面向磁浮轨道的多源点云数据的混合滤波方法

doi:10.19678/j.issn.1000-3428.0067725

摘要/Abstract

摘要：

在磁浮轨道的仿真数据处理过程中, 磁浮轨道点云数据的滤波提取是重要环节之一, 实际应用应根据待提取的磁浮数据特性, 采用高效的滤波方法。磁浮轨道的点云数据对象主要包括由无人机(UAV)倾斜摄影获取的磁浮轨道的图像数据并经过三维重建后形成的稠密点云数据、由手持式激光雷达扫描磁浮轨道获取的激光点云数据。根据这两种点云的数据特性, 考虑磁浮轨道四周复杂场景的点云环境, 分别对两种点云进行混合滤波。首先, 对激光点云数据采用八叉树下采样方法, 有效降低了点云数据的数量级, 节省了运行时间。然后, 分别对激光点云与稠密点云数据采用布料模拟滤波(CSF)方法, 过滤了地平面点云数据, 保留了非地面点云数据; 采用统计离群点去除(SOR)滤波方法, 筛除了大量离群点; 根据磁浮轨道特征, 采用直通滤波过滤了坐标范围外的点云数据。实验结果表明, 在不影响磁浮轨道结构的前提下, 对于采用八叉树下采样方法的激光点云数据和没有采用八叉树下采样的稠密点云数据, 该方法的滤波率分别为86.15%和64.76%, 经混合滤波后的两种点云数据的结构近似, 点云数量处于同一数量级, 为磁浮轨道点云特征提取等后续任务提供了有效保障。

关键词: 磁浮轨道, 多源点云数据, 八叉树下采样, 布料模拟滤波, 统计离群点去除滤波

Abstract:

In the simulation data processing of maglev tracks, the filtering and extraction of maglev track point cloud data is an important link. Thus, practical applications should adopt an efficient filtering method according to the characteristics of the maglev data to be extracted. The point cloud data objects of the maglev track primarily include the image data of the maglev track, which is obtained by Unmanned Aerial Vehicle (UAV) oblique photography and formed into dense point cloud data after 3D reconstruction, and the laser point cloud data, which is obtained by handheld lidar scanning of the maglev track. Based on the data characteristics of these point clouds and considering the complex scenes around the maglev track, the two types of point clouds are mixed and filtered. First, the octree downsampling method is used for laser point cloud data, which effectively reduces the order of magnitude of the point cloud data and saves running time. The Cloth Simulation Filtering (CSF) method is then used on the laser point cloud and dense point cloud data to filter the ground plane point cloud and retain the non-ground point cloud data, respectively. A Statistical Outlier Removal (SOR) filtering method is used to screen a large number of outliers. Based on the characteristics of the maglev track, point clouds outside the coordinate range are filtered through straight-through filtering. On the premise of not changing the structure of the maglev track, the experimental results show that the filtering rates of the proposed method are 86.15% and 64.76% for the octree-downsampled laser point cloud data and the dense point cloud data without octree downsampling, respectively. These two point cloud datasets have similar structural ranges after hybrid filtering and a number of point clouds of the same order of magnitude, which can be effective for methods such as feature extraction of point clouds in maglev orbits.

Key words: maglev track, multisource point cloud data, octree downsampling, Cloth Simulation Filtering(CSF), Statistical Outlier Removal (SOR) filtering

张玉鑫, 张雷, 欧冬秀. 面向磁浮轨道的多源点云数据的混合滤波方法[J]. 计算机工程, 2024, 50(9): 54-62.

ZHANG Yuxin, ZHANG Lei, OU Dongxiu. Hybrid Filtering Method for Multisource Point Cloud Data of Maglev Tracks[J]. Computer Engineering, 2024, 50(9): 54-62.

https://www.ecice06.com/CN/Y2024/V50/I9/54

图/表 16

图1 多源点云数据混合滤波方法框架

Fig.1 Framework of hybrid filtering method for multisource point cloud data

图2 八叉树拓扑结构简易示意图

Fig.2 Simplified schematic diagram of the octree topology

图3 CSF示意图

Fig.3 Schematic diagram of CSF

图4 激光扫描仪采集的点云数据示意图

Fig.4 Schematic diagram of point cloud data collected by laser scanner

图5 同济大学嘉定校区1.5 km磁浮交通试验线全线

Fig.5 Entire 1.5 km maglev traffic test line of Jiading Campus of Tongji University

图6 磁浮轨道梁物理结构的无人机图像

Fig.6 Unmanned aerial vehicle images of the physical structure of the maglev track beam

图7 磁浮三维重建稠密点云整体效果

Fig.7 Overall effect of maglev 3D reconstruction of dense point clouds

图8 局部桥墩与桥梁俯视图

Fig.8 Top view of local bridge piers and beams

图9 CSF后的激光点云局部图

Fig.9 Local view of laser point cloud after CSF

图10 混合滤波后的激光点云示意图

Fig.10 Schematic diagram of laser point cloud after hybrid filtering

图11 稠密点云SOR滤波对比

Fig.11 Comparison of SOR filtering of dense point clouds

图12 稠密点云道岔结构SOR滤波对比

Fig.12 Comparison of SOR filtering of dense point cloud turnout structure

图13 一条分割后的道岔轨道稠密点云示意图

Fig.13 Schematic diagram of dense point cloud of a split turnout track

图14 稠密点云CSF对比

Fig.14 Comparison of CSF of dense point clouds

图15 混合滤波后的稠密点云示意图

Fig.15 Schematic diagram of dense point clouds after hybrid filtering

参考文献 39

1	XU Y S, TONG X H, STILLA U. Voxel-based representation of 3D point clouds: methods, applications, and its potential use in the construction industry. Automation in Construction, 2021, 126, 103675. doi: 10.1016/j.autcon.2021.103675
2	DIGNE J, DE FRANCHIS C. The bilateral filter for point clouds. Image Processing On Line, 2017, 7, 278- 287. doi: 10.5201/ipol.2017.179&restype=unixref&xml=\|Nat Cell Biol\|\|10\|\|295\|2008\|\|\|
3	TOMASI C, MANDUCHI R. Bilateral filtering for gray and color images[C]//Proceedings of the 6th International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 1998: 839-846.
4	CHEN B H, TSENG Y S, YIN J L. Gaussian-adaptive bilateral filter. IEEE Signal Processing Letters, 2020, 27, 1670- 1674. doi: 10.1109/LSP.2020.3024990
5	KURUP A M, BOS J. DSOR: a scalable statistical filter for removing falling snow from LiDAR point clouds in severe winter weather[EB/OL]. [2023-04-11]. http://arxiv.org/abs/2109.07078v2.
6	ESMEIDE A, NARVAEZ L. Point cloud denoising using robust principal component analysis[EB/OL]. [2023-04-11]. https://www.semanticscholar.org/paper/Point-cloud-denoising-using-robust-principal-Narv%C3%A1ez-Narv%C3%A1ez/8f7cf639363210f64305e2edc7c10d0f0d6ff83e.
7	JENKE P, WAND M, BOKELOH M, et al. Bayesian point cloud reconstruction. Computer Graphics Forum, 2006, 25(3): 379- 388. doi: 10.1111/j.1467-8659.2006.00957.x
8	李倩, 向海昀, 张玉婷, 等. 结合高斯滤波与MASK的G-MASK人脸对抗攻击. 计算机工程, 2024, 50(2): 308- 316. URL
	LI Q, XIANG H Y, ZHANG Y T, et al. G-MASK facial adversarial attack combining Gaussian filtering and MASK. Computer Engineering, 2024, 50(2): 308- 316. URL
9	GAO W, TEDRAKE R. FilterReg: robust and efficient probabilistic point-set registration using Gaussian filter and twist parameterization[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D. C., USA: IEEE Press, 2019: 11087-11096.
10	ALQADI Z A, BARAKAT M T. A case study to improve the quality of median filter. International Journal of Computer Science and Mobile Computing, 2021, 10(11): 19- 28. doi: 10.47760/ijcsmc.2021.v10i11.004
11	CAI S S, ZHANG W M, LIANG X L, et al. Filtering airborne LiDAR data through complementary cloth simulation and progressive TIN densification filters. Remote Sensing, 2019, 11(9): 1037. doi: 10.3390/rs11091037
12	YANG A X, WU Z Y, YANG F L, et al. Filtering of airborne LiDAR bathymetry based on bidirectional cloth simulation. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 163, 49- 61. doi: 10.1016/j.isprsjprs.2020.03.004
13	PATRICK S, WEINMANN M, KLEIN R. Incomplete Gamma kernels: generalizing locally optimal projection operators[EB/OL]. [2023-04-11]. https://arxiv.org/abs/2205.01087.
14	FAIGENBAUM-GOLOVIN S, LEVIN D. Manifold reconstruction and denoising from scattered data in high dimension. Journal of Computational and Applied Mathematics, 2023, 421, 114818. doi: 10.1016/j.cam.2022.114818
15	顾砾, 季怡, 刘纯平. 基于多模态特征融合的三维点云分类方法. 计算机工程, 2021, 47(2): 279- 284. URL
	GU L, JI Y, LIU C P. Classification method of three-dimensional point cloud based on multiple modal feature fusion. Computer Engineering, 2021, 47(2): 279- 284. URL
16	ZHANG W M, QI J B, WAN P, et al. An easy-to-use airborne LiDAR data filtering method based on cloth simulation. Remote Sensing, 2016, 8(6): 501. doi: 10.3390/rs8060501
17	XIONG B, JIANG W Z, LI D K, et al. Voxel grid-based fast registration of terrestrial point cloud. Remote Sensing, 2021, 13(10): 1905. doi: 10.3390/rs13101905
18	SHI W J, XU J W, ZHU D C, et al. RGB-D semantic segmentation and label-oriented voxel grid fusion for accurate 3D semantic mapping. IEEE Transactions on Circuits and Systems for Video Technology, 2022, 32(1): 183- 197. doi: 10.1109/TCSVT.2021.3056726
19	POUX F, BILLEN R. Voxel-based 3D point cloud semantic segmentation: unsupervised geometric and relationship featuring vs deep learning methods. ISPRS International Journal of Geo-Information, 2019, 8(5): 213. doi: 10.3390/ijgi8050213
20	FAIRFIELD N, KANTOR G, WETTERGREEN D. Real-time SLAM with octree evidence grids for exploration in underwater tunnels. Journal of Field Robotics, 2007, 24(1/2): 3- 21.
21	杨铭, 张晓丽, 霍朗宁, 等. 森林区域机载LiDAR点云数据的改进八叉树滤波算法研究. 北京林业大学学报, 2018, 40(11): 102- 111. URL
	YANG M, ZHANG X L, HUO L N, et al. Improved octree filtering algorithm of airborne LiDAR data in forest environment. Journal of Beijing Forestry University, 2018, 40(11): 102- 111. URL
22	BUJÁN S, CORDERO M, MIRANDA D. Hybrid overlap filter for LiDAR point clouds using free software. Remote Sensing, 2020, 12(7): 1051. doi: 10.3390/rs12071051
23	ŠTULAR B, LOZIĆ E. Comparison of filters for archaeology-specific ground extraction from airborne LiDAR point clouds. Remote Sensing, 2020, 12(18): 3025. doi: 10.3390/rs12183025
24	CHEN C F, CHANG B T, LI Y Y, et al. Filtering airborne LiDAR point clouds based on a scale-irrelevant and terrain-adaptive approach. Measurement, 2021, 171, 108756. doi: 10.1016/j.measurement.2020.108756
25	唐超, 左文涛, 李小飞. 结合修剪均值与高斯加权中值滤波的图像去噪算法. 计算机工程, 2021, 47(9): 210- 216. URL
	TANG C, ZUO W T, LI X F. Image denoising algorithm combining trimmed mean and Gaussian weighted median filtering. Computer Engineering, 2021, 47(9): 210- 216. URL
26	ZAMAN F, WONG Y P, NG B Y. Density-based denoising of point cloud[EB/OL]. [2023-04-11]. http://arxiv.org/abs/1602.05312.
27	KUMAR A, SODHI S S. Comparative analysis of Gaussian filter, median filter and denoise autoenocoder[C]//Proceedings of the 7th International Conference on Computing for Sustainable Global Development. Washington D. C., USA: IEEE Press, 2020: 45-51.
28	PAULY M, GROSS M. Spectral processing of point-sampled geometry[C]//Proceedings of the 28th Annual Conference on Computer Graphics and Interactive Techniques. New York, USA: ACM Press, 2001: 379-386.
29	CLARENZ U, RUMPF M, TELEA A. Fairing of point based surfaces[C]//Proceedings of International Conference on Computer Graphics. Washington D. C., USA: IEEE Press, 2004: 600-603.
30	熊嘉阳, 邓自刚. 高速磁悬浮轨道交通研究进展. 交通运输工程学报, 2021, 21(1): 177- 198. URL
	XIONG J Y, DENG Z G. Research progress of high-speed maglev rail transit. Journal of Traffic and Transportation Engineering, 2021, 21(1): 177- 198. URL
31	蔡文锋, 张威风, 冯洋, 等. 提速情况下磁浮轨道结构振动响应及传递特性研究. 铁道标准设计, 2023, 67(3): 55- 60. URL
	CAI W F, ZHANG W F, FENG Y, et al. Study on the vibration response and transfer characteristics of maglev track structures under speed-up. Railway Standard Design, 2023, 67(3): 55- 60. URL
32	蔡文涛, 王春江, 滕念管, 等. 超高速磁浮轨道梁体系的跨平台耦合振动分析. 上海交通大学学报, 2021, 55(10): 1228- 1236. URL
	CAI W T, WANG C J, TENG N G, et al. Analysis of cross-platform coupling vibration of ultra-high-speed maglev track beam system. Journal of Shanghai Jiao Tong University, 2021, 55(10): 1228- 1236. URL
33	李洪鲁. 基于图像的高速磁浮长定子轨道异常状况的检测研究[D]. 长沙: 国防科技大学, 2019.
	LI H L. Image-based detection of abnormal condition for high-speed maglev long stator track[D]. Changsha: National University of Defense Technology, 2019. (in Chinese)
34	黄玉龙. 基于视频图像的管道裂纹缺陷检测方法研究[D]. 西安: 西安理工大学, 2018.
	HUANG Y L. Research on pipeline crack detection based on video image[D]. Xi'an: Xi'an University of Technology, 2018. (in Chinese)
35	姚连璧, 刘昊, 孙向东, 等. 一种磁浮滑行面点云数据提取方法: CN112581579B[P]. 2022-11-18.
	YAO L B, LIU H, SUN X D, et al. A method for extracting point cloud data of maglev sliding surface: CN112581579B[P]. 2022-11-18. (in Chinese)
36	SCHÖNBERGER J L, FRAHM J M. Structure-from-motion revisited[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Washington D. C., USA: IEEE Press, 2016: 4104-4113.
37	SCHÖNBERGER J L, ZHENG E L, FRAHM J M, et al. Pixelwise view selection for unstructured multi-view stereo[C]//Proceedings of the European Conference on Computer Vision. Berlin, Germany: Springer, 2016: 501-518.
38	陈武, 姜三, 李清泉, 等. 无人机影像增量式运动恢复结构研究进展. 武汉大学学报(信息科学版), 2022, 47(10): 1662- 1674. URL
	CHEN W, JIANG S, LI Q Q, et al. Research progress on incremental motion recovery structures for unmanned aircraft images. Journal of Wuhan University (Information Science Edition), 2022, 47(10): 1662- 1674. URL
39	FURUKAWA Y, PONCE J. Accurate, dense, and robust multiview stereopsis. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(8): 1362- 1376.

选择文件类型/文献管理软件名称

选择包含的内容