Tunnel face monitoring is of great significance for ensuring construction safety and project quality. However, traditional monitoring methods relying on manual measurement and on-site observation are inefficient and limited in accuracy. To improve the monitoring level, a rapid digitization method of tunnel face combined with deep learning was proposed. By constructing a homography transformation model of the tunnel face, applying a feature matching algorithm based on deep learning and a fast adaptive joint feature extraction algorithm based on multi-scale information, rapid and accurate joint feature extraction independent of camera parameters and shooting angles was achieved. Real-time monitoring was also accomplished with the help of fast joint detection and fusion methods. The experimental results show that this method can effectively extract the global joint information of the tunnel face, significantly improve the data acquisition efficiency and reduce the operation difficulty. In conclusion, this method provides a new idea and approach for tunnel engineering monitoring and effectively guarantees the safety of tunnel construction.
BAI Yu, HAO Yi-ren, CHEN Wei, PEI Shao-kang, WANG He, WANG Ji-chao, FANG Hao.
Fast and Robust Algorithm for Extracting Global Joint Features on Tunnel Face[J].
Science Technology and Engineering, 2026, 26(3): 1157-1165 DOI:10.12404/j.issn.1671-1815.2408157
一些方法使用三维重建模型进行掌子面数字化监测。董鑫等[6]中通过研究隧道掌子面摄影图像,采用SURF(speeded up robust features),分析和处理图像,提取围岩结构面信息包括岩体层理、节理及断层等,利用三维重建技术,建立隧道围岩结构面三维结构模型。段兴龙等[7]以甘肃省某在建高速公路隧道衬砌断面检测工程实践为例,研究了基于三维激光扫描技术的隧道衬砌断面检测方法,利用一次扫描即可准确建立隧道三维矢量模型。陶雄等[8]中通过某隧道项目的实例分析,探究了三维激光扫描精度在隧道断面检测中的应用。徐晓雅等[9]中针对山岭隧道地上地下一体化三维建模效率低、数据融合难度大的问题,提出一套基于建筑信息模型+地理信息系统+倾斜摄影技术的建模方法,实现隧道工程信息的集成与共享。
还有一些方法使用深度学习进行掌子面表面节理检测,取得了不错的效果。彭磊等[10]提出了一种基于Mask R-CNN(Mask regions with convolutional neural networks)的隧道掌子面节理图像智能识别分割算法,可直接用于检测隧道掌子面图片中的节理目标。高洪飞等[11]中通过引入Swin Transformer网络提高了算法对掌子面图像节理特征的提取能力,采用迁移学习降低训练成本,加速模型收敛。鲁冠宏等[12]中提出了一种基于深度学习的隧道衬砌裂缝检测网络YOLOv5-CT(YOLOv5 CBAM Transformer),可以实现隧道衬砌裂缝实时检测。张运波等[13]采用先进的EfficientNet网络作为主干网络,加强Mask R-CNN算法获取特征信息的能力,提升了识别精度。孟馨等[14]中联合隧道散射地震成像(tunnel seismic tomography, TST)偏移图像、地质分区与物探指标数据集训练模型,建立了一种工程地质环境评价及围岩等级预测方法。
LiuHong, LiuPeng, HePeng. Robust evaluation method for surrounding rock classification of super-large section tunnels and its application[J]. Highway, 2017, 62(9): 8-11.
ZhaoHuijun, MaYuehui, LiuNingning, et al. Measurement of tunnel surrounding rock deformation based on digital image processing[J]. Modern Electronic Technology, 2018, 41(21): 76-79.
WangPingrang, HuangHongwei, XueYadong. Real-time and rapid identification of tunnel lining cracks based on image structure information[J]. Science Technology and Engineering, 2022, 22(29): 13075-13082.
YuJun, LinZhibin, LiYuanhai. Deformation analysis of surrounding rock in deep tunnels based on transparent rock mass[J]. Journal of Underground Space and Engineering, 2017, 13(4): 943-949.
LuoJia, LiuDagang. Non-contact measurement method for surrounding rock deformation based on image processing technology[J]. Computer Application and Software, 2013, 30(7): 32-34.
DongXin, XiongZiming, HaoYiqing, et al. Research on identifying structural planes in subsea tunnel surrounding rocks based on photogrammetry[J]. Port Engineering Technology, 2017, 54(4): 100-104.
XuXiaoya, WangZhangqiong, LiLeilie, et al. Integrated three-dimensional modeling method of the above-ground and underground parts of mountain tunnels[J]. Science Technology and Engineering, 2024, 24(8): 3373-3380.
PengLei, ZhouChun, HuFeng, et al. Intelligent detection and segmentation of joint surfaces in tunnel headings based on deep learning[J]. Yangtze River, 2023, 54(10): 243-250.
GaoHongfei, YangRensong, ZhangYunbo, et al. Intelligent identification method and application of tunnel heading face joints based on the enhanced Mask RCNN algorithm[J]. Industrial Construction, 2023, 53(S2): 583-587.
LuGuanhong, LüChengshun, TianJun, et al. Detection of highway tunnel lining cracks based on improved YOLOv5[J]. Science Technology and Engineering, 2025, 25(7): 2997-3006.
ZhangYunbo, LeiMingfeng, XiaoYongzhuo, et al. Intelligent identification method of tunnel heading face joints based on convolutional neural network[J]. China Journal of Highway and Transport, 2024, 37(7): 35-45.
MengXin, QinYongjun, XieLiangfu, et al. Prediction of surrounding rock grade of the long Dongtianshan Tunnel based on deep learning[J]. Science Technology and Engineering, 2022, 22(25): 11219-11226.
[29]
LindebergT. Scale-space theory: a basic tool for analyzing structures at different scales[J]. Journal of Applied Statistics, 1994, 21(1/2): 225-270.
[30]
DetoneD, MalisiewiczT, RabinovichA, et al. SuperPoint:self-supervised interest point detection and description[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Salt Lake City: IEEE, 2018: 18-22.
[31]
FischlerM A, BollesR C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography[J]. Commun ACM, 1981, 24: 381-395.