Multi scale fusion point cloud denoising method based on improved statistical filtering
-
摘要:
随着矿山智能化的发展，三维激光扫描技术也逐步应用到煤矿巷道监测中。由于井下煤矿巷道复杂的环境因素影响，利用三维激光扫描技术所获取的点云数据通常包含一系列噪点，会对后续的点云数据处理产生不利影响。研究采用曲率与统计滤波联合优化的方法，提出了一种基于改进统计滤波的多尺度融合点云去噪方法。首先，利用KD-tree构建待去噪巷道原始点云拓扑结构，在原始点云中每个点处构建邻域，并计算邻域内每2个点之间的平均欧氏距离和每个点的曲率；然后，设置曲率阈值对点云进行区分，将原始点云中的曲率值与预设的大尺度噪声阈值及其他噪声阈值进行比较，筛选得到初始去噪点云；之后，由于初始去噪点云中还可能包含部分离散点，再利用半径滤波去除较少的离散点，完成点云去噪；最后，在瓦斯煤层爆炸实验室模拟巷道和防空洞拱形巷道试验环境中对所提出的点云去噪方法进行试验研究。结果表明：所提出的基于改进统计滤波的多尺度融合点云去噪方法，能够在尽可能保留巷道表面点的情况下有效去除噪声点，并且通过与传统点云去噪方法的试验对比发现，传统统计滤波去噪后的巷道点云出现不同程度的空洞，已无法维持巷道结构，而本研究所提出的曲率与统计滤波的联合优化方法依旧保留着完整的巷道结构，在实验室模拟巷道中特征点保留率能够提高4.624%，在拱形模拟巷道中特征点保留率能够提高10.27%，煤矿巷道现场试验证明，该方法弥补了传统点云去噪方法的不足，更能满足矿山巷道监测的需求。
Abstract:With the development of intelligent mining, 3D laser scanning technology has gradually been applied to coal mine tunnel monitoring. However, due to the complex environmental factors in underground coal mines, the point cloud data obtained through 3D laser scanning usually contains a series of noise points, which adversely affect subsequent point cloud data processing. This study proposes a multi-scale fusion point cloud denoising method based on improved statistical filtering using a joint optimization approach of curvature and statistical filtering. First, a KD-tree is used to construct the topological structure of the raw point cloud in the tunnel to be denoised. For each point in the raw point cloud, a neighborhood is constructed, and the average Euclidean distance between every two points in the neighborhood as well as the curvature of each point are calculated. Then, a curvature threshold is set to differentiate the points, comparing the curvature values of the raw point cloud with large-scale noise thresholds and other noise thresholds to filter out an initial denoised point cloud. Afterward, since the initial denoised point cloud may still contain some scattered points, a radius filter is applied to remove the fewer scattered points, completing the point cloud denoising process. Finally, the proposed point cloud denoising method is experimentally studied in a simulated tunnel environment of gas and coal dust explosion laboratory and arched tunnel. The results show that the proposed multi-scale fusion point cloud denoising method based on an improved statistical filter can effectively remove noise points while retaining the surface points of the tunnel as much as possible. Through experimental comparison with traditional point cloud denoising methods, it was found that the tunnel point cloud denoised by traditional statistical filtering had varying degrees of voids and could no longer maintain the tunnel structure. However, the joint optimization method of curvature and statistical filtering proposed in this paper still retains the complete tunnel structure. In laboratory simulated tunnel, the feature point retention rate can be increased by 4.624%, and in arch simulated tunnel, the feature point retention rate can be increased by 10.27%. Field tests in coal mine roadway have proven that this method can compensate for the shortcomings of traditional point cloud denoising methods and better meet the requirements of coal mine roadway monitoring.
-
-
![]()
图 4 巷道表面点与噪声点邻域曲率示意
Figure 4. Schematic diagram of curvature in area of roadway surface points and noise points
![]()
图 9 曲率邻域参数与阈值参数对去噪效果影响
Figure 9. Influence of curvature domain parameters and threshold parameters on denoising performance
![]()
图 12 实验室巷道不同去噪方法效果对比
Figure 12. Comparison of effectiveness of different denoising methods in experimental tunnels
![]()
图 13 拱形巷道不同去噪方法效果对比
Figure 13. Comparison of effectiveness of different denoising methods in arched tunnel
表 1 传统方法与本方法去噪结果统计
Table 1 Statistical analysis of denoising results between traditional method and this method
去噪前后 传统方法 改进方法 φ=1 φ=0.6 φ1=1，
φ2=2φ1=0.6，
φ2=2去噪前总点数 1 264 153 1 264 153 1 264 153 1 264 153 去噪后总点数 1 219 989 1 162 313 1 236 985 1 219 533 去噪前行人点数 25 907 25 907 25 907 25 907 去除行人点数 23 035 25 459 22 698 25 490 行人去噪率/% 88.914 98.271 87.613 98.390 去噪前其他点数 1 238 246 1 238 246 1 238 246 1 238 246 去噪后其他点数 1 217 117 1 161 865 1 233 776 1 219 116 其他点保留率/% 98.294 93.832 99.639 98.455 去噪精度 0.964 59 0.918 98 0.978 03 0.964 23 特征保留率/% 87.004 86.817 87.183 87.165 
下载: 导出CSV
表 2 拱形巷道去噪结果对比统计
Table 2 Comparison and statistics of denoising results for arched tunnel
去噪前后 传统方法 改进方法 φ=1 φ=0.6 φ1=1，
φ2=2φ1=0.6，
φ2=2去噪前总点数 2 035 766 2 035 766 2 035 766 2 035 766 去噪后总点数 1 848 536 1 763 765 1 976 178 1 969 499 去噪前行人噪点 33 351 33 351 33 351 33 351 去除行人噪点 33 186 33 351 33 057 33 266 行人去噪率/% 99.505 100.000 99.118 99.745 去噪前其他点数 2 002 415 2 002 415 2 002 415 2 002 415 去噪后其他点数 1 848 371 1 763 765 1 975 884 1 969 414 其他点保留率/% 92.307 88.082 98.675 98.352 去噪精度 0.90709 0.86548 0.96976 0.96648 特征保留率/% 93.413 93.546 93.853 93.932
下载: 导出CSV
-
[1] 吴一全,陈慧娴,张耀. 基于深度学习的三维点云处理方法研究进展[J]. 中国激光,2024,51(5):0509001. doi: 10.3788/CJL230924 WU Yiquan,CHEN Huixian,ZHANG Yao. Review of 3D point cloud processing methods based on deep learning[J]. Chinese Journal of Lasers,2024,51(5):0509001. doi: 10.3788/CJL230924
[2] 杨林,马宏伟,王岩. 基于激光惯性融合的煤矿井下移动机器人SLAM算法[J]. 煤炭学报,2022,47(9):3523−3534. YANG Lin,MA Hongwei,WANG Yan. LiDAR-Inertial SLAM for mobile robot in underground coal mine[J]. Journal of China Coal Society,2022,47(9):3523−3534.
[3] 肖俊,石光田. 三维点云去噪技术[J]. 中国科学院大学学报,2023,40(5):577−595. XIAO Jun,SHI Guangtian. Three-dimensional point cloud denoising[J]. Journal of University of Chinese Academy of Sciences,2023,40(5):577−595.
[4] 亓玉浩,关士远. 基于激光SLAM的综采工作面实时三维建图方法[J]. 工矿自动化,2022,48(11):139−144. QI Yuhao,GUAN Shiyuan. Real-time 3D mapping method of fully mechanized working face based on laser SLAM[J]. Journal of Mine Automation,2022,48(11):139−144.
[5] 魏恋欢,刘善军,杨天鸿,等. 联合LiDAR DEM与时序SAR数据的露天矿特大型滑坡监测[J]. 东北大学学报(自然科学版),2021,42(12):1753−1760. WEI Lianhuan,LIU Shanjun,YANG Tianhong,et al. Collaborative monitoring of open pit mine induced large scale landslide with LiDAR DEM and time series SAR data[J]. Journal of Northeastern University (Natural Science),2021,42(12):1753−1760.
[6] SI L,WANG Z B,LIU P,et al. A novel coal–rock recognition method for coal mining working face based on laser point cloud data[J]. IEEE Transactions on Instrumentation and Measurement,2021,70:2514118.
[7] 马宏伟,苏浩,薛旭升,等. 煤矿井下移动机器人激光标靶定位方法研究[J]. 煤炭科学技术,2024,52(11):60−73. doi: 10.12438/cst.2024-1008 MA Hongwei,SU Hao,XUE Xusheng,et al. Research on laser target positioning method for underground mobile robot in coal mine[J]. Coal Science and Technology,2024,52(11):60−73. doi: 10.12438/cst.2024-1008
[8] KUKUTSCH R,KAJZAR V,KONICEK P,et al. Possibility of convergence measurement of gates in coal mining using terrestrial 3D laser scanner[J]. Journal of Sustainable Mining,2015,14(1):30−37. doi: 10.1016/j.jsm.2015.08.005
[9] KAJZAR V,KUKUTSCH R,WACLAWIK P,et al. Innovative approach to monitoring coal pillar deformation and roof movement using 3D laser technology[J]. Procedia Engineering,2017,191:873−879. doi: 10.1016/j.proeng.2017.05.256
[10] 刘晓阳,胡乔森,李慧娟. 基于三维激光扫描技术的巷道顶板监测研究[J]. 中国煤炭,2017,43(7):81−84,107. doi: 10.3969/j.issn.1006-530X.2017.07.021 LIU Xiaoyang,HU Qiaosen,LI Huijuan. Research on coal mine roof monitoring based on three-dimensional laser scanning technology[J]. China Coal,2017,43(7):81−84,107. doi: 10.3969/j.issn.1006-530X.2017.07.021
[11] 郭良林,周大伟,张德民,等. 基于激光点云的巷道变形监测及支护研究[J]. 煤矿安全,2020,51(8):178−183. GUO Lianglin,ZHOU Dawei,ZHANG Demin,et al. Research on deformation monitoring and supporting of tunnel based on laser point cloud[J]. Safety in Coal Mines,2020,51(8):178−183.
[12] 余国锋,郑群,陈昊喆. 三维可视化激光扫描技术在深井软岩巷道中的应用[J]. 中国矿山工程,2022,51(5):37−43. YU Guofeng,ZHENG Qun,CHEN Haozhe. Application of 3D visual laser scanning technology in deep softrock roadway[J]. China Mine Engineering,2022,51(5):37−43.
[13] 戴文祥,陈雷,闫鹏飞,等. 基于三维激光扫描的煤矿巷道形变监测方法[J]. 工矿自动化,2023,49(10):61−67,95. DAI Wenxiang,CHEN Lei,YAN Pengfei,et al. A deformation monitoring method for coal mine roadway based on 3D laser scanning[J]. Journal of Mine Automation,2023,49(10):61−67,95.
[14] ZHOU S T,LIU X L,WANG C Y,et al. Non-iterative denoising algorithm based on a dual threshold for a 3D point cloud[J]. Optics and Lasers in Engineering,2020,126:105921. doi: 10.1016/j.optlaseng.2019.105921
[15] 焦亚男,马杰,钟斌斌. 一种基于尺度变化的点云并行去噪方法[J]. 武汉大学学报(工学版),2021,54(3):277−282. JIAO Yanan,MA Jie,ZHONG Binbin. Point cloud parallel de-noising algorithms based on scale change[J]. Engineering Journal of Wuhan University,2021,54(3):277−282.
[16] 郭昌龙,夏振平,李超超,等. 结合改进半径滤波和局部平面拟合的点云去噪算法[J]. 激光与光电子学进展,2024,61(12):204−212. GUO Changlong,XIA Zhenping,LI Chaochao,et al. Point cloud denoising algorithm combined with improved radius filtering and local plane fitting[J]. Laser & Optoelectronics Progress,2024,61(12):204−212.
[17] 武晴晴,靳洁. 小波变换技术的三维激光扫描点云数据分类和去噪系统[J]. 自动化与仪器仪表,2023(6):35−39. WU Qingqing,JIN Jie. Classification and denoising system for three-dimensional laser scanning point cloud data using wavelet transform technology[J]. Automation & Instrumentation,2023(6):35−39.
[18] LIU Z,XIAO X W,ZHONG S S,et al. A feature-preserving framework for point cloud denoising[J]. Computer-Aided Design,2020,127:102857. doi: 10.1016/j.cad.2020.102857
[19] MA X,OIN C,YOU H,et al. Rethinking network design andlocal geometry in point cloud:A simple residual mlp frame-work[J]. ArXiv,2022(4):1−6.
[20] CHEN H H,WEI Z Y,LI X Z,et al. RePCD-net:Feature-aware recurrent point cloud denoising network[J]. International Journal of Computer Vision,2022,130(3):615−629. doi: 10.1007/s11263-021-01564-7
[21] WANG X T,FAN X P,ZHAO D B. PointFilterNet:A filtering network for point cloud denoising[J]. IEEE Transactions on Circuits and Systems for Video Technology,2023,33(3):1276−1290. doi: 10.1109/TCSVT.2022.3207789
[22] REN Z Y,KIM M,LIU F,et al. TIGER:Time-varying denoising model for 3D point cloud generation with diffusion process[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway,NJ:IEEE,2024:9462−9471.
[23] LIU Z,ZHAO Y W,ZHAN S J,et al. PCDNF:Revisiting learning-based point cloud denoising via joint normal filtering[J]. IEEE Transactions on Visualization and Computer Graphics,2024,30(8):5419−5436.
[24] 侯广哲,秦贵和,梁艳花. 基于下采样的自监督点云去噪方法[J]. 吉林大学学报(理学版),2024,62(1):100−105. HOU Guangzhe,QIN Guihe,LIANG Yanhua. Self-supervised point cloud denoising method based on downsampling[J]. Journal of Jilin University (Science Edition),2024,62(1):100−105.
[25] 郑理科,王健,李志远,等. 一种局部最优邻域法向量估算的巷道点云去噪方法[J]. 测绘科学,2023,48(4):140−148,171. ZHENG Like,WANG Jian,LI Zhiyuan,et al. A denoising method of roadway point cloud based on local optimal neighborhood normal vector estimation[J]. Science of Surveying and Mapping,2023,48(4):140−148,171.
[26] 连忠文,任助理,郝英豪,等. 基于区域生长的非结构巷道点云去噪方法[J]. 工矿自动化,2024,50(3):48−55. LIAN Zhongwen,REN Zhuli,HAO Yinghao,et al. A point cloud denoising method for unstructured roadways based on regional growth[J]. Journal of Mine Automation,2024,50(3):48−55.
[27] 任助理,袁瑞甫,王李管,等. 复杂地下巷道场景三维点云两阶段去噪方法[J/OL]. 煤炭科学技术,2024:1−15[2025−03−25]. https://kns.cnki.net/KCMS/detail/detail.aspx?filename=MTKJ2024032000C&dbname=CJFD&dbcode=CJFQ. REN Zhuli,YUAN Ruifu,WANG Liguan,et al. Two-stage denoising method for 3D point cloud in complex underground roadway scene[J/OL]. Coal Science and Technology,2024:1−15[2025−03−25]. https://kns.cnki.net/KCMS/detail/detail.aspx?filename=MTKJ2024032000C&dbname=CJFD&dbcode=CJFQ.
[28] 托雷,康志忠,谢远成,等. 利用三维点云数据的地铁隧道断面连续截取方法研究[J]. 武汉大学学报(信息科学版),2013,38(2):171−175,185. TUO Lei,KANG Zhizhong,XIE Yuancheng,et al. Continuously vertical section abstraction for deformation monitoring of subway tunnel based on terrestrial point clouds[J]. Geomatics and Information Science of Wuhan University,2013,38(2):171−175,185.
[29] XU X Y,YANG H,NEUMANN I. Time-efficient filtering method for three-dimensional point clouds data of tunnel structures[J]. Advances in Mechanical Engineering,2018,10(5):1687814018773159.
[30] ZHUA N N,JIAA Y H,LUO L. Tunnel point cloud filtering method based on elliptic cylindrical model[J]. ISPRS - International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2016,41:735−740.
[31] 吴姝,王涛,崔英花,等. 一种参数自适应的统计滤波降噪算法[J]. 应用激光,2024,44(9):124−132. WU Shu,WANG Tao,CUI Yinghua,et al. A statistical filtering noise reduction algorithm based on parameter adaptive[J]. Applied Laser,2024,44(9):124−132.
计量
- 文章访问数: 58
- HTML全文浏览量: 9
- PDF下载量: 28
