Bulletin of Surveying and Mapping ›› 2023, Vol. 0 ›› Issue (10): 145-149.doi: 10.13474/j.cnki.11-2246.2023.0310

Previous Articles     Next Articles

Application of line segment detector algorithm in extracting building contour lines from airborne LiDAR

LIU Jianxing1,2, HU Quanyi2, LE Meiyan1,2, SHU Sihong1,2   

  1. 1. Jiangxi Nuclear Industry Surveying and Mapping Institute Group Co., Ltd., Nanchang 330038, China;
    2. Jiangxi Nuclear Industry Geological Survey, Nanchang 330038, China
  • Received:2023-04-12 Published:2023-10-28

Abstract: To address the problem of time-consuming and low accuracy of extracting building contour lines from airborne LiDAR point cloud data, this paper proposes an (line segment detector)LSD-based method for extracting building contour lines from airborne LiDAR. The method first rasterizes the classified building point clouds to obtain the binary image.Then, the binary image is expanded and eroded to eliminate the voids generated by the rasterization. Finally, the LSD algorithm is used for straight line detection to obtain the regular building contour lines. Using the measured data for verification, the proposed method can detect the building contour lines with sub-pixel accuracy, which is about 50 times more efficient than the traditional Canny algorithm.

Key words: binary image, LSD algorithm, building contour lines, airborne LiDAR, point cloud

CLC Number: