测绘通报 ›› 2020, Vol. 0 ›› Issue (5): 55-58.doi: 10.13474/j.cnki.11-2246.2020.0145

• 学术研究 • 上一篇    下一篇

矿区地表彩色点云的自动分类

蔡来良1, 宋德云1, 魏峰远1, 薛渊2, 舒前进3   

  1. 1. 河南理工大学, 河南 焦作 454150;
    2. 山西晋煤集团坪上煤业有限公司, 山西 晋城 048000;
    3. 中国矿业大学, 江苏 徐州 221116
  • 收稿日期:2019-10-31 修回日期:2020-01-03 出版日期:2020-05-25 发布日期:2020-06-02
  • 作者简介:蔡来良(1983-),男,博士,讲师,主要研究方向为开采沉陷、点云处理。E-mail:cll@hpu.edu.cn
  • 基金资助:
    国家自然科学基金(41701597;U1810203);中国博士后科学基金(2018M642746)

Color point cloud classification of mine surface

CAI Lailiang1, SONG Deyun1, WEI Fengyuan1, XUE Yuan2, SHU Qianjin3   

  1. 1. School of Surveying and Land Information Engineering, Jiaozuo 454150, China;
    2. Shanxi Jinmei Group Pingshang Coal Industry Co., Ltd., Jincheng 048000, China;
    3. College of Mechanics and Civil Engineering, Xuzhou 221116, China
  • Received:2019-10-31 Revised:2020-01-03 Online:2020-05-25 Published:2020-06-02

摘要: 以矿区的彩色三维激光点云数据为研究对象,提出了矿区点云快速自动分类及目标提取的方法。首先根据彩色点云的RGB值计算HSV空间中的H值,根据各地物间H值的差异,分别对地面点与非地面点根据地物颜色先验值进行点的提取。然后对提取的点进行聚类计算,利用各类地物点云在空间分布上的显著差异,采用分层截面投影,由投影点最小包围盒的长宽比及面积比对矿区地物点云进行自动分类与提取。最后以Riegl VZ-1000扫描仪采集的某矿区地表点云数据为试验对象,验证本文算法的可行性和实用性。

关键词: 地面激光扫描技术, RGB, HSV, 点云分类, 分层截面

Abstract: Taking the color 3D laser point cloud data of the mining area as the research object, a fast automatic classification and target extraction method for the point cloud of the mining area is proposed. Firstly, the H value in HSV space is calculated according to the RGB value of color point cloud. According to the difference of H value among different objects, the points of ground point and non ground point are extracted according to the prior value of ground object color. Then, the extracted points are clustered and calculated. Using the significant difference in spatial distribution of all kinds of ground object point clouds, layered cross-section projection is adopted, and the ratio of length to width of the minimum bounding box of projection points is used In this paper. Finally the RIEGL vz-1000 scanner is used as the experimental object to verify the feasibility and practicability of this algorithm.

Key words: ground laser scanning technology, RGB, HSV, point cloud classification, stratified section

中图分类号: