融合残差与多尺度特征的U-Net泥页岩裂缝分割

陈子浩; 贾鹏飞; 屈文航; 王勇

doi:10.16152/j.cnki.xdxbzr.2023-05-008

您当前的位置：

首页 >

文章列表页 >

融合残差与多尺度特征的U-Net泥页岩裂缝分割

人工智能交叉应用 | 更新时间：2023-10-30

- 融合残差与多尺度特征的U-Net泥页岩裂缝分割
- U-Net-fusion segmentation of crack on mud shale using residual and multi-scale features
- 西北大学学报（自然科学版） 2023年53卷第5期页码：761-770
- 作者机构：
  
  1.西北大学　地质学系/大陆动力学国家重点实验室，陕西　西安　710069
  2.中国科学院武汉岩土力学研究所　岩土力学与工程国家重点实验室，湖北　武汉　430071
- 作者简介：
  
  陈子浩，男，从事图像处理与深度学习研究，804864150@qq.com。
  贾鹏飞，男，副教授，从事浅层气地质灾害与防治研究，pengfeijia@nwu.edu.cn。
- 基金信息：
  
  国家自然科学基金(51979269);国家重大仪器研制项目(52127815)
- DOI：10.16152/j.cnki.xdxbzr.2023-05-008
  中图分类号：
扫描看全文
陈子浩, 贾鹏飞, 屈文航, 等. 融合残差与多尺度特征的U-Net泥页岩裂缝分割[J]. 西北大学学报（自然科学版）, 2023,53(5):761-770.

CHEN Zihao, JIA Pengfei, QU Wenhang, et al. U-Net-fusion segmentation of crack on mud shale using residual and multi-scale features[J]. Journal of Northwest University (Natural Science Edition), 2023,53(5):761-770.
陈子浩, 贾鹏飞, 屈文航, 等. 融合残差与多尺度特征的U-Net泥页岩裂缝分割[J]. 西北大学学报（自然科学版）, 2023,53(5):761-770. DOI： 10.16152/j.cnki.xdxbzr.2023-05-008.

CHEN Zihao, JIA Pengfei, QU Wenhang, et al. U-Net-fusion segmentation of crack on mud shale using residual and multi-scale features[J]. Journal of Northwest University (Natural Science Edition), 2023,53(5):761-770. DOI： 10.16152/j.cnki.xdxbzr.2023-05-008.

摘要

泥页岩CT图像裂缝分割是获取裂缝信息的关键环节，对揭示裂缝空间展布规律和明确储层特征具有重要意义。针对传统图像处理方法无法自主分割泥页岩裂缝，以及泥页岩CT图像中裂缝形态复杂多样，尺度变化大，细小裂缝与周围岩体的灰度值相似导致分割效率低的问题，通过改进U-Net网络模型，提出一种融合残差网络和多尺度特征的卷积神经网络MCS-Net，用于实现高效的泥页岩裂缝分割。所提网络包括特征提取、特征融合和预测输出3部分，特征提取部分使用残差模块代替常规卷积，充分获取浅层特征信息和高层语义信息；特征融合部分采用多尺度卷积融合模块获取裂缝多尺度特征信息，随后逐层上采样还原图像分辨率并准确定位裂缝位置；预测输出部分最终生成泥页岩二元裂缝分割图。实验在构建的泥页岩裂缝数据集MudshaleCrack上进行测试，并与6种主流的深度学习语义分割网络进行对比。结果表明，所提的MCS-Net优于对比网络，评价指标IoU为85.32%，F1值为92.56%，与改进前的U-Net相比，IoU和F1值分别提升了0.042 8和0.090 3，证明了所提方法对泥页岩裂缝分割的可行性。

Abstract

Crack segmentation in CT images of mud shale is a key link in obtaining crack information, which is of great significance in revealing the spatial distribution of cracks and clarifying the reservoir bed characteristics. Aiming at the problems that traditional image processing methods are unable to segment mud shale cracks autonomously, as well as the low segmentation efficiency due to the complexity and variety of crack morphology in CT images of mud shale, large scale changes, and the similarity of gray scale values between the small cracks and the surrounding rock body, a convolutional neural network MCS-Net, which integrates residual network and multiscale features, is proposed to realize the highly efficient segmentation of mud shale cracks through the improvement of the U-Net network model. The proposed network consists of three parts: feature extraction, special fusion and prediction output. The feature extraction part uses the residual module instead of the conventional convolution to fully obtain the shallow feature information and the high-level semantic information; the feature fusion part adopts the multi-scale convolutional fusion module to obtain the multi-scale feature information of the cracks, and then up-sampling layer-by-layer to restore the resolution of the image and accurately locate the cracks; and the prediction output part ultimately generates a binary mud shale crack segmentation map. The experiments are tested on the constructed mudshale crack dataset MudshaleCrack and compared with six mainstream deep learning semantic segmentation networks. The results show that the proposed MCS-Net outperforms the comparison networks, with an evaluation metric IoU of 85.32% and an F1 value of 92.56%, which improves the IoU and F1 values by 0.042 8 and 0.090 3, respectively, when compared with the pre-improvement U-Net, which proves the feasibility of the proposed method for mudshale crack segmentation.

关键词

泥页岩CT图像裂缝分割U-Net模型残差网络多尺度特征

Keywords

CT image of mud shalecrack segmentationU-Net modelresidual networkmulti-scale feature

references

包友书. 陆相泥页岩在水平地应力作用下裂缝的多样性：以济阳坳陷古近系泥页岩为例[J]. 石油学报, 2019, 40(7): 777-785.

BAO Y S. Fracture diversity of continental shale under horizontal geostress: A casestudy of the Paleogene shale in Jiyang depression[J]. Acta Petrolei Sinica, 2019, 40(7): 777-785.

郭小波, 黄志龙, 柳波, 等. 马朗凹陷芦草沟组泥页岩储层微观孔隙特征及地质意义[J]. 西北大学学报(自然科学版), 2014, 44(1): 88-95.

GUO X B, HUANG Z L, LIU B, et al. Microscopic characteristics and geological significance of Lucaogou Formation shale reservoir in Malang Sag[J]. Journal of Northwest University(Natural Science Edition), 2014, 44(1): 88-95.

何强, 李凤霞, 史爱萍, 等. 基于三维CT重构的油页岩复杂水力裂缝网络分形表征[J]. 油气地质与采收率, 2021, 28(5): 116-123.

HE Q, LI F X, SHI A P, et al. Fractal characterization of complex hydraulic fracture networks of oil shalevia 3D CT reconstruction[J]. Petroleum Geology and Recovery Efficiency, 2021, 28(5): 116-123.

KHERADMANDI N, MEHRANFAR V. A critical review and comparative study on image segmentation-based techniques for pavement crack detection[J]. Construction and Building Materials, 2022, 321: 126162.

WEIJERMARS R, WANG J, PHAM T. Borehole failure mechanisms in naturally fractured formations[J]. Rock Mechanics and Rock Engineering, 2022, 55(5): 3001-3022.

蒋凯. 印度尼西亚M气田泥页岩井壁失稳机理及对策[J]. 中国海上油气, 2021, 33(2): 142-147.

JIANG K. Mechanisms and countermeasures of mud-shale wellbore instability in Mgas field, Indonesia[J]. China Offshore Oil and Gas, 2021, 33(2): 142-147.

彭文泉. 潍北凹陷孔店组二段上亚段页岩气储层压裂可行性研究[J]. 地质学报, 2019, 93(S1): 45-52.

PENG W Q. Feasibility study on shale gas reservoir fracturing in the upper member of the Second Member of Kongdian Formation in the Weibei sag[J]. Acta Geologica Sinica, 2019, 93(S1): 45-52.

TIAN Y X, LIU X J, LUO P Y, et al. Testing of drilling mud filter cake for low permeability micro-fracture plugging performance in shale rocks[J]. Journal of Petroleum Exploration and Production Technology, 2022, 12(12): 3289-3302.

CAO Y, WU Q, JIN Z J, et al. Characterization of microstructures in lacustrine organic-rich shale using micro-CT images: Qingshankou formation in Songliao Basin[J]. Energies, 2022, 15(18): 6712.

LI B K, NIE X, CAI J C, et al. U-Net model for multi-component digital rock modeling of shales based on CT and QEMSCAN images[J]. Journal of Petroleum Science and Engineering, 2022, 216: 110734.

REZAEI F, IZADI H, MEMARIAN H, et al. The effectiveness of different thresholding techniques in segmenting micro CT images of porous carbonates to estimate porosity[J]. Journal of Petroleum Science and Engineering, 2019, 177: 518-527.

DENG L W, WANG X F, AI L. A new fracture image segmentation method based on MSA-k clustering algorithm[C]//2020 39th Chinese Control Conference(CCC). Shenyang: IEEE, 2020: 2125-2129.

田志, 肖立志, 廖广志, 等. 基于沉积过程的数字岩石建模方法研究[J]. 地球物理学报, 2019, 62(1): 248-259.

TIAN Z, XIAO L Z, LIAO G Z, et al. Study on digital rock reconstruction method based on sedimentological process[J]. Chinese Journal of Geophysics, 2019, 62(1): 248-259.

宋天明, 滕奇志. 一种用于岩心序列图像轮廓提取的改进Snake算法[J]. 光电子·激光, 2011, 22(2): 302-306.

SONG T M, TENG Q Z. Improved Snake algorithm for contour extracting of rock core sequential images[J]. Journal of Optoelectronics·Laser, 2011, 22(2): 302-306.

邹永宁, 杨瑞娜, 罗骁, 等. 石油岩芯CT图像裂缝分割算法研究[J]. 强激光与粒子束, 2016, 28(5): 167-172.

ZOU Y N, YANG R N, LUO X, et al. Crack detection algorithm for oil rock core industrial CT image[J]. High Power Laser and Particle Beams, 2016, 28(5): 167-172.

LI Z W, ZHANG G Y. Fracture segmentation method based on contour evolution and gradient direction consistency in sequence of coal rock CT images[J]. Mathematical Problems in Engineering, 2019, 2019: 1-8.

SHELHAMER E, LONG J, DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4): 640-651.

NOH H, HONG S, HAN B. Learning deconvolution network for semantic segmentation[C]//2015 IEEE International Conference on Computer Vision(ICCV). Santiago: IEEE, 2015: 1520-1528.

RONNEBERGER O, FISCHER P, BROX T. U-net: Convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2015: 234-241.

BADRINARAYANAN V, KENDALL A, CIPOLLA R. SegNet: A deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12): 2481-2495.

DAI J F, LI Y, HE K M, et al. R-FCN: Object detection via region-based fully convolutional networks[EB/OL]. (2016-06-21)[2023-01-20]. https://arxiv.org/abs/1605.06409https://arxiv.org/abs/1605.06409.

ZHOU Z W, SIDDIQUEE M M R, TAJBAKHSH N, et al. UNet＋＋: Redesigning skip connections to exploit multiscale features in image segmentation[J]. IEEE Transactions on Medical Imaging, 2020, 39(6): 1856-1867.

CHEN J N, LU Y Y, YU Q H, et al. TransUNet: Transformers make strong encoders for medical image segmentation[EB/OL]. (2021-02-08)[2023-01-20]. https://arxiv.org/abs/2102.04306https://arxiv.org/abs/2102.04306.

HYUN S, LEE J S, JEON S, et al. Pixel-level crack detection in X-ray computed tomography image of granite using deep learning[J]. Tunnel and Underground Space, 2019, 29(3): 184-196.

TIAN W, CHENG K, LIU Q, et al. Meso-structure segmentation of concrete CT image based on mask and regional convolution neural network[J]. Materials & Design, 2021, 208: 109919.

LU F L, FU C C, ZHANG G Y, et al. Adaptive multi-scale feature fusion based U-net for fracture segmentation in coal rock images[J]. Journal of Intelligent & Fuzzy Systems, 2022, 42(4): 3761-3774.

HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Las Vegas: IEEE, 2016: 770-778.

HE K M, ZHANG X Y, REN S Q, et al. Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification[C]//2015 the IEEE International Conference on Computer Vision(ICCV). Santiago: IEEE, 2016: 1026-1034.

CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(4): 834-848.

HOWARD A G, ZHU M L, CHEN B, et al. MobileNets: Efficient convolutional neural networks for mobile vision applications[EB/OL]. (2017-04-17)[2023-01-20]. https://arxiv.org/abs/1704.04861https://arxiv.org/abs/1704.04861.

谢飞, 周炀, 管子玉, 等. 基于可变形卷积的小肠间质瘤多模型联合检测[J]. 西北大学学报(自然科学版), 2021, 51(1): 16-23.

XIE F, ZHOU Y, GUAN Z Y, et al. Multi-model joint detection of gastrointestinal stromal tumor with deformable convolution[J]. Journal of Northwest University (Natural Science Edition), 2021, 51(1): 16-23.

BUADES A, COLL B, MOREL J M. A non-local algorithm for image denoising[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR′05). San Diego: IEEE, 2005: 60-65.

王珏, 罗姗, 邹永宁. 邻域加窗的非局部均值CT成像去噪方法[J]. 计算机工程与应用, 2015, 51(9): 168-172.

WANG J, LUO S, ZOU Y N. Non-local means of CT image de-noising method based on neighborhood window[J]. Computer Engineering and Applications, 2015, 51(9): 168-172.

KANG J, LI N Y, ZHAO L Q, et al. Construction of complex digital rock physics based on full convolution network[J]. Petroleum Science, 2022, 19(2): 651-662.

OKTAY O, SCHLEMPER J, LE FOLGOC L, et al. Attention U-net: Learning where to look for the pancreas[EB/OL]. (2018-05-20)[2023-01-20]. https://arxiv.org/abs/1804.03999https://arxiv.org/abs/1804.03999.

ALOM M Z, YAKOPCIC C, TAHA T M, et al. Nuclei segmentation with recurrent residual convolutional neural networks based U-Net (R2U-net)[C]//NAECON 2018-IEEE National Aerospace and Electronics Conference.Dayton: IEEE, 2018: 228-233.

HUANG H M, LIN L F, TONG R F, et al. UNet 3＋: A full-scale connected Unet for medical imagesegmentation[C]//ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing.Barcelona: IEEE, 2020: 1055-1059.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据