Terrain Feature Extraction Based On Airborne LiDAR Point Cloud Data

The current process of globalization is accelerating,and infrastructure has gained a new definition.Spatial and temporal information has become an important infrastructure resource.The surveying and mapping industry has undergone tremendous changes,from small surveying and mapping for drawing topographic maps to intelligent surveying and mapping for serving geospatial information.Airborne Li DAR measurement technology continues to develop and is widely used in the surveying and mapping industry,greatly improving the efficiency of obtaining geospatial information and further improving the accuracy of spatial position coordinates.The point cloud filtering algorithm is a necessary work done to obtain real ground points from the original point cloud data.The improvement of the accuracy of ground point filtering algorithms is beneficial for obtaining accurate ground points,providing favorable guarantees for post processing and analyzing terrain information of ground point clouds.Ridge(valley)lines are characteristic lines of mountains.Extracting ridge(valley)lines based on point cloud data is beneficial for the development of geomorphological research and analysis,further simplifying the storage space of point cloud data,and promoting the further development of DEM data.The main work and contributions of this article are as follows:(1)In response to the problem of low accuracy in extracting ground points in densely vegetated areas,this paper proposes a method for extracting ground points from dense vegetation point clouds based on the RANSAC algorithm.Firstly,a virtual grid is introduced to segment point cloud data and calculate the grid index for each point;Then the slope values of all points in each grid are obtained,and the local slope statistics are carried out.The slope threshold is calculated based on the combination of K-means clustering and normal distribution for preliminary slope filtering;Then extract real ground points based on random sampling consistency.(2)In the process of extracting ridge(valley)points,this paper proposes a method of using cubic polynomials for local area surface fitting.The point generated by the intersection of the normal section in the main direction of the ridge(valley)line and the cubic polynomial local area surface fitting to obtain the curvature extreme value of the region’s vertical section is used as the ridge(valley)feature point of the local area.(3)In the process of generating ridge(valley)lines,this paper proposes a gradually growing method for generating ridge(valley)lines.Firstly,construct a Kd-tree to screen accurate ridge(valley)feature points;Then generate the main and branch feature lines of the ridge(valley)line in batches;In order to improve the efficiency of generating ridge(valley)lines,the starting point is connected to the growth point during the generation process,and whether it is a feature point is determined based on the angle between the main direction of the starting point and the starting edge,and the curvature threshold is used to determine whether it is a feature point;In the refinement processing of ridges(valleys),for cases where the gap between feature points is too large,segmentation between two points is performed and horizontal projection is performed.Perform surface fitting in each segment,extract supplementary feature points,and sketch the basic contour of the mountain.