建筑立面点云对偶深度图像中窗户提取方法
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金项目(No.41361077;No.41561085);江西省自然科学基金项目(No.20161BAB203091)资助


Window extraction method in dual depth image of point cloud of building façade
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    提出了一种地面激光点云数据的建筑物立面窗户提取方法。针对窗户包含点云的情况,首先对建筑立面点云进行特征分析及无效点去除,由于窗户、展台、及其他凹凸特征物与建筑立面的墙面存在深度差,可基于此采用距离加权倒数IDW内插方法生成建筑面片的对偶深度图像,再经阈值分割、中值滤波平滑处理,形态学滤波等,找到窗户内的点云及边界。而对窗户空洞情况,可采用构建TIN进行边界提取,由于窗户边界点构成的边长远远大于窗户周边点,可基于此找到窗户边界点。通过对实测点云数据进行实验,将提取的效果与实际情况作对比,结果表明,该方法能有效对窗户进行提取。

    Abstract:

    A window extraction method of building facades from ground laser point cloud data is proposed.In view of the situation that the windows contain point clouds,the feature analysis and invalid point removal of point clouds on building facades are carried out first.Because the depth difference between windows,exhibition stands and other concave and convex features and the wall surface of building facades exists,dual depth images of building facades can be generated by distance weighted reciprocal IDW interpolation method based on these features,and then processed by threshold segmentation,median filtering smoothing and morphological filtering,etc.,find the point cloud and boundary in the window.For window voids,the TIN can be used for boundary extraction.Since the edge length of the window boundary points is much larger than the peripheral points of the window,the window boundary points can be found based on this.By experimenting with the measured point cloud data,the extracted effect is compared with the actual situation.The results show that the method can effectively extract the window.

    参考文献
    相似文献
    引证文献
引用本文

邹纪伟,刘德儿,杨鹏,刘靖钰,李瑞雪,冀炜臻.建筑立面点云对偶深度图像中窗户提取方法[J].激光与红外,2020,50(7):875~881
ZOU Ji-wei, LIU De-er, YANG Peng, LIU Jing-yu, LI Rui-xue, JI Wei-zhen. Window extraction method in dual depth image of point cloud of building façade[J]. LASER & INFRARED,2020,50(7):875~881

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2020-08-05
  • 出版日期: