Abstract:Identifying individual trees and delineating their canopy structures from the forest point cloud data acquired by an airborne LiDAR (Light Detection And Ranging) has significant implications in forestry inventory. Once accurately identified, tree structural attributes such as tree height, crown diameter, canopy based height and diameter at breast height can be derived. This paper focuses on a novel computationally efficient method to adaptively calibrate the kernel bandwidth of a computational scheme based on mean shift-a non-parametric probability density-based clustering technique-to segment the 3D (three-dimensional) forest point clouds and identify individual tree crowns. The basic concept of this method is to partition the 3D space over each test plot into small vertical units (irregular columns containing 3D spatial features from one or more trees) first, by using a fixed bandwidth mean shift procedure and a small square grouping technique, and then rough estimation of crown sizes for distinct trees within a unit, based on an original 2D (two-dimensional) incremental grid projection technique, is applied to provide a basis for dynamical calibration of the kernel bandwidth for an adaptive mean shift procedure performed in each partition. The adaptive mean shift-based scheme, which incorporates our proposed bandwidth calibration method, is validated on 10 test plots of a dense, multi-layered evergreen broad-leaved forest located in South China. Experimental results reveal that this approach can work effectively and when compared to the conventional point-based approaches (e.g., region growing, k-means clustering, fixed bandwidth or multi-scale mean shift), its accuracies are relatively high: it detects 86 percent of the trees ("recall") and 92 percent of the identified trees are correct ("precision"), showing good potential for use in the area of forest inventory.

Airborne LiDAR (Light Detection And Ranging) remote sensing for individual tree-level forest inventory necessitates proper extraction of individual trees and accurate measurement of tree structural parameters. Due to the inadequate tree finding capability offered by LiDAR technology and the complex patterns of forest canopies, significant omission and commission errors occur frequently in the segmentation results. Aimed at error reduction and accuracy refinement, this paper presents a novel adaptive mean shift-based clustering scheme aided by a tree trunk detection technique to segment individual trees and estimate tree structural parameters based solely on the airborne LiDAR data. Tree trunks are detected by analyzing points' vertical histogram to detach all potential crown points and then clustering the separated trunk points according to their horizontal mutual distances. The detected trunk information is used to adaptively calibrate the kernel bandwidth of the mean shift procedure in the fine segmentation stage by applying an original 2D (two-dimensional) estimation of individual crown diameters. Trunk detection results and LiDAR point clusters generated by the adaptive mean shift procedures serve as mutual references for final detection of individual trees. Experimental results show that a combination of adaptive mean shift clustering and detected tree trunk can provide a significant performance improvement in individual tree-level forest measurement. Compared with conventional clustering techniques, the trunk detection-aided mean shift clustering approach can detect 91.1% of the trees ("recall") with a higher tree positioning accuracy (the mean positioning error is reduced by 33%) in a multi-layered coniferous and broad-leaved mixed forest in South China, and 93.5% of the identified trees are correct ("precision"). The tree detection brings the estimation of structural parameters for individual trees up to an accuracy level: −2.2% mean relative error and 5.8% relative RMSE (Root Mean Square Error) for tree height and 0.6% mean relative error and 21.9% relative RMSE for crown diameter, respectively.

Segmenting individual trees from the forest point cloud has significant implications in forestry inventory. This paper presents a novel computational scheme to segment and cluster the 3D point cloud data acquired by an airborne LiDAR. The scheme employs a mean shift-based iterative procedure on the data sets in a defined complex multimodal feature space to cluster points with similar modes together. Experimental results reveal that the proposed scheme can work effectively and the average accuracy of tree detection (88.6%) can meet the requirements of forest inventory.

No abstract