测绘通报 ›› 2019, Vol. 0 ›› Issue (4): 26-31.doi: 10.13474/j.cnki.11-2246.2019.0107

• 学术研究 • 上一篇    下一篇

利用三维激光扫描数据进行建筑物立面点云分割算法分析

贺亦峰1, 胡荣2, 邹进贵1   

  1. 1. 武汉大学测绘学院, 湖北 武汉 430079;
    2. 文华学院, 湖北 武汉 430074
  • 收稿日期:2018-06-17 修回日期:2018-12-27 出版日期:2019-04-25 发布日期:2019-05-07
  • 通讯作者: 胡荣。E-mail:554731250@qq.com E-mail:554731250@qq.com
  • 作者简介:贺亦峰(1995-),男,硕士生,主要研究方向为三维激光点云数据处理。E-mail:heyifeng@whu.edu.cn
  • 基金资助:

    国家自然科学基金(41674005)

Analysis on point cloud of building facade segmentation algorithms based on 3D laser scanning data

HE Yifeng1, HU Rong2, ZOU Jingui1   

  1. 1. School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China;
    2. Wenhua College, Wuhan 430074, China
  • Received:2018-06-17 Revised:2018-12-27 Online:2019-04-25 Published:2019-05-07

摘要:

三维激光扫描技术在当今社会的应用越来越广泛,但由于点云数据量大,处理效率低下,如何快速高效地将大量点云数据进行重建与识别成为解决问题的关键。点云分割技术能够将立面点云中的特征信息与背景点云分离开来,为地物特征信息的提取和识别工作提供了重要的技术支持。本文通过编程实现了多种点云分割算法,对建筑物立面进行分割处理,详细分析了不同算法的分割精度及适用范围。

关键词: 点云分割, 欧氏聚类, 区域增长, RANSAC算法, 分割精度分析

Abstract:

The application of 3D laser scanning technology is becoming more and more widespread in current society. However, due to the large amount of point cloud data, how to reconstruct and identify a large number of point cloud data quickly and efficiently is the key to solve the problem. The point cloud segmentation technology can separate the feature information in the facade point cloud from the background point cloud, providing important technical support for the extraction and identification of feature information. In this paper, a variety of point cloud segmentation algorithms are programmed to segment the facades of buildings, and the segmentation accuracy and applicable range of different algorithms are analyzed in detail.

Key words: point cloud segmentation, Euclidean cluster, region growing, RANSAC algorithm, segmentation accuracy analysis

中图分类号: