测绘通报 ›› 2019, Vol. 0 ›› Issue (1): 34-38.doi: 10.13474/j.cnki.11-2246.2019.0007

• 学术研究 • 上一篇    下一篇

机载激光点云中高压电塔自动识别方法

刘洋, 杨必胜, 梁福逊   

  1. 武汉大学测绘遥感信息工程国家重点实验室, 湖北 武汉 430079
  • 收稿日期:2018-03-14 修回日期:2018-04-20 出版日期:2019-01-25 发布日期:2019-02-14
  • 作者简介:刘洋(1993-),男,硕士生,主要从事三维激光点云数据处理工作。E-mail:lyang@whu.edu.cn
  • 基金资助:
    国家自然科学基金重点项目(41531177);国家杰出青年科学基金(41725005)

An automatically recognizing method for transmission towers from ALS point cloud

LIU Yang, YANG Bisheng, LIANG Fuxun   

  1. State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China
  • Received:2018-03-14 Revised:2018-04-20 Online:2019-01-25 Published:2019-02-14

摘要: 提出了一种基于格网特征的机载激光点云高压电塔自动识别方法。首先对机载激光点云数据进行滤波去噪处理;然后对点云数据进行规则格网化特征分析获得高压电塔粗识别区域;最后对粗识别区域进行外接邻域网格线性特征悬空点集检测以确定电塔识别结果,并以分层切片法分析获取电塔平面中心坐标。采用大型无人机实际线路巡检获取的机载点云数据对本文算法进行验证,试验结果表明本算法可实现高压电塔的快速自动识别,对无人机电力巡检智能诊断具有一定的促进作用。

关键词: 输电线路, 机载点云, 格网特征, 高压电塔, 自动识别

Abstract: In this paper,a method is proposed to recognize the transmission high voltage towers automatically in ALS point cloud.Firstly,we denoise the ALS data with statistical method.Secondly,we calculate and analyze 2D grid features to extract the rough area of transmission towers.Finally,we recognize the transmission towers from rough tower areas by confirming the bounding grids with linear feature,and calculate the 2D tower center by hierarchical slicing analysis.The proposed method is validated with the transmission line ALS data obtained by the actual inspection of the large-scale unmanned helicopter.Experiment shows that the proposed method recognizes the transmission towers automatically and effectively,which means the proposed method can be applied to processing actual transmission inspection ALS data.

Key words: transmission line, ALS point cloud, grid feature, high voltage tower, automatical recognization

中图分类号: