An Individual Tree Segmentation Method Based on Watershed Algorithm and Three-Dimensional Spatial Distribution Analysis From Airborne LiDAR Point Clouds

Abstract

Accurate individual tree segmentation is an important basis for the subsequent calculation and analysis of forestry parameters. However, rasterized canopy height model based methods often suffer from 3-D information loss due to the interpolation operation. Therefore, this article proposes an individual tree segmentation method based on the marker-controlled watershed algorithm and 3-D spatial distribution analysis from airborne Li-DAR point clouds. First, based on the potential tree apices derived from the local maxima filtering, the marker-controlled watershed segmentation algorithm is conducted to obtain the coarse point clusters. Then, within the principal component analysis defined local coordinate reference framework, a multidirectional 3-D spatial profile analysis is performed on each point cluster to refine the potential tree apex positions. Finally, the refined potential tree apex positions are used as a prior of K-means clustering to achieve the coarse-to-fine individual tree segmentation. Comparative experiments were conducted on the public NEWFOR dataset to evaluate the proposed method. Results indicate that the proposed method is efficient and robust for segmenting individual trees.

Lidar Boosts 3D Ecological Observations and Modelings: A Review and Perspective

“…Individual plant segmentation, i.e., delineating individual plants from the lidar point cloud, has therefore become the prerequisite first step, and the procedures can be generally divided into two categories: canopy height model (CHM)-based methods and point cloud-based methods. CHM-based methods rely on gaps among plants in lidarderived CHMs and use image segmentation techniques (e.g., watershed segmentation) to delineate the boundary of individual trees [88]- [91]. Point cloud-based methods segment points of each individual plant directly from lidar point clouds [92]- [95].…”

A Crown Morphology-Based Approach to Individual Tree Detection in Subtropical Mixed Broadleaf Urban Forests Using UAV LiDAR Data

“…The smoother parameters include sliding window size and standard deviation of Gaussian smoother and sliding window size of average smoother. The values of smoother parameters were selected following other related studies [3,30] and are listed in Table 1.…”

“…However, extensive field work needs to be performed to collect data for obtaining the a priori information and inaccurate a priori information may affect the tuning of the sliding window size. Smoothing techniques such as Gaussian filtering are usually employed to correct surface irregularities [29,30]. Nevertheless, the optimal smoothing parameter values are also difficult to determine and prior knowledge is required [31].…”

Hierarchical Instance Recognition of Individual Roadside Trees in Environmentally Complex Urban Areas from UAV Laser Scanning Point Clouds

“…To fully take advantages of the well-developed image-processing algorithms in tree crowns delineation, the height variation of canopy height model (CHM) is employed to find treetops. According to the statement of [6], the CHM-based tree detection algorithms (such as watershed analysis [7], spatial wavelet analysis [8], and template matching [9]) are fast and efficient but the existing problems are also obvious: it changes the shape of the tree crown to a certain extent, reducing the original data information thanks to a variety of canopy sizes. Some studies directly segmented individual trees on original point clouds since the increased spatial resolution of ALS (airborne laser scanning) point cloud.…”