Individual Tree Crown Segmentation Directly from UAV-Borne LiDAR Data Using the PointNet of Deep Learning

Abstract

Accurate individual tree crown (ITC) segmentation from scanned point clouds is a fundamental task in forest biomass monitoring and forest ecology management. Light detection and ranging (LiDAR) as a mainstream tool for forest survey is advancing the pattern of forest data acquisition. In this study, we performed a novel deep learning framework directly processing the forest point clouds belonging to the four forest types (i.e., the nursery base, the monastery garden, the mixed forest, and the defoliated forest) to realize the ITC segmentation. The specific steps of our approach were as follows: first, a voxelization strategy was conducted to subdivide the collected point clouds with various tree species from various forest types into many voxels. These voxels containing point clouds were taken as training samples for the PointNet deep learning framework to identify the tree crowns at the voxel scale. Second, based on the initial segmentation results, we used the height-related gradient information to accurately depict the boundaries of each tree crown. Meanwhile, the retrieved tree crown breadths of individual trees were compared with field measurements to verify the effectiveness of our approach. Among the four forest types, our results revealed the best performance for the nursery base (tree crown detection rate r = 0.90; crown breadth estimation R2 > 0.94 and root mean squared error (RMSE) < 0.2m). A sound performance was also achieved for the monastery garden and mixed forest, which had complex forest structures, complicated intersections of branches and different building types, with r = 0.85, R2 > 0.88 and RMSE < 0.6 m for the monastery garden and r = 0.80, R2 > 0.85 and RMSE < 0.8 m for the mixed forest. For the fourth forest plot type with the distribution of crown defoliation across the woodland, we achieved the performance with r = 0.82, R2 > 0.79 and RMSE < 0.7 m. Our method presents a robust framework inspired by the deep learning technology and computer graphics theory that solves the ITC segmentation problem and retrieves forest parameters under various forest conditions.

Combining Multiple Geospatial Data for Estimating Aboveground Biomass in North Carolina Forests

“…The specific method of estimation of tree heights from LiDAR data we use was chosen for its proven wide applicability and use. While other methods of tree height estimation might be better suited under specific situations [35][36][37], there is no indication that alternatively derived tree height estimates would have altered our main conclusion about the disagreement in the CDL and the NLCD impacting the area-total AGFB estimates.…”

“…The canopy height information combined with stand density or canopy cover has been used for biomass estimation [33,34]. The challenges and performance of the methods to detect and delineate individual trees from LiDAR data are a subject of continuous research [35][36][37]. In this study, we apply a widely used method proposed by [38].…”

Estimation of Northern Hardwood Forest Inventory Attributes Using UAV Laser Scanning (ULS): Transferability of Laser Scanning Methods and Comparison of Automated Approaches at the Tree- and Stand-Level

“…From an application point of view, collecting ULS data in leaf-off hardwood stands supports a wide array of tree-level analyses, such as selective logging [118], allometric model development [119], tree stem modeling [62], tree species identification [120], health and vigor evaluation [121], or tree competition [122]. One aspect that seems particularly important to explore further is the use of deep learning analysis for species identification from bottom-up ITD trees from high-density ULS [123]. Integrating ULS data with TLS data could also be further studied for estimating the biomass of individual trees.…”

“…Integrating methods that can delineate or merge trees based on the analysis of similarities between segments (e.g., [113,117]) and their assignment to a specific canopy layer also present some potential for processing high-density point clouds (e.g., [20,127]). Last, including some evaluation criteria that are based on dendrometric criteria and machine learning (e.g., [123,128]) should also be further explored to really benefit from gleaning the full information that is available in high density ULS data.…”

Mago Approach for Semantic Segmentation: The Case Study of Uavid Benchmark Dataset

“…Semantic segmentation is a Computer Vision technique (Förstner and Wrobel, 2016) that aims to the recognition and the comprehension of the content of an image at the pixel level. This approach is widely used in remote sensing applications, especially in the analysis of urban scenarios (Ajmar et al, 2019;Huang et al, 2019, Schmitz et al, 2019, Zhou et al, 2019 or in the delineation of forest trees (Chen et al, 2021;Sothe et al, 2020;Kempf et al, 2019). The segmentation approach could be based on imagery (Marmanis et al, 2018) or three-dimensional models (Ao et al, 2019), as well as on the combination of both 2D and 3D information (Ding et al, 2019).…”