Quantification of occlusions influencing the tree stem curve retrieving from single-scan terrestrial laser scanning data

Background: The stem curve of standing trees is an essential parameter for accurate estimation of stem volume.This study aims to directly quantify the occlusions within the single-scan terrestrial laser scanning(TLS) data,evaluate its correlation with the accuracy of the retrieved stem curves, and subsequently, to assess the capacity of single-scan TLS to estimate stem curves.Methods: We proposed an index, occlusion rate, to quantify the occlusion level in TLS data. We then analyzed three influencing factors for the occlusion rate: the percentage of basal area near the scanning center, the scanning distance and the source of occlusions. Finally, we evaluated the effects of occlusions on stem curve estimates from single-scan TLS data.Results: The results showed that the correlations between the occlusion rate and the stem curve estimation accuracies were strong(r = 0.60–0.83), so was the correlations between the occlusion rate and its influencing factors(r = 0.84–0.99). It also showed that the occlusions from tree stems were the main factor of the low detection rate of stems, while the non-stem components mainly influenced the completeness of the retrieved stem curves.Conclusions: Our study demonstrates that the occlusions significantly affect the accuracy of stem curve retrieval from the single-scan TLS data in a typical-size(32 m × 32 m) forest plot. However, the single-scan mode has the capacity to accurately estimate the stem curve in a small forest plot(< 10 m × 10 m) or a plot with a lower occlusion rate, such as less than 35% in our tested datasets. The findings from this study are useful for guiding the practice of retrieving forest parameters using single-scan TLS data.

Cloth simulation-based construction of pitfree canopy height models from airborne LiDAR data

Background:The universal occurrence of randomly distributed dark holes(i.e.,data pits appearing within the tree crown)in LiDAR-derived canopy height models(CHMs)negatively affects the accuracy of extracted forest inventory parameters.Methods:We develop an algorithm based on cloth simulation for constructing a pit-free CHM.Results:The proposed algorithm effectively fills data pits of various sizes whilst preserving canopy details.Our pitfree CHMs derived from point clouds at different proportions of data pits are remarkably better than those constructed using other algorithms,as evidenced by the lowest average root mean square error(0.4981 m)between the reference CHMs and the constructed pit-free CHMs.Moreover,our pit-free CHMs show the best performance overall in terms of maximum tree height estimation(average bias=0.9674 m).Conclusion:The proposed algorithm can be adopted when working with different quality LiDAR data and shows high potential in forestry applications.

Improving the estimation of canopy cover from UAV-LiDAR data using a pit-free CHM-based method

Accurate and rapid estimation of canopy cover(CC)is crucial for many ecological and environmental models and for forest management.Unmanned aerial vehicle-light detecting and ranging(UAV-LiDAR)systems represent a promising tool for CC estimation due to their high mobility,low cost,and high point density.However,the CC values from UAV-LiDAR point clouds may be underestimated due to the presence of large quantities of within-crown gaps.To alleviate the negative effects of within-crown gaps,we proposed a pit-free CHM-based method for estimating CC,in which a cloth simulation method was used to fill the within-crown gaps.To evaluate the effect of CC values and withincrown gap proportions on the proposed method,the performance of the proposed method was tested on 18 samples with different CC values(40−70%)and 6 samples with different within-crown gap proportions(10−60%).The results showed that the CC accuracy of the proposed method was higher than that of the method without filling within-crown gaps(R^(2)=0.99 vs 0.98;RMSE=1.49%vs 2.2%).The proposed method was insensitive to within-crown gap proportions,although the CC accuracy decreased slightly with the increase in withincrown gap proportions.

Building segmentation and modeling from airborne LiDAR data

Due to the high accuracy and fast acquisition speed offered by airborne Light Detection and Ranging(LiDAR)technology,airborne LiDAR point clouds have been widely used in three-dimensional building model reconstruction.This paper presents a novel approach to segment building roofs from point clouds using a Gaussian mixture model in which buildings are represented by a mixture of Gaussians(MoG).The Expectation-Maximization(EM)algorithm with the minimum description length(MDL)principle is employed to obtain the optimal parameters of the MoG model for separating building roofs.To separate complete planar building roofs,coplanar Gaussian components are merged according to their distances to the corresponding planes.In addition,shape analysis is utilized to remove nonplanar objects caused by trees and irregular artifacts.Building models are obtained by combining segmented planar roofs,topological relationships,and regularized building boundaries.Roof intersection segments and points are derived by the segmentation results,and a rasterbased regularization method is employed to obtain geometrically correct and regular building models.Experimental results suggest that the segmentation method is able to separate building roofs with high accuracy while maintaining correct topological relationships among roofs.

Estimation of Larch Growth at the Stem,Crown,and Branch Levels Using Ground-Based LiDAR Point Cloud

Tree growth is an important indicator of forest health and can reflect changes in forest structure.Traditional tree growth estimates use easy-to-measure parameters,including tree height,diameter at breast height,and crown diameter,obtained via forest in situ measurements,which are labor intensive and time consuming.Some new technologies measure the diameter of trees at different positions to monitor the growth trend of trees,but it is difficult to take into account the growth changes at different tree levels.The combination of terrestrial laser scanning and quantitative structure modeling can accurately estimate tree structural parameters nondestructively and has the potential to estimate tree growth from different tree levels.In this context,this paper estimates tree growth from stem-,crown-,and branch-level attributes observed by terrestrial laser scanning.Specifically,tree height,diameter at breast height,stem volume,crown diameter,crown volume,and first-order branch volume were used to estimate the growth of 55-year-old larch trees in Saihanba of China,at the stem,crown,and branch levels.The experimental results showed that tree growth is mainly reflected in the growth of the crown,i.e.,the growth of branches.Compared to onedimensional parameter growth(tree height,diameter at breast height,or crown diameter),three-dimensional parameter growth(crown,stem,and first-order branch volumes)was more obvious,in which the absolute growth of the first-order branch volume is close to the stem volume.Thus,it is necessary to estimate tree growth at different levels for accurate forest inventory.