Automatic Tree Detection Research Articles

Terrestrial laser scanning vs. manual methods for assessing complex forest stand structure: a comparative analysis on plenter forests

In continuous cover forestry, plenter silviculture is regarded as an elaborated system for optimizing the sustainable production of high-quality timber maintaining a constant but heterogeneous canopy. Its complexity necessitates high silvicultural expertise and a detailed assessment of forest stand structural variables. Terrestrial laser scanning (TLS) can offer reliable techniques for long-term tree mapping, volume calculation, and stand variables assessment in complex forest structures. We conducted surveys using both automated TLS and conventional manual methods (CMM) on two plots with contrasting silvicultural regimes within the Black Forest, Germany. Variations in automated tree detection and stand variables were greater between different TLS surveys than with CMM. TLS detected an average of 523 tree stems per hectare, while CMM counted 516. Approximately 9.6% of trees identified with TLS were commission errors, with 6.5% of CMM trees being omitted using TLS. Basal area per hectare was slightly higher in TLS (38.9 m3) than in CMM (38.2 m3). However, CMM recorded a greater standing volume (492.7 m3) than TLS (440.5 m3). The discrepancy in stand volume between methods was primarily due to TLS underestimating tree height, especially for taller trees. DBH bias was minor at 1 cm between methods. Repeated TLS inventories successfully matched an average of 424 tree positions per hectare. While TLS adequately characterizes complex plenter forest structures, we propose enhancing this methodology with personal laser scanning to optimize crown coverage and efficiency and direct volume measurements for increased accuracy of wood volume estimations. Additionally, utilizing 3D point cloud data-derived metrics, such as structural complexity indices, can further enhance plenter forest management.

Estimation of dendrometric characteristics in city parks according to data from UAVs and ground-based LiDAR

The possibilities of using ground-based LiDAR and processing of aerial photography data to assess the characteristics of individual trees in urban parks are shown. The actual data obtained during the examination of the object by these methods made it possible to analyze the state of the plantation and identify dendro parameters. It has been established that the methods of automatic search for trees using 3D point clouds can be successfully used in artificial plantations. The UAV data processing method made it possible to correctly detect in automatic mode about 64% of the park’s trees (343 trees), while the number of false positives and the number of missed trees was quite high (111 and 195 trees, respectively), which was associated with a large proportion deciduous trees in the park. The weighted average of the quality of automatic tree detection in the park was 0.69. Ground-based LiDAR data in manual mode made it possible to detect all trees, determine their heights, crown diameter and trunk diameter at a height of 1.3 m, as well as identify tree species and condition category (using panoramic images). To increase the correctness of tree detection by 3D point clouds, the methods used need to be improved.

Benchmarking laser scanning and terrestrial photogrammetry to extract forest inventory parameters in a complex temperate forest

National forest inventories (NFI) are important for the assessment of the state and development of forests. Traditional NFIs often rely on statistical sampling approaches as well as expert assessment which may suffer from observer bias and may lack robustness for time series analysis. Over the course of the last decade, close-range remote sensing techniques such as terrestrial and mobile laser scanning became ever more established for the assessment of three-dimensional (3D) forest structure. With the ongoing trend to make the systems smaller, easier to use and more efficient, the pathway is being opened for an operational inclusion of such devices within the framework of an NFI to support the traditional field assessment. Close-range remote sensing could potentially speed up field inventory work as well as increase the area in which certain parameters are assessed. Benchmarks are needed to evaluate the performance of different close-range remote sensing devices and approaches, both in terms of efficiency as well as accuracy. In this study we evaluate the performance of two terrestrial (TLS), one handheld mobile (PLS) and two drone based (UAVLS) laser scanning systems to detect trees and extract the diameter at breast height (DBH) in three plots with a steep gradient in tree and understorey vegetation density. As a novelty, we also tested the acquisition of 3D point-clouds using a low-cost action camera (GoPro) in conjunction with the Structure from Motion (SfM) technique and compared its performance with those of the more costly LiDAR devices. Among the many parameters evaluated in traditional NFIs, the focus of the performance evaluation of this study is set on the automatic tree detection and DBH extraction. The results showed that TLS delivers the highest tree detection rate (TDR) of up to 94.6% under leaf-off and up to 82% under leaf-on conditions and a relative RMSE (rRMSE) for the DBH extraction between 2.5 and 9%, depending on the undergrowth complexity. The tested PLS system (leaf-on) achieved a TDR of up to 80% with an rRMSE between 3.7 and 5.8%. The tested UAVLS systems showed lowest TDR of less than 77% under leaf-off and less than 37% under leaf-on conditions. The novel GoPro approach achieved a TDR of up to 53% under leaf-on conditions. The reduced TDR can be explained by the reduced area coverage due to the chosen circular acquisition path taken with the GoPro approach. The DBH extraction performance on the other hand is comparable to those of the LiDAR devices with an rRMSE between 2 and 9%. Further benchmarks are needed in order to fully assess the applicability of these systems in the framework of an NFI. Especially the robustness under varying forest conditions (seasonality) and over a broader range of forest types and canopy structure has to be evaluated.

UAV-Based Characterization of Tree-Attributes and Multispectral Indices in an Uneven-Aged Mixed Conifer-Broadleaf Forest

Unmanned aerial vehicles (UAVs) have contributed considerably to forest monitoring. However, gaps in the knowledge still remain, particularly for natural forests. Species diversity, stand heterogeneity, and the irregular spatial arrangement of trees provide unique opportunities to improve our perspective of forest stands and the ecological processes that occur therein. In this study, we calculated individual tree metrics, including several multispectral indices, in order to discern the spectral reflectance of a natural stand as a pioneer area in Mexican forests. Using data obtained by UAV DJI 4, and in the free software environments OpenDroneMap and QGIS, we calculated tree height, crown area, number of trees and multispectral indices. Digital photogrammetric procedures, such as the ForestTools, Structure from Motion and Multi-View Stereo algorithms, yielded results that improved stand mapping and the estimation of stand attributes. Automated tree detection and quantification were limited by the presence of overlapping crowns but compensated by the novel stand density mapping and estimates of crown attributes. Height estimation was in line with expectations (R2 = 0.91, RMSE = 0.36) and is therefore a useful parameter with which to complement forest inventories. The diverse spectral indices applied yielded differential results regarding the potential vegetation activity present and were found to be complementary to each other. However, seasonal monitoring and careful estimation of photosynthetic activity are recommended in order to determine the seasonality of plant response. This research contributes to the monitoring of natural forest stands and, coupled with accurate in situ measurements, could refine forest productivity parameters as a strategy for the validity of results. The metrics are reliable and rapid and could serve as model inputs in modern inventories. Nevertheless, increased efforts in the configuration of new technologies and algorithms are required, including full consideration of the costs implied by their adoption.

Comparison of Classical Methods and Mask R-CNN for Automatic Tree Detection and Mapping Using UAV Imagery

Detecting and mapping individual trees accurately and automatically from remote sensing images is of great significance for precision forest management. Many algorithms, including classical methods and deep learning techniques, have been developed and applied for tree crown detection from remote sensing images. However, few studies have evaluated the accuracy of different individual tree detection (ITD) algorithms and their data and processing requirements. This study explored the accuracy of ITD using local maxima (LM) algorithm, marker-controlled watershed segmentation (MCWS), and Mask Region-based Convolutional Neural Networks (Mask R-CNN) in a young plantation forest with different test images. Manually delineated tree crowns from UAV imagery were used for accuracy assessment of the three methods, followed by an evaluation of the data processing and application requirements for three methods to detect individual trees. Overall, Mask R-CNN can best use the information in multi-band input images for detecting individual trees. The results showed that the Mask R-CNN model with the multi-band combination produced higher accuracy than the model with a single-band image, and the RGB band combination achieved the highest accuracy for ITD (F1 score = 94.68%). Moreover, the Mask R-CNN models with multi-band images are capable of providing higher accuracies for ITD than the LM and MCWS algorithms. The LM algorithm and MCWS algorithm also achieved promising accuracies for ITD when the canopy height model (CHM) was used as the test image (F1 score = 87.86% for LM algorithm, F1 score = 85.92% for MCWS algorithm). The LM and MCWS algorithms are easy to use and lower computer computational requirements, but they are unable to identify tree species and are limited by algorithm parameters, which need to be adjusted for each classification. It is highlighted that the application of deep learning with its end-to-end-learning approach is very efficient and capable of deriving the information from multi-layer images, but an additional training set is needed for model training, robust computer resources are required, and a large number of accurate training samples are necessary. This study provides valuable information for forestry practitioners to select an optimal approach for detecting individual trees.

Mathematically optimized trajectory for terrestrial close-range photogrammetric 3D reconstruction of forest stands

Terrestrial close-range photogrammetry offers a low-cost method of three-dimensional (3D) reconstruction of forest stands that provides automatically processable 3D data that can be used to evaluate inventory parameters of forest stands and individual trees. However, fundamental methodological problems in image acquisition and processing remain. This study enhances the methodology of photogrammetric Structure from Motion reconstruction of forest stands by determining the best photographer's trajectory for image acquisition. The study comprises 1) mathematical optimization of the route in a square grid using integer programming, 2) evaluation of point clouds derived from sequences of real photographs, simulating different trajectories, and 3) verification on real trajectories. In a forest research plot, we established a 1 m square grid of 625 (i.e., 25 × 25) photographic positions, and at each position, we captured 16 photographs in uniformly spaced directions. We adopted real tree positions and diameters, and the coordinates of the photographic positions, including orientation angles of captured images, were recorded. We then formulated an integer programming optimization model to find the most efficient trajectory that provided coverage of all sides of all trees with sufficient counts of images. Subsequently, we used the 10,000 captured images to produce image subsets simulating image sequences acquired during the photographer's movement along 84 different systematic trajectories of seven patterns based on either parallel lines or concentric orbits. 3D point clouds derived from the simulated image sequences were evaluated for their suitability for automatic tree detection and estimation of diameters at breast height. The results of the integer programming model indicated that the optimal trajectory consisted of parallel line segments if the camera is pointed forward – in the travel direction, or concentric orbits if the camera is pointed to a side – perpendicular to the travel direction. With point clouds derived from the images of the simulated trajectories, the best diameter estimates on automatically detected trees were achieved with trajectories consisting of parallel lines in two perpendicular directions where each line was passed in both opposite directions. For efficient image acquisition, resulting in point clouds of reasonable quality with low counts of images, a trajectory consisting of concentric orbits, including the plot perimeter with the camera pointed towards the plot center, proved to be the best. Results of simulated trajectories were verified with the photogrammetric reconstruction of the forest stand based on real trajectories for six patterns. The mathematical optimization was consistent with the results of the experiment, which indicated that mathematical optimization may represent a valid tool for planning trajectories for photogrammetric 3D reconstruction of scenes in general.

Three-dimensional tree monitoring in urban cities using automatic tree detection method with mobile LiDAR data

Three-dimensional tree monitoring in urban cities using automatic tree detection method with mobile LiDAR data

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.

Comparison of 3D Point Clouds Obtained by Terrestrial Laser Scanning and Personal Laser Scanning on Forest Inventory Sample Plots

In forest inventory, trees are usually measured using handheld instruments; among the most relevant are calipers, inclinometers, ultrasonic devices, and laser range finders. Traditional forest inventory has been redesigned since modern laser scanner technology became available. Laser scanners generate massive data in the form of 3D point clouds. We have developed a novel methodology to provide estimates of the tree positions, stem diameters, and tree heights from these 3D point clouds. This dataset was made publicly accessible to test new software routines for the automatic measurement of forest trees using laser scanner data. Benchmark studies with performance tests of different algorithms are welcome. The dataset contains co-registered raw 3D point-cloud data collected on 20 forest inventory sample plots in Austria. The data were collected by two different laser scanning systems: (1) A mobile personal laser scanner (PLS) (ZEB Horizon, GeoSLAM Ltd., Nottingham, UK) and (2) a static terrestrial laser scanner (TLS) (Focus3D X330, Faro Technologies Inc., Lake Mary, FL, USA). The data also contain digital terrain models (DTMs), field measurements as reference data (ground-truth), and the output of recent software routines for the automatic tree detection and the automatic stem diameter measurement.

3D MAP SYSTEM FOR TREE MONITORING IN HONG KONG USING GOOGLE STREET VIEW IMAGERY AND DEEP LEARNING

Abstract. In densely built urban areas such as Hong Kong, the positive effect of urban trees is to help maintain high environmental and social sustainability for the city while unmanaged trees lead to negative effects such as accidents, outbreaks of pests and diseases. The public awareness of urban tree population has been increasing and preserving all the benefits offered by trees, a continuous monitoring concept would be required. In this work, an efficient 3D map system for tree inventory in Hong Kong is presented to the based on automated tree detection from publicly available Google street view (GSV) panorama images. First, Convolutional Neural Networks (CNNs) based object detector and classifier – YOLOv3 with pretrained model is adopted to learn GSV images to detect tree objects. GSV depth image has been utilized to decode depth values of each GSV panorama image and will provide accurate information to calculate the tree geographic position. A “field of view” filter was designed to remove duplicated tree detection within the overlapped areas followed by spatial clustering applied to further increase the tree localization accuracy. The average distance between the detected trees and ground truth data was achieved within 3 meters for selected roads used for the experiment. Second, a 3D Map platform prototype for facilitating the urban tree monitoring and management was developed. Currently, there is no true 3D platform for interpreting the results of tree records in Hong Kong city areas. With the help of webGL technology, contemporary browsers are able to show 3D buildings, terrain and other scene components together with the obtained tree records in an open source 3D GIS platform, the level of visualization is enhanced as all the detected trees are placed on the 3D digital terrain model. Consequently, it is easy for end-users to know the actual position of the trees and their distribution.

Detection, Segmentation, and Model Fitting of Individual Tree Stems from Airborne Laser Scanning of Forests Using Deep Learning

Accurate measurements of the structural characteristics of trees such as height, diameter, sweep and taper are an important part of forest inventories in managed forests and commercial plantations. Both terrestrial and aerial LiDAR are currently employed to produce pointcloud data from which inventory metrics can be determined. Terrestrial/ground-based scanning typically provides pointclouds resolutions of many thousands of points per m 2 from which tree stems can be observed and inventory measurements made directly, whereas typical resolutions from aerial scanning (tens of points per m 2 ) require inventory metrics to be regressed from LiDAR variables using inventory reference data collected from the ground. Recent developments in miniaturised LiDAR sensors are enabling aerial capture of pointclouds from low-flying aircraft at high-resolutions (hundreds of points per m 2 ) from which tree stem information starts to become directly visible, enabling the possibility for plot-scale inventories that do not require access to the ground. In this paper, we develop new approaches to automated tree detection, segmentation and stem reconstruction using algorithms based on deep supervised machine learning which are designed for use with aerially acquired high-resolution LiDAR pointclouds. Our approach is able to isolate individual trees, determine tree stem points and further build a segmented model of the main tree stem that encompasses tree height, diameter, taper, and sweep. Through the use of deep learning models, our approach is able to adapt to variations in pointcloud densities and partial occlusions that are particularly prevalent when data is captured from the air. We present results of our algorithms using high-resolution LiDAR pointclouds captured from a helicopter over two Radiata pine forests in NSW, Australia.

Automatic Tree Detection from Three-Dimensional Images Reconstructed from 360° Spherical Camera Using YOLO v2

It is important to grasp the number and location of trees, and measure tree structure attributes, such as tree trunk diameter and height. The accurate measurement of these parameters will lead to efficient forest resource utilization, maintenance of trees in urban cities, and feasible afforestation planning in the future. Recently, light detection and ranging (LiDAR) has been receiving considerable attention, compared with conventional manual measurement techniques. However, it is difficult to use LiDAR for widespread applications, mainly because of the costs. We propose a method for tree measurement using 360° spherical cameras, which takes omnidirectional images. For the structural measurement, the three-dimensional (3D) images were reconstructed using a photogrammetric approach called structure from motion. Moreover, an automatic tree detection method from the 3D images was presented. First, the trees included in the 360° spherical images were detected using YOLO v2. Then, these trees were detected with the tree information obtained from the 3D images reconstructed using structure from motion algorithm. As a result, the trunk diameter and height could be accurately estimated from the 3D images. The tree detection model had an F-measure value of 0.94. This method could automatically estimate some of the structural parameters of trees and contribute to more efficient tree measurement.

Detecting Trees in Street Images via Deep Learning With Attention Module

Although object detection techniques have been widely employed in various practical applications, automatic tree detection is still a difficult challenge, especially for street-view images. In this article, we propose a unified end-to-end trainable network for automatic street tree detection based on a state-of-the-art deep learning-based object detector. We tackle low illumination and heavy occlusion conditions in tree detection, which have not been extensively studied until now, due to clear challenges. Existing generic object detectors cannot be directly applied to this task due to aforementioned challenges. To address these issues, we first present a simple, yet effective image brightness adjustment method to handle low illuminance cases. Moreover, we propose a novel loss and a tree part-attention module to reduce false detections caused by heavy occlusion, inspired by the previously proposed occlusion-aware region-convolutional neural network (R-CNN) work. We train and evaluate several versions of the proposed network and validate the importance of each component. It is demonstrated that the resulting framework, part attention network for tree detection (PANTD), can efficiently detect trees in street-view images. The experimental results show that our approach achieves high accuracy and robustness under various conditions.

Automated tree detection from 3D lidar images using image processing and machine learning.

Trees in 3D images obtained from lidar were automatically extracted in the presence of other objects that were not trees. We proposed a method combining 3D image processing and machine learning techniques for this automatic detection. Consequently, tree detection could be done with 95% accuracy. First, the objects in the 3D images were segmented one by one; then, each of the segmented objects was projected onto 2D images. Finally, the 2D image was classified into "tree" and "not tree" using a one-class support vector machine, and trees in the 3D image were successfully extracted.

Terrestrial Structure from Motion Photogrammetry for Deriving Forest Inventory Data

The measurements of tree attributes required for forest monitoring and management planning, e.g., National Forest Inventories, are derived by rather time-consuming field measurements on sample plots, using calipers and measurement tapes. Therefore, forest managers and researchers are looking for alternative methods. Currently, terrestrial laser scanning (TLS) is the remote sensing method that provides the most accurate point clouds at the plot-level to derive these attributes from. However, the demand for even more efficient and effective solutions triggers further developments to lower the acquisition time, costs, and the expertise needed to acquire and process 3D point clouds, while maintaining the quality of extracted tree parameters. In this context, photogrammetry is considered a potential solution. Despite a variety of studies, much uncertainty still exists about the quality of photogrammetry-based methods for deriving plot-level forest attributes in natural forests. Therefore, the overall goal of this study is to evaluate the competitiveness of terrestrial photogrammetry based on structure from motion (SfM) and dense image matching for deriving tree positions, diameters at breast height (DBHs), and stem curves of forest plots by means of a consumer grade camera. We define an image capture method and we assess the accuracy of the photogrammetric results on four forest plots located in Austria and Slovakia, two in each country, selected to cover a wide range of conditions such as terrain slope, undergrowth vegetation, and tree density, age, and species. For each forest plot, the reference data of the forest parameters were obtained by conducting field surveys and TLS measurements almost simultaneously with the photogrammetric acquisitions. The TLS data were also used to estimate the accuracy of the photogrammetric ground height, which is a necessary product to derive DBHs and tree heights. For each plot, we automatically derived tree counts, tree positions, DBHs, and part of the stem curve from both TLS and SfM using a software developed at TU Wien (Forest Analysis and Inventory Tool, FAIT), and the results were compared. The images were oriented with errors of a few millimetres only, according to checkpoint residuals. The automatic tree detection rate for the SfM reconstruction ranges between 65% and 98%, where the missing trees have average DBHs of less than 12 cm. For each plot, the mean error of SfM and TLS DBH estimates is −1.13 cm and −0.77 cm with respect to the caliper measurements. The resulting stem curves show that the mean differences between SfM and TLS stem diameters is at maximum −2.45 cm up to 3 m above ground, which increases to almost +4 cm for higher elevations. This study shows that with the adopted image capture method, terrestrial SfM photogrammetry, is an accurate solution to support forest inventory for estimating the number of trees and their location, the DBHs and stem curve up to 3 m above ground.

Comparing Terrestrial Laser Scanning (TLS) and Wearable Laser Scanning (WLS) for Individual Tree Modeling at Plot Level

This study presents a comparison between the use of wearable laser scanning (WLS) and terrestrial laser scanning (TLS) devices for automatic tree detection with an estimation of two dendrometric variables: diameter at breast height (DBH) and total tree height (TH). Operative processes for data collection and automatic forest inventory are described in detail. The approach used is based on the clustering of points belonging to each individual tree, the isolation of the trunks, the iterative fitting of circles for the DBH calculation and the computation of the TH of each tree. TLS and WLS point clouds were compared by the statistical analysis of both estimated forest dendrometric parameters and the possible presence of bias. Results show that the apparent differences in point density and relative precision between both 3D forest models do not affect tree detection and DBH estimation. Nevertheless, tree height estimation using WLS appears to be affected by the limited scanning range of the WLS used in this study. TH estimations for trees below a certain height are equivalent using WLS or TLS, whereas TH of taller trees is clearly underestimated using WLS.

From Google Maps to a fine-grained catalog of street trees

Up-to-date catalogs of the urban tree population are of importance for municipalities to monitor and improve quality of life in cities. Despite much research on automation of tree mapping, mainly relying on dedicated airborne LiDAR or hyperspectral campaigns, tree detection and species recognition is still mostly done manually in practice. We present a fully automated tree detection and species recognition pipeline that can process thousands of trees within a few hours using publicly available aerial and street view images of Google MapsTM. These data provide rich information from different viewpoints and at different scales from global tree shapes to bark textures. Our work-flow is built around a supervised classification that automatically learns the most discriminative features from thousands of trees and corresponding, publicly available tree inventory data. In addition, we introduce a change tracker that recognizes changes of individual trees at city-scale, which is essential to keep an urban tree inventory up-to-date. The system takes street-level images of the same tree location at two different times and classifies the type of change (e.g., tree has been removed). Drawing on recent advances in computer vision and machine learning, we apply convolutional neural networks (CNN) for all classification tasks. We propose the following pipeline: download all available panoramas and overhead images of an area of interest, detect trees per image and combine multi-view detections in a probabilistic framework, adding prior knowledge; recognize fine-grained species of detected trees. In a later, separate module, track trees over time, detect significant changes and classify the type of change. We believe this is the first work to exploit publicly available image data for city-scale street tree detection, species recognition and change tracking, exhaustively over several square kilometers, respectively many thousands of trees. Experiments in the city of Pasadena, California, USA show that we can detect >70% of the street trees, assign correct species to >80% for 40 different species, and correctly detect and classify changes in >90% of the cases.

Automated tree detection and crown delineation using airborne laser scanner data in heterogeneous East-Central Europe forest with different species mix

Identifying tree locations is a basic step in the derivation of other tree parameters using remote sensing techniques, particularly when using airborne laser scanning. There are several techniques for identifying tree positions. In this paper, we present a raster-based method for determining tree position and delineating crown coverage. We collected data from nine research plots that supported different mixes of species. We applied a raster-based method to raster layers with six different spatial resolutions and used terrestrial measurement data as reference data. Tree identification at a spatial resolution of 1.5 m was demonstrated to be the most accurate, with an average identification ratio (IR) of 95% and average detection ratio of 68% being observed. At a higher spatial resolution of 0.5 m, IR was overestimated by more than 600%. At a lower spatial resolution of 3 m, IR was underestimated at less than 44% of terrestrial measurements. The inventory process was timed to enable evaluation of the time efficiency of automatic methods.

An International Comparison of Individual Tree Detection and Extraction Using Airborne Laser Scanning

The objective of the “Tree Extraction” project organized by EuroSDR (European Spatial data Research) and ISPRS (International Society of Photogrammetry and Remote Sensing) was to evaluate the quality, accuracy, and feasibility of automatic tree extraction methods, mainly based on laser scanner data. In the final report of the project, Kaartinen and Hyyppä (2008) reported a high variation in the quality of the published methods under boreal forest conditions and with varying laser point densities. This paper summarizes the findings beyond the final report after analyzing the results obtained in different tree height classes. Omission/Commission statistics as well as neighborhood relations are taken into account. Additionally, four automatic tree detection and extraction techniques were added to the test. Several methods in this experiment were superior to manual processing in the dominant, co-dominant and suppressed tree storeys. In general, as expected, the taller the tree, the better the location accuracy. The accuracy of tree height, after removing gross errors, was better than 0.5 m in all tree height classes with the best methods investigated in this experiment. For forest inventory, minimum curvature-based tree detection accompanied by point cloud-based cluster detection for suppressed trees is a solution that deserves attention in the future.

An application for tree detection using satellite imagery and vegetation data

Virtual reconstruction of large landscapes from satellite imagery can be a time-consuming task due to the number of objects that must be extracted. Poor image resolution and noise hinder automatic detection processes and thus must be corrected by the user. This paper describes an application that allows the user to guide the automatic detection of trees from satellite imagery and spatial vegetation data. The requirements of the system are specified and an architecture that satisfies these constraints is presented. The resulting application provides an intuitive computer-aided method for the selection and classification of trees.