基于单线激光雷达的道路区域分割  

Road Area Segmentation Based on Single Line Lidar

作　　者：李宏 赵礼刚[1] 张浩傑 LI Hong;ZHAO Ligang;ZHANG Haojie(College of Mechanical Engineering,Jiangsu University of Science and Technology,Zhenjiang Jiangsu 212100,China)

机构地区：[1]江苏科技大学机械工程学院,江苏镇江212100

出　　处：《机床与液压》2024年第21期57-63,共7页Machine Tool & Hydraulics

摘　　要：针对单线激光雷达在进行道路区域分割时出现的识别不准确以及精度问题,研究一种基于最大熵原理提取自适应阈值,并通过滑动窗口法实现道路区域分割的方法。利用中值滤波对原始点云数据进行预处理,以减少噪声和异常值产生的影响;结合最大熵原理提取的自适应阈值和滑动窗口算法完成对道路区域点云的分割;最后基于概率论的方法确定道路边界点云。通过对3种不同类型的结构化道路进行实验,得到路面、障碍物以及路沿的识别准确率依次为96.88%、86.82%、95.50%,证明了该方法的有效性和适用性。Aiming at the problem of recognition inaccuracy and accuracy of single-line lidar in road area segmentation,a method of extracting adaptive threshold based on maximum entropy principle and realizing road area segmentation by sliding window method was studied.The median filter was used to preprocess the original point cloud data to reduce the impact of noise and outliers.Then,the adaptive threshold extracted from the maximum entropy principle and the sliding window algorithm were used to segment the point cloud in the road area.Finally,the road boundary point cloud was determined based on probability theory.Experiments on three different types of structured roads show that the recognition accuracy of road surface,obstacle and curb is 96.88%,86.82%and 95.50%,which proves the effectiveness and applicability of this method.

关 键 词：单线激光雷达 最大熵原理 移动窗口 道路区域分割 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]