基于特征点提取的RANSAC-ICP三维点云配准方法

doi:10.6040/j.issn.1672-3961.0.2023.102

摘要/Abstract

摘要： 针对现有点云配准算法中易出现的误匹配、迭代时间长、精度低等问题,提出一种基于特征点提取的随机采样一致性与迭代最近点的三维点云配准方法,设计体素滤波降采样、关键特征点提取和几何特征描述、改进的随机采样一致性和点到面的迭代最近点算法框架。在降采样的基础上,提取点云关键几何特征点,并进行关键点邻域描述。采用四点对的随机采样一致性算法和点到面的迭代最近点算法分别进行点云粗配准与精细配准;采用K维树方法加速迭代,奇异值分解求解最优变换矩阵,最终实现三维点云配准。利用激光雷达及配套设备,模拟无人车辆采集点云数据,并选取不同时间间隔的点云图像,引入均方根误差和运算时间指标,验证算法的性能。试验结果表明:在粗配准阶段,配准速度相较于采样一致性算法和四点一致性算法,配准速度平均提高78.44%和61.02%,在处理100帧以下的数据时,配准误差在10 cm范围内;在精配准阶段,配准误差较粗配准、正态分布变换算法、传统的迭代最近点算法分别降低5.11、4.94和0.53 cm,配准时间较传统的迭代最近点算法平均提高33.06%。

关键词: 激光SLAM, 点云配准, 关键点提取, 随机采样一致性, 迭代最近点

中图分类号:

TN958.98

李岩,张子毅,王建柱. 基于特征点提取的RANSAC-ICP三维点云配准方法[J]. 山东大学学报 (工学版), 2024, 54(5): 144-154.

LI Yan, ZHANG Ziyi, WANG Jianzhu. RANSAC-ICP 3D point cloud registration method based on feature point extraction[J]. Journal of Shandong University(Engineering Science), 2024, 54(5): 144-154.

参考文献

[1] 康俊民, 赵祥模, 徐志刚. 无人车行驶环境特征分类方法[J]. 交通运输工程学报, 2016, 16(6): 140-148. KANG Junmin, ZHAO Xiangmo, XU Zhigang. Classification method of running environment features for unmanned vehicle[J]. Journal of Traffic and Transportation Engineering, 2016, 16(6): 140-148.
[2] 王润民, 朱宇, 赵祥模, 等. 自动驾驶测试场景研究进展[J]. 交通运输工程学报, 2021, 21(2): 21-37. WANG Runmin, ZHU Yu, ZHAO Xiangmo, et al. Research progress on test scenario of autonomous driving[J]. Journal of Traffic and Transportation Engineering, 2021, 21(2): 21-37.
[3] TREJOS Kevin, RINCÓN Laura, BOLANOS Miguel, et al. 2D SLAM algorithms characterization, calibration, and comparison considering cose rrror, map accuracy as well as CPU and memory usage[J]. Sensors, 2022, 22(18): 6903-6940.
[4] 吴建清, 宋修广. 同步定位与建图技术发展综述[J]. 山东大学学报(工学版), 2021, 51(5): 16-31. WU Jianqing, SONG Xiuguang. Review on development of simultaneous localization and mapping technology[J]. Journal of Shandong University(Engineering Science), 2021, 51(5): 16-31.
[5] 高扬, 曹王欣, 夏洪垚, 等. 低可见度环境下基于同步定位与构图的无人驾驶汽车定位算法[J]. 交通运输工程学报, 2022, 22(3): 251-262. GAO Yang, CAO Wangxin, XIA Hongyao, et al. Driverless vehicle positioning algorithm based on simultaneous positioning and mapping in low-visibility environment[J]. Journal of Traffic and Transportation Engineering, 2022, 22(3): 251-262.
[6] WU Jianqing, XU Hao, SUN Renjuan, et al. Road boundary-enhanced automatic background filtering for roadside LiDAR sensors[J]. IEEE Intelligent Transpor-tation Systems Magazine, 2021, 14(4): 60-72.
[7] 杨佳琪, 张世坤, 范世超, 等. 多视图点云配准算法综述[J]. 华中科技大学学报(自然科学版), 2022, 50(11): 16-34. YANG Jiaqi, ZHANG Shikun, FAN Shichao, et al. Survey on multi-view point cloud registration algorithm[J]. Journal of Huazhong University of Science and Technology(Natural Science Edition), 2022, 50(11): 16-34.
[8] MARTÍN Bueno, HENSE González-Jorge, JOAQUÍN Martínez-Sánchez, et al. Automatic point cloud coarse registration using geometric keypoint descriptors for indoor scenes[J]. Automation in Construction, 2017, 81: 134-148.
[9] 刘江, 张旭, 朱继文. 一种基于K-D树优化的ICP三维点云配准方法[J]. 测绘工程, 2016, 25(6): 15-18. LIU Jiang, ZHANG Xu, ZHU Jiwen. ICP three-dimensional point cloud registration based on K-D tree optimization[J]. Engineering of Surveying and Mapping, 2016, 25(6): 15-18.
[10] 康俊民, 赵祥模, 杨荻. 二维激光雷达数据角点特征的提取[J]. 交通运输工程学报, 2018, 18(3): 228-238. KANG Junmin, ZHAO Xiangmo, YANG Di. Corner feature extraction of 2D lidar data[J]. Journal of Traffic and Transportation Engineering, 2018, 18(3): 228-238.
[11] 蓝秦隆. 基于改进GA-SA的点云配准算法研究[D]. 武汉:武汉大学, 2021. LAN Qinlong. Research on point cloud registration algorithm based on improved GA-SA[D]. Wuhan: Wuhan University, 2021.
[12] ZHANG Ji, SINGH Sanjiv. Low-drift andreal-time Lidar odometry and mapping[J]. Autonomous Robots, 2017, 41(2): 401-416.
[13] 伍梦琦, 李中伟, 钟凯, 等. 基于几何特征和图像特征的点云自适应拼接方法[J]. 光学学报, 2015, 35(2): 237-244. WU Mengqi, LI Zhongwei, ZHONG Kai, et al. Adaptive point cloud registration method based on geometric features and photometric features[J]. Acta Optica Sinica, 2015, 35(2): 237-244.
[14] ZHANG Jing, YU Keping, WEN Zheng, et al. 3D reconstruction for motion blurred images using deep learning-based intelligent systems[J]. CMC-computers Materials & Continua, 2021, 66(2): 2087-2104.
[15] DERPANIS Konstantinos. Overview of the RANSAC algorithm[J]. Image Rochester NY, 2010, 4(1): 2-3.
[16] 王鹏, 朱睿哲, 孙长库. 基于改进的RANSAC的场景分类点云粗配准算法[J]. 激光与光电子学进展, 2020, 57(4): 312-320. WANG Peng, ZHU Ruizhe, SUN Changku. Point cloud coarse registration algorithm with scene classification based on improved RANSAC[J]. Laser & Optoelectronics Progress, 2020, 57(4): 312-320.
[17] WU Qinghua, LIU Jiacheng, GAO Can, et al. Improved RANSAC point cloud spherical target detection and parameter estimation method based on principal curvature constraint[J]. Sensors, 2022, 22(15): 5850.
[18] 梁涛, 韩峰, 陈国栋. 基于连续点云数据的既有铁路轨面信息快速提取算法设计[J]. 铁道科学与工程学报, 2021, 18(10): 2544-2551. LIANG Tao, HAN Feng, CHEN Guodong. Algorithm design for fast extraction of rail-surface information for existing railway based on continuous point cloud data[J]. Journal of Railway Science and Engineering, 2021, 18(10): 2544-2551.
[19] MOHAMAD Mustafa, AHMED Mirza Tahir, RAPPAPORT David, et al. Super generalized 4pcs for 3d registration[C] //2015 International Conference on 3D Vision. Lyon, France: IEEE, 2015: 598-606.
[20] 石雪飞, 徐梓齐, 朱荣, 等. 基于三维模型重构技术的公路预制构件尺寸检验评价方法[J]. 交通运输工程学报, 2021, 21(2): 66-81. SHI Xuefei, XU Ziqi, ZHU Rong, et al. Dimensional inspection and evaluation method of highway prefabricated components based on3D model reconstruction technology[J]. Journal of Traffic and Transportation Engineering, 2021, 21(2): 66-81.
[21] 周勇, 吕琛, 侯福金, 等. 基于坐标转换的多路侧激光雷达数据配准方法[J]. 山东大学学报(工学版), 2022, 52(6): 41-49. ZHOU Yong, LÜ Chen, HOU Fujin, et al. Datafusion method of multi roadside LiDAR based on coordinate transformation[J]. Journal of Shandong University(Engineering Science), 2022, 52(6): 41-49.
[22] 李慧慧, 刘超, 陶远. 一种改进的ICP激光点云精确配准方法[J]. 激光杂志, 2021, 42(1): 84-87. LI Huihui, LIU Chao, TAO Yuan. A laser point cloud precise registration method with improved ICP[J]. Laser Journal, 2021, 42(1): 84-87.
[23] ZHANG Juyong, YAO Yuxin, DENG Bailin. Fast and robust iterative closest point[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(7): 3450-3466.
[24] GUAN Wei, LI Wentao, REN Yan. Point cloud registration based on improved ICP algorithm[C] //2018 Chinese Control and Decision Conference(CCDC). Shenyang, China: IEEE, 2018: 1461-1465.
[25] CHEN Yang, MEDIONI Gérard. Object modelling by registration of multiple range images[J]. Image and Vision Computing, 1992, 10(3): 145-155.
[26] SEGALV Aleksandr, HAEHNEL Dirk, THRUN Sebastian. Generalized-icp[J]. Robotics: Science and Systems, 2009, 2(4): 435-443.
[27] DU Shaoyi, LIU Juan, ZHANG Chunjia, et al. Probability iterative closest point algorithm for mD point set registration with noise[J]. Neurocomputing, 2015, 157: 187-198.
[28] WU Jianqing, LÜ Chen, YUE Hongya. Grid-based lane identification with roadside LiDAR data[J]. International Journal of Sensor Networks, 2022, 38(2): 85-96.
[29] WU Jianqing, TIAN Yuan, XU Hao, et al. Automatic ground points filtering of roadside LiDAR data using a channel-based filtering algorithm[J]. Optics & Laser Technology, 2019, 115: 374-383.
[30] RUSU Radu Bogdan, BLODOW Nico, BEETZ Michael. Fast point feature histograms(FPFH)for 3D registration[C] //2009 IEEE International Conference on Robotics and Automation. Kobe, Japan: IEEE, 2009: 3212-3217.
[31] 张晗, 康国华, 张琪, 等. 基于改进SAC-IA算法的激光点云粗配准[J]. 航天控制, 2019, 37(5): 67-74. ZHANG Han, KANG Guohua, ZHANG Qi, et al. Laser point cloud coarse registration based on improved SAC-IA algorithm[J]. Aerospace Control, 2019, 37(5): 67-74.
[32] LI Lin, YANG Fan, ZHU Haihong, et al. An improved RANSAC for 3D point cloud plane segmentation based on normal distribution transformation cells[J]. Remote Sensing, 2017, 9(5): 433-449.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed