Xuebo Yang

Plot-scale aboveground biomass density (AGBD) from airborne LiDAR is commonly estimated using machine-learning models or parametric area-based models. Machine-learning models can achieve high accuracy but provide limited structural interpretation, whereas parametric models are transparent but mainly empirical. Here we develop an Interception–Profile–Height Biomass Index (IPH-BI) that connects plot-scale biomass to individual-tree allometry through a radiative-transfer and Beer–Lambert derivation. IPH-BI represents biomass using three LiDAR-observable components: canopy interception, relative vertical profile, and canopy height.We tested the three components against co-registered terrestrial and UAV laser scanning (TLS+ULS) references on 104 plots. The first-return canopy cover observable was correlated with the TLS+ULS reference (Pearson r = 0.784). Pairwise differences among ULS first-return profiles closely matched those among the corresponding TLS+ULS profiles (Pearson r = 0.90). The fitted plot-level height exponent (≈ 1.08) was smaller than the value propagated from tree-level allometry (≈ 2.32), indicating that tree-level scaling does not transfer directly to the plot scale. We evaluated our model using 180 plots from three airborne LiDAR data sources covering forests in Guangxi, Hebei, and Yunnan, China, and Virginia and Florida, USA. Across these plots, IPH-BI achieved R2 = 0.541 and an RMSE of 39.7 Mg/ha. Its accuracy approached that of a 36-feature random forest (R2 = 0.587, RMSE = 37.6 Mg/ha) and exceeded that of a log-linear area-based model (R2 = 0.401, RMSE = 45.3 Mg/ha). Holding out an entire data source reduced the accuracy of all models. For IPH-BI, however, re-fitting a single parameter using five plots from the new source recovered substantially more predictive accuracy than the corresponding recalibrations of the random forest and area-based model. IPH-BI was less sensitive than the random forest to point-density loss and remained stable across vertical-bin thicknesses. These results show that IPH-BI provides an interpretable, testable, and readily recalibratable framework for plot-scale airborne LiDAR biomass estimation.

Fine three-dimensional (3D) reconstruction of real forest scenes can provide a reference for forestry digitization and forestry resource management applications. Airborne LiDAR technology can provide valuable data for large-area forest scene reconstruction. This paper proposes a 3D reconstruction method for complex forest scenes with trees, shrubs, and grass, based on airborne LiDAR point clouds. First, forest vertical distribution characteristics are used to segment tree, shrub, and ground–grass points from an airborne LiDAR point cloud. For ground–grass points, a ground–grass grid model is constructed. For tree points, a method based on hierarchical canopy point fitting is proposed to construct a trunk model, and a crown model is constructed with the 3D α-shape algorithm. For shrub points, a shrub model is directly constructed based on the 3D α-shape algorithm. Finally, tree, shrub, and ground–grass models are spatially combined to achieve the reconstruction of real forest scenes. Taking six forest plots located in Hebei, Yunnan, and Guangxi provinces in China and Baden-Württemberg in Germany as study areas, experimental results show that the accuracy of individual tree segmentation reaches 87.32%, the accuracy of shrub segmentation reaches 60.00%, the height accuracy of the grass model is evaluated with an RMSE < 0.15 m, the volume accuracy of shrub and tree models is assessed with R² > 0.848 and R² > 0.904, respectively. Furthermore, we compared the model constructed in this study with simplified point cloud and voxel models. The results demonstrate that the proposed modeling approach can meet the demand for the high-accuracy and lightweight modeling of large-area forest scenes.

Multiple scattering of laser ray leads to distance calculation error and accumulated intensity error in LiDAR waveform. This study quantitatively assesses multiple scattering in LiDAR canopy waveform and explores its relationship with sensor configurations, forest structure, and leaf biochemistry. Airborne LiDAR measurements were conducted to verify presence of multiple scattering in spaceborne LiDAR waveforms. A Monte Carlo-based radiative transfer model, DART, was applied to simulate laser multiple scattering. The simulated results were compared with the actual measurements and further revealed how forest characteristics and sensor parameters influence multiple scattering. Results show that: 1) multiple scattering contributes significantly (∼30% of canopy subwaveform and ∼18% of total waveform) when there are sufficient scatterers in large LiDAR footprint, and the scatterers possess a reasonable spatial distribution and a strong albedo at the laser wavelength; 2) contribution of multiple scattering is typically less than 5% for small-footprint LiDAR; 3) multiple scattering of large-footprint LiDAR increases with forest leaf area index and fractional coverage, while also varying with laser pointing angle, crown shape, and leaf angle distribution; 4) leaf biochemical contents affect multiple scattering only at their sensitive wavelength. These findings help to quantitatively explain errors in the simulation, validation, and interpretation of LiDAR signal.

Three-dimensional (3D) point cloud registration is a prerequisite for scene reconstruction and 3D object recognition in computer vision and remote sensing. Numerous previous studies have presented a series of point cloud registration algorithms with diverse efficiencies and accuracies. However, registering point clouds with repetitive structures is still challenging. In this study, we propose the Global Planar Structural Constraint Optimal-based algorithm (GPSCO), which is specifically designed to handle registration of repetitive structures. Its novelty lies in establishing the geometric constraint of multiple planes and registering based on the global optimal geometric constraint. The specific algorithm involves clustering the parallel planes into plane groups, estimating matching scores between plane groups, and selecting three corresponding pairs of nonparallel plane groups to form the plane structural constraint. The transformation matrix determined in the case of the optimal structural constraint is taken as the final result. The two terrestrial LiDAR datasets (HS1 and HS2) of real scenes with repetitive structures were collected to evaluate the GPSCO algorithm. Additionally, the GPSCO algorithm is validated on four public benchmark datasets, such as Whu-Park, Whu-Campus, ETH-Hauptgebaude, and ETH-Stairs. The registration results demonstrate that the GPSCO algorithm achieves 95.65%, 86.36%, 100%, 100%, 100%, and 88.89% successful registration rates on the six datasets, respectively, and significantly outperforms the existing methods on HS1 and HS2 with repetitive structures. The corresponding datasets and code are available at [https://github.com/fog223/GPSCO].

Light detection and ranging, or LiDAR, is an advanced active remote sensing technology developed in the last 30 years to measure variable distances to the Earth. This book explains the fundamental concepts of LiDAR technology and its extended spaceborne, airborne, terrestrial, mobile, and unmanned aerial vehicle (UAV) platforms. It addresses the challenges of massive LiDAR data intelligent processing, LiDAR software engineering, and in-depth applications. The theory and algorithms are integrated with multiple applications in a systematic way and with step-by-step instructions. Written for undergraduate and graduate students and practitioners in the field of LiDAR remote sensing, this book is a much-needed comprehensive resource. FEATURES Explains the fundamentals of LiDAR remote sensing, including theory, techniques, methods, and applications Highlights the dissemination and popularization of LiDAR remote sensing technology in the last decade Includes new advances in LiDAR data processing and applications Introduces new technologies such as spaceborne LiDAR and photon-counting LiDAR Provides multiple LiDAR application cases regarding topography mapping, forest investigation, power line inspection, building modeling, automatic driving, crop monitoring, indoor navigation, cultural heritage conservation, and underwater mapping This book is written for graduate and upper-level undergraduate students taking courses in remote sensing, geography, photogrammetric engineering, laser techniques, surveying and mapping, geographic information systems (GIS), forestry, and resources and environmental protection. It is also a comprehensive resource for researchers and scientists interested in learning techniques for collecting LiDAR remote sensing data and processing, analyzing, and managing LiDAR data for applications in forestry, surveying and mapping, cultural relic protection, and digital products. Chapters 1 and 2 of this book are freely available as a downloadable Open Access PDF at https://www.taylorfrancis.com under a Creative Commons Attribution-Non Commercial-No Derivatives (CC-BY-NC-ND) 4.0 license.

LiDAR data processing is the basis of subsequent extensive applications. It aims to provide high-precision three-dimensional data or attribute information of targets for applications. In practice, the processing methods of LiDAR discrete point cloud data, full-waveform data, and photon-counting data are different due to the different detection and digitization methods. This chapter introduces the processing flows and key techniques for these three LiDAR data types in detail. The key steps of LiDAR discrete point cloud data processing include point cloud denoising, point cloud filtering, and point cloud classification. LiDAR full-waveform data processing steps involve waveform denoising, waveform decomposition, waveform deconvolution, and waveform characteristic parameter extraction. LiDAR photon-counting data processing steps involve photon-counting denoising and photon-counting classification.

LiDAR remote sensing is listed as one of the three cutting-edge techniques in the field of Earth observation, along with imaging spectroscopy and imaging radar. In the past 30 years, new types of LiDAR, such as full-waveform, photon counting, and multi-spectral/hyperspectral LiDAR, have been introduced one after another. The manufacturing of the LiDAR system, the data processing software, and the applications have flourished, further promoting the application fields and market development potential of LiDAR remote sensing. Based on systematic analyses of previous research progress of LiDAR remote sensing, this chapter briefly introduces the future development trend of LiDAR, including LiDAR sensor development, LiDAR data processing in the era of big data, and comprehensive applications of LiDAR.

LiDAR remote sensing is widely applied in many industries related to economic and social development. This chapter introduces several typical LiDAR applications and specific methods in detail. For topographic mapping, the surface model and the contour line can be generated from the LiDAR data. For forestry investigation, LiDAR remote sensing is widely used for extractions of individual tree parameters and forest stand parameters, tree species classification, and forest parameter mapping at the regional scale. For power line inspection, key steps include transmission corridor point cloud classification, three-dimensional (3D) modeling of power lines and power towers, and safety analysis. For digital cities, building 3D models can be constructed from the LiDAR measurements through the steps of building point cloud extraction, roof segmentation, contour extraction, and model generation. Other applications of LiDAR are also presented, including environmental perception and autonomous positioning in automatic driving, leaf inclination and fraction of absorbed photosynthetically active radiation (FPAR) estimations in crop monitoring, digital archaeological and cultural heritage conservation, indoor 3D modeling and navigation, and underwater mapping.

Principles of LiDAR remote sensing are the theoretical basis for subsequent LiDAR data acquisition, processing, and applications. They involve expertise in many fields, such as space science, computer science, earth science, machinery, and electronics. This chapter first introduces the ranging and radiation principles of LiDAR. They are expressed by the LiDAR ranging equation and the LiDAR radiative transfer equation, respectively. For LiDAR systems with different platforms (spaceborne/airborne/terrestrial LiDAR) and different methods of detection and digitization (full waveform/discrete return/photon counting), their operating principles are described as well. Finally, the influence mechanism of the external environment (i.e., atmosphere and sunlight) on the LiDAR signal is briefly introduced.

LiDAR data acquisition is the prerequisite and foundation of LiDAR data processing and application. This chapter describes in detail the acquisition of LiDAR data on different platforms, i.e., satellite, aircraft, and terrestrial platforms. Airborne LiDAR data acquisition includes flight plan preparation, flight implementation, and data preprocessing. Terrestrial LiDAR data acquisition includes preparation before scanning, scanning implementation, and data collection. These are helpful for practitioners to conduct the measurement experiments of airborne or terrestrial LiDAR properly. For spaceborne LiDAR, the platform, system parameters, data products, and data download methods of some typical systems (e.g., ICESat-1, ICESat-2, and GEDI) are introduced.