Individual Tree Detection Research Articles

Early detection of bark beetle infestation in Norway spruce forests of Central Europe using Sentinel-2

In the past decade, massive outbreaks of bark beetles (Ips spp.) have caused large-scale decline of coniferous-dominated, prevailingly managed forests of Central Europe. Timely detection of newly infested trees is important for minimizing economic losses and effectively planning forest management activities to stop or at least slow outbreaks. With the advancement of Copernicus services, a pair of Sentinel-2 satellites provides a unique remote sensing data source of multi-temporal observations in high spatial resolution on the scale of individual forest stands (although not allowing for individual tree detection). This study investigates the potential for using seasonal trajectories of Sentinel-2 bands and selected vegetation indices in early detection of bark beetle infestation (so–called green-attack stage detection) in Norway spruce monoculture forests in the Czech Republic. Spectral trajectories of nine bands and six vegetation indices were constructed for the 2018 vegetation season using 14 satellite observations from April to November to distinguish four infestation classes. We used a random forest algorithm to classify healthy (i.e., stands not infested) and infested trees with various trajectories of decay. The seasonal trajectories of vegetation indices separated the infestation classes better than did the individual bands. Among the most promising vegetation indices we identified the tasselled cap wetness (TCW) component and normalized difference index constructed from near and shortwave infrared bands. Analysing the inter-annual change of the indices was more promising for early detection than is single-date classification. It achieved 96% classification accuracy on day of year 291 for the tested data set.The algorithm for early detection of tree infestation based on the assessment of seasonal changes in TCW was applied on the time series of Sentinel-2 observations from 2019 and its outputs were verified using field observations of forest conditions conducted on 80 spruce forest plots (located in spruce monoculture stands). The overall accuracy of 78% was achieved for the separation of healthy and green-attack classes. Our study highlights the great potential of multi-temporal remote sensing and the use of shortwave infrared wavelengths for early detection of spruce forest decline caused by bark beetle infestation.

Precision Agriculture Workflow, from Data Collection to Data Management Using FOSS Tools: An Application in Northern Italy Vineyard

In the past decades, technology-based agriculture, also known as Precision Agriculture (PA) or smart farming, has grown, developing new technologies and innovative tools to manage data for the whole agricultural processes. In this framework, geographic information, and spatial data and tools such as UAVs (Unmanned Aerial Vehicles) and multispectral optical sensors play a crucial role in the geomatics as support techniques. PA needs software to store and process spatial data and the Free and Open Software System (FOSS) community kept pace with PA’s needs: several FOSS software tools have been developed for data gathering, analysis, and restitution. The adoption of FOSS solutions, WebGIS platforms, open databases, and spatial data infrastructure to process and store spatial and nonspatial acquired data helps to share information among different actors with user-friendly solutions. Nevertheless, a comprehensive open-source platform that, besides processing UAV data, allows directly storing, visualising, sharing, and querying the final results and the related information does not exist. Indeed, today, the PA’s data elaboration and management with a FOSS approach still require several different software tools. Moreover, although some commercial solutions presented platforms to support management in PA activities, none of these present a complete workflow including data from acquisition phase to processed and stored information. In this scenario, the paper aims to provide UAV and PA users with a FOSS-replicable methodology that can fit farming activities’ operational and management needs. Therefore, this work focuses on developing a totally FOSS workflow to visualise, process, analyse, and manage PA data. In detail, a multidisciplinary approach is adopted for creating an operative web-sharing tool able to manage Very High Resolution (VHR) agricultural multispectral-derived information gathered by UAV systems. A vineyard in Northern Italy is used as an example to show the workflow of data generation and the data structure of the web tool. A UAV survey was carried out using a six-band multispectral camera and the data were elaborated through the Structure from Motion (SfM) technique, resulting in 3 cm resolution orthophoto. A supervised classifier identified the phenological stage of under-row weeds and the rows with a 95% overall accuracy. Then, a set of GIS-developed algorithms allowed Individual Tree Detection (ITD) and spectral indices for monitoring the plant-based phytosanitary conditions. A spatial data structure was implemented to gather the data at canopy scale. The last step of the workflow concerned publishing data in an interactive 3D webGIS, allowing users to update the spatial database. The webGIS can be operated from web browsers and desktop GIS. The final result is a shared open platform obtained with nonproprietary software that can store data of different sources and scales.

A Crown Morphology-Based Approach to Individual Tree Detection in Subtropical Mixed Broadleaf Urban Forests Using UAV LiDAR Data

Owing to the complex forest structure and large variation in crown size, individual tree detection in subtropical mixed broadleaf forests in urban scenes is a great challenge. Unmanned aerial vehicle (UAV) light detection and ranging (LiDAR) is a powerful tool for individual tree detection due to its ability to acquire high density point cloud that can reveal three-dimensional crown structure. Tree detection based on a local maximum (LM) filter, which is applied on a canopy height model (CHM) generated from LiDAR data, is a popular method due to its simplicity. However, it is difficult to determine the optimal LM filter window size and prior knowledge is usually needed to estimate the window size. In this paper, a novel tree detection approach based on crown morphology information is proposed. In the approach, LMs are firstly extracted using a LM filter whose window size is determined by the minimum crown size and then the crown morphology is identified based on local Gi* statistics to filter out LMs caused by surface irregularities contained in CHM. The LMs retained in the final results represent treetops. The approach was applied on two test sites characterized by different forest structures using UAV LiDAR data. The sensitivity of the approach to parameter setting was analyzed and rules for parameter setting were proposed. On the first test site characterized by irregular tree distribution and large variation in crown size, the detection rate and F-score derived by using the optimal combination of parameter values were 72.9% and 73.7%, respectively. On the second test site characterized by regular tree distribution and relatively small variation in crown size, the detection rate and F-score were 87.2% and 93.2%, respectively. In comparison with a variable-size window tree detection algorithm, both detection rates and F-score values of the proposed approach were higher.

Linking urban tree inventories to remote sensing data for individual tree mapping

Urban trees provide a range of vital social and environmental services. Currently, inventories of individual urban trees are conducted in-situ by professional arborists. Such an approach to urban tree inventories means they are challenging to maintain and only capture information describing trees on accessible land. Whilst remote sensing approaches have shown the potential to derive individual tree attributes, these studies rarely make use of existing inventory information. In this study, we present a method to parameterise an algorithm for individual tree detection and delineation from airborne remote sensing data. The approach uses existing inventory data as training information firstly for the detection of the canopy area and secondly to parameterise a marker-based watershed segmentation algorithm. In this parameterisation, crown segmentation shape, as well as features derived from the remote sensing data, are used to determine if a segment contains one or more trees. If a segment contains more than one tree, it is split with the number of markers increased until each segment includes only one tree. The approach was evaluated within three distinct urban areas: the central business district, an urban park and a residential area, to be determined. Commission and omission errors ranged between 11% and 27% across the three regions, with commission typically caused by land covers on private land that were unaccounted for in the training processes. In all areas the overall tree count was within two per cent of that defined by reference information. The accurate tree count produced by this approach suggests it has the potential to be adopted by government agencies for routine tree inventory maintenance.

Prediction error aggregation behaviour for remote sensing augmented forest inventory approaches

Abstract In this study we investigated the behaviour of aggregate prediction errors in a forest inventory augmented with multispectral Airborne Laser Scanning and airborne imagery. We compared an Area-Based Approach (ABA), Edge-tree corrected ABA (EABA) and Individual Tree Detection (ITD). The study used 109 large 30 × 30 m sample plots, which were divided into four 15 × 15 m subplots. Four different levels of aggregation were examined: all four subplots (quartet), two diagonal subplots (diagonal), two edge-adjacent subplots (adjacent) and subplots without aggregation. We noted that the errors at aggregated levels depend on the selected predictor variables, and therefore, this effect was studied by repeating the variable selection 200 times. At the subplot level, EABA provided the lowest mean of root mean square error ($\overline{\mathrm{RMSE}}$) values of 200 repetitions for total stem volume (EABA 21.1 percent, ABA 23.5 percent, ITD 26.2 percent). EABA also fared the best for diagonal and adjacent aggregation ($\overline{\mathrm{RMSE}}$: 17.6 percent, 17.4 percent), followed by ABA ($\overline{\mathrm{RMSE}}$: 19.3 percent, 18.2 percent) and ITD ($\overline{\mathrm{RMSE}}$: 21.8, 21.9 percent). Adjacent subplot errors of ABA were less correlated than errors of diagonal subplots, which resulted also in clearly lower RMSEs for adjacent subplots. This appears to result from edge tree effects, where omission and commission errors cancel for trees leaning from one subplot into the other. The best aggregate performance was achieved at the quartet level, as expected from fundamental properties of variance. ABA and EABA had similar RMSEs at the quartet level ($\overline{\mathrm{RMSE}}$ 15.5 and 15.3 percent), with poorer ITD performance ($\overline{\mathrm{RMSE}}$ 19.4 percent).

Can a Remote Sensing Approach with Hyperspectral Data Provide Early Detection and Mapping of Spatial Patterns of Black Bear Bark Stripping in Coast Redwoods?

The prevalence of black bear (Ursus americanus) bark stripping in commercial redwood (Sequoia sempervirens (D. Don) Endl.) timber stands has been increasing in recent years. This stripping is a threat to commercial timber production because of the deleterious effects on redwood tree fitness. This study sought to unveil a remote sensing method to detect these damaged trees early and map their spatial patterns. By developing a timely monitoring method, forest timber companies can manipulate their timber harvesting routines to adapt to the consequences of the problem. We explored the utility of high spatial resolution UAV-collected hyperspectral imagery as a means for early detection of individual trees stripped by black bears. A hyperspectral sensor was used to capture ultra-high spatial and spectral information pertaining to redwood trees with no damage, those that have been recently attacked by bears, and those with old bear damage. This spectral information was assessed using the Jeffries-Matusita (JM) distance to determine regions along the electromagnetic spectrum that are useful for discerning these three-health classes. While we were able to distinguish healthy trees from trees with old damage, we were unable to distinguish healthy trees from recently damaged trees due to the inherent characteristics of redwood tree growth and the subtle spectral changes within individual tree crowns for the time period assessed. The results, however, showed that with further assessment, a time window may be identified that informs damage before trees completely lose value.

Influence of Agisoft Metashape Parameters on UAS Structure from Motion Individual Tree Detection from Canopy Height Models

Applications of unmanned aerial systems for forest monitoring are increasing and drive a need to understand how image processing workflows impact end-user products’ accuracy from tree detection methods. Increasing image overlap and making acquisitions at lower altitudes improve how structure from motion point clouds represents forest canopies. However, only limited testing has evaluated how image resolution and point cloud filtering impact the detection of individual tree locations and heights. We evaluate how Agisoft Metashape’s build dense cloud Quality (image resolution) and depth map filter settings influence tree detection from canopy height models in ponderosa pine forests. Finer resolution imagery with minimal filtering provided the best visual representation of vegetation detail for trees of all sizes. These same settings maximized tree detection F-score at >0.72 for overstory (>7 m tall) and >0.60 for understory trees. Additionally, overstory tree height bias and precision improve as image resolution becomes finer. Overstory and understory tree detection in open-canopy conifer systems might be optimized using the finest resolution imagery that computer hardware enables, while applying minimal point cloud filtering. The extended processing time and data storage demands of high-resolution imagery must be balanced against small reductions in tree detection performance when down-scaling image resolution to allow the processing of greater data extents.

A Density-Based Algorithm for the Detection of Individual Trees from LiDAR Data

Nowadays, LiDAR is widely used for individual tree detection, usually providing higher accuracy in coniferous stands than in deciduous ones, where the rounded-crown, the presence of understory vegetation, and the random spatial tree distribution may affect the identification algorithms. In this work, we propose a novel algorithm that aims to overcome these difficulties and yield the coordinates and the height of the individual trees on the basis of the point density features of the input point cloud. The algorithm was tested on twelve deciduous areas, assessing its performance on both regular-patterned plantations and stands with randomly distributed trees. For all cases, the algorithm provides high accuracy tree count (F-score > 0.7) and satisfying stem locations (position error around 1.0 m). In comparison to other common tools, the algorithm is weakly sensitive to the parameter setup and can be applied with little knowledge of the study site, thus reducing the effort and cost of field campaigns. Furthermore, it demonstrates to require just 2 points·m−2 as minimum point density, allowing for the analysis of low-density point clouds. Despite its simplicity, it may set the basis for more complex tools, such as those for crown segmentation or biomass computation, with potential applications in forest modeling and management.

Growing status observation for oil palm trees using Unmanned Aerial Vehicle (UAV) images

For both the positive economic benefit and the negative ecological impact of the rapid expansion of oil palm plantations in tropical developing countries, it is significant to achieve accurate detection for oil palm trees in large-scale areas. Especially, growing status observation and smart oil palm plantation management enabled by such accurate detections would improve plantation planning, oil palm yield, and reduce manpower and consumption of fertilizer. Although existing studies have already reached a high accuracy in oil palm tree detection, rare attention has been paid to automated observation of each single oil palm tree’s growing status. Nowadays, with its high spatial resolution and low cost, Unmanned Aerial Vehicle (UAV) has become a promising tool for monitoring the growing status of individual oil palms. However, the accuracy is still a challenging issue because of the extreme imbalance and high similarity between different classes. In this paper, we propose a Multi-class Oil PAlm Detection approach (MOPAD) to reap both accurate detection of oil palm trees and accurate monitoring of their growing status. Based on Faster RCNN, MOPAD combines a Refined Pyramid Feature (RPF) module and a hybrid class-balanced loss module to achieve satisfying observation of the growing status for individual oil palms. The former takes advantage of multi-level features to distinguish similar classes and detect small oil palms, and the latter effectively resolves the problem of extremely imbalanced samples. Moreover, we elaborately analyze the distribution of different kinds of oil palms, and propose a practical workflow for detecting oil palm vacancy. We evaluate MOPAD using three large-scale UAV images photographed in two sites in Indonesia (denoted by Site 1 and Site 2), containing 363,877 oil palms of five categories: healthy palms, dead palms, mismanaged palms, smallish palms and yellowish palms. Our proposed MOPAD achieves an F1-score of 87.91% (Site 1) and 99.04% (Site 2) for overall oil palm tree detection, and outperforms other state-of-the-art object detection methods by a remarkable margin of 10.37–17.09% and 8.14%-21.32% with respect to the average F1-score for multi-class oil palm detection in Site 1 and Site 2, respectively. Our method demonstrates excellent potential for individual oil palm tree detection and observation of growing status from UAV images, leading to more precise and efficient management of oil palm plantations.

Assessment of Above-Ground Carbon Storage by Urban Trees Using LiDAR Data: The Case of a University Campus

The biomass represented by urban trees is important for urban decision-makers, green space planners, and managers seeking to optimize urban ecosystem services. Carbon storage by urban trees is one of these services. Suitable methods for assessing carbon storage by urban trees are being explored. The latest technologies in remote sensing and data analyses can reduce data collection costs while improving accuracy. This paper introduces an assessment approach that combines ground measurements with unmanned aerial vehicle-based light detection and ranging (LiDAR) data to estimate carbon storage by urban trees. Methods underpinning the approach were tested for the case of the Vancouver campus of the University of British Columbia (UBC), Canada. The study objectives were (1) to test five automated individual tree detection (AITD) algorithms and select one on the basis of the highest segmentation accuracy, (2) to develop a model to estimate the diameter at breast height (DBH), and (3) to estimate and map carbon storage over the UBC campus using LiDAR heights, estimated DBHs, and an existing tree-level above-ground carbon estimation model. Of the segmentation algorithms tested, the Dalponte AITD had the highest F score of 0.83. Of the five CW thresholds (th) tested in the DBH estimation model, we chose one resulting in the lowest Akaike’s information criterion, the highest log-likelihood, and the lowest root-mean-squared error (19.55 cm). Above-ground carbon was estimated for each tree in the study area and subsequently summarized, resulting in an estimated 5.27 kg C·m−2 over the main campus of UBC, Vancouver. The approach could be used in other urban jurisdictions to obtain essential information on urban carbon storage in support of urban landscape governance, planning, and management.

Assessing tree height and density of a young forest using a consumer unmanned aerial vehicle (UAV)

Accurate, cost-effective monitoring and management of young forests is important for future stand quality. There is a critical need for a rapid assessment tool for forest monitoring and management. This study uses a low-cost unmanned aerial vehicle (UAV) to complete a tree height and tree density assessment in a newly forested Chinese fir (Cunninghamia lanceolata (Lamb) Hook) planting (15 sample plots), Shunchang County, Fujian, China (1.12 ha). Images obtained from a Phantom4-Multispectral UAV were used to generate a digital surface model (DSM) with DJI Terra software (0.02 m spatial resolution). Based on the DSM, the individual trees were identified and the height of each corresponding tree was determined. The impacts of factors related to individual tree detection (ITD) and tree height accuracy were also analyzed. For the tree-level, the highest accuracy of ITD for Chinese fir was 98.93% (F-score = 98.93%). Remotely sensed individual tree heights produced an R2 value of 0.89, RMSE value of 0.22 m when compared to a field survey. At the stand-level, tree height assessment yielded R2 = 0.95, RMSE = 0.12 m, and tree density assessment yielded R2 = 0.99, RMSE = 48 tree ha−1. The results highlight that UAVs can successfully monitor forest parameters and hold great potential as a supplement or substitute tool in field inventory.

Определение характеристик смешанных древостоев по данным аэрофотосъёмки с применением беспилотного летательного аппарата (БПЛА)

Nowadays, due to the rapid development of lightweight unmanned aerial vehicles (UAV), remote sensing systems of ultra-high resolution have become available to many researchers. Conventional ground-based measurements for assessing tree stand attributes can be expensive, as well as time- and labor-consuming. Here, we assess whether remote sensing measurements with lightweight UAV can be more effective in comparison to ground survey methods in the case of temperate mixed forests. The study was carried out at the Prioksko-Terrasny Biosphere Nature Reserve (Moscow region, Russia). This area belongs to a coniferous-broad-leaved forest zone. Our field works were carried out on the permanent sampling plot of 1 ha (100×100 m) established in 2016. The coordinates of the plot center are N 54.88876°, E 37.56273° in the WGS 84 datum. All trees with DBH (diameter at breast height) of at least 6 cm (779 trees) were mapped and measured during the ground survey in 2016 (See Fig. 1 and Table 1). Mapping was performed with Laser Technology TruPulse 360B angle and a distance meter. First, polar coordinates of each tree trunk were measured, and then, after conversion to the cartesian coordinates, the scheme of the stand was validated onsite. Species and DBH were determined for each tree. For each living tree, we detected a social status class (according to Kraft). Also for living trees, we measured the tree height and the radii of the crown horizontal projection in four cardinal directions. A lightweight UAV Phantom 4 (DJI-Innovations, Shenzhen, China) equipped with an integrated camera of 12Mp sensor was used for aerial photography in this study. Technical parameters of the camera are available in Table 2. The aerial photography was conducted on October 12, 2017, from an altitude of 68 m. The commonly used mosaic flight mode was used with 90% overlapping both for side and front directions. We applied Agisoft Metashape software for orthophoto mosaic image and dense point cloud building. The canopy height model (CHM) was generated with lidR package in R. We used lasground() function and cloth simulation filter for classification of ground points. To create a normalized dataset with the ground at 0, we used spatial interpolation algorithm tin based on a Delaunay triangulation, which performs a linear interpolation within each triangle, implemented in the lasnormilise() function. CHM was generated according to the pit-free algorithm based on the computation of a set of classical triangulations at different heights. The location and height of individual trees were automatically detected by the function FindTreesCHM() from the package rLIDAR in R. The algorithm implemented in this function is local maximum with fixed window size. Accuracy assessment of automatically detected trees (in QGIS software) was performed through visual interpretation of orthophoto mosaic and comparison with ground survey data. The number of correctly detected trees, omitted by the algorithm and not existing but detected trees were counted. As a result of aerial photography, 501 images were obtained. During these data processing with the Metashape, dense point cloud of 163.7 points / m2 was generated. CHM with 0.5 m resolution was calculated. According to the individual-tree detection algorithm, 241 trees were found automatically (See Fig. 2A). The total accuracy of individual tree detection was 73.9%. Coniferous trees (Pinus sylvestris and Picea abies) were successfully detected (86.0% and 100%, respectively), while results for birch (Betula spp.) required additional treatment. The algorithm correctly detected only 58.2% of birch trees due to false-positive trees (See Fig. 2B and Table 3). These results confirm the published literature data obtained for managed tree stands. Tree heights retrieved from the UAV were well-matched to ground-based method results. The mean tree heights retrieved from the UAV and ground surveys were 25.0±4.8 m (min 8.2 m, max 32.9 m) and 25.3±5.2 m (min 5.9 m, max 34.0 m), respectively (no significant difference, p-value=0.049). Linear regression confirmed a strong relationship between the estimated and measured heights (y=k*x, R2 =0.99, k=0.98) (See Fig. 3A). Slightly larger differences in heights estimated by the two methods were found for birch and pine; for spruce, the differences were smaller (See Fig. 3B and Table 4). We believe that ground measurements of birch and pine height are less accurate than for spruce due to different crown shapes of these trees. So, our results suggested that UAV data can be used for tree stand attributes estimation, but automatically obtained data require validation.

Moving to Automated Tree Inventory: Comparison of UAS-Derived Lidar and Photogrammetric Data with Manual Ground Estimates

Unmanned aircraft systems (UAS) have advanced rapidly enabling low-cost capture of high-resolution images with cameras, from which three-dimensional photogrammetric point clouds can be derived. More recently UAS equipped with laser scanners, or lidar, have been employed to create similar 3D datasets. While airborne lidar (originally from conventional aircraft) has been used effectively in forest systems for many years, the ability to obtain important tree features such as height, diameter at breast height, and crown dimensions is now becoming feasible for individual trees at reasonable costs thanks to UAS lidar. Getting to individual tree resolution is crucial for detailed phenotyping and genetic analyses. This study evaluates the quality of three three-dimensional datasets from three sensors—two cameras of different quality and one lidar sensor—collected over a managed, closed-canopy pine stand with different planting densities. For reference, a ground-based timber cruise of the same pine stand is also collected. This study then conducted three straightforward experiments to determine the quality of the three sensors’ datasets for use in automated forest inventory: manual mensuration of the point clouds to (1) detect trees and (2) measure tree heights, and (3) automated individual tree detection. The results demonstrate that, while both photogrammetric and lidar data are well-suited for single-tree forest inventory, the photogrammetric data from the higher-quality camera is sufficient for individual tree detection and height determination, but that lidar data is best. The automated tree detection algorithm used in the study performed well with the lidar data, detecting 98% of the 2199 trees in the pine stand, but fell short of manual mensuration within the lidar point cloud, where 100% of the trees were detected. The manually-mensurated heights in the lidar dataset correlated with field measurements at r = 0.95 with a bias of −0.25 m, where the photogrammetric datasets were again less accurate and precise.

Individual tree detection and crown delineation from Unmanned Aircraft System (UAS) LiDAR in structurally complex mixed species eucalypt forests

Estimates of forest stocking density per hectare (NHa) are important in characterising ecological conditions and assessing changes in forest dynamics after disturbances due to pyrogenic, anthropogenic and biotic factors. We use Unmanned Aircraft Systems (UAS) LiDAR with mean point density of 1485 points m−2 across 39 flight sites to develop a bottom-up approach for individual tree and crown delineation (ITCD). The ITCD algorithm was evaluated across mixed species eucalypt forests (MSEF) using 2790 field measured stem locations across a broad range of dominant eucalypt species with randomly leaning trunks and highly irregular intertwined canopy structure. Two top performing ITCD algorithms in benchmarking studies resulted in poor performance when optimised to our plot data (mean Fscore: 0.61 and 0.62), which emphasises the challenge posed for ITCD in the structurally complex conditions of MSEF. To address this, our novel bottom-up ITCD algorithm uses kernel densities to stratify the vegetation profile and differentiate understorey from the rest of the vegetation. For vegetation above understorey, the ITCD algorithm adopted a novel watershed clustering procedure on point density measures within horizontal slices. A Principal Component Analysis (PCA) procedure was then applied to merge the slice-specific clusters into trunks, branches, and canopy clumps, before a voxel connectivity procedure clustered these biomass segments into overstorey trees. The segmentation process only requires two parameters to be calibrated to site-specific conditions across 39 MSEF sites using a Shuffled Complex Evolution (SCE) optimiser. Across the 39 field sites, the ITCD algorithm had mean Fscore of 0.91, True Positive (TP) trees represented 85% of measured trees and predicted plot-level stocking (NP) averaged 94% of actual stocking (NOb). As a representation of plot-level basal area (BA), TP trees represented 87% of BA, omitted trees represented slightly smaller trees and made up 8% of BA, and a further 5% of BA had commission error. Spatial maps of NHa using 0.5 m grid-cells showed that omitted trees were more prevalent in high density forest stands, and that 63% of grid-cells had a perfect estimate of NHa, whereas a further 31% of the grid-cells overestimate or underestimate one tree within the search window. The parsimonious modelling framework allows for the two calibrated site-specific parameters to be predicted (R2: 0.87 and 0.66) using structural characteristics of vegetation clusters within sites. Using predictions of these two site-specific parameters across all sites results in mean FScore of 0.86 and mean TP of 0.77, under circumstances where no ground observations were required for calibration. This approach generalises the algorithm across new UAS LiDAR data without undertaking time-consuming ground measurements within tall eucalypt forests with complex vegetation structure.

UAV Laser Scans Allow Detection of Morphological Changes in Tree Canopy

High-resolution laser scans from unmanned aerial vehicles (UAV) provide a highly detailed description of tree structure at the level of fine branches. Apart from ultrahigh spatial resolution, unmanned aerial laser scanning (ULS) can also provide high temporal resolution due to its operability and flexibility during data acquisition. We examined the phenomenon of bending branches of dead trees during one year from ULS multi-temporal data. In a multi-temporal series of three ULS datasets, we detected a synchronized reversible change in the inclination angles of the branches of 43 dead trees in a stand of blue spruce (Picea pungens Engelm.). The observed phenomenon has important consequences for both tree physiology and forest remote sensing (RS). First, the inclination angle of branches plays a crucial role in solar radiation interception and thus influences the total photosynthetic gain. The ability of a tree to change the branch position has important ecophysiological consequences, including better competitiveness across the site. Branch shifting in dead trees could be regarded as evidence of functional mycorrhizal interconnections via roots between live and dead trees. Second, we show that the detected movement results in a significant change in several point cloud metrics often utilized for deriving forest inventory parameters, both in the area-based approach (ABA) and individual tree detection approaches, which can affect the prediction of forest variables. To help quantify its impact, we used point cloud metrics of automatically segmented individual trees to build a generalized linear model to classify trees with and without the observed morphological changes. The model was applied to a validation set and correctly identified 86% of trees that displayed branch movement, as recorded by a human observer. The ULS allows for the study of this phenomenon across large areas, not only at individual tree levels.

Individual Tree Parameters Estimation for Chinese Fir (Cunninghamia lanceolate (Lamb.) Hook) Plantations of South China Using UAV Oblique Photography: Possibilities and Challenges

Chinese fir (Cunninghamia lanceolate (Lamb.) Hook) individual tree parameters extraction is important for scientific forest management. However, high-precision parameters extraction by field investigation or spaceborne optical remote sensing is difficult when the forest is dense and the terrain is complex. This article proposes a framework for extracting individual tree parameters by combining low-cost and high-efficiency unmanned aerial vehicle-based oblique photogrammetry with manned airborne light detection and ranging data and explores the influence of spatial resolution on the accuracy of parameter extraction. The variable window filtering (VWF) was used to detect an individual tree. The marker-controlled watershed segmentation (MCWS) and seed region growing algorithms were used to delineate the crown. The individual tree detection using VWF based on 1 m resolution achieves precision of 80%. For the crown delineation, it is more accurate based on the 0.25 m resolution using MCWS algorithm with the detection accuracy of 65%. The results show that the proposed framework can effectively detect the tree and delineate the crown under complex terrain conditions and the optimal resolution for different parameter extraction is determined, which has important guiding significance to determine the flight parameters and reduce unnecessary data processing.

Multi-objective forest planning at tree-level combining mixed integer programming and airborne laser scanning

The development of forest planning at tree-level is nowadays feasible due to the advances in laser scanning. Tree-level data remains still not fully utilized in the context of tree selection mechanisms and harvest-scheduling models. Using spatial optimization with tree-level data can shift the situation. The research presented a decision-support model set based on mixed integer programming (MIP) to select trees based on economical and spatial goals. Trees were represented as polygons to conduct spatial optimization and account for perimeter boundary minimization while establishing adjacency constraints. The proposed model framework method was tested in a Stone pine (Pinus pinea L.) stand of almost 4400 trees in Spain where individual tree detection with airborne laser scanning data was conducted. The MIP-based tree selections were presented using different priorities for single and multi-objective optimizations to demonstrate the potential of spatial goals to make forest management planning more efficient when controlling the spatial arrangement of solutions. The results showed the general MIP formulation is feasible and all presented combinatorial problems reached global optimality fast. The use of MIP formulations to schedule tree-level decisions has positive cascading effects towards operational forest planning while expanding the current frontiers of tree-level forest planning based on heuristic optimization.

Individual tree detection using template matching of multiple rasters derived from multispectral airborne laser scanning data

ABSTRACT Multispectral airborne laser scanning (MS-ALS) provides information about 3D structure as well as the intensity of the reflected light and is a promising technique for acquiring forest information. Data from MS-ALS have been used for tree species classification and tree health evaluation. This paper investigates its potential for individual tree detection (ITD) when using intensity as an additional metric. To this end, rasters of height, point density, vegetation ratio, and intensity at three wavelengths were used for template matching to detect individual trees. Optimal combinations of metrics were identified for ITD in plots with different levels of canopy complexity. The F-scores for detection by template matching ranged from 0.94 to 0.73, depending on the choice of template derivation and raster generalization methods. Using intensity and point density as metrics instead of height increased the F-scores by up to 14% for the plots with the most understorey trees.

Individual Tree Attribute Estimation and Uniformity Assessment in Fast-Growing Eucalyptus spp. Forest Plantations Using Lidar and Linear Mixed-Effects Models

Fast-growing Eucalyptus spp. forest plantations and their resultant wood products are economically important and may provide a low-cost means to sequester carbon for greenhouse gas reduction. The development of advanced and optimized frameworks for estimating forest plantation attributes from lidar remote sensing data combined with statistical modeling approaches is a step towards forest inventory operationalization and might improve industry efficiency in monitoring and managing forest resources. In this study, we first developed and tested a framework for modeling individual tree attributes in fast-growing Eucalyptus forest plantation using airborne lidar data and linear mixed-effect models (LME) and assessed the gain in accuracy compared to a conventional linear fixed-effects model (LFE). Second, we evaluated the potential of using the tree-level estimates for determining tree attribute uniformity across different stand ages. In the field, tree measurements, such as tree geolocation, species, genotype, age, height (Ht), and diameter at breast height (dbh) were collected through conventional forest inventory practices, and tree-level aboveground carbon (AGC) was estimated using allometric equations. Individual trees were detected and delineated from lidar-derived canopy height models (CHM), and crown-level metrics (e.g., crown volume and crown projected area) were computed from the lidar 3-D point cloud. Field and lidar-derived crown metrics were combined for ht, dbh, and AGC modeling using an LME. We fitted a varying intercept and slope model, setting species, genotype, and stand (alone and nested) as random effects. For comparison, we also modeled the same attributes using a conventional LFE model. The tree attribute estimates derived from the best LME model were used for assessing forest uniformity at the tree level using the Lorenz curves and Gini coefficient (GC). We successfully detected 96.6% of the trees from the lidar-derived CHM. The best LME model for estimating the tree attributes was composed of the stand as a random effect variable, and canopy height, crown volume, and crown projected area as fixed effects. The %RMSE values for tree-level height, dbh, and AGC were 8.9%, 12.1%, and 23.7% for the LFE model and improved to 7.3%, 7.1%, and 13.6%, respectively, for the LME model. Tree attributes uniformity was assessed with the Lorenz curves and tree-level estimations, especially for the older stands. All stands showed a high level of tree uniformity with GC values approximately 0.2. This study demonstrates that accurate detection of individual trees and their associated crown metrics can be used to estimate Ht, dbh, and AGC stocks as well as forest uniformity in fast-growing Eucalyptus plantations forests using lidar data as inputs to LME models. This further underscores the high potential of our proposed approach to monitor standing stock and growth in Eucalyptus—and similar forest plantations for carbon dynamics and forest product planning.

Assessment of an improved individual tree detection method based on local-maximum algorithm from unmanned aerial vehicle RGB imagery in overlapping canopy mountain forests

ABSTRACT Low consumer-grade cameras attached to small unmanned aerial vehicles (UAV) can easily acquire high spatial resolution images, leading to convenient forest monitoring at small-scales for forest managers. However, most studies were carried out in the low canopy density and flat ground plantations to detect individual trees. We selected overlapping canopy plantation in mountainous area in the eastern of China and acquired high spatial resolution UAV RGB images to detect individual trees. A total of 402 reference trees were located in three rectangle plots (900 m2). To enhance the confidence of the tested individual tree detection method, clear-cutting and Real-Time Kinematic (RTK) were used to obtain the truth values in the plots. A novel method for semi-automatic individual tree detection was proposed based on a local-maximum algorithm and UAV-derived DSM data (LAD) in this study. The detection accuracy of LAD was compared with commonly used methods based on UAV-derived orthophoto images, local-maximum algorithm (LAO), object-oriented feature segmentation (OFS), multiscale segmentation technique (MST) and manual visual interpretation (MVI). The overall accuracy (OA (%) decreased in the order of LAD (84.5%) > MST (69.1%) > OFS (65.1%) > MVI (64.1%) > LAO (59.1%). LAD had only 15.5%s omission errors (OM (%), which was less than half of the other four methods in comparison. It was noteworthy that MVI had 35.9% OM %, which revealed that MVI should be used carefully as the truth value. LAD showed similar repeated detection error (RP (%) and completely wrong detection (CW (%), while the other four methods had obviously higher CW % than the RP %. From our results, it can be concluded that the proposed LAD method may help improving the accuracy of individual tree detection to an acceptable accuracy (>80%) in dense mountain forests, and has practical advantages in future research direction to assess tree attributes from UAV RGB image.