摘要
城市道路覆盖面广,容易忽略道路轮廓的分散性,LiDAR点云检测可以对其轮廓进行分割,很大程度上解决了提取效率低的问题。提出基于LiDAR点云检测的城市道路轮廓提取技术。采用空间区域网格扫描技术进行城市道路光学图像采集,对采集的城市道路光学图像进行区域性融合滤波处理,提取城市道路光学图像的灰度像素特征量,采用模糊信息融合方法进行城市道路光学图像信息融合,构建城市道路光学图像的三维点云特征大数据分布模型,结合LiDAR点云分布式检测的方法进行城市道路的轮廓特征分割,采用帧点最大灰度级特征集检测和多参数信息融合方法,实现城市道路轮廓提取和点云检测。仿真表明,在实验次数为100次时,所提方法提取时间为12 s;样本数为2000个时,所提方法提取精度为1。充分证明采用该方法进行城市道路轮廓提取的时间较短、提取准确度较高,空间规划识别能力较强,具有很好的参数优化能力。
The coverage of urban roads is wide,and it is easy to ignore the dispersion of road contours.LiDAR point cloud detection can segment their contours,which largely solves the problem of low extraction efficiency.This paper proposes a technology for extracting urban road contours based on LiDAR point cloud detection.The spatial area grid scanning technology is adopted for urban road optical image acquisition,the regional fusion filtering is performed on the collected urban road optical image,the gray pixel feature values of urban road optical image are extracted,and the fuzzy information fusion method is used for urban road optical image information fusion,a 3D point cloud feature big data distribution model of urban road optical images is constructed,the LiDAR point cloud distributed detection method is combined for contour feature segmentation of urban roads,the frame point maximum gray level feature set detection and multi-parameter information fusion methods are applied to realize the contour extraction and point cloud detection of urban roads.The simulation shows that when the number of experiments is 100,the extraction time of the proposed method is 12 s.When the number of samples is 2000,the extraction accuracy of the proposed method is 1.It is fully proved that this method can be used to extract urban road contours in a short time with high extraction accuracy,strong spatial planning recognition ability,and good parameter optimization ability.
作者
彭译萱
陈晨
ZHU Junle
PENG Yixuan;CHEN Chen;ZHU Junle(The Guangdong University of Technology,Guangzhou 510090,China;Sun Yat-sen University,Guangzhou 510090,China;Shenzhen THS Hi-tech Co.,Ltd.,Shenzhen Guangdong 518000,China)
出处
《激光杂志》
北大核心
2020年第10期135-138,共4页
Laser Journal
基金
国家自然科学青年基金项目(No.51708126)
教育部人文社会科学研究青年基金项目(No.17YJC760002)
广东工业大学校内青年项目(No.17ZS0039)。
