A segment joint multi-measure interaction index for shield tunnel recognition in RMLS point clouds

doi:10.11947/j.AGCS.2026.20250546

Abstract

Abstract:

Segment joints are critical weak points affecting the structural safety of shield tunnels, and their accurate recognition is essential for structural health monitoring. Rail-borne mobile laser scanning (RMLS) provides a high-quality data foundation for the recognition of segment joints. However, noise, occlusions, and uneven point cloud density in tunnels reduce the accuracy and robustness of existing methods. To address these limitations, this study proposes a shield tunnel segment joint recognition method based on the segment joint multi-measure interaction (SJMI) index. The method first removes clearly non-lining points using a cross-sectional ellipse fitting approach. Next, a SJMI is constructed by combining a bilateral wedge-fall geometric measure with an intensity residual measure, quantitatively capturing the distinctive characteristics of segment joints. Finally, the constructed SJMI index is used to sequentially recognize circumferential and longitudinal joints with high accuracy. Experiments on real RMLS point cloud data demonstrate that the proposed SJMI-based method effectively captures segment joint features, improves the recognition of actual joints, and suppresses noise and false positives. Quantitative evaluation shows that for circumferential joints, the IoU, recall, and precision reach 90.21%, 90.32%, and 99.84%, respectively, while for longitudinal joints, the corresponding metrics are 94.52%, 95.89%, and 98.23%. Compared with the current best-performing method, the IoU of circumferential and longitudinal joints increased by 2.25 percentage points and 7.96 percentage points, respectively, clearly demonstrating the significant accuracy advantage of the proposed method. In addition, the SJMI-based approach maintains strong robustness in complex tunnel environments and provides a reliable technical basis for the digital operation and maintenance of shield tunnels, segment joint anomaly detection, and structural health assessment.

Key words: shield tunnel, segment joint, rail-borne mobile laser scanning, point cloud, multi-measure interaction, clustering

CLC Number:

P237

Ze YOU, Liying WANG, Yiwei YU, Zhiwei QIN, Chunxi XIE, Yimo GENG, Xinao LI. A segment joint multi-measure interaction index for shield tunnel recognition in RMLS point clouds[J]. Acta Geodaetica et Cartographica Sinica, 2026, 55(6): 1101-1115.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

URL: http://xb.chinasmp.com/EN/10.11947/j.AGCS.2026.20250546

http://xb.chinasmp.com/EN/Y2026/V55/I6/1101

Figures/Tables 19

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Fig. 8

Fig. 9

Fig. 10

Fig. 11

Fig. 12

Tab. 1

Fig. 13

Fig. 14

Fig. 15

Tab. 2

Fig. 16

Tab. 3

References 30

[1]	LI Xinyu, ZHANG Dingli, HOU Yanjuan. Analysis of shield tunnel ground deformation characteristics and affecting factors in water-rich soft stratum: a case study on the section tunnel of Tianjin metro line 6[J]. Applied Sciences, 2022, 12(12): 6208.
[2]	FENG Yong, FENG Shijin, ZHANG Xiaolei, et al. A two-step deep learning-based framework for metro tunnel lining defect recognition[J]. Tunnelling and Underground Space Technology, 2024, 150: 105832.
[3]	刘新根, 陈莹莹, 刘学增. 激光扫描盾构隧道断面变形快速检测[J]. 交通运输工程学报, 2021, 21(2): 107-116.
	LIU Xingen, CHEN Yingying, LIU Xuezeng. Laser scanning-based rapid detection of deformation of shield tunnel section[J]. Journal of Traffic and Transportation Engineering, 2021, 21(2): 107-116.
[4]	NIU Yapeng, XIE Yongli, ZHANG Hongguang, et al. Study on deformation characteristics of segment joints of the immersed tunnel in Hong Kong-Zhuhai-Macau bridge[J]. Advances in Materials Science and Engineering, 2022: 8099308.
[5]	CAMARA M, WANG Liying, YOU Ze. Three-dimensional point cloud displacement analysis for tunnel deformation detection using mobile laser scanning[J]. Applied Sciences, 2025, 15(2): 625.
[6]	CAMARA M, WANG Liying, YOU Ze. Tunnel cross-section deformation monitoring based on mobile laser scanning point cloud[J]. Sensors, 2024, 24(22): 7192.
[7]	余建伟, 廖江海, 陈智鹏, 等. 基于断面基准面点云模型的隧道管片分割与断面提取[J]. 测绘通报, 2021(10): 159-162.
	YU Jianwei, LIAO Jianghai, CHEN Zhipeng, et al. A tunnel segment segmentation and section extraction based on point cloud model of cross-section datum plane[J]. Bulletin of Surveying and Mapping, 2021(10): 159-162.
[8]	YU Anbin, MEI Wensheng, HAN Mulin. Deep learning based method of longitudinal dislocation detection for metro shield tunnel segment[J]. Tunnelling and Underground Space Technology, 2021, 113: 103949.
[9]	朱家松, 郑澳, 雷占占, 等. 基于改进Yolov5的地铁隧道附属设施与衬砌表观病害检测方法[J]. 铁道科学与工程学报, 2023, 20(3): 1008-1019.
	ZHU Jiasong, ZHENG Ao, LEI Zhanzhan, et al. Metro tunnel accessorial facilities and lining diseases detection method based on improved Yolov5[J]. Journal of Railway Science and Engineering, 2023, 20(3): 1008-1019.
[10]	LI Zhonghao, MA Enlin, LAI Jinxing, et al. Tunnel deformation prediction during construction: an explainable hybrid model considering temporal and static factors[J]. Computers & Structures, 2024, 294: 107276.
[11]	CUI Hao, LI Jian, MAO Qingzhou, et al. STSD: a large-scale benchmark for semantic segmentation of subway tunnel point cloud[J]. Tunnelling and Underground Space Technology, 2024, 150: 105829.
[12]	YE Zehao, LIN Wei, FARAMARZI A, et al. SAM4Tun: no-training model for tunnel lining point cloud component segmentation[J]. Tunnelling and Underground Space Technology, 2025, 158: 106401.
[13]	王兵海. 基于PointNet的地铁盾构隧道点云典型要素分割[J]. 铁道建筑技术, 2022(12): 159-163.
	WANG Binghai. Segmentation of typical elements in metro shield tunnel point clouds based on PointNet[J]. Railway Construction Technology, 2022(12): 159-163.
[14]	黄星霖, 张星, 李清泉, 等. 多尺度与注意力聚合的隧道点云语义分割方法[J]. 测绘科学, 2024, 49(7): 56-65.
	HUANG Xinglin, ZHANG Xing, LI Qingquan, et al. Multi-scale attention aggregation graph convolution for tunnel point cloud semantic segmentation[J]. Science of Surveying and Mapping, 2024, 49(7): 56-65.
[15]	LI Jincheng, ZHANG Zhenxin, SUN Haili, et al. GL-Net: semantic segmentation for point clouds of shield tunnel via global feature learning and local feature discriminative aggregation[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2023, 199: 335-349.
[16]	JI Ankang, ZHOU Yunxiang, ZHANG Limao, et al. Semi-supervised learning-based point cloud network for segmentation of 3D tunnel scenes[J]. Automation in Construction, 2023, 146: 104668.
[17]	ZHOU Yunxiang, JI Ankang, ZHANG Limao, et al. Attention-enhanced sampling point cloud network (ASPCNet) for efficient 3D tunnel semantic segmentation[J]. Automation in Construction, 2023, 146: 104667.
[18]	ZHANG Zhiwei, ZHENG Gang, CHENG Xuesong, et al. Analytical approach for longitudinal deformation of shield tunnels considering bending-shear-torsional effects of circumferential joints[J]. Tunnelling and Underground Space Technology, 2024, 152: 105946.
[19]	ZHI Zhiyang, CHANG Bingtao, LI Yuan, et al. P-CSF: polar coordinate cloth simulation filtering algorithm for multi-type tunnel point clouds[J]. Tunnelling and Underground Space Technology, 2025, 155: 106144.
[20]	CUI Hao, REN Xiaochun, MAO Qingzhou, et al. Shield subway tunnel deformation detection based on mobile laser scanning[J]. Automation in Construction, 2019, 106: 102889.
[21]	巩健. 盾构隧道环内错台移动扫描检测方法研究[J]. 铁道勘察, 2026, 52(2): 17-22.
	GONG Jian. Research on mobile scanning detection method for intra-ring segment offsets in shield tunnels[J]. Railway Investigation and Surveying, 2026, 52(2): 17-22.
[22]	林康, 石波, 杨密, 等. 基于激光点云环缝环号识别的盾构隧道病害定位方法[J]. 铁道科学与工程学报, 2023, 20(12): 4835-4847.
	LIN Kang, SHI Bo, YANG Mi, et al. Disease location method of shield tunnel based on laser point cloud ring seam ring number identification[J]. Journal of Railway Science and Engineering, 2023, 20(12): 4835-4847.
[23]	傅静雅. 面向盾构隧道的移动LiDAR点云强度校正及应用研究[D]. 赣州: 江西理工大学, 2023.
	FU Jingya. Research on intensity correction and application of moving LiDAR point cloud for shield tunnel[D]. Ganzhou: Jiangxi University of Science and Technology, 2023.
[24]	易炼, 殷晓广, 张同刚. 基于螺栓孔特征的盾构隧道管片接缝识别[J]. 铁道标准设计, 2025, 69(2): 111-118.
	YI Lian, YIN Xiaoguang, ZHANG Tonggang. Segment seam detection of shield tunnels based on bolt hole feature[J]. Railway Standard Design, 2025, 69(2): 111-118.
[25]	WANG Liying, YOU Ze, FENG Yong, et al. Extraction and location of subway shield tunnel segment joints from RMLS point clouds[J]. International Journal of Digital Earth, 2025, 18: 2528627.
[26]	YOU Ze, WANG Liying. Dislocation detection of shield tunnel segments under non-uniform deformation conditions using RMLS point clouds[J]. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2025, X-1/W2-2025: 195-204.
[27]	SHEN Yueqian, WANG Jinguo, WANG Jinhu, et al. Methodology for extraction of tunnel cross-sections using dense point cloud data[J]. Journal of Geodesy and Geoinformation Science, 2021, 4(2): 56-71.
[28]	WANG Liying, YOU Ze. JCFI: a composite index for RMLS-based shield tunnel segment joint recognition[C]//Proceedings of 2026 ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Toronto: IEEE, 2026.
[29]	ZHU Xinge, ZHOU Hui, WANG Tai, et al. Cylindrical and asymmetrical 3D convolution networks for LiDAR segmentation[C]//Proceedings of 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville: IEEE, 2021: 9934-9943.
[30]	CHOY C, GWAK J Y, SAVARESE S. 4D spatio-temporal convnets: minkowski convolutional neural networks[C]//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach: IEEE, 2019: 3075-3084.

算法	环缝IoU	纵缝IoU
SJMI	90.21	94.52
PointNet	70.88	64.47
Cylinder3D	77.52	44.57
MinkUNet	79.31	59.88
基于单一测度的局部极值方法^[25]	87.96	86.56
GL-Net	79.69	71.09

算法	训练/计算测度时间/h	测试/识别接缝时间/min	总时间/h
SJMI	1.20	6.8	1.31
PointNet	2.46	8.6	2.60
Cylinder3D	4.05	10.6	4.23
MinkUNet	2.65	1.2	2.67
基于单一测度的局部极值方法^[25]	1.68	12.2	1.88
GL-Net	3.07	12.0	3.27