Explore the vast range of titles available.

Tropical forests concentrate the largest diversity of species on the planet and play a key role in maintaining environmental processes. Due to the importance of those forests, there is growing interest in mapping their components and getting information at an individual tree level to conduct reliable satellite-based forest inventory for biomass and species distribution qualification. Individual tree crown information could be manually gathered from high resolution satellite images; however, to achieve this task at large-scale, an algorithm to identify and delineate each tree crown individually, with high accuracy, is a prerequisite. In this study, we propose the application of a convolutional neural network—Mask R-CNN algorithm—to perform the tree crown detection and delineation. The algorithm uses very high-resolution satellite images from tropical forests. The results obtained are promising—the R e c a l l , P r e c i s i o n , and F 1 score values obtained were were 0.81 , 0.91 , and 0.86 , respectively. In the study site, the total of tree crowns delineated was 59,062 . These results suggest that this algorithm can be used to assist the planning and conduction of forest inventories. As the algorithm is based on a Deep Learning approach, it can be systematically trained and used for other regions.

The automatic detection of tree crowns and estimation of crown areas from remotely sensed information offer a quick approach for grasping the dynamics of forest ecosystems and are of great significance for both biodiversity and ecosystem conservation. Among various types of remote sensing data, unmanned aerial vehicle (UAV)-acquired RGB imagery has been increasingly used for tree crown detection and crown area estimation; the method has efficient advantages and relies heavily on deep learning models. However, the approach has not been thoroughly investigated in deciduous forests with complex crown structures. In this study, we evaluated two widely used, deep-learning-based tree crown detection and delineation approaches (DeepForest and Detectree2) to assess their potential for detecting tree crowns from UAV-acquired RGB imagery in an alpine, temperate deciduous forest with a complicated species composition. A total of 499 digitized crowns, including four dominant species, with corresponding, accurate inventory data in a 1.5 ha study plot were treated as training and validation datasets. We attempted to identify an effective model to delineate tree crowns and to explore the effects of the spatial resolution on the detection performance, as well as the extracted tree crown areas, with a detailed field inventory. The results show that the two deep-learning-based models, of which Detectree2 (F1 score: 0.57) outperformed DeepForest (F1 score: 0.52), could both be transferred to predict tree crowns successfully. However, the spatial resolution had an obvious effect on the estimation accuracy of tree crown detection, especially when the resolution was greater than 0.1 m. Furthermore, Dectree2 could estimate tree crown areas accurately, highlighting its potential and robustness for tree detection and delineation. In addition, the performance of tree crown detection varied among different species. These results indicate that the evaluated approaches could efficiently delineate individual tree crowns in high-resolution optical images, while demonstrating the applicability of Detectree2, and, thus, have the potential to offer transferable strategies that can be applied to other forest ecosystems.

Tree species surveys are crucial to forest resource management and can provide references for forest protection policy making. The traditional tree species survey in the field is labor-intensive and time-consuming, supporting the practical significance of remote sensing. The availability of high-resolution satellite remote sensing data enable individual tree species (ITS) recognition at low cost. In this study, the potential of the combination of such images and a convolutional neural network (CNN) to recognize ITS was explored. Firstly, individual tree crowns were delineated from a high-spatial resolution WorldView-3 (WV3) image and manually labeled as different tree species. Next, a dataset of the image subsets of the labeled individual tree crowns was built, and several CNN models were trained based on the dataset for ITS recognition. The models were then applied to the WV3 image. The results show that the distribution maps of six ITS offered an overall accuracy of 82.7% and a kappa coefficient of 0.79 based on the modified GoogLeNet, which used the multi-scale convolution kernel to extract features of the tree crown samples and was modified for small-scale samples. The ITS recognition method proposed in this study, with multi-scale individual tree crown delineation, avoids artificial tree crown delineation. Compared with the random forest (RF) and support vector machine (SVM) approaches, this method can automatically extract features and outperform RF and SVM in the classification of six tree species.

The ability to automatically delineate individual tree crowns using remote sensing data opens the possibility to collect detailed tree information over large geographic regions. While individual tree crown delineation (ITCD) methods have proven successful in conifer-dominated forests using Light Detection and Ranging (LiDAR) data, it remains unclear how well these methods can be applied in deciduous broadleaf-dominated forests. We applied five automated LiDAR-based ITCD methods across fifteen plots ranging from conifer- to broadleaf-dominated forest stands at Harvard Forest in Petersham, MA, USA, and assessed accuracy against manual delineation of crowns from unmanned aerial vehicle (UAV) imagery. We then identified tree- and plot-level factors influencing the success of automated delineation techniques. There was relatively little difference in accuracy between automated crown delineation methods (51–59% aggregated plot accuracy) and, despite parameter tuning, none of the methods produced high accuracy across all plots (27—90% range in plot-level accuracy). The accuracy of all methods was significantly higher with increased plot conifer fraction, and individual conifer trees were identified with higher accuracy (mean 64%) than broadleaf trees (42%) across methods. Further, while tree-level factors (e.g., diameter at breast height, height and crown area) strongly influenced the success of crown delineations, the influence of plot-level factors varied. The most important plot-level factor was species evenness, a metric of relative species abundance that is related to both conifer fraction and the degree to which trees can fill canopy space. As species evenness decreased (e.g., high conifer fraction and less efficient filling of canopy space), the probability of successful delineation increased. Overall, our work suggests that the tested LiDAR-based ITCD methods perform equally well in a mixed temperate forest, but that delineation success is driven by forest characteristics like functional group, tree size, diversity, and crown architecture. While LiDAR-based ITCD methods are well suited for stands with distinct canopy structure, we suggest that future work explore the integration of phenology and spectral characteristics with existing LiDAR as an approach to improve crown delineation in broadleaf-dominated stands.

Tropical forests are a major component of the global carbon cycle and home to two-thirds of terrestrial species. Upper-canopy trees store the majority of forest carbon and can be vulnerable to drought events and storms. Monitoring their growth and mortality is essential to understanding forest resilience to climate change, but in the context of forest carbon storage, large trees are underrepresented in traditional field surveys, so estimates are poorly constrained. Aerial photographs provide spectral and textural information to discriminate between tree crowns in diverse, complex tropical canopies, potentially opening the door to landscape monitoring of large trees. Here we describe a new deep convolutional neural network method, Detectree2, which builds on the Mask R-CNN computer vision framework to recognize the irregular edges of individual tree crowns from airborne RGB imagery. We trained and evaluated this model with 3797 manually delineated tree crowns at three sites in Malaysian Borneo and one site in French Guiana. As an example application, we combined the delineations with repeat lidar surveys (taken between 3 and 6 years apart) of the four sites to estimate the growth and mortality of upper-canopy trees. Detectree2 delineated 65 000 upper-canopy trees across 14 km2 of aerial images. The skill of the automatic method in delineating unseen test trees was good (F1 score = 0.64) and for the tallest category of trees was excellent (F1 score = 0.74). As predicted from previous field studies, we found that growth rate declined with tree height and tall trees had higher mortality rates than intermediate-size trees. Our approach demonstrates that deep learning methods can automatically segment trees in widely accessible RGB imagery. This tool (provided as an open-source Python package) has many potential applications in forest ecology and conservation, from estimating carbon stocks to monitoring forest phenology and restoration.Python package available to install at https://github.com/PatBall1/ Detectree2.

Mapping forest types and tree species at regional scales to provide information for ecologists and forest managers is a new challenge for the remote sensing community. Here, we assess the potential of a U‐net convolutional network, a recent deep learning algorithm, to identify and segment (1) natural forests and eucalyptus plantations, and (2) an indicator of forest disturbance, the tree species Cecropia hololeuca, in very high resolution images (0.3 m) from the WorldView‐3 satellite in the Brazilian Atlantic rainforest region. The networks for forest types and Cecropia trees were trained with 7611 and 1568 red‐green‐blue (RGB) images, respectively, and their dense labeled masks. Eighty per cent of the images were used for training and 20% for validation. The U‐net network segmented forest types with an overall accuracy >95% and an intersection over union (IoU) of 0.96. For C. hololeuca, the overall accuracy was 97% and the IoU was 0.86. The predictions were produced over a 1600 km2 region using WorldView‐3 RGB bands pan‐sharpened at 0.3 m. Natural and eucalyptus forests compose 79 and 21% of the region's total forest cover (82 250 ha). Cecropia crowns covered 1% of the natural forest canopy. An index to describe the level of disturbance of the natural forest fragments based on the spatial distribution of Cecropia trees was developed. Our work demonstrates how a deep learning algorithm can support applications such as vegetation, tree species distributions and disturbance mapping on a regional scale. In this paper, we have assessed the potential of a deep learning algorithm, the U‐net model, to identify and segment (1) natural forest and eucalyptus plantation, and (2) a tree species indicator of past forest disturbance (Cecropia hololeuca) using Red‐Green‐Blue WorldView‐3 images at 0.3 m of spatial resolution. The overall accuracies of both forest types and C. hololeuca segmentations were above 95%. The method was therefore used to map forest types and all individuals of C. hololeuca in a 1600 km² region of fragmented Atlantic Forest near São Paulo, Brazil. From the C. hololeuca occurrence and distribution in the fragments, we derived a new disturbance metric. Our results show that this method is very promising for applications such as tree species or vegetation mapping.

For the accurate and automatic mapping of forest stands based on very-high-resolution satellite imagery and digital orthophotos, precise object detection at the individual tree level is necessary. Currently, supervised deep learning models are primarily applied for this task. To train a reliable model, it is crucial to have an accurate tree crown annotation dataset. The current method of generating these training datasets still relies on manual annotation and labeling. Because of the intricate contours of tree crowns, vegetation density in natural forests and the insufficient ground sampling distance of the imagery, manually generated annotations are error-prone. It is unlikely that the manually delineated tree crowns represent the true conditions on the ground. If these error-prone annotations are used as training data for deep learning models, this may lead to inaccurate mapping results for the models. This study critically validates manual tree crown annotations on two study sites: a forest-like plantation on a cemetery and a natural city forest. The validation is based on tree reference data in the form of an official tree register and tree segments extracted from UAV laser scanning (ULS) data for the quality assessment of a training dataset. The validation results reveal that the manual annotations detect only 37% of the tree crowns in the forest-like plantation area and 10% of the tree crowns in the natural forest correctly. Furthermore, it is frequent for multiple trees to be interpreted in the annotation as a single tree at both study sites.

Deep learning models in the field of individual tree detection and crown delineation (ITDCD) rely on large and high-quality annotation datasets to produce accurate predictions. Training data or annotations for most ITDCD studies are collected through manual labeling. Manual labeling, especially for complex structures like tree crowns, is a time-consuming process that often results in error-prone annotations. Error-prone annotations, in turn, can lead to significant errors in the predictions of deep learning models. Semi- or fully-automated training data creation shows the potential to make the creation process more efficient and ensure high quality of the training dataset. In this work, we present a methodology for generating semi-automated synthetic training data for deep learning-based ITDCD applications. Furthermore, a systematic criteria-based - validity, efficiency, variety and scalability - comparison is conducted between the manual and synthetic training data creation methods to structurally and practically illustrate the advantages and disadvantages of the two approaches. Overall, the semi-automated synthetic data approach outperforms manual labeling in terms of validity, efficiency, and scalability; once the algorithm is implemented, it rapidly generates arbitrarily large, high-quality, reproducible tree crown annotation datasets. In contrast, a manual creation approach shows its advantages as an efficient way to create small, low-quality datasets (e.g., for fine-tuning a pre-trained model) compared to developing a semi-automated method from scratch.

In current individual tree crown (ITC) delineation methods for high-resolution multispectral imagery, either a spectral band or a brightness component of the multispectral image is employed in delineation with reference to edges or shapes of crowns, whereas spectra of tree crowns are seldom taken into account. Such methods normally perform well in coniferous forests with obvious between-crown shadows, but fail in dense deciduous or mixed forests, in which tree crowns are close to each other, between-crown shadows and boundaries are unobvious, whereas adjacent tree crowns may be of distinguishable spectra. In order to effectively delineate crowns in dense deciduous or mixed forests, a new ITC delineation method using both brightness and spectra of the image is proposed in this study. In this method, a morphological gradient map of the image is first generated, treetops of multi-scale crowns are extracted from the gradient map and refined regarding the spectral differences between neighboring crowns, the gradient map is segmented using a watershed approach with treetops as markers, and the resulting segmentation map is refined to yield a crown map. Evaluated on images of a rainforest and a deciduous forest, the proposed method more accurately delineated adjacent broad-leaved tree crowns with similar brightness but different spectra than the other two typical ITC delineation algorithms, achieving a delineation accuracy of up to 76% in the rainforest and 63% in the deciduous forest.

Multispectral imaging using unmanned aerial systems (UAS) enables rapid and accurate detection of pest insect infestations, which are an increasing threat to midlatitude natural forests. Pest detection at the level of an individual tree is of particular importance in mixed forests, where it enables a sensible forest management approach. In this study, we propose a method for individual tree crown delineation (ITCD) followed by feature extraction to detect a bark beetle disturbance in a mixed urban forest using a photogrammetric point cloud (PPC) and a multispectral orthomosaic. An excess green index (ExG) threshold mask was applied before the ITCD to separate targeted coniferous trees from deciduous trees and backgrounds. The individual crowns of conifer trees were automatically delineated as (i) a full tree crown using marker-controlled watershed segmentation (MCWS), Dalponte2016 (DAL), and Li 2012 (LI) region growing algorithms or (ii) a buffer (BUFFER) around a treetop from the masked PPC. We statistically compared selected spectral and elevation features extracted from automatically delineated crowns (ADCs) of each method to reference tree crowns (RTC) to distinguish between the forest disturbance classes and two tree species. Moreover, the effect of PPC density on the ITCD accuracy and feature extraction was investigated. The ExG threshold mask application resulted in the excellent separability of targeted conifer trees and the increasing shape similarity of ADCs compared to RTC. The results revealed a strong effect of PPC density on treetop detection and ITCD. If the PPC density is sufficient (>10 points/m2), the ADCs produced by DAL, MCWS, and LI methods are comparable, and the extracted feature statistics of ADCs insignificantly differ from RTC. The BUFFER method is less suitable for detecting a bark beetle disturbance in the mixed forest because of the simplicity of crown delineation. It caused significant differences in extracted feature statistics compared to RTC. Therefore, the point density was found to be more significant than the algorithm used. We conclude that automatic ITCD methods may constitute a substitute for the time-consuming manual tree crown delineation in tree-based bark beetle disturbance detection and sanitation of individual infested trees using the suggested methodology and high-density (>20 points/m2, 10 points/m2 minimum) PPC.