A method of reconstructing building outlines from lidar original point cloud

Abstract

The present invention provides a kind of method of rebuilding LiDAR original point cloud building contour line, comprises on the basis of LiDAR discrete point cloud, constructs triangular grid according to the principle of Delaunay triangular net; Extract two points with relatively large elevation values ​​in the triangle and store them in the form of point pairs; then use the point pairs as "data chain nodes", and form a ring data chain by connecting the "data chain nodes" to complete each protrusion Clustering of object edge points; finally count the number of points in each protruding object point set, and filter out non-building edge point sets; add each point to the fitting data one by one according to the connection order and refit, and use the newly added point The distance from the fitted straight line to the previous step is used as a judgment standard, and finally the intersection points of adjacent fitted sides are connected to obtain the outline of the building; the method of the present invention has high speed, high efficiency and accuracy, and solves the discreteness and Non-characteristic and regular problems of contour line reconstruction.

Description

technical field [0001] The invention relates to a method for extracting the outline of the outer boundary of a building, in particular to a method for reconstructing the outline of the building's roof in the discrete point cloud data obtained by the airborne LiDAR. Background technique [0002] Building outline information has been widely used in many military and civilian related fields, especially in the establishment of urban digital models and planning management, the design of traffic lines, the survey of power lines, the detection and analysis of terrain, forest supervision, etc. As well as the field of precision target strike and damage effect assessment in modern combat plays an important role. [0003] LiDAR is a laser radar detection system that uses the Global Positioning System (GPS, Global Position System) and the Inertial Measurement Unit (IMU, Inertial Measurement Unit) to perform laser scanning to obtain three-dimensional information of the target. The data ...

AI Technical Summary

Problems solved by technology

The three-dimensional information of the target space acquired by LiDAR technology is discrete point data, and there is no close connection between points. The relationship between the adjacent points that should be the most relevant cannot be determined in the data.

[0007] 2. Non-characteristic problems of target data

For the original discrete point data, it is difficult to use existing methods to quickly extract features to obtain the target area

[0008] 3. Dealing with the regularity of the results

Images