基于无人机和计算机视觉技术的隧道三维重建

doi:10.11996/JG.j.2095-302X.2026040882

图学学报 ›› 2026, Vol. 47 ›› Issue (4): 882-894.DOI: 10.11996/JG.j.2095-302X.2026040882

基于无人机和计算机视觉技术的隧道三维重建

冯勇¹^,²^,³, 张晓磊¹^,²(), 吕瑞鑫¹, 朱书林¹

¹ 同济大学地下建筑与工程系，上海 200092
² 同济大学城市交通研究院，上海 201804
³ 同济大学道路与交通工程教育部重点实验室，上海 201804

收稿日期:2025-12-25 接受日期:2026-04-09 出版日期:2026-08-31 发布日期:2026-08-31
通讯作者:张晓磊，E-mail：Xiaolei_Zhang@tongji.edu.cn
基金资助:
国家自然科学基金(42372335);上海市科委项目(21DZ1204400);同济大学项目(kh0023020242373);上海市教委项目(kz0023020250157);中央高校基本科研业务费专项资金

3D reconstruction of tunnels based on UAV and computer vision techniques

FENG Yong¹^,²^,³, ZHANG Xiaolei¹^,²(), LV Ruixin¹, ZHU Shulin¹

¹ Department of Geotechnical Engineering, Tongji University, Shanghai 200092, China
² Urban Mobility Institute, Tongji University, Shanghai 201804, China
³ MOE Key Laboratory of Road and Traffic Engineering, Tongji University, Shanghai 201804, China

Received:2025-12-25 Accepted:2026-04-09 Published:2026-08-31 Online:2026-08-31
Contact: ZHANG Xiaolei，E-mail：Xiaolei_Zhang@tongji.edu.cn
Supported by:
National Natural Science Foundation of China(42372335);Shanghai Municipal Science and Technology Commission(21DZ1204400);Tongji University(kh0023020242373);Shanghai Municipal Education Commission(kz0023020250157);Fundamental Research Funds for the Central Universities

摘要/Abstract

摘要：

隧道三维重建技术可广泛应用于结构病害检测、运维状态可视化展示、结构安全状态评估，以及数字孪生模型构建。针对现有基于单目视觉的隧道三维重建技术存在的自动化程度低、建模精度不足、建模效率欠佳等缺陷，本研究提出了耦合无人机和计算机视觉的隧道三维重建技术。在硬件端，集成了隧道巡检无人机，机体结构采用碳纤维和铝合金材料，配备光流传感器、激光雷达、三轴陀螺仪、三轴加速度计、电子罗盘以及数字气压计等传感器，基于多传感器融合策略在GNSS受限的隧道环境内实现厘米级定位与导航，搭载高性能光电吊舱，实现高清视频实时采集和传输。在算法端，集成了隧道三维重建算法框架，具体包括隧道视频截取与分帧、基于SfM技术解算相机位姿和稀疏点云、通过Gipuma算法生成深度图、基于TGV优化算法进行表面模型重构、利用纹理重建技术构建隧道纹理模型。试验结果表明，隧道巡检无人机可在复杂的隧道环境内稳定飞行，并采集高质量的视频数据。以视频帧为数据源，公路隧道三维重构模型的均方根重投影误差为1.52像素，最大重投影误差为25.44像素，网格数量为668 088。进一步探究了视频帧数量与尺寸对三维重建的影响：采用99张视频帧作为原始数据，可在建模精度与效率之间取得良好平衡；当计算资源充足时，建议先对原始视频帧进行上采样，再执行SfM，然后使用原始尺寸深度图进行表面模型重建。将本研究的建模算法框架与传统方案进行对比，结果证明该算法具有更高的计算精度与效率，且对计算机硬件依赖性较低。此外，在地铁隧道场景中的泛化性试验验证了该技术具有良好的工程可拓展性。其设备和算法可为自动、精准、高效的隧道三维重建提供支撑。

关键词: 隧道无人机, 三维重建, 运动恢复结构, 深度图, 隐式建模

Abstract:

Tunnel 3D reconstruction technology can be widely applied in structural defect detection, operation and maintenance state visualization, structural safety assessment, and digital twin model construction. To address the shortcomings of existing monocular vision-based tunnel 3D reconstruction technologies, such as low automation, insufficient modeling accuracy, and poor efficiency, a tunnel 3D reconstruction technique integrating drones and computer vision was proposed. On the hardware end, a tunnel inspection drone was developed, featuring a body structure made of carbon fiber and aluminum alloy, equipped with optical flow sensors, LiDAR, a three-axis gyroscope, a three-axis accelerometer, an electronic compass, and a digital barometer. With a multi-sensor fusion strategy, centimeter-level positioning and navigation were achieved in GNSS-restricted tunnel environments. The drone was also equipped with a high-performance electro-optical pod to enable real-time high-definition video acquisition and transmission. On the algorithm end, a tunnel 3D reconstruction algorithmic framework was designed, comprising video segmentation and frame extraction, camera pose and sparse point-cloud estimation using SfM, depth-map generation via the Gipuma algorithm, surface-model reconstruction based on TGV optimization, and tunnel-texture model construction using texture reconstruction techniques. Experimental results demonstrated that the tunnel inspection drone could stably operate in complex tunnel environments, capturing high-quality video data. Using video frames as the data source, the reconstructed highway tunnel 3D model achieved a root mean square reprojection error of 1.52 pixels, a maximum reprojection error of 25.44 pixels, and a total of 668 088 mesh faces. Further analysis of the impact of video-frame count and resolution on 3D reconstruction revealed that using 99 video frames as the input data achieved a balance between modeling accuracy and efficiency. When computational resources were abundant, it was recommended to upsample the original video frames before performing SfM and then use the original-sized depth maps for surface-model reconstruction. Additionally, a comparison between the proposed modeling framework and traditional solutions showed that the proposed methods exhibit higher computational efficiency and precision, as well as lower dependency on hardware resources. Furthermore, generalization tests in metro tunnels verify its engineering scalability. The proposed equipment and algorithms provide robust support for automated, accurate, and efficient tunnel 3D reconstruction.

Key words: tunnel drone, 3D reconstruction, structure from motion, depth map, implicit modeling

中图分类号:

U456.3

冯勇, 张晓磊, 吕瑞鑫, 朱书林. 基于无人机和计算机视觉技术的隧道三维重建[J]. 图学学报, 2026, 47(4): 882-894.

FENG Yong, ZHANG Xiaolei, LV Ruixin, ZHU Shulin. 3D reconstruction of tunnels based on UAV and computer vision techniques[J]. Journal of Graphics, 2026, 47(4): 882-894.

图/表 23

图1 隧道巡检无人机((a) 整体外观与关键部件；(b) 俯视图；(c) 侧视图)

Fig. 1 Tunnel inspection drone ((a) Overall appearance and key components; (b) Top view; (c) Side view)

表1 隧道巡检无人机的技术参数

Table 1 Technical parameters of the tunnel inspection drone

无人机参数	单位	取值
尺寸	mm	≤990(含四向可碰撞保护罩)
无刷电机数量	个	4
空机重量	g	≥2 800(含四向可碰撞保护罩)
最大起飞重量	g	≥4 900(含电池及保护罩)
最大飞行时间	min	≥15
最大飞行速度	m/s	2.0
最大上升速度	m/s	1.0
工作温度	℃	-20~50
最大操控距离	m	≥300(无遮挡)

表2 激光雷达的主要参数

Table 2 Main parameters of the laser radar

激光雷达参数	单位	描述
激光波长	nm	905
量程	m	40(目标物体反射率10%) 70(目标物体反射率80%)
近处盲区	m	0.1
视场角	°	360(水平) -7~52(竖直)
测距随机误差	cm	≤2(@10 m) ≤3(@0.2 m)
角度随机误差	°	≤0.15
点云输出	点/秒	200 000
尺寸	mm	65×65×60
重量	g	265

图2 隧道巡检无人机的关键配件((a) 光电吊舱；(b) 手持地面制遥控站)

Fig. 2 Key components of the tunnel inspection drone ((a) Electro-optical gimbal; (b) Handheld ground-based remote control station)

表3 吊舱的核心技术参数

Table 3 Core technical parameters of the pod

光电吊舱参数		单位	描述
云台参数	角度抖动量	°	±0.01
	可控俯仰转动范围	°	-135~+45
	可控水平转动范围	°	-160~+160
	横滚转动范围	°	-30~+30
相机参数	镜头焦距	mm	5.15~47.38(±5%)
	光圈	F	1.8~2.5
	视场	°	无变倍：对角79.5，水平71.5 10×光学变倍：对角7.7，水平6.7
	TF卡录视频分辨率	—	2 K (2 560×1 440)@30 fps 1 080 p(1 920×1 080)@30 fps 720 p(1 280×720)@30 fps
	视频存储码率	Mbps	12
	视频文件格式	—	MP4

图3 隧道三维重建的技术路线

Fig. 3 Technical framework for 3D reconstruction of tunnels

图4 隧道重构过程中的距离场示意图

Fig. 4 Schematic diagram of the distance field during the tunnel reconstruction process

图5 用于案例研究的公路隧道

Fig. 5 The highway tunnel used for the case study

图6 公路隧道内开展无人机飞行试验

Fig. 6 Drone flight tests in the highway tunnel

图7 图像之间的特征点匹配示例((a) 第1与第2视频帧；(b) 第3与第4视频帧)

Fig. 7 Example of feature point matching between images ((a) The first and the second video frame; (b) The third and the fourth video frame)

图8 相机轨迹的解算结果

Fig. 8 Camera trajectory solving results

图9 SfM产生的隧道稀疏点云((a) 高噪声点云；(b) 低噪声点云)

Fig. 9 Tunnel sparse point clouds generated by the SfM ((a) Noisy point cloud; (b) Denoised point cloud)

图10 基于Gipuma算法生成的深度图示例

Fig. 10 Example of depth map generated based on Gipuma algorithm

图11 基于TGV优化重构的隧道表面模型

Fig. 11 Tunnel surface model reconstructed via the TGV optimization

图12 去噪后的隧道表面模型

Fig. 12 Tunnel surface model after denoising

图13 隧道网格模型局部放大示意图

Fig. 13 Schematic diagram of the tunnel mesh model with local enlargement

图14 隧道纹理模型的侧视图

Fig. 14 Side view of the tunnel texture model

图15 隧道纹理模型的内部视图

Fig. 15 Internal view of the tunnel texture model

表4 不同输入帧数的三维重建指标

Table 4 3D reconstruction metrics for different input frame counts

序号	图像数量	RMSRE/pix	MRE/pix	全过程用时/s	稀疏点云数量/个	表面模型面数/个	重建内存用量/GB	纹理映射内存用量/MB
1	49	1.54	21.62	66	9 156	547 706	1.05	381.53
2	99	1.52	25.44	102	18 559	668 088	1.51	485.70
3	200	1.47	30.16	148	36 849	705 928	1.56	670.67
4	400	10.51	47.92	279	67 069	719 148	1.70	957.99

表5 不同图像尺寸下隧道三维重建过程的各项指标

Table 5 Various indicators of tunnel 3D reconstruction process under different image sizes

序号	图像尺寸-SfM过程	图像尺寸-表面模型重构	RMSRE/pix	MRE/pix	全过程用时/s	稀疏点云数量/个	表面模型面数/个	重建内存用量/MB	纹理映射内存用量/MB
1	原始尺寸	原始尺寸	1.52	25.44	102	18 559	668 088	1 546.24	485.70
2	原始尺寸	四分之一原始尺寸	1.52	25.44	69	18 559	204 275	502.85	145.29
3	原始尺寸	十六分之一原始尺寸	1.52	25.44	67	18 559	81 474	221.45	65.42
4	四倍原始尺寸	原始尺寸	1.16	27.33	124	54 846	724 797	1 597.44	1 010.38
5	四倍原始尺寸	四分之一原始尺寸	1.16	27.33	94	54 846	218 754	630.92	295.64
6	四倍原始尺寸	十六分之一原始尺寸	1.16	27.33	66	54 846	84 236	283.55	189.16
7	四分之一原始尺寸	原始尺寸	1.81	22.24	87	3 756	510 036	524.48	449.82
8	四分之一原始尺寸	四分之一原始尺寸	1.81	22.24	56	3 756	138 517	343.82	137.79
9	四分之一原始尺寸	十六分之一原始尺寸	1.81	22.24	53	3 756	52 115	245.59	70.44

图16 基于传统方案生成的隧道表面模型和纹理模型

Fig. 16 Tunnel surface model and texture model generated based on conventional schemes

图17 隧道局部区域被护栏遮挡

Fig. 17 Local tunnel area occluded by a guardrail

图18 基于本文方案生成的地铁隧道纹理模型

Fig. 18 Metro tunnel texture model generated based on the schemes proposed in this study

参考文献 22

[1]	洪开荣, 冯欢欢. 中国公路隧道近10年的发展趋势与思考[J]. 中国公路学报, 2020, 33(12): 62-76. DOI
	HONG K R, FENG H H. Development trends and views of highway tunnels in China over the past decade[J]. China Journal of Highway and Transport, 2020, 33(12): 62-76 (in Chinese).
[2]	陈湘生, 徐志豪, 包小华, 等. 中国隧道建设面临的若干挑战与技术突破[J]. 中国公路学报, 2020, 33(12): 1-14. DOI
	CHEN X S, XU Z H, BAO X H, et al. Challenges and technological breakthroughs in tunnel construction in China[J]. China Journal of Highway and Transport, 2020, 33(12): 1-14 (in Chinese).
[3]	《中国公路学报》编辑部. 中国交通隧道工程学术研究综述·2022[J]. 中国公路学报, 2022, 35(4): 1-40. DOI
	Editorial Department of China Journal of Highway and Transport. Review on China’s traffic tunnel engineering research: 2022[J]. China Journal of Highway and Transport, 2022, 35(4): 1-40 (in Chinese).
[4]	韩兴博, 叶飞, 王永东, 等. 劣化可靠度下的隧道衬砌结构预防性养护决策[J]. 中国公路学报, 2021, 34(1): 104-115. DOI
	HAN X B, YE F, WANG Y D, et al. Predictive maintenance decision-making of tunnel lining structure based on deterioration reliability[J]. China Journal of Highway and Transport, 2021, 34(1): 104-115 (in Chinese).
[5]	FENG Y, FENG S J, ZHANG X L, et al. A two-step deep learning-based framework for metro tunnel lining defect recognition[J]. Tunnelling and Underground Space Technology, 2024, 150: 105832. DOI URL
[6]	FENG S J, FENG Y, ZHANG X L, et al. Deep learning with visual explanations for leakage defect segmentation of metro shield tunnel[J]. Tunnelling and Underground Space Technology, 2023, 136: 105107. DOI URL
[7]	FENG Y, ZHANG X L, FENG S J, et al. Improved SOLOv2 detection method for shield tunnel lining water leakages[J]. Journal of Intelligent Construction, 2023, 1(1): 1-15. DOI URL
[8]	FENG Y, ZHANG X L, FENG S J, et al. Intelligent segmentation and quantification of tunnel lining cracks via computer vision[J]. Structural Health Monitoring, 2025, 24(3) 1896-1926. DOI URL
[9]	MA D, FANG H Y, WANG N N, et al. A low-cost 3D reconstruction and measurement system based on structure-from-motion (SFM) and multi-view stereo (MVS) for sewer pipelines[J]. Tunnelling and Underground Space Technology, 2023, 141: 105345. DOI URL
[10]	徐青山. 基于块匹配和深度学习的多视图立体视觉方法研究[D]. 武汉: 华中科技大学, 2021.
	XU Q S. Research on PatchMatch and deep learning for multi-view stereo[D]. Wuhan: Huazhong University of Science and Technology, 2021 (in Chinese).
[11]	李子欣. 基于Metashape的移动机器人三维地面环境的重构[D]. 北京: 北京交通大学, 2019.
	LI Z X. Reconstruction of 3D navigation environment for a mobile robot based on Metashape[D]. Beijing: Beijing Jiaotong University, 2019 (in Chinese).
[12]	葛海明, 张维, 王小龙, 等. 基于SfM的城市电缆隧道三维重建方法优化研究[J]. 图学学报, 2023, 44(3): 540-550. DOI
	GE H M, ZHANG W, WANG X L, et al. Optimization of 3D reconstruction method for urban cable tunnel based on SfM method[J]. Journal of Graphics, 2023, 44(3): 540-550 (in Chinese).
[13]	SCHÖNBERGER J L, FRAHM J M. Structure-from-motion revisited[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE Press, 2016: 4104-4113.
[14]	史沛哲. 基于三维重建的盾构隧道病害检测与评估方法研究[D]. 上海: 同济大学, 2022.
	SHI P Z. Research on defect inspection and evaluation method of shield tunnel based on 3D reconstruction[D]. Shanghai: Tongji University, 2022 (in Chinese).
[15]	XUE Y D, ZHANG S, ZHOU M L, et al. Novel SfM-DLT method for metro tunnel 3D reconstruction and visualization[J]. Underground Space, 2021, 6(2): 134-141. DOI URL
[16]	XUE Y D, SHI P Z, JIA F, et al. 3D reconstruction and automatic leakage defect quantification of metro tunnel based on SfM-Deep learning method[J]. Underground Space, 2022, 7(3): 311-323. DOI URL
[17]	FANG K, DONG A, TANG H M, et al. 3D tunnel reconstruction and visualization through multi-smartphone photogrammetry[J]. Measurement, 2023, 223: 113764. DOI URL
[18]	GALLIANI S, LASINGER K, SCHINDLER K. Massively parallel multiview stereopsis by surface normal diffusion[C]// 2015 IEEE International Conference on Computer Vision. New York: IEEE Press, 2015: 873-881.
[19]	POLIARNYI N. Out-of-core surface reconstruction via global TGV minimization[C]// 2021 IEEE/CVF International Conference on Computer Vision. New York: IEEE Press, 2021: 5621-5630.
[20]	LU Y J, WANG S, FAN S S, et al. Image-based 3D reconstruction for multi-scale civil and infrastructure projects: a review from 2012 to 2022 with new perspective from deep learning methods[J]. Advanced Engineering Informatics, 2024, 59: 102268. DOI URL
[21]	LOWE D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2): 91-110. DOI
[22]	FISCHLER M A, BOLLES R C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM, 1981, 24(6): 381-395. DOI URL

基于无人机和计算机视觉技术的隧道三维重建

3D reconstruction of tunnels based on UAV and computer vision techniques

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 23

参考文献 22

相关文章 15

编辑推荐

Metrics

本文评价

[1]	欧阳泽洪, 沈旭昆, 任曦, 胡勇, 黄勇. 基于共视引导的大规模场景运动恢复结构[J]. 图学学报, 2026, 47(4): 695-703.
[2]	王紫威, 王录涛, 李桉同, 沈艳. 单目深度模糊感知估计的少视图三维高斯重建[J]. 图学学报, 2026, 47(4): 704-713.
[3]	周雪杨, 沈旭昆, 胡勇. 基于神经场逆渲染的三维重建研究综述[J]. 图学学报, 2026, 47(3): 449-471.
[4]	廖健康, 张严辞. 针对过度重建问题的频率强度高斯泼溅算法[J]. 图学学报, 2026, 47(3): 564-575.
[5]	任皓, 李少波, 弓茂, 王博. 基于特征点引导干扰物识别的神经辐射场重建[J]. 图学学报, 2026, 47(1): 111-119.
[6]	潘宇轩, 金锐, 刘雨, 张琳. 基于生成模型的无监督多视点立体视觉网络[J]. 图学学报, 2026, 47(1): 29-38.
[7]	王道累, 丁子健, 杨君, 郑劭恺, 朱瑞, 赵文彬. 基于体素网格特征的NeRF大场景重建方法[J]. 图学学报, 2025, 46(3): 502-509.
[8]	黄志勇, 佘雅丽, 华喜锋, 向梦丽, 杨晨龙, 丁妥君. DCSplat：一种深度约束的稀疏视角三维重建方法[J]. 图学学报, 2025, 46(3): 510-519.
[9]	胡悦, 孙智达, 黄惠. 面向无人机路径规划的可视分析系统[J]. 图学学报, 2025, 46(3): 655-665.
[10]	孙禾衣, 李艺潇, 田希, 张松海. 结合程序内容生成与扩散模型的图像到三维瓷瓶生成技术[J]. 图学学报, 2025, 46(2): 332-344.
[11]	周伟, 苍慜楠, 程浩宗. 基于AR技术的文物数字化三维图像重建方法[J]. 图学学报, 2025, 46(2): 369-381.
[12]	邱佳新, 宋倩云, 徐丹. 基于改进神经辐射场的民族舞蹈重建方法[J]. 图学学报, 2025, 46(2): 415-424.
[13]	熊超, 王云艳, 罗雨浩. 特征对齐与上下文引导的多视图三维重建[J]. 图学学报, 2024, 45(5): 1008-1016.
[14]	黄家晖, 穆太江. 动态三维场景重建研究综述[J]. 图学学报, 2024, 45(1): 14-25.
[15]	石敏, 王炳祺, 李兆歆, 朱登明. 一种带高光处理的无缝纹理映射方法[J]. 图学学报, 2024, 45(1): 148-158.