首页 > 论文 > 中国激光 > 46卷 > 7期(pp:710002--1)

基于深度图的三维激光雷达点云目标分割方法

Target Segmentation Method for Three-Dimensional LiDAR Point Cloud Based on Depth Image

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

三维激光雷达广泛应用在智能车系统中,点云目标分割是智能车环境感知中的关键技术。针对目前三维激光雷达点云目标分割算法实时性和准确性不高的问题,提出一种基于深度图的点云目标快速分割方法。将点云数据表示为深度图,建立深度图与点云数据的映射关系。利用激光雷达扫描线的角度阈值去除地面点云数据,结合深度图和自适应参数改进的DBSCAN(Density-Based Spatial Clustering of Applications with Noise)算法对非地面点云进行聚类分割。实验结果表明该方法相对于传统聚类算法在时间效率上有很大的提升,且能较好地降低欠分割错误率,分割准确度提升10%,达到了85.02%。

Abstract

Point cloud target segmentation is the key to perceive targets for a smart car using three-dimensional (3D) LiDAR. Aiming at the problems of poor real-time and low accuracy of the existing in 3D LiDAR point cloud target segmentation algorithms, an approach based on a depth map is proposed in this paper to realize fast and accurate segmentation for point cloud target segmentation. The original data are transformed into a depth map, and the mapping relationship between point cloud data and a depth map is established. After removing the ground point cloud data by using the angle threshold of the LiDAR scanning line, the non-ground point cloud is clustered and segmented by the improved DBSCAN(Density-Based Spatial Clustering of Applications with Noise) algorithm combined with the depth map and the adaptive parameters. Experimental results show that the proposed method has a significant improvement in time efficiency compared with the traditional clustering algorithms. Moreover, the under-segment error rate is decreased while the segmentation accuracy is increased by 10% to 85.02%.

Newport宣传-MKS新实验室计划
补充资料

DOI:10.3788/CJL201946.0710002

所属栏目:遥感与传感器

基金项目:重庆市技术创新与应用示范项目、重庆邮电大学人才引进项目、教育部-中国移动科研基金;

收稿日期:2019-01-11

修改稿日期:2019-03-11

网络出版日期:2019-07-01

作者单位    点击查看

范小辉:重庆邮电大学通信与信息工程学院, 重庆 400065重庆邮电大学电子信息与网络工程研究院, 重庆 400065
许国良:重庆邮电大学电子信息与网络工程研究院, 重庆 400065
李万林:重庆邮电大学电子信息与网络工程研究院, 重庆 400065
王茜竹:重庆邮电大学电子信息与网络工程研究院, 重庆 400065
常亮亮:重庆邮电大学通信与信息工程学院, 重庆 400065重庆邮电大学电子信息与网络工程研究院, 重庆 400065

联系人作者:许国良(xugl@cqupt.edu.cn)

备注:重庆市技术创新与应用示范项目、重庆邮电大学人才引进项目、教育部-中国移动科研基金;

【1】Huang L, Chen S Y, Zhang J F et al. Real-time motion tracking for indoor moving sphere objects with a LiDAR sensor. Sensors. 17(9), (2017).

【2】Douillard B, Underwood J, Vlaskine V et al. A pipeline for the segmentation and classification of 3D point clouds. ∥Khatib O, Kumar V, Sukhatme G. Experimental robotics. Berlin, Heidelberg: Springer. 585-600(2014).

【3】Lu X H, Yao J, Tu J G, Remote Sensing, Spatial Information Sciences et al. Pairwise linkage for point cloud segmentation[J]. III-. 3, 201-208(2016).

【4】Klasing K, Wollherr D and Buss M. A clustering method for efficient segmentation of 3D laser data. [C]∥2008 IEEE International Conference on Robotics and Automation, May 19-23, 2008, Pasadena, CA, USA. New York: IEEE. 4043-4048(2008).

【5】Luo Z, Habibi S and Mohrenschildt M. LiDAR based real time multiple vehicle detection and tracking. World Academy of Science, Engineering and Technology, International Journal of Computer, Electrical, Automation, Control and Information Engineering. 10(6), 1125-1132(2016).

【6】Huang G, Wu C Z, Lü N C et al. A study of laser radar object detection based on improved DBSCAN algorithm. Journal of Transport Information and Safety. 33(3), 23-28(2015).
黄钢, 吴超仲, 吕能超 等. 基于改进DBSCAN算法的激光雷达目标物检测方法. 交通信息与安全. 33(3), 23-28(2015).

【7】Asvadi A, Premebida C, Peixoto P et al. 3D Lidar-based static and moving obstacle detection in driving environments: an approach based on voxels and multi-region ground planes. Robotics and Autonomous Systems. 83, 299-311(2016).

【8】Yang F, Zhu Z, Gong X J et al. Real-time dynamic obstacle detection and tracking using 3D Lidar. Journal of Zhejiang University(Engineering Science). 46(9), 1565-1571(2012).
杨飞, 朱株, 龚小谨 等. 基于三维激光雷达的动态障碍实时检测与跟踪. 浙江大学学报(工学版). 46(9), 1565-1571(2012).

【9】Ye G. Multi-target detection and tracking algorithm for autonomous driving car based on a 3D lidar in urban traffic environment. Beijing: Beijing Institute of Technology. (2016).
叶刚. 城市环境基于三维激光雷达的自动驾驶车辆多目标检测及跟踪算法研究. 北京: 北京理工大学. (2016).

【10】B?rcs A, Nagy B and Benedek C. Fast 3-D urban object detection on streaming point clouds. ∥Agapito L, Bronstein M, Rother C. Lecture notes in computer science. Cham: Springer. 8926, 628-639(2015).

【11】Moosmann F, Pink O and Stiller C. Segmentation of 3D lidar data in non-flat urban environments using a local convexity criterion. [C]∥2009 IEEE Intelligent Vehicles Symposium, June 3-5, 2009, Xi''an, China. New York: IEEE. 215-220(2009).

【12】Hui Z Y, Cheng P G, Guan Y L et al. Review on airborne LiDAR point cloud filtering. Laser & Optoelectronics Progress. 55(6), (2018).
惠振阳, 程朋根, 官云兰 等. 机载LiDAR点云滤波综述. 激光与光电子学进展. 55(6), (2018).

【13】Chu P, Cho S, Sim S et al. A fast ground segmentation method for 3D point cloud. Journal of Information Processing Systems. 13(3), 491-499(2017).

【14】Huang Z W, Liu F and Hu G W. Improved method for LiDAR point cloud data filtering based on hierarchical pseudo-grid. Acta Optica Sinica. 37(8), (2017).
黄作维, 刘峰, 胡光伟. 基于多尺度虚拟格网的LiDAR点云数据滤波改进方法. 光学学报. 37(8), (2017).

【15】Wang X, Wang J Q, Li K Q et al. Fast segmentation of 3-D point clouds for intelligent vehicles. Journal of Tsinghua University (Science and Technology). 54(11), 1440-1446(2014).
王肖, 王建强, 李克强 等. 智能车辆3-D点云快速分割方法. 清华大学学报(自然科学版). 54(11), 1440-1446(2014).

【16】Bogoslavskyi I and Stachniss C. Efficient online segmentation for sparse 3D laser scans. PFG-Journal of Photogrammetry, Remote Sensing and Geoinformation Science. 85(1), 41-52(2017).

【17】Yuan X and Zhao C X. A laser point cloud clustering algorithm for robot navigation. Robot. 33(1), 90-96(2011).
袁夏, 赵春霞. 一种应用于机器人导航的激光点云聚类算法. 机器人. 33(1), 90-96(2011).

【18】Moosmann F. Interlacing self-localization, moving object tracking and mapping for 3D range sensors. (2013).

引用该论文

Xiaohui Fan, Guoliang Xu, Wanlin Li, Qianzhu Wang, Liangliang Chang. Target Segmentation Method for Three-Dimensional LiDAR Point Cloud Based on Depth Image[J]. Chinese Journal of Lasers, 2019, 46(7): 0710002

范小辉, 许国良, 李万林, 王茜竹, 常亮亮. 基于深度图的三维激光雷达点云目标分割方法[J]. 中国激光, 2019, 46(7): 0710002

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF