基于随机抽样一致性算法的车辆轮胎点云提取方法  

Vehicle Tire Point Cloud Extraction Method Based on Random Sampling Consistency Algorithm

作　　者：赵诚 陈嘉平 李春晓 陈迎新 贾克斌[2] ZHAO Cheng;CHEN Jia-ping;LI Chun-xiao;CHEN Ying-xin;JIA Ke-bin(Vehicle Testing Engineering Research Institute of China,Beijing 102100 China;Beijing University of Technology,Beijing 100124 China)

机构地区：[1]中机科(北京)车辆检测工程研究院有限公司,北京102100 [2]北京工业大学,北京100124

出　　处：《自动化技术与应用》2023年第1期14-16,25,共4页Techniques of Automation and Applications

摘　　要：针对激光雷达获取的车辆底盘轮廓点云中轮胎的特征提取与分割问题,提出了一种基于随机抽样一致性算法的车辆轮胎点云提取方法。为了提高轮胎提取的准确性,首先采用随机抽样一致性算法对目标车辆点云进行平面提取,然后对提取的平面点云进行K-means聚类,剔除离群点,分割出实际的连续平面点云;最后通过采用随机抽样一致性算法对剩余点云进行轮胎提取。为了验证提取方法的有效性,通过计算机仿真的方法,生成车辆底盘轮廓点云,对该仿真数据进行轮胎特征的提取与分割。结果表明,本文提出的方法具有良好的分割提取效果。Aiming at the problem of feature extraction and segmentation of tires in vehicle contour, a method of vehicle tire point cloud extraction based on random sampling consensus algorithm is proposed. In order to improve the accuracy of tire extraction, firstly,the plane point cloud of the target vehicle is extracted by using the random sampling consensus algorithm, and then the extracted plane point cloud is clustered by K-means to remove outliers and segment the actual continuous plane point cloud. Finally, the tire is extracted from the remaining point cloud by using the random sampling consensus algorithm. In order to verify the effectiveness of the extraction method, the vehicle contour point cloud is generated by computer simulation, and noise is added to the point cloud. The simulation results show that the proposed method has good tire-extraction effect.

关 键 词：激光雷达 车辆轮胎特征提取 随机抽样一致性 K-MEANS聚类 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]