A method for 3D model reconstruction of large-diameter rescue wells based on multiple cameras

Authors

• GU Hairong, National Engineering Research Center of Highway Maintenance Equipment, Chang’an University, Xi’an 710064, China;Key Laboratory of Intelligent Manufacturing of Construction Machinery, Anhui Jianzhu University, Hefei 230601, ChinaFollow
• WANG Boyang, National Engineering Research Center of Highway Maintenance Equipment, Chang’an University, Xi’an 710064, China
• YANG Wenjuan, School of Water Conservancy and Environment, Chang’an University, Xi’an 710064, ChinaFollow
• TONG Yubo, National Engineering Research Center of Highway Maintenance Equipment, Chang’an University, Xi’an 710064, China
• WANG Jiaxi, National Engineering Research Center of Highway Maintenance Equipment, Chang’an University, Xi’an 710064, China
• SUN Lishun, National Engineering Research Center of Highway Maintenance Equipment, Chang’an University, Xi’an 710064, China

Abstract

Objective Surface drilling for mine rescue represents a critical technique in the emergency rescue system against mine disasters. To accurately assess the passability of large-diameter rescue wells during rescue operations, it is essential to build precise 3D models of the rescue wells for rapid reconstruction of rescue scenarios. Methods Based on the theory of multi-camera-based 3D model reconstruction, this study investigated the method of arranging four Intel D435i cameras within a large-diameter rescue well. Accordingly, a multi-camera-based 3D model reconstruction system for a large-diameter rescue well was established, involving the joint calibration of the camera array and the selection of optimal feature extraction and matching algorithms of well wall images. Using the large-diameter rescue well simulation platform, an experimental study was conducted on 3D model reconstruction of large-diameter rescue wells. The fitting of the shaft diameter using point cloud data enabled the passability of large-diameter rescue wells to be accurately determined. Results and Conclusions The proposed joint calibration method for the four-camera array significantly improved calibration accuracy, with errors below 0.5 pixels. With the optimized arrangement scheme, the multi-camera system could rapidly acquire the well wall images under a lift speed of up to 0.16 m/s. The algorithm combination, including the optimal structure from motion (SFM) algorithm for camera position and posture estimation, the scale invariant feature transform (SIFT) algorithm for information extraction of well wall textures, and the fast library for approximate nearest neighbors (FLANN) algorithm for mismatched point pair removal, enabled the efficient matching of well wall images and rapid 3D model reconstruction, thereby enhancing the realism and visualization effect of the 3D rescue well model. Comparison with the radius of the actual large-diameter rescue well revealed errors of circumferential point cloud radii below 1.37%. Furthermore, the whole model yielded point cloud radius errors of less than 4%, with 94% of the errors falling within 2.5%. The results of this study provide a scientific basis for assessing the passability of large-diameter rescue wells during emergency rescue operations against mine disasters.

Keywords

large-diameter rescue well, 3D model reconstruction, four-camera array, camera calibration, deformation detection, emergency rescue against mine disasters

DOI

10.12363/issn.1001-1986.25.07.0488

Recommended Citation

GU Hairong, WANG Boyang, YANG Wenjuan, et al. (2026) "A method for 3D model reconstruction of large-diameter rescue wells based on multiple cameras," Coal Geology & Exploration: Vol. 54: Iss. 6, Article 24.
DOI: 10.12363/issn.1001-1986.25.07.0488
Available at: https://cge.researchcommons.org/journal/vol54/iss6/24

Reference

[1] 许超,姚宁平,姜磊,等. 矿山救援大直径多级跟管钻进关键工艺参数[J]. 煤田地质与勘探,2025,53(4):235−242 XU Chao,YAO Ningping,JIANG Lei,et al. Critical technical parameters of large–diameter multi–stage casing drilling for mine rescue[J]. Coal Geology & Exploration,2025,53(4):235−242

[2] 顾海荣,罗佳,高子渝,等. 基于深度相机的大直径救援井三维模型重建研究[J]. 煤田地质与勘探,2023,51(5):188−197 GU Hairong,LUO Jia,GAO Ziyu,et al. 3D model reconstruction of large–diameter rescue well based on depth camera[J]. Coal Geology & Exploration,2023,51(5):188−197

[3] 江佳斌. 基于机器视觉的定位及缺陷识别智能检测技术研究与应用[D]. 杭州：浙江大学，2020. JIANG Jiabin. Key technology research and application in intelligent object positioning and element defects detection based on machine vision[D]. Hangzhou：Zhejiang University，2020.

[4] BUJAKOWSKI W,BIELEC B,MIECZNIK M,et al. Reconstruction of geothermal boreholes in Poland[J]. Geothermal Energy,2020,8(1):10.

[5] NIE Mingyu，SUN Guangmin，ZHANG Yadong，et al. Research on localization and three–dimensional reconstruction of borehole wall fractures based on ultrasonic imaging data[C]//2023 8th International Conference on Intelligent Computing and Signal Processing (ICSP). Xi’an：IEEE，2023：238–242.

[6] 董秀军. 三维激光扫描技术及其工程应用研究[D]. 成都：成都理工大学，2007. DONG Xiujun. The three–dimensional laser scanning technique and research on its engineering application[D]. Chengdu：Chengdu University of Technology，2007.

[7] RUTKOWSKI W,LIPECKI T. Use of the iPhone 13 pro LiDAR scanner for inspection and measurement in the mineshaft sinking process[J]. Remote Sensing,2023,15(21):5089.

[8] WALTON G,DELALOYE D,DIEDERICHS M S. Development of an elliptical fitting algorithm to improve change detection capabilities with applications for deformation monitoring in circular tunnels and shafts[J]. Tunnelling and Underground Space Technology,2014,43:336−349.

[9] 肖杰杰. 基于单目深度估计的三维重建算法研究[D]. 邯郸：河北工程大学，2024. XIAO Jiejie. Research on 3D reconstruction algorithm based on monocular depth estimation[D]. Handan：Hebei University of Engineering，2024.

[10] 罗佳. 基于深度相机的矿山救援井三维模型重建方法研究[D]. 西安：长安大学，2022. LUO Jia. Research on 3D model reconstruction method of mine rescue channel based on depth camera[D]. Xi’an：Chang’an University，2022.

[11] 查剑林. 井筒内壁成图系统关键技术研究[D]. 徐州：中国矿业大学，2015. ZHA Jianlin. Study on the key technology for shaft wall mapping system[D]. Xuzhou：China University of Mining and Technology，2015.

[12] 潘泉,于昕,程咏梅,等. 信息融合理论的基本方法与进展[J]. 自动化学报,2003,29(4):599−615 PAN Quan,YU Xin,CHENG Yongmei,et al. Essential methods and progress of information fusion theory[J]. Acta Automatica Sinica,2003,29(4):599−615

[13] 赵晓通. 基于RGB–D和Lidar传感器的煤矿救援竖井变形状态检测方法研究[D]. 西安：长安大学，2021. ZHAO Xiaotong. Deformation detection method of coal mine rescue shaft based on RGB–D and Lidar sensors[D]. Xi’an：Chang’an University，2021.

[14] 邱怡乐. 基于线结构光扫描的金属件表面缺陷检测[D]. 成都：电子科技大学，2020. QIU Yile. Surface defect detection based on line–structured laser scanning[D]. Chengdu：University of Electronic Science and Technology of China，2020.

[15] 黄鹏程,江剑宇,杨波. 双目立体视觉的研究现状及进展[J]. 光学仪器,2018,40(4):81−86 HUANG Pengcheng,JIANG Jianyu,YANG Bo. Research status and progress of binocular stereo vision[J]. Optical Instruments,2018,40(4):81−86

[16] 黄理达,吴林煌. 基于几何特征的图像畸变校正方法[J]. 电子技术应用,2025,51(9):68−72 HUANG Lida,WU Linhuang. Image distortion correction method based on geometric features[J]. Application of Electronic Technique,2025,51(9):68−72

[17] 王玲玲. 基于双目立体视觉的水下三维重建[D]. 杭州：浙江大学，2011. WANG Lingling. Underwater 3D reconstruction based on binocular stereo vision[D]. Hangzhou：Zhejiang University，2011.

[18] 陈朝伟,石林,项德贵. 长宁–威远页岩气示范区套管变形机理及对策[J]. 天然气工业,2016,36(11):70−75 CHEN Zhaowei,SHI Lin,XIANG Degui. Mechanism of casing deformation in the Changning–Weiyuan national shale gas project demonstration area and countermeasures[J]. Natural Gas Industry,2016,36(11):70−75

[19] 桂德竹. 基于组合宽角相机低空影像的城市建筑物三维模型构建研究[D]. 徐州：中国矿业大学，2010. GUI Dezhu. Study on construction of 3D building based on wide–angle and combine camera images from UAV[D]. Xuzhou：China University of Mining and Technology，2010.

[20] 员方. 柱表面三维变形高精度测量系统与分析方法[D]. 南京：东南大学，2022. YUAN Fang. High precision measurement system and analysis method for three–dimensional deformation of cylindrical surface[D]. Nanjing：Southeast University，2022.

[21] 陈月. 基于SIFT特征检测与匹配的快速图像拼接方法研究[D]. 长春：吉林大学，2017. CHEN Yue. Study on fast image stitching method based on SIFT feature detection and matching[D]. Changchun：Jilin University，2017.

[22] 赵璐璐,耿国华,周明全,等. 一种用于三维重建的彩色Sift准稠密匹配算法[J]. 计算机应用研究,2012,29(9):3543−3546 ZHAO Lulu,GENG Guohua,ZHOU Mingquan,et al. Quasi–dense matching based on color Sift algorithm for 3D reconstruction[J]. Application Research of Computers,2012,29(9):3543−3546

[23] 娄天禺. 基于上下文相关性的视觉特征表示及其应用[D]. 北京：北京交通大学，2012. LOU Tianyu. Contextual visual feature representation and application[D]. Beijing：Beijing Jiaotong University，2012.

[24] 吴金津. 改进的SIFT算法及其在图像拼接中的应用[D]. 株洲：湖南工业大学，2015. WU Jinjin. Improved SIFT algorithm and its application in image mosaic[D]. Zhuzhou：Hunan University of Technology，2015.

[25] 朱鸣镝,陈婵,孙东岳. 基于特征点匹配的改进模板匹配算法[J]. 计算机与数字工程,2022,50(2):377−382 ZHU Mingdi,CHEN Chan,SUN Dongyue. Improved template matching algorithm based on feature point matching[J]. Computer & Digital Engineering,2022,50(2):377−382

[26] 曾攀. 点面特征融合的室内视觉定位方法研究[D]. 南京：东南大学，2020. ZENG Pan. Research on indoor visual localization method based on fusion of point–plane feature[D]. Nanjing：Southeast University，2020.

[27] 崔宾阁,孟翱翔. 基于最近邻有向图的遥感图像快速分割算法[J]. 计算机科学,2013,40(10):274−278 CUI Binge,MENG Aoxiang. Fast remote sensing image segmentation algorithm based on nearest neighbor direct graph[J]. Computer Science,2013,40(10):274−278

[28] 陈立潮,张秀琴,潘理虎,等. 煤矿考勤系统中人脸识别算法的研究[J]. 工矿自动化,2015,41(4):69−73 CHEN Lichao,ZHANG Xiuqin,PAN Lihu,et al. Research of face recognition algorithm of coal mine workers attendance system[J]. Industry and Mine Automation,2015,41(4):69−73

[29] 肖腾. 基于影像关联图优化的鲁棒SfM方法研究[D]. 武汉：武汉大学，2021. XIAO Teng. Research on robust structure from motion via optimizing view–graph[D]. Wuhan：Wuhan University，2021.

[30] 查正东. 弱纹理区域的无人机影像快速三维重建方法研究[D]. 西安：长安大学，2022. ZHA Zhengdong. Research on fast map reconstruction method based on UAV image[D]. Xi’an：Chang’an University，2022.

[31] 王刚. 多视觉信息融合的放射性区域重建[D]. 绵阳：西南科技大学，2021. WANG Gang. Radioactive area reconstruction with multi–visual information fusion[D]. Mianyang：Southwest University of Science and Technology，2021.

[32] 程亚彬,张宏伟,郭子路. 基于最小二乘法拟合圆的矿车车厢铆接孔检测技术研究[J]. 煤矿机械,2021,42(12):168−171 CHENG Yabin,ZHANG Hongwei,GUO Zilu. Research on riveting hole detection technology of tramcar carriage based on least square method fitting circle[J]. Coal Mine Machinery,2021,42(12):168−171