scholarly journals Machine Learning Approaches for Estimating Forest Stand Height Using Plot-Based Observations and Airborne LiDAR Data

Forests ◽  
2018 ◽  
Vol 9 (5) ◽  
pp. 268 ◽  
Author(s):  
Junghee Lee ◽  
Jungho Im ◽  
Kyungmin Kim ◽  
Lindi Quackenbush
Keyword(s):  
Machine Learning ◽  
Forest Stand ◽  
Airborne Lidar ◽  
Lidar Data ◽  
Learning Approaches ◽  
Stand Height ◽  
Airborne Lidar Data

scholarly journals Using low density LiDAR data to map Mediterranean forest characteristics by means of an area-based approach and height threshold analysis

2016 ◽  
pp. 103 ◽  
Author(s):  
J. Guerra-Hernández ◽  
M. Tomé ◽  
E. González-Ferreiro
Keyword(s):  
Regression Models ◽  
Deciduous Forest ◽  
Coniferous Forest ◽  
Mixed Forest ◽  
Forest Stand ◽  
Airborne Lidar ◽  
Mediterranean Forest ◽  
Lidar Data ◽  
Low Density ◽  
Airborne Lidar Data

<p>This study reports progress in forest inventory methods involving the use of low density airborne LiDAR data and an area-based approach (ABA). It also emphasizes the usefulness of the Spanish countrywide LiDAR dataset for mapping forest stand attributes in Mediterranean stone pine forest characterized by complex orography. Lowdensity airborne LiDAR data (0.5 first returns m<sup><span lang="EN-US">–2</span></sup>) was used to develop individual regression models for a set of forest stand variables in different types of forest. LiDAR data is now freely available for most of the Spanish territory and is provided by the Spanish National Aerial Photography Program (Plan Nacional de Ortofotografía Aérea, PNOA). The influence of height thresholds (MHT: Minimun Height Threshold and BHT: Break Height Threshold) used in extracting LiDAR metrics was also investigated. The best regression models explained 61-85%, 67-98% and 74-98% of the variability in ground-truth stand height, basal area and volume, respectively. The magnitude of error for predicting structural vegetation parameters was higher in closed deciduous and mixed forest than in the more homogeneous coniferous stands. Analysis of height thresholds (HT) revealed that these parameters were not particularly important for estimating several forest attributes in the coniferous forest; nevertheless, substantial differences in volume modelling were observed when the height thresholds (MHT and BHT) were increased in complex structural vegetation (mixed and deciduous forest). A metric-by-metric analysis revealed that there were significant differences in most of the explanatory variables computed from different height thresholds (HBT and MHT).The best models were applied to the reference stands to yield spatially explicit predictions about the forest resources. Reliable mapping of biometric variables was implemented to facilitate effective and sustainable management strategies and practices in Mediterranean Forest ecosystems.</p>

scholarly journals A Forest Attribute Mapping Framework: A Pilot Study in a Northern Boreal Forest, Northwest Territories, Canada

2018 ◽  
Vol 10 (9) ◽  
pp. 1338 ◽  
Author(s):  
Craig Mahoney ◽  
Ron Hall ◽  
Chris Hopkinson ◽  
Michelle Filiatrault ◽  
Andre Beaudoin ◽  
...  
Keyword(s):  
Vegetation Type ◽  
Scale Up ◽  
Mean Difference ◽  
Airborne Lidar ◽  
Lidar Data ◽  
Laser Altimeter ◽  
Regional Estimates ◽  
Stand Height ◽  
Crown Closure ◽  
Airborne Lidar Data

A methods framework is presented that utilizes field plots, airborne light detection and ranging (LiDAR), and spaceborne Geoscience Laser Altimeter System (GLAS) data to estimate forest attributes over a 20 Mha area in Northern Canada. The framework was implemented to scale up forest attribute models from field data to intersecting airborne LiDAR data, and then to GLAS footprints. GLAS data were sequentially filtered and submitted to the k-nearest neighbour (k-NN) imputation algorithm to yield regional estimates of stand height and crown closure at a 30 m resolution. Resulting outputs were assessed against independent airborne LiDAR data to evaluate regional estimates of stand height (mean difference = −1 m, RMSE = 5 m) and crown closure (mean difference = −5%, RMSE = 9%). Additional assessments were performed as a function of dominant vegetation type and ecoregion to further evaluate regional products. These attributes form the primary descriptive structure attributes that are typical of forest inventory mapping programs, and provide insight into how they can be derived in northern boreal regions where field information and physical access is often limited.

scholarly journals Cloth simulation-based construction of pit-free canopy height models from airborne LiDAR data

2020 ◽  
Vol 7 (1) ◽  
Author(s):  
Wuming Zhang ◽  
Shangshu Cai ◽  
Xinlian Liang ◽  
Jie Shao ◽  
Ronghai Hu ◽  
...  
Keyword(s):  
Tree Height ◽  
Point Clouds ◽  
Airborne Lidar ◽  
Canopy Height ◽  
Lidar Data ◽  
Cloth Simulation ◽  
Simulation Based ◽  
Maximum Tree Height ◽  
Airborne Lidar Data ◽  
Better Than

Abstract 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 pit-free 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.

Sign in / Sign up

Export Citation Format

Share Document