Segmentation of micro-cracks in fractured coal based on convolutional neural network
-
摘要: 深部开采导致的煤岩裂隙结构的延伸、贯通及扩展是引起煤岩破坏的关键因素。裂隙结构的准确表征是分析煤岩破坏机理及开展煤与瓦斯绿色共采的核心要素之一。本研究采用Waifu2x卷积神经网络模型对裂隙煤岩CT图像进行处理，获得分辨率更高的CT图像。并提出“裂隙煤岩标注七步处理法”，提高了标记效率，采用图像增强技术扩充训练集，满足了模型对于训练集数量和质量的要求。采用训练后的U-Net卷积神经网络进行图像分割，提取煤岩内裂隙信息。通过对比发现，本方法所获得的裂隙连通性、开度及分布与真实CT图像更为接近，所提取的裂隙4个量化指标均优于其他提取方法。研究为裂隙煤岩物理力学行为及机理分析提供基础。Abstract: The extension, penetration and expansion of coal and rock fracture structure caused by deep mining are the key factors that cause coal and rock damage.Therefore, the accurate characterization of the fracture structure is one of the core elements for understanding the coal and rock failure mechanism and developing coal and gas green co-mining.In this study, the Waifu2x convolutional neural network model was used to process the fractured coal and rock CT images to obtain CT images with higher resolution.And put forward the "seven-step processing method of fissure coal and rock labeling", which improves the efficiency of labeling, and uses image enhancement technology to expand the training set, which meets the model's requirements for the quantity and quality of the training set.The trained U-Net convolutional neural network is used for image segmentation to extract the fracture information in the coal and rock.Through comparison, it is found that the fracture connectivity, opening and distribution obtained by this method are closer to the real CT images, and the four quantitative indicators of the extracted fractures are better than other extraction methods.This study can provide a research foundation for the physical and mechanical behavior and mechanism analysis of fractured coal.
-
图 1 本研究所用U-Net网络模型示意图
Figure 1. Schematic diagram of u-net network model used in this study
表 1 环境硬件配置
Table 1. Environment hardware configuration
项目 系统 CPU 内存 GPU 参数 Windows 10专业版 Intel i7-7700HQ 16GB GTX1060 6G 
下载: 导出CSV
表 2 环境软件配置
Table 2. Environment software configuration
项目 Python Keras Tensorflow Cudatoolkit 版本 3.6 2.1.6 1.14.0 10.0.130
下载: 导出CSV
表 3 不同算法裂隙提取准确性评价
Table 3. Accuracy evaluation of different crack extraction algorithms
评价指标 U-Net卷积神经网络 全局阈值分割 局部阈值分割 Weka机器学习 PA 0.981 0.976 0.942 0.973 MPA 0.868 0.771 0.785 0.809 MIoU 0.797 0.729 0.619 0.730 FWIoU 0.966 0.956 0.916 0.954
下载: 导出CSV
-
[1] 孙旭东, 张蕾欣, 张博. 碳中和背景下我国煤炭行业的发展与转型研究[J]. 中国矿业, 2021, 30(2): 1-6. doi: 10.3969/j.issn.1009-105x.2021.02.001Sun Xudong, Zhang Leixin, Zhang Bo. Research on the coal industry development and transition in China under the background of carbon neutrality[J]. China Mining Magazine, 2021, 30(2): 1-6. doi: 10.3969/j.issn.1009-105x.2021.02.001 [2] 谢和平, 高峰, 鞠杨. 深部岩体力学研究与探索[J]. 岩石力学与工程学报, 2015, 34(11): 2161-2178. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX201511001.htmXie Heping, Gao Feng, Ju Yang. Research and development of rock mechanics in deep ground engineering[J]. Chinese Journal of Rock Mechanics and Engineering, 2015, 34(11): 2161-2178. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX201511001.htm [3] 谢和平, 周宏伟, 薛东杰, 等. 我国煤与瓦斯共采: 理论、技术与工程[J]. 煤炭学报, 2014, 39(8): 1391-1397. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201408003.htmXie Heping, Zhou Hongwei, Xue Dongjie, et al. Theory, technology and engineering of simultaneous exploitation of coal and gas in China[J]. Journal of China Coal Society, 2014, 39(8): 1391-1397. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201408003.htm [4] 袁亮. 我国深部煤与瓦斯共采战略思考[J]. 煤炭学报, 2016, 41(1): 1-6. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201601002.htmYuan Liang. Strategic thinking of simultaneous exploitation of coal and gas in deep mining[J]. Journal of China Coal Society, 2016, 41(1): 1-6. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201601002.htm [5] 郑江韬, 王海峰, 蔡长鹏, 等. 开采扰动后裂隙煤岩内气体渗流行为研究[J]. 矿业研究与开发, 2020, 40(11): 101-105. https://www.cnki.com.cn/Article/CJFDTOTAL-KYYK202011018.htmZheng Jiangtao, Wang Haifeng, Cai Changpeng, et al. Study on gas percolation behavior in the fractured coal after mining disturbance[J]. Mining Research and Development, 2020, 40(11): 101-105. https://www.cnki.com.cn/Article/CJFDTOTAL-KYYK202011018.htm [6] Ju Y, Zhang Q G, Yang Y M, et al. An experimental investigation on the mechanism of fluid flow through single rough fracture of rock[J]. Science China Technological Sciences, 2013, 56(8): 2070-2080. doi: 10.1007/s11431-013-5274-6 [7] 谢和平, 高峰, 周宏伟, 等. 煤与瓦斯共采中煤层增透率理论与模型研究[J]. 煤炭学报, 2013, 38(7): 1101-1108. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201307001.htmXie Heping, Gao Feng, Zhou Hongwei, et al. On theoretical and modeling approach to mining-enhanced permeability for simultaneous exploitation of coal and gas[J]. Journal of China Coal Society, 2013, 38(7): 1101-1108. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201307001.htm [8] 姜耀东, 潘一山, 姜福兴, 等. 我国煤炭开采中的冲击地压机理和防治[J]. 煤炭学报, 2014, 39(2): 205-213. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201402001.htmJiang Yaodong, Pan Yishan, Jiang Fuxing, et al. State of the art review on mechanism and prevention of coal bumps in China[J]. Journal of China Coal Society, 2014, 39(2): 205-213. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201402001.htm [9] Zhao Y X, Liu S M, Zhao G F, et al. Failure mechanisms in coal: dependence on strain rate and microstructure[J]. Journal of Geophysical Research: Solid Earth, 2014, 119(9): 6924-6935. doi: 10.1002/2014JB011198 [10] Ju Y, Xi C D, Zhang Y, et al. Laboratory in situ CT observation of the evolution of 3D fracture networks in coal subjected to confining pressures and axial compressive loads: a novel approach[J]. Rock Mechanics and Rock Engineering, 2018, 51(11): 3361-3375. doi: 10.1007/s00603-018-1459-4 [11] Voorn M, Exner U, Rath A. Multiscale Hessian fracture filtering for the enhancement and segmentation of narrow fractures in 3D image data[J]. Computers & Geosciences, 2013, 57: 44-53. [12] Voorn M, Exner U, Barnhoorn A, et al. Porosity, permeability and 3D fracture network characterisation of dolomite reservoir rock samples[J]. Journal of Petroleum Science and Engineering, 2015, 127: 270-285. doi: 10.1016/j.petrol.2014.12.019 [13] Iassonov P, Gebrenegus T, Tuller M. Segmentation of X-ray computed tomography images of porous materials: a crucial step for characterization and quantitative analysis of pore structures[J]. Water Resources Research, 2009, 45(9): W09415. [14] Kaestner A, Lehmann E, Stampanoni M. Imaging and image processing in porous media research[J]. Advances in Water Resources, 2008, 31(9): 1174-1187. [15] Garfi G, John C M, Berg S, et al. The sensitivity of estimates of multiphase fluid and solid properties of porous rocks to image processing[J]. Transport in Porous Media, 2020, 131(3): 985-1005. [16] Alhammadi A M, Gao Y, Akai T, et al. Pore-scale X-ray imaging with measurement of relative permeability, capillary pressure and oil recovery in a mixed-wet micro-porous carbonate reservoir rock[J]. Fuel, 2020, 268: 117018. [17] Lu X, Armstrong R T, Mostaghimi P. High-pressure X-ray imaging to interpret coal permeability[J]. Fuel, 2018, 226: 573-582. [18] Zhang G L, Ranjith P G, Liang W G, et al. Stress-dependent fracture porosity and permeability of fractured coal: an in situ X-ray tomography study[J]. International Journal of Coal Geology, 2019, 213: 103279. [19] Ronneberger O, Fischer P, Brox T. U-net: convolutional networks for biomedical image segmentation[C]// Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015. Munich, Germany, 2015. [20] 周飞燕, 金林鹏, 董军. 卷积神经网络研究综述[J]. 计算机学报, 2017, 40(6): 1229-1251. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJX201706001.htmZhou Feiyan, Jin Linpeng, Dong Jun. Review of convolutional neural network[J]. Chinese Journal of Computers, 2017, 40(6): 1229-1251. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJX201706001.htm [21] 南方哲, 钱育蓉, 行艳妮, 等. 基于深度学习的单图像超分辨率重建研究综述[J]. 计算机应用研究, 2020, 37(2): 321-326. https://www.cnki.com.cn/Article/CJFDTOTAL-JSYJ202002001.htmNan Fangzhe, Qian Yurong, Xing Yanni, et al. Survey of single image super resolution based on deep learning[J]. Application Research of Computers, 2020, 37(2): 321-326. https://www.cnki.com.cn/Article/CJFDTOTAL-JSYJ202002001.htm [22] 汪媛, 楚亚蕴. 多指标指纹图像分割方法研究[J]. 阜阳师范学院学报: 自然科学版, 2011, 28(2): 46-49.Wang Yuan, Chu Yayun. Research on fingerprint image segmentation method based on multi-factors[J]. Journal of Fuyang Teachers College: Natural Science, 2011, 28(2): 46-49. [23] 王嫣然, 陈清亮, 吴俊君. 面向复杂环境的图像语义分割方法综述[J]. 计算机科学, 2019, 46(9): 36-46. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJA201909007.htmWang Yanran, Chen Qingliang, Wu Junjun. Research on image semantic segmentation for complex environments[J]. Computer Science, 2019, 46(9): 36-46. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJA201909007.htm [24] 曹清洁, 史再峰, 张嘉平, 等. 分区域多标准的全参考图像质量评价算法[J]. 天津大学学报: 自然科学与工程技术版, 2019, 52(6): 625-630. https://www.cnki.com.cn/Article/CJFDTOTAL-TJDX201906009.htmCao Qingjie, Shi Zaifeng, Zhang Jiaping, et al. Sub-regional and multiple criteria full-reference image quality assessment[J]. Journal of Tianjin University: Science and Technology, 2019, 52(6): 625-630. https://www.cnki.com.cn/Article/CJFDTOTAL-TJDX201906009.htm [25] 田萱, 王亮, 丁琪. 基于深度学习的图像语义分割方法综述[J]. 软件学报, 2019, 30(2): 440-468. https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201902014.htmTian Xuan, Wang Liang, Ding Qi. Review of image semantic segmentation based on deep learning[J]. Journal of Software, 2019, 30(2): 440-468. https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201902014.htm [26] 景庄伟, 管海燕, 臧玉府, 等. 基于深度学习的点云语义分割研究综述[J]. 计算机科学与探索, 2021, 15(1): 1-26. https://www.cnki.com.cn/Article/CJFDTOTAL-KXTS202101001.htmJing Zhuangwei, Guan Haiyan, Zang Yufu, et al. Survey of point cloud semantic segmentation based on deep learning[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(1): 1-26. https://www.cnki.com.cn/Article/CJFDTOTAL-KXTS202101001.htm [27] Otsu N. A threshold selection method from gray-level histograms[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1979, 9(1): 62-66. [28] Phansalkar N, More S, Sabale A, et al. Adaptive local thresholding for detection of nuclei in diversity stained cytology images[C]// 2011 International Conference on Communications and Signal Processing. February 10-12, 2011. Kerala, India. IEEE, 2011: 218-220. [29] Arganda-Carreras I, Kaynig V, Rueden C, et al. Trainable Weka Segmentation: a machine learning tool for microscopy pixel classification[J]. Bioinformatics, 2017, 33(15): 2424-2426. -
点击查看大图
图(6) / 表(3)
计量
- 文章访问数: 397
- HTML全文浏览量: 64
- PDF下载量: 67
- 被引次数: 0