Method for extracting single tree information from LiDAR point cloud in layered clustering mode

Abstract

The invention discloses a method for extracting single tree information from LiDAR in a layered clustering mode and belongs to the scope of LiDAR point cloud data processing and information extraction. According to the key technical point, 1, the point cloud is horizontally layered; 2, all the layers of point cloud are classified in a k mean value clustering mode, and different attribute values are provided for all the layers of point cloud; 3, the point cloud meeting preset conditions is fused; 4, the single tree positions, the tree height, the crown breadth and other information are extracted from the point cloud with the same attribute values, and forest stand structure information is calculated through the single tree information. The method has the key advantages that extraction of floor stratum single trees is achieved, and the number of accurately extracted forest stand forests reaches 80% or above; the function of extracting the single tree information from LiDAR and calculating the forest stand structure is achieved; no excessive requirement on the point cloud density is generated, and it is proved that the point cloud of 2p / m<2> can complete relevant work. The method can be applied to the field of LiDAR inversion information, is especially suitable for the forest stand condition with the complex structure, and can accurately extract the single tree positions, the tree height, the crown breadth, the forest stand structure parameters and other relevant information.

Description

[0001] 1. Technical field [0002] The invention relates to a method for extracting individual tree information from a laser radar (LiDAR) point cloud, in particular to a method for extracting understory trees in a complex stand structure, suitable for data processing and information extraction of airborne laser radar point clouds, The invention belongs to the technical field of lidar point cloud data processing. [0003] 2. Technical Background [0004] LiDAR (Light Detection And Ranging, LiDAR) is an active remote sensing technology that measures the distance between the sensor and the target through the laser emitted by the sensor. Due to the high penetrability of LiDAR, applying LiDAR technology to forestry remote sensing can obtain a large amount of point cloud data containing forest structure information; by segmenting the whole point cloud into individual trees, it is possible to further obtain individual tree-scale forest stands Information will be replaced by fast and...

AI Technical Summary

Problems solved by technology

[0007] (1) Under the conditions of high forest stand density and complex forest layer structure, it is difficult to extract the lower forest trees, and the forest trees that can be extracted are roughly about 20-60% of the actual number; how to extract more forest trees has become one of the challenges;

[0008] (2) Part of the individual tree information is missing, resulting in an unsound inversion function of the stand structure

Images