-
摘要: 煤储层裂隙的精细描述对于煤层气资源的勘探开发具有重要意义，裂隙的尺寸、走向、分布密度等直接影响煤层的渗透性，准确识别和提取煤层中的裂隙信息是揭示体积压裂过程中裂缝网络形成与扩展机理的关键。传统的裂隙提取方法往往依赖人工标注和基于一定图像处理技术的特征提取，这些方法在精度和效率上存在明显不足。提出一种基于Trans-UNet网络和μCT图像的煤裂隙提取方法，Trans-UNet结合了Transformer模块和卷积神经网络(CNN)的优点，不仅具备全局特征提取能力，还能够捕捉图像中的局部细节特征，大幅提高了分割精度和网络的鲁棒性。首先对煤样μCT图像进行预处理，包括使用差值法提高图像质量、使用数据增强技术扩大样本数量等。随后，利用Trans-UNet网络对处理后的图像进行分割，提取裂隙特征，并比较不同神经网络模型的分割结果。结果表明，提出的方法在数据集上表现出优越性能，Trans-UNet模型在煤裂隙提取上的准确性(Accuracy)、精确度(Precision)、F1分数(F1-Score)和交并比(IoU)分别达到91.3%、89.5%、89.8%和84.0%，相较于U-Net、U-Net++等其他多种智能模型有显著提升。结合μCT图像的细粒度特征，将Trans-UNet网络应用于煤裂隙提取任务，是一种高效且准确的解决方案，为煤层气勘探开发领域的相关图像处理任务提供了新的思路。
-
关键词:
- Trans-UNet /
- μCT图像 /
- 煤裂隙 /
- 图像分割 /
- 深度学习
Abstract: The detailed description of coal fractures is of great significance for the exploration and development of coalbed methane resources. The size, orientation, and distribution density of fractures directly affect the permeability of the coal seam. Accurate identification and extraction of fracture information in coal seams are crucial for revealing the mechanisms of fracture network formation and propagation during hydraulic fracturing. Traditional fracture extraction methods often rely on manual labeling and feature extraction based on certain image processing techniques, which have significant limitations in terms of accuracy and efficiency. This paper proposes a coal fracture extraction method based on Trans-UNet and μCT images. Trans-UNet combines the advantages of Transformer modules and Convolutional Neural Network (CNN), possessing both global feature extraction capability and the ability to capture local details in images, significantly improving segmentation accuracy and network robustness. First, the μCT images of coal samples are pre-processed, including using interpolation methods to improve image quality and data augmentation techniques to increase the number of samples. Subsequently, the processed images are segmented using the Trans-UNet network to extract fracture features, and the segmentation results of different neural network models are compared. The results show that the proposed method outperforms other models on the dataset. The Trans-UNet model achieves an Accuracy of 91.3%, Precision of 89.5%, F1-score of 89.8%, and Intersection over Union (IoU) of 84.0%, significantly outperforming other intelligent models such as U-Net and U-Net++. Combining the fine-grained features of μCT images, the application of the Trans-UNet network to coal fracture extraction tasks is an efficient and accurate solution, providing new insights for image processing tasks in related fields.-
Keywords:
- Trans-UNet /
- μCT images /
- coal fractures /
- image segmentation /
- deep learning
-
-
[1] JIN Yi，ZHENG Junling，DONG Jiabin，et al. Fractal topography and complexity assembly in multifractals[J]. Fractals，2022，30(3)：2250052.
[2] 李国永，姚艳斌，王辉，等. 鄂尔多斯盆地神木-佳县区块深部煤层气地质特征及勘探开发潜力[J]. 煤田地质与勘探，2024，52(2)：70-80. LI Guoyong，YAO Yanbin，WANG Hui，et al. Deep coalbed methane resources in the Shenmu-Jiaxian Block，Ordos Basin，China：Geological characteristics and potential for exploration and exploitation[J]. Coal Geology & Exploration，2024，52(2)：70-80.
[3] 李斌，杨帆，张红杰，等. 神府区块深部煤层气高效开发技术研究[J]. 煤田地质与勘探，2024，52(8)：57-68. LI Bin，YANG Fan，ZHANG Hongjie，et al. Technology for efficient production of deep coalbed methane in the Shenfu Block[J]. Coal Geology & Exploration，2024，52(8)：57-68.
[4] MOORE T A. Coalbed methane：A review[J]. International Journal of Coal Geology，2012，101：36-81.
[5] 叶桢妮，侯恩科，段中会，等. 不同煤体结构煤的孔隙-裂隙分形特征及其对渗透性的影响[J]. 煤田地质与勘探，2019，47(5)：70-78. YE Zhenni，HOU Enke，DUAN Zhonghui，et al. Fractal characteristics of pores and microfractures of coals with different structure and their effect on permeability[J]. Coal Geology & Exploration，2019，47(5)：70-78.
[6] 施雷庭，赵启明，任镇宇，等. 煤岩裂隙形态对渗流能力影响数值模拟研究[J]. 油气藏评价与开发，2023，13(4)：424-432. SHI Leiting，ZHAO Qiming，REN Zhenyu，et al. Numerical simulation study on the influence of coal rock fracture morphology on seepage capacity[J]. Petroleum Reservoir Evaluation and Development，2023，13(4)：424-432.
[7] 李学博，刘春春，张武昌，等. 高煤阶煤裂隙发育特征对煤层气开发的影响：以沁水盆地南部郑庄区块为例[J]. 中国煤层气，2022，19(3)：12-15. LI Xuebo，LIU Chunchun，ZHANG Wuchang，et al. Influence of development characteristics of high-rank coal fractures on coalbed methane development：Taking Zhengzhuang Block in South Qinshui Basin as an example[J]. China Coalbed Methane，2022，19(3)：12-15.
[8] 王昱，宋晓夏，胡咤咤，等. 西山煤田屯兰区块煤层气低产井的增产改造措施及效果分析[J/OL]. 河南理工大学学报(自然科学版)，2024：1-13[2025-02-13]. http://kns.cnki.net/kcms/detail/41.1384.N.20240821.1707.002.html. WANG Yu，SONG Xiaoxia，HU Zhazha，et al. Measures and effect analysis of stimulation and rehabilitation of low-production coalbed methane wells in Tunlan Block，Xishan Coalfield[J/OL]. Journal of Henan Polytechnic University(Natural Science)，2024：1-13[2025-02-13]. http://kns.cnki.net/kcms/detail/41.1384.N.20240821.1707.002.html.
[9] 王跃鹏，孙正财，刘向君，等. 煤层割理结构及其对井壁稳定的影响研究[J]. 油气藏评价与开发，2020，10(4)：45-52. WANG Yuepeng，SUN Zhengcai，LIU Xiangjun，et al. Study on cleat structure and its influence on wellbore stability in coal seams[J]. Reservoir Evaluation and Development，2020，10(4)： 45-52.
[10] 胡秋嘉，刘世奇，毛崇昊，等. 基于X-ray CT与FIB-SEM的无烟煤孔裂隙发育特征[J]. 煤矿安全，2021，52(9)：10-15. HU Qiujia，LIU Shiqi，MAO Chonghao，et al. Characteristics of pores and fractures in anthracite coal based on X-ray CT and FIB-SEM[J]. Safety in Coal Mines，2021，52(9)：10-15.
[11] KETCHAM R A. Computational methods for quantitative analysis of three-dimensional features in geological specimens[J]. Geosphere，2005，1(1)：32-41.
[12] 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.
[13] DENG Hang，FITTS J P，PETERS C A. Quantifying fracture geometry with X-ray tomography：Technique of Iterative Local Thresholding (TILT) for 3D image segmentation[J]. Computational Geosciences，2016，20(1)：231-244.
[14] GOLAB A，WARD C R，PERMANA A，et al. High-resolution three-dimensional imaging of coal using microfocus X-ray computed tomography，with special reference to modes of mineral occurrence[J]. International Journal of Coal Geology，2013，113：97-108.
[15] WILDENSCHILD D，SHEPPARD A P. X-ray imaging and analysis techniques for quantifying pore-scale structure and processes in subsurface porous medium systems[J]. Advances in Water Resources，2013，51：217-246.
[16] RAMANDI H L，IRTZA S，SIROJAN T，et al. FracDetect：A novel algorithm for 3D fracture detection in digital fractured rocks[J]. Journal of Hydrology，2022，607：127482.
[17] HU Zhazha，LU Shuangfang，KLAVER J，et al. An integrated imaging study of the pore structure of the Cobourg limestone：A potential nuclear waste host rock in Canada[J]. Minerals，2021，11(10)：1042.
[18] MATHEWS J P，CAMPBELL Q P，XU Hao，et al. A review of the application of X-ray computed tomography to the study of coal[J]. Fuel，2017，209：10-24.
[19] TAN Jianquan，ZHOU Wenrui，LIN Ling，et al. A review of semantic medical image segmentation based on different paradigms[J]. International Journal on Semantic Web and Information Systems (IJSWIS)，2024，20(1)：1-25.
[20] KNACKSTEDT M A，LATHAM S，MADADI M，et al. Digital rock physics：3D imaging of core material and correlations to acoustic and flow properties[J]. The Leading Edge，2009，28(1)：28-33.
[21] CNUDDE V，BOONE M N. High-resolution X-ray computed tomography in geosciences：A review of the current technology and applications[J]. Earth-Science Reviews，2013，123：1-17.
[22] LEI Lian，YANG Qiliang，YANG Ling，et al. Deep learning implementation of image segmentation in agricultural applications：A comprehensive review[J]. Artificial Intelligence Review，2024，57(6)：149.
[23] 尹艺晓，马金刚，张文凯，等. 从U-Net到Transformer：混合模型在医学图像分割中的应用进展[J/OL]. 激光与光电子学进展，2024：1-38[2025-02-13]. http://kns.cnki.net/kcms/detail/31.1690.TN.20240612.0920.042.html. YIN Yixiao，MA Jingang，ZHANG Wenkai，et al. From U-Net to Transformer：Progress in the application of hybrid models in medical image segmentation[J/OL]. Laser & Optoelectronics Progress，2024：1-38[2025-02-13]. http://kns.cnki.net/kcms/detail/31.1690.TN.20240612.0920.042.html.
[24] 王登科，房禹，魏建平，等. 基于深度学习的煤岩Micro-CT裂隙智能提取与应用[J]. 煤炭学报，2024，49(8)：3439-3452. WANG Dengke，FANG Yu，WEI Jianping，et al. Intelligent extraction of Micro-CT fissures in coal based on deep learning and its application[J]. Journal of China Coal Society，2024，49(8)：3439-3452.
[25] 郝天轩，徐新革，赵立桢. 煤岩裂隙图像识别方法研究[J]. 工矿自动化，2023，49(10)：68-74. HAO Tianxuan，XU Xinge，ZHAO Lizhen. Research on image recognition methods for coal rock fractures[J]. Journal of Mine Automation，2023，49(10)：68-74.
[26] 郑江韬，齐子豪，刘佳存，等. 基于卷积神经网络的煤岩微裂隙提取方法[J]. 矿业科学学报，2022，7(6)：680-688. ZHENG Jiangtao，QI Zihao，LIU Jiacun，et al. Segmentation of micro-cracks in fractured coal based on convolutional neural network[J]. Journal of Mining Science and Technology，2022，7(6)：680-688.
[27] YU Jinxia，WU Chengyi，LI Yingying，et al. Intelligent identification of coal crack in CT images based on deep learning[J]. Computational Intelligence and Neuroscience，2022，2022(1)：7092436.
[28] LU Fengli，FU Chengcai，ZHANG Guoying，et al. Convolution neural network based on fusion parallel multiscale features for segmenting fractures in coal-rock images[J]. Journal of Electronic Imaging，2020，29(2)：23008.
[29] KARIMPOULI S，TAHMASEBI P，SAENGER E H. Coal cleat/fracture segmentation using convolutional neural networks[J]. Natural Resources Research，2020，29(3)：1675-1685.
[30] VASWANI A，SHAZEER N，PARMAR N，et al. Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems，2017：6000-6010.
[31] RAMANDI H L，ARMSTRONG R T，MOSTAGHIMI P. Micro-CT image calibration to improve fracture aperture measurement[J]. Case Studies in Nondestructive Testing and Evaluation，2016，6：4-13.
[32] SUN Haoran. A review of 3D-2D registration methods and applications based on medical images[J]. Highlights in Science，Engineering and Technology，2023，35：200-224.
[33] 孙书魁，范菁，孙中强，等. 基于深度学习的图像数据增强研究综述[J]. 计算机科学，2024，51(1)：150-167. SUN Shukui，FAN Jing，SUN Zhongqiang，et al. Survey of image data augmentation techniques based on deep learning[J]. Computer Science，2024，51(1)：150-167.
[34] 章展熠，张宝荃，王周立，等. 多茶类CNN图像识别的数据增强优化及类激活映射量化评价[J]. 茶叶科学，2023，43(3)：411-423. ZHANG Zhanyi，ZHANG Baoquan，WANG Zhouli，et al. Data enhancement optimization and class activation mapping quantitative evaluation for CNN image recognition of multiple tea categories[J]. Journal of Tea Science，2023，43(3)：411-423.
[35] 王气洪，贾洪杰，黄龙霞，等. 联合数据增强的语义对比聚类[J]. 计算机研究与发展，2024，61(6)：1511-1524. WANG Qihong，JIA Hongjie，HUANG Longxia，et al. Semantic contrastive clustering with federated data augmentation[J]. Journal of Computer Research and Development，2024，61(6)：1511-1524.
[36] SU Run，ZHANG Deyun，LIU Jinhuai，et al. MSU-Net：Multi-scale U-Net for 2D medical image segmentation[J]. Frontiers in Genetics，2021，12：639930.
[37] SHAN Liang，HU Bin，CHEN Long，et al. Detecting COVID-19 on CT images with impulsive-backpropagation neural networks[C]//202234th Chinese Control and Decision Conference (CCDC)，2022：2797-2803.
[38] HE Kaiming，ZHANG Xiangyu，REN Shaoqing，et al. Delving deep into rectifiers：Surpassing human-level performance on image net classification[C]//Proceedings of the IEEE International Conference on Computer Vision，2015：1026-1034.
[39] 张成业，李飞跃，李军，等. 基于DeepLabv3+与GF-2高分辨率影像的露天煤矿区土地利用分类[J]. 煤田地质与勘探，2022，50(6)：94-103. ZHANG Chengye，LI Feiyue，LI Jun，et al. Recognition of land use on open-pit coal mining area based on DeepLabv3+ and GF-2 high-resolution images[J]. Coal Geology & Exploration，2022，50(6)：94-103.
计量
- 文章访问数: 19
- HTML全文浏览量: 0
- PDF下载量: 5
下载:
