基于多源点热扩散的隧道裂缝几何特征计算

doi:10.19678/j.issn.1000-3428.0066468

摘要/Abstract

摘要：

裂缝几何参数分析为评估隧道的健康状况提供了理论依据。为了提升裂缝参数计算的准确性，提出基于多源点热扩散的隧道裂缝几何特征计算算法。基于U-Net神经网络得到裂缝分割掩膜，设计以裂缝边缘为多源点的热扩散方法来计算裂缝区域的测地距离场，获取裂缝参数。提取裂缝边缘为源点进行热扩散，通过求解热流方程和梯度场找到距离增加方向，求解泊松方程得到裂缝区域的测地距离场，根据测地距离极值计算裂缝长度。该算法充分利用了裂缝复杂的边缘信息以及裂缝骨架特征，计算得到的测地距离场可进一步应用于计算裂缝宽度。基于测地距离场得到的裂缝骨架具有光滑、无多余分叉的优点，且裂缝宽度计算过程中避免了多次迭代和误差累积，减少了计算量。实验结果表明，该算法在测试集上的平均骨架匹配度为92.84%，相比于Zhang细化算法和Hilditch细化算法分别提升了2.57和1.41个百分点，断点和分叉数量也明显少于同类算法，适用于裂缝狭长弯曲、边缘复杂多变的几何特征计算。

关键词: 隧道裂缝, 多源点热扩散, 测地距离, 裂缝分割, 裂缝参数

Abstract:

The analysis of the geometric parameters of cracks provides a theoretical basis for evaluating tunnel health. To improve the accuracy of the crack parameter calculation, a calculation algorithm based on a multi-source thermal method is proposed for tunnel crack geometric parameters. A crack segmentation mask is obtained based on a U-Net neural network, and a thermal method using crack edges as multiple sources is designed to calculate the geodesic distance of the crack. Thus, the crack edge is extracted as the source point for thermal diffusion. The direction of the distance increase is determined by solving the thermal flow equation and gradient field. Finally, the Poisson equation is solved to obtain the geodesic distance field in the crack area. On this basis, the crack length and width are calculated according to the extreme values of the geodesic distance. This algorithm fully utilizes the complex edge information and crack skeleton characteristics, and the calculated geodesic distance field can be applied to calculate the crack width. Crack skeletons obtained based on the geodesic distance field have the advantages of smoothness and no extra branches, and the width calculation process avoids multiple iterations and error accumulation, thereby reducing computational cost. The experimental results show that the average skeleton matching degree of this algorithm on the test set is 92.84%, which is 2.57 and 1.41 percentage points higher than that of Zhang's and Hilditch's thinning algorithms, respectively. Additionally, the number of breakpoints and branches is significantly less than that of similar algorithms, which is suitable for the geometric characteristic calculation of cracks with long, curved shapes and complex edges.

Key words: tunnel crack, multi-source thermal diffusion, geodesic distance, crack segmentation, crack parameter

闫星志, 刘向阳, 杨鹏宇. 基于多源点热扩散的隧道裂缝几何特征计算[J]. 计算机工程, 2023, 49(12): 294-303.

Xingzhi YAN, Xiangyang LIU, Pengyu YANG. Calculation of Geometric Characteristics of Tunnel Cracks Based on Multi-source Thermal Diffusion[J]. Computer Engineering, 2023, 49(12): 294-303.

https://www.ecice06.com/CN/Y2023/V49/I12/294

图/表 19

图1 U-Net结构

Fig.1 U-Net structure

图2 测地距离示意图

Fig.2 Schematic diagram of the geodesic distance

图3 多源点热扩散过程

Fig.3 Process of multi-source thermal diffusion

图4 裂缝区域的距离场热力图

Fig.4 Thermal diagram of distance field in crack area

图5 裂缝几何特征计算流程

Fig.5 Procedure of crack geometric characteristic calculation

图6 多连通域处理过程

Fig.6 Process of multi-connected domain processing

图7 最小外接矩阵

Fig.7 Minimum circumscribed matrix

图8 扩散方向

Fig.8 Diffusion direction

图9 测地距离场三维示意图

Fig.9 3D schematic diagram of geodesic distance field

图10 根据裂缝方向的局部极值提取

Fig.10 Local extremum extraction according to crack direction

图11 裂缝初始骨架

Fig.11 Initial skeleton of cracks

图12 裂缝骨架提取过程

Fig.12 Process of crack skeleton extraction

图13 裂缝分割效果

Fig.13 Crack segmentation effect

图14 性能指标计算结果

Fig.14 Performance index calculation results

图15 裂缝骨架提取效果比较

Fig.15 Comparison of crack skeleton extraction effect

图16 实际拍摄的裂缝图像

Fig.16 Actual captured crack images

参考文献 28

1	中华人民共和国交通运输部. 2021年交通运输行业发展统计公报[EB/OL]. [2022-07-12]. http://www.gov.cn/xinwen/2022-05/25/content_5692174.htm.
	Ministry of Transport of the People's Republic of China. Statistical bulletin of transportation industry development in 2021[EB/OL]. [2022-07-12]. http://www.gov.cn/xinwen/2022-05/25/content_5692174.htm. (in Chinese)
2	蒋雅君, 陶磊, 刘莎, 等. 运营公路隧道结构预防性养护经济性评价方法探讨. 隧道建设, 2020, 40 (S2): 332- 339. URL
	JIANG Y J, TAO L, LIU S, et al. Discussion on economic evaluation method of preventive maintenance of operating highway tunnel structure. Tunnel Construction, 2020, 40 (S2): 332- 339. URL
3	刘文斌, 王世伟, 秦之富. 公路隧道经常检查及养护工作探讨. 隧道建设, 2019, 39 (S1): 486- 491. URL
	LIU W B, WANG S W, QIN Z F. Discussion on regular inspection and maintenance of highway tunnels. Tunnel Construction, 2019, 39 (S1): 486- 491. URL
4	瞿中, 王彩云. 基于注意力机制和轻量级空洞卷积的混凝土路面裂缝检测. 计算机科学, 2023, 50 (2): 192- 196. URL
	QU Z, WANG C Y. Crack detection of concrete pavement based on attention mechanism and lightweight dilated convolution. Computer Science, 2023, 50 (2): 192- 196. URL
5	于海洋, 景鹏, 张文涛, 等. 基于残差与注意力机制的道路裂缝检测U-Net改进模型. 计算机工程, 2023, 49 (6): 265- 273. URL
	YU H Y, JING P, ZHANG W T, et al. Improved U-Net model for road crack detection based on residual and attention mechanism. Computer Engineering, 2023, 49 (6): 265- 273. URL
6	ZHANG Q Y, BARRI K, BABANAJAD S K, et al. Real-time detection of cracks on concrete bridge decks using deep learning in the frequency domain. Engineering, 2021, 7 (12): 1786- 1796. doi: 10.1016/j.eng.2020.07.026
7	ZHANG J, DUAN F Q, ZHOU M Q, et al. Stable and realistic crack pattern generation using a cracking node method. Frontiers of Computer Science, 2018, 12 (4): 777- 797. doi: 10.1007/s11704-016-5511-9
8	LIU S G, CHEN D. Computer simulation of batik printing patterns with cracks. Textile Research Journal, 2015, 85 (18): 1972- 1984. doi: 10.1177/0040517514561919
9	ALI R, CHUAH J H, ABU TALIP M S, et al. Automatic pixel-level crack segmentation in images using fully convolutional neural network based on residual blocks and pixel local weights. Engineering Applications of Artificial Intelligence, 2021, 104, 104391. doi: 10.1016/j.engappai.2021.104391
10	ALI R, CHUAH J H, TALIP M S A, et al. Structural crack detection using deep convolutional neural networks. Automation in Construction, 2022, 133, 103989. doi: 10.1016/j.autcon.2021.103989
11	ZHANG Y, YUEN K V. Crack detection using fusion features-based broad learning system and image processing. Computer-Aided Civil and Infrastructure Engineering, 2021, 36 (12): 1568- 1584. doi: 10.1111/mice.12753
12	张振海, 贾争满, 季坤. 基于改进的Otsu法的地铁隧道裂缝识别方法研究. 重庆交通大学学报(自然科学版), 2022, 41 (1): 84- 90. URL
	ZHANG Z H, JIA Z M, JI K. Crack identification method of subway tunnel based on improved Otsu method. Journal of Chongqing Jiaotong University(Natural Science), 2022, 41 (1): 84- 90. URL
13	王建锋, 邱雨, 刘水宙. 基于图像补偿的隧道衬砌裂缝检测方法. 浙江大学学报(工学版), 2022, 56 (7): 1404- 1415. URL
	WANG J F, QIU Y, LIU S Z. Tunnel lining crack detection method based on image compensation. Journal of Zhejiang University(Engineering Science), 2022, 56 (7): 1404- 1415. URL
14	ZHANG L, YANG F, ZHANG Y D, et al. Road crack detection using deep convolutional neural network[C]//Proceedings of IEEE International Conference on Image Processing. Washington D. C., USA: IEEE Press, 2016: 3708-3712.
15	彭雨诺, 刘敏, 万智, 等. 基于改进YOLO的双网络桥梁表观病害快速检测算法. 自动化学报, 2022, 48 (4): 1018- 1032. URL
	PENG Y N, LIU M, WAN Z, et al. A dual deep network based on the improved YOLO for fast bridge surface defect detection. Acta Automatica Sinica, 2022, 48 (4): 1018- 1032. URL
16	周颖, 刘彤. 基于计算机视觉的混凝土裂缝识别. 同济大学学报(自然科学版), 2019, 47 (9): 1277- 1285. URL
	ZHOU Y, LIU T. Computer vision-based crack detection and measurement on concrete structure. Journal of Tongji University(Natural Science), 2019, 47 (9): 1277- 1285. URL
17	NISHIYAMA S, MINAKATA N, KIKUCHI T, et al. Improved digital photogrammetry technique for crack monitoring. Advanced Engineering Informatics, 2015, 29 (4): 851- 858.
18	刘娟红, 孟翔, 段品佳, 等. 基于MATLAB的混凝土裂缝宽度计算方法研究. 材料导报, 2022, 36 (6): 84- 89. URL
	LIU J H, MENG X, DUAN P J, et al. Study on calculation method of concrete crack width based on MATLAB. Materials Reports, 2022, 36 (6): 84- 89. URL
19	朱磊, 李东彪, 闫星志, 等. 基于改进Mask R-CNN深度学习算法的隧道裂缝智能检测方法. 图学学报, 2023, 44 (1): 177- 183. URL
	ZHU L, LI D B, YAN X Z, et al. Intelligent detection method of tunnel cracks based on improved Mask R-CNN deep learning algorithm. Journal of Graphics, 2023, 44 (1): 177- 183. URL
20	ZHANG T Y, SUEN C Y. A fast parallel algorithm for thinning digital patterns. Communications of the ACM, 1984, 27 (3): 236- 239.
21	RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation[C]//Proceedings of International Conference on Medical Image Computing and Computer-Assisted Intervention. Berlin, Germany: Springer, 2015: 234-241.
22	CRANE K, WEISCHEDEL C, WARDETZKY M. The heat method for distance computation. Communications of the ACM, 2017, 60 (11): 90- 99.
23	SETHIAN J A, VLADIMIRSKY A. Fast methods for the Eikonal and related Hamilton-Jacobi equations on unstructured meshes. Proceedings of the National Academy of Sciences of the United States of America, 2000, 97 (11): 5699- 5703.
24	ABEYWICKRAMA T, CHEEMA M A, TANIAR D. K-nearest neighbors on road networks. Proceedings of the VLDB Endowment, 2016, 9 (6): 492- 503.
25	SHI Y, CUI L M, QI Z Q, et al. Automatic road crack detection using random structured forests. IEEE Transactions on Intelligent Transportation Systems, 2016, 17 (12): 3434- 3445.
26	NASERI M, HEIDARI S, GHEIBI R, et al. A novel quantum binary images thinning algorithm: a quantum version of the Hilditch's algorithm. Optik, 2017, 131, 678- 686.
27	WANG P F, ZHAO F, MA S W. Skeleton extraction method based on distance transform[C]//Proceedings of the 11th International Conference on Electronic Measurement & Instruments. Washington D. C., USA: IEEE Press, 2014: 519-523.
28	苏辰耀, 刘向阳. 基于热方法的骨架提取算法. 计算机与现代化, 2022, (3): 59- 63. URL
	SU C Y, LIU X Y. Skeleton extraction algorithm based on thermal method. Computer and Modernization, 2022, (3): 59- 63. URL

选择文件类型/文献管理软件名称

选择包含的内容