vision and Applications, 2001, 13(2): 61-69. doi: 10.1007/PL00013273
[7]	叶润青, 牛瑞卿, 张良培, 等. 基于图像分类的矿物含量测定及精度评价[J]. 中国矿业大学学报, 2011, 40(5): 810-815. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGKD201105024.htm YE Runqing, NIU Ruiqing, ZHANG Liangpei, et al. Mineral contents determination and accuracy evaluation based on classification of petrographic images[J]. Journal of China University of Mining & Technology, 2011, 40(5): 810-8152. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGKD201105024.htm
[8]	ALIGHOLI S, KHAJAVI R, RAZMARA M. Automated mineral identification algorithm using optical properties of crystals[J]. Computers and Geosciences, 2015, 85: 175-183. doi: 10.1016/j.cageo.2015.09.014
[9]	MAITRE J, BOUCHARD K, BÉDARD L P. Mineral grains recognition using computer vision and machine learning[J]. Computers & Geosciences, 2019, 130: 84-93.
[10]	李培军. 用ASETR图像和地统计学纹理进行岩性分类. 矿物岩石, 2004, 24(3): 116-120. https://www.cnki.com.cn/Article/CJFDTOTAL-KWYS200403014.htm LI Peijun. Lithological discrimination using ASTER image and geostatistical texture[J]. Journal of Mineralogy and Petrology, 2004, 24(3): 116-120. https://www.cnki.com.cn/Article/CJFDTOTAL-KWYS200403014.htm
[11]	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. https://www.cnki.com.cn/Article/CJFDTOTAL-KXJS201418050.htm GUO Chao, LIU Ye. Recognition of rock images based on Multiple Color Spaces[J]. Science Technology and Engineering, 2014, 14(18): 247-251, 255. https://www.cnki.com.cn/Article/CJFDTOTAL-KXJS201418050.htm
[12]	杨艳梅, 柳娜, 程国建, 等. 基于Spark平台的岩石图像聚类分析[J]. 西安石油大学学报: 自然科学版, 2016, 31(6): 114-118. https://www.cnki.com.cn/Article/CJFDTOTAL-XASY201606018.htm YANG Yanmei, LIU Na, CHENG Guojian, et al. Clustering analysis of rock images based on Spark platform[J]. Journal of Xi'an Shiyou University: Natural Science Edition, 2016, 31(6): 114-118. https://www.cnki.com.cn/Article/CJFDTOTAL-XASY201606018.htm
[13]	MARMO R, AMODIO S, TAGLIAFERRI R, et al. Textural identification of carbonate rocks by image processing and neural network: methodology proposal and examples[J]. Computers and Geosciences, 2005, 31(5): 649-659. doi: 10.1016/j.cageo.2004.11.016
[14]	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. https://www.cnki.com.cn/Article/CJFDTOTAL-KXJS201331015.htm 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. https://www.cnki.com.cn/Article/CJFDTOTAL-KXJS201331015.htm
[15]	刘曦阳. 图像识别技术在古生物化石图像上的应用[D]. 吉林大学, 2018. LIU Xiyang. The application of image recognition technology in paleontological fossil images[D]. Jilin university, 2018.
[16]	岳翔, 呼和, 贾建忠. 一种基于深度学习的有孔虫化石识别方法[J]. 电脑知识与技术, 2019, 15(27): 173+178. https://www.cnki.com.cn/Article/CJFDTOTAL-DNZS201927076.htm YUE Xiang, HU He, JIA Jianzhong. A method for the identification of foraminifera fossils based on deep learning[J]. Computer Knowledge and Technology, 2019, 15(27): 173+178. https://www.cnki.com.cn/Article/CJFDTOTAL-DNZS201927076.htm
[17]	HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). 2016: 770-778.