Paper
22 December 2021 A novel background filtering method for LiDAR based on point cloud ground line segmentation
Xiuguang Song, Zhaoyou Ma, Chunyu Zhou, Fei Cheng, Xucai Zhuang, Jianqing Wu
Author Affiliations +
Proceedings Volume 12058, Fifth International Conference on Traffic Engineering and Transportation System (ICTETS 2021); 120582J (2021) https://doi.org/10.1117/12.2619737
Event: 5th International Conference on Traffic Engineering and Transportation System (ICTETS 2021), 2021, Chongqing, China
Abstract
Accurate road information is important for modern intelligent transportation systems. Light Detection and Ranging (LiDAR) is an ideal equipment to collect microscope traffic data which is helpful to analyze traffic behavior and operation mechanism. Background filtering is the key to obtaining road target information. To reduce the calculation, the Region of Interest (ROI) area is defined and only ground lines and targets are kept in the ROI area. The traditional random sample consensus (RANSAC) algorithm filters out the ground lines by establishing a plane model. The performance of the RANSAC is limited. To extract the target point clouds more accurately, this paper proposed a method of ground segmentation based on polynomial fitting of third order surface and multiple error control. Using this method, ground lines recognition accuracy can reach 97.1%, which was about 24.74% higher than that of traditional algorithm.
© (2021) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xiuguang Song, Zhaoyou Ma, Chunyu Zhou, Fei Cheng, Xucai Zhuang, and Jianqing Wu "A novel background filtering method for LiDAR based on point cloud ground line segmentation", Proc. SPIE 12058, Fifth International Conference on Traffic Engineering and Transportation System (ICTETS 2021), 120582J (22 December 2021); https://doi.org/10.1117/12.2619737
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Clouds

LIDAR

Roads

Detection and tracking algorithms

Error analysis

Magnetic sensors

Cameras

Back to Top