Single tree modeling method and device and storage medium

Abstract

The invention discloses a single tree modeling method and device and a storage medium. The method comprises the following steps: carrying out individual tree clipping on forest point cloud data of a target forest region, and extracting part of individual tree point cloud data as a sample data set; training the sample data set by utilizing a pre-designed 2.5 D Manhattan distance model to obtain a target forest region model; inputting all the single tree point cloud data into the target forest region model to obtain a segmentation threshold of a target forest region; and extracting target tree data sets one by one according to the segmentation threshold. By utilizing the single tree model constructed by the method, the reference form of the tree species can be obtained by inputting all realsingle tree sample point clouds into the single tree model, the parameter threshold required by the algorithm is extracted from the form, and the de-empirical thresholding is achieved.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a single tree modeling method, device and storage medium. Background technique [0002] In the past ten years, predecessors have proposed many methods of measuring the number of individual plants through laser-like big data. These data directly indicate the possibility of realizing single tree segmentation from laser radar data, and also highlight the possibility of realizing single tree segmentation. Segmentation is challenging. The commonality of the early single tree segmentation algorithms is that they all used the canopy height model (CHM), such as the local maximum focus filter algorithm proposed by Popescu and the watershed cutting algorithm applied by Mei and Koch. However, the generation of CHM requires: 1. Separate the ground points and vegetation points first; 2. Then generate a digital elevation model (digital elevation model, DEM) and a digital surface model (digital ...

AI Technical Summary

Problems solved by technology

The fixed threshold will cause the accuracy of the algorithm proposed by Li et al. to be unable to maintain a high level when performing single-tree segmentation in other forest areas

In the algorithm proposed by Ferraz et al., they try to extract a small number of reliable single trees as training samples for statistics through CHM, and obtain a broadband threshold suitable for local application. However, there are errors in the single tree model obtained through CHM, which will increase the fitting finesse and difficulty

Images