Tree Architecture Research Articles

An FPGA-Based YOLOv5 Accelerator for Real-Time Industrial Vision Applications.

The You Only Look Once (YOLO) object detection network has garnered widespread adoption in various industries, owing to its superior inference speed and robust detection capabilities. This model has proven invaluable in automating production processes such as material processing, machining, and quality inspection. However, as market competition intensifies, there is a constant demand for higher detection speed and accuracy. Current FPGA accelerators based on 8-bit quantization have struggled to meet these increasingly stringent performance requirements. In response, we present a novel 4-bit quantization-based neural network accelerator for the YOLOv5 model, designed to enhance real-time processing capabilities while maintaining high detection accuracy. To achieve effective model compression, we introduce an optimized quantization scheme that reduces the bit-width of the entire YOLO network-including the first layer-to 4 bits, with only a 1.5% degradation in mean Average Precision (mAP). For the hardware implementation, we propose a unified Digital Signal Processor (DSP) packing scheme, coupled with a novel parity adder tree architecture that accommodates the proposed quantization strategies. This approach efficiently reduces on-chip DSP utilization by 50%, offering a significant improvement in performance and resource efficiency. Experimental results show that the industrial object detection system based on the proposed FPGA accelerator achieves a throughput of 808.6 GOPS and an efficiency of 0.49 GOPS/DSP for YOLOv5s on the ZCU102 board, which is 29% higher than a commercial FPGA accelerator design (Xilinx's Vitis AI).

Allometric equations for biomass and carbon pool estimation in short rotation Pinus radiata stands of the Western Cape, South Africa

AbstractThe quantification of tree growth and carbon storage over time is an important task for sustainable forest management and carbon sequestration projects. For the South African short-rotation Pinus radiata (D. Don) forests, this knowledge is lacking. We developed allometric equations and compared the estimated weights to previously published biomass studies and we used Dirichlet Regression (DR) modelling to ensure additivity of the component proportions. The biomass components and their contribution to carbon storage depend strongly on forest structure and mean tree size but also on-site conditions and tree architecture. Our first two hypotheses were that the (1) best model for stemwood (SW), bark and total mass will include the combined variable DBH2H and (2) that the DR will yield statistically similar estimates for all components when compared to the best models. Our third hypothesis was that allometric equations developed for sites with high resource availability (e.g. wet, fertile sites) will yield biased estimates when extrapolated to sites with lower levels of resource availability (drier and/or infertile sites). The results indicated that DBH2H was the best variable to describe SW, bark and total mass and the DR yield similar estimates for all component proportions when compared to the best models. There were strong similarities in the SW and total mass of independent test sites in comparison to the SW and total mass of this study but greater variability in the bark, needle and branch mass. This can be associated to site and seasonal differences as well as variability in tree architecture brought about by different silvicultural operations on individual sites. Previously developed equations by other authors for sites with high resource availability overpredicted the SW and total mass of the models developed in this study. Our set of additive component equations performed well even when applied to sites of similar productivity over a climate gradient. The presented new equations bridge the gap in knowledge where allometric equations for short rotation Radiata pine stands are lacking.

Tree architecture and leaf morphometry of Qualea grandiflora Mart. (Vochysiaceae) in a fragment of dry forest and pasture in the State of Goiás, Brazil

Investigating the functional traits of plant communities is extremely important for conserving habitats and plant species. This study aimed to evaluate the tree architecture and leaf morphometry of Qualea grandiflora (Vochysiaceae) in a fragment of semideciduous dry forest and pasture in the Boa Esperança settlement, municipality of Piracanjuba, State of Goiás, Brazil. Environmental, tree architecture, and foliar morphometry data were collected for seven trees in the semideciduous dry forest and five in the pasture area. We used the Shapiro-Wilk test to assess the normality of the collected data. The student t-test was used to compare the means of the functional traits between the two environments (forest and pasture). The average value of canopy cover was 85% for the semideciduous dry forest and 0% for the pasture. The mean values for leaf length and width were lower in the pasture. There was no significant difference in the petiole length in the two evaluated environments and for none of the functional traits related to the tree architecture. However, individuals of Qualea grandiflora grow better in the seasonal semideciduous dry forest, probably as a response to less stressful environmental conditions for this species. Thus, for sustainable exploitation by the local community, it is recommended that the tree matrices be chosen within the semideciduous dry forest area.

Differential warming at crown scale impacts walnut primary growth onset and secondary growth rate.

Trees are exposed to significant spatio-temporal thermal variations, which can induce intracrown discrepancies in the onset and dynamics of primary and secondary growth. In recent decades, an increase in late winter and early spring temperatures has been observed, potentially accelerating bud break, cambial activation and their potential coordination. Intracrown temperature heterogeneities could lead to asymmetric tree shapes unless there is a compensatory mechanism at the crown level. An original warming experiment was conducted on young Juglans regia trees in a greenhouse. From February to August, the average temperature difference during the day between warmed and control parts was 4°C. The warming treatment advanced the date of budbreak significantly, by up to 14 days. Warming did not alter secondary growth resumption but increased growth rates, leading to higher xylem cell production (twice as many) and to an increase in radial increment (+80% compared to control). Meristems resumptions were asynchronous without coordination in response to temperature. Buds on warmed branches began to swell two weeks prior to cambial division, which was one week earlier than on control branches. A difference in carbon and water remobilisation at the end of bud ecodormancy was noted under warming. Overall, our results argue for a lack of compensatory mechanisms at the crown scale, which may lead to significant changes in tree architecture in response to intra-crown temperature heterogeneities.

L1-Tree: A novel algorithm for constructing 3D tree models and estimating branch architectural traits using terrestrial laser scanning data

Branch architecture provides crucial information for the understanding of plant trait variability and the adaptive strategies employed by trees in response to their environment. High-fidelity terrestrial laser scanning (TLS) data provide an accurate, efficient, and non-destructive means for constructing three-dimensional (3D) tree models and estimating architectural traits. However, the complex canopy structure of trees in natural forests and the presence of occlusion in TLS data pose significant challenges to achieving this goal. In this study, we present a novel algorithm, L1-Tree, for the construction of 3D tree models and the estimation of architectural traits from TLS data. This algorithm is grounded in the L1-Median algorithm and integrates a tree skeleton optimization procedure that considers the structural characteristics of tree branches. By comparing modeling results and manually derived branch traits for 24 trees of 24 species, we found that the L1-Tree algorithm achieved precision, recall, and F-score values of 0.94 for branch identification, coefficient of determination, root-mean-squared error, and normalized root-mean-squared error of 0.998, 0.068 m, and 0.3 % for branch length estimation, and a respective value of 0.958, 0.257 cm and 0.9 % for branch radius estimation. Additionally, the branch identification accuracy and accuracy in branch architectural trait estimation remained satisfactory across branch orders. Compared to established 3D tree model construction algorithms (e.g., TreeQSM), our L1-Tree algorithm demonstrated a superior capability in handling noisy environments and data gaps, making it a robust tool for TLS data-based tree architecture studies. Leaf-wood separation emerged as a crucial step influencing the performance of the L1-Tree algorithm. We observed significant drop in branch identification accuracy when using an automatic leaf-wood separation algorithm as input, highlighting the urgent need to develop effective leaf-wood separation algorithms to generate high-quality wood point clouds for tree architecture studies.

Occurrence of Wetness on the Fruit Surface Modeled Using Spatio-Temporal Temperature Data from Sweet Cherry Tree Canopies

Typically, fruit cracking in sweet cherry is associated with the occurrence of free water at the fruit surface level due to direct (rain and fog) and indirect (cold exposure and dew) mechanisms. Recent advances in close range remote sensing have enabled the monitoring of the temperature distribution with high spatial resolution based on light detection and ranging (LiDAR) and thermal imaging. The fusion of LiDAR-derived geometric 3D point clouds and merged thermal data provides spatially resolved temperature data at the fruit level as LiDAR 4D point clouds. This paper aimed to investigate the thermal behavior of sweet cherry canopies using this new method with emphasis on the surface temperature of fruit around the dew point. Sweet cherry trees were stored in a cold chamber (6 °C) and subsequently scanned at different time intervals at room temperature. A total of 62 sweet cherry LiDAR 4D point clouds were identified. The estimated temperature distribution was validated by means of manual reference readings (n = 40), where average R2 values of 0.70 and 0.94 were found for ideal and real scenarios, respectively. The canopy density was estimated using the ratio of the number of LiDAR points of fruit related to the canopy. The occurrence of wetness on the surface of sweet cherry was visually assessed and compared to an estimated dew point (Ydew) index. At mean Ydew of 1.17, no wetness was observed on the fruit surface. The canopy density ratio had a marginal impact on the thermal kinetics and the occurrence of wetness on the surface of sweet cherry in the slender spindle tree architecture. The modelling of fruit surface wetness based on estimated fruit temperature distribution can support ecophysiological studies on tree architectures considering resilience against climate change and in studies on physiological disorders of fruit.

The Little Book of Trees

From the publisher: Packed with surprising facts, this delightful and gorgeously designed book will beguile any nature lover. Expertly written and beautifully illustrated throughout with color photographs and original color artwork, The Little Book of Trees is an accessible and enjoyable mini-reference about the world’s trees, with examples drawn from across the globe. It fits an astonishing amount of information in a small package, covering a wide range of topics—from tree anatomy, diversity, and architecture to habitat and conservation. It also includes curious facts and a section on trees in myths, folklore, and modern culture from around the world. The result is an irresistible guide to the amazing lives of trees.

LED streetlamps alter tree architecture, downregulate the photosynthetic process and alter the sugar metabolism of Populus alba L.

The escalating issue of light pollution in urban environments poses multifaceted challenges, affecting not only the nocturnal sky but also exerting intricate influences on plant physiology. This study delves into the physiological responses of an urban ornamental tree, Populus alba L. clone DI-1, to varying intensities of streetlamp LED night lighting (NL), a shift from traditional High-Pressure Sodium lamps. The investigation “sheds light” on the molecular pathways underlying observed physiological regulations, aiming to offer a comprehensive understanding of the manifold effects of NL on this tree species. NL altered tree architecture, i.e. increased branch length and diameters, underscoring the dynamic response of trees to nocturnal artificial lighting conditions. Regarding tree physiology, the NL-triggered net CO2 assimilation (Pn) during the night resulted in limitations in stomatal conductance during daylight hours. This led to a reduction in Pn, particularly during dawn, hindering the quantum yield for the reduction of end acceptors of PSI. Changes in chlorophyll a-to-b proportion and overall concentration, electron transport chain, and gene expression further highlight the intricate interplay between NL and plant metabolic regulation. Notably, the increased gene expression of sugar transporters in both NL trees suggested a responsive shift in sugar and starch metabolism. This was reflected in the absence of a starch accumulation during daylight hours in NL leaves. The study emphasizes the need for a holistic approach to urban lighting, considering its profound impact on photosynthesizing citizens. These findings highlight the pressing need for the development of innovative lighting spectra with reduced impact on plant physio-chemistry while ensuring visibility for citizen safety.

Development of a Precise Tree Structure from LiDAR Point Clouds

Abstract. A precise tree structure that represents the distribution of tree stem, branches, and leaves is crucial for accurately capturing the full representation of a tree. Light Detection and Ranging (LiDAR)-based three-dimensional (3D) point clouds (PCs) capture the geometry of scanned objects including forests stands and individual trees. PCs are irregular, unstructured, often noisy, and contaminated by outliers. Researchers have struggled to develop methods to separate leaves and wood without losing the tree geometry. This paper proposes a solution that employs only the spatial coordinates (x, y, z) of the PC. The new algorithm works as a filtering approach, utilizing multi-scale neighborhood-based geometric features (GFs) e.g., linearity, planarity, and verticality to classify linear (wood) and non-linear (leaf) points. This involves finding potential wood points and coupling them with an octree-based segmentation to develop a tree architecture. The main contributions of this paper are (i) investigating the potential of different GFs to split linear and non-linear points, (ii) introducing a novel method that pointwise classifies leaf and wood points, and (iii) developing a precise 3D tree structure. The performance of the new algorithm has been demonstrated through terrestrial laser scanning PCs. For a Scots pine tree, the new method classifies leaf and wood points with an overall accuracy of 97.9%.

Treegraph: tree architecture from terrestrial laser scanning point clouds

AbstractAccurate quantification of tree architecture is critical to interpreting the growth, health and functioning of trees and forests. Terrestrial laser scanning (TLS) offers millimetre‐level point cloud data, but current approaches to 3D tree reconstruction from TLS point clouds primarily focus on retrieving total volume at tree scale for aboveground biomass (AGB) estimation. Few methods have been designed specifically to provide tree architectural properties, including branch‐level morphology and topology, rather than AGB; derived topological traits have tended to be a compromise, and of secondary importance to volume. We present Treegraph, a new approach explicitly designed to retrieve the architectural traits of trees at multiple scales, from the whole tree scale down to individual branches and internodes, using TLS data with limited assumptions about tree form. It provides morphological traits such as branch length and diameter, alongside topological traits including parent–daughter connections of branches and internodes, furcation (branching) number and branch order. We compare Treegraph‐derived morphological and topological traits with manual measurements of branches from eight destructively harvested trees, yielding RMSE values of 0.60 m (5.96%) for branch length, 2.99 cm (33.45%) for branch diameter, 0.46 (19.38%) for furcation number and 0.08 m (33.16%) for internode length, respectively. In a broader application to 603 trees from tropical, temperate and urban forests, we demonstrate that the derived morphological and topological traits support testing of structure‐related metabolic scaling theories. Testing branches over 10 cm in diameter across 18 657 branching nodes shows that Treegraph‐derived branch‐level scaling exponents deviate from WBE predictions, exhibiting area‐preserving behaviour while displaying asymmetry in length and diameter of daughter branches. Available as open‐source Python software, Treegraph provides fine‐level branching network information, promoting improved insights into tree structure and function. This data‐driven approach reduces the need for empirical heuristic parameters, which has the potential for advancing large‐scale ecological studies on tree architecture.

An analysis of natural variation in Pinus pinaster through the lens of systems biology

Pinus pinaster is a main species in Mediterranean forests, being naturally distributed through the Mediterranean basin, covering regions with a variety of geoclimatic conditions. This distribution in different environments leads, by natural selection, to a natural variation within the species that manifests at phenotypic level in provenances with different growth features and overall tree architectures. Studying P. pinaster natural variation is necessary to understand the genetic heritage of the species and can provide valuable information for information-based decision-making regarding forest management and breeding programs. In this paper, we analyzed the natural variation in needles and buds from three provenances from contrasting geoclimatic locations using a common garden approach and proteomics. The integration of the proteomics data with tree growth-related parameters, geoclimatic features at provenances original locations, and sample-matched metabolomics data previously generated provided novel knowledge on metabolism rearrangements related to secondary metabolism and associated to growth features and the adaptation to light and UV-B radiation intensities.

Stormwise: Innovative Forest Management to Promote Storm Resistance in Roadside Forests

Abstract A growing proportion of forested landscapes are interspersed with human infrastructure, such as utility lines and roads, increasing the potential for tree-failure consequences due to storms and other causes. Utilities and other institutions have strong incentives to reduce such interactions and allocate substantial resources to risk reduction, but trees and forests in densely populated landscapes also provide significant amenities to society. We present a generalized framework for “Stormwise” forest management, focused on resistance of roadside forests to storms, based on elements of arboricultural and silvicultural practice and tree biomechanics and architecture. We detail results of a multidisciplinary research program focused on management outcomes, opportunities and barriers to implementation, and allocation of investment based on physical and social landscape characteristics. We discuss initial findings, the potential for widespread adoption of resilience-focused management in roadside and infrastructure-adjacent forests, and the importance of such work, considering a changing climate. Study Implications: The research and implementation program we detail here illustrates the potential for “Stormwise” forest management to reduce storm-damage recovery costs, result in fewer and shorter-duration power and transportation interruptions, and allow for low-investment ground-based management in future forest entries. We illustrate a multifaceted, interdisciplinary research program that links the geospatial, social, and biophysical components of understanding forest infrastructure systems. We illustrate how implementation of Stormwise management has the potential to benefit stakeholders such as residents and utilities, offset the substantial economic costs of tree-related power outages, and reduce societal disruptions associated with interactions between trees and infrastructure during storms.

Parametrization of biological assumptions to simulate growth of tree branching architectures.

Modeling and simulating the growth of the branching of tree species remains a challenge. With existing approaches, we can reconstruct or rebuild the branching architectures of real tree species, but the simulation of the growth process remains unresolved. First, we present a tree growth model to generate branching architectures that resemble real tree species. Secondly, we use a quantitative morphometric approach to infer the shape similarity of the generated simulations and real tree species. Within a functional-structural plant model, we implement a set of biological parameters that affect the branching architecture of trees. By modifying the parameter values, we aim to generate basic shapes of spruce, pine, oak and poplar. Tree shapes are compared using geometric morphometrics of landmarks that capture crown and stem outline shapes. Five biological parameters, namely xylem flow, shedding rate, proprioception, gravitysense and lightsense, most influenced the generated tree branching patterns. Adjusting these five parameters resulted in the different tree shapes of spruce, pine, oak, and poplar. The largest effect was attributed to gravity, as phenotypic responses to this effect resulted in different growth directions of gymnosperm and angiosperm branching architectures. Since we were able to obtain branching architectures that resemble real tree species by adjusting only a few biological parameters, our model is extendable to other tree species. Furthermore, the model will also allow the simulation of structural tree-environment interactions. Our simplifying approach to shape comparison between tree species, landmark geometric morphometrics, showed that even the crown-trunk outlines capture species differences based on their contrasting branching architectures.

Energy Efficient Predictive Maintenance in Resource Constraint Industrial IoT devices using Smoothed Mish Activation Function

Traditionally, Industrial IoT devices collect sensor data to a cloud platform where ML/DL processing is to be done. Edge computing has the advantage of reducing latency, improved battery performance, safe transmission and reducing vulnerability. These benefits are particularly significant considering the limited resources on IoT devices, such as only a few kilobytes of RAM, and the critical importance of energy savings in industrial applications. Arduino Uno uses 2kB of RAM and 32 kB of read only flash memory. Optimal performance is required for ML inference on EDGE devices to get good accuracies and low latency. This research paper is based on memory optimization in resource constrained devices, focusing on a novel approach that combines the Bonsai tree architecture with a smoothed Mish activation function. While the Bonsai tree architecture has been explored in prior works, this research contributes to the refinement of the activation function, a crucial component in achieving superior test accuracy within limited memory constraints. Although many algorithms have been developed with good accuracy but can’t be used in optimal power consumption with low latency and resource constraint like fewer RAM and ROM. A smooth Mish activation function in Bonsai tree algorithm is proposed to enhance accuracy with lowest ML model size, where Bonsai tree is a single,shallow,sparse tree where data is projected in low dimension space. Lowest ML model size consumes least power. We observed the smoothed Mish activation function in the Bonsai tree algorithm outperform with low latency and power consumption. This paper aims to elucidate the methodology, experimental results, and implications of the modified activation function, shedding light on its potential to revolutionize memory-efficient edge computing in the Industrial IoT network.

BT-CNN: a balanced binary tree architecture for classification of brain tumour using MRI imaging.

Deep learning is a very important technique in clinical diagnosis and therapy in the present world. Convolutional Neural Network (CNN) is a recent development in deep learning that is used in computer vision. Our medical investigation focuses on the identification of brain tumour. To improve the brain tumour classification performance a Balanced binary Tree CNN (BT-CNN) which is framed in a binary tree-like structure is proposed. It has a two distinct modules-the convolution and the depthwise separable convolution group. The usage of convolution group achieves lower time and higher memory, while the opposite is true for the depthwise separable convolution group. This balanced binarty tree inspired CNN balances both the groups to achieve maximum performance in terms of time and space. The proposed model along with state-of-the-art models like CNN-KNN and models proposed by Musallam et al., Saikat et al., and Amin et al. are experimented on public datasets. Before we feed the data into model the images are pre-processed using CLAHE, denoising, cropping, and scaling. The pre-processed dataset is partitioned into training and testing datasets as per 5 fold cross validation. The proposed model is trained and compared its perforarmance with state-of-the-art models like CNN-KNN and models proposed by Musallam et al., Saikat et al., and Amin et al. The proposed model reported average training accuracy of 99.61% compared to other models. The proposed model achieved 96.06% test accuracy where as other models achieved 68.86%, 85.8%, 86.88%, and 90.41% respectively. Further, the proposed model obtained lowest standard deviation on training and test accuracies across all folds, making it invariable to dataset.

Large deformation of trees in a strong wind

Understanding the dynamic response of trees to a strong wind is crucial to alleviating the damages and losses of tree forests caused by windstorms. Previous studies have elucidated small deformations of a broad range of tree species, yet mathematical models that can accurately pinpoint the location and severity of wind-induced damages are still lacking. To bridge this gap, the present work puts forward the first geometrically accurate, physics-based model to capture large deformation of trees induced by a strong wind. The proposed model incorporates a branched tree architecture to represent the topology of a realistic tree. The large, dynamic deformation of the trunk and branches of the tree is described by a system of nonlinear partial differential equations derived from structural theories rooted in classical mechanics. The aerodynamic loading is quantified based on the instantaneous relative velocity between the wind and the tree. The proposed model could successfully reproduce the effects of aerodynamic damping by capturing the nonlinear interplays between the wind and the tree, in contrast to prior analytical models that relied on empirical assumptions of the damping coefficients. The model further reveals a reduction in the drag force due to wind-induced shape reconfiguration, which is enhanced exponentially as a function of the wind speed. The proposed modeling framework could aid in the development of novel forest management strategies to mitigate wind-induced economic and environmental losses.

Stochastic modelling of development and biomass allocation: Computation applied to architecture of young mahogany trees (Khaya senegalensis Desr. A. Juss), a native African savannah emblematic agroforestry species

The architectural plasticity forms observed in trees is a result of meristem functioning, which generates new organs and branches, and adjusts growth processes in response to heterogeneous climatic adaptations that affect biomass allocation. Analyzing this plasticity should enable the selection of adapted individuals for optimizing successful cropping systems. Mahogany tree (Khaya senegalensis) is a rhythmically growing indigenous agroforestry tree that is heavily exploited for its multiple uses. Understanding its growth characteristics, as well as the complexity of its structure (randomness, rhythmicity, etc. of Mahogany tree), could facilitate its conservation and sustainable management. This study aims to model the architecture and physiology of young mahogany trees based on field data using an organ-level structural–functional model called 'GreenLab', which is founded on source-sink relationships. Ninety trees aged 6, 12, and 24 months were measured in the field. Development was calculated using a dual-scale automaton based on the Monte Carlo process, while biomass production and its distribution to different plant organs (source and sink) were calibrated using Pressler's law using Markov chains. Meristematic activity laws, combined with organ sinks (D: Demand), photosynthesis (Q: Supply), and organic series (Q/D: Trophic pressure), were employed to simulate individual architecture. The results demonstrate that the model realistically and flexibly describes topological development and replicates biomass production and allocation processes for rhythmically growing trees. This model will enable the identification of mahogany ideotypes suited for enhancing agroforestry cropping systems based on this species and several other threatened species. These findings introduce and thus lay the groundwork for a computational plant model tailored to the needs of agroforestry from a novel perspective, offering new avenues for agronomic and forestry applications in West Africa and Côte d'Ivoire.

Fast CU size decision and intra-prediction mode decision method for H.266/VVC

AbstractH.266/Versatile Video Coding (VVC) is the most recent video coding standard developed by the Joint Video Experts Team (JVET). The quad-tree with nested multi-type tree (QTMT) architecture that improves the compression performance of H.266/VVC is introduced. Moreover, H.266/VVC contains a greater number of intra-prediction modes than H.265/High Efficiency Video Coding (HEVC), totalling 67. However, these lead to extremely the coding computational complexity. To cope with the above issues, a fast intra-coding unit (CU) size decision method and a fast intra-prediction mode decision method are proposed in this paper. Specifically, the trained Support Vector Machine (SVM) classifier models are utilized for determining CU partition mode in a fast CU size decision scheme. Furthermore, the quantity of intra-prediction modes added to the RDO mode set decreases in a fast intra-prediction mode decision scheme based on the improved search step. Simulation results illustrate that the proposed overall algorithm can decrease 55.24% encoding runtime with negligible BDBR.

Unearthing Current Knowledge Gaps in Our Understanding of Tree Stability: Review and Bibliometric Analysis

Forest preservation and management are paramount for sustainable mitigation of climate change, timber production, and the economy. However, the potential of trees and forests to provide these benefits to the ecosystem is hampered by natural phenomena such as windthrow and anthropogenic activities. The aim of the current research was to undertake a critical thematic review (from 1983 to 2023) informed by a bibliometric analysis of existing literature on tree stability. The results revealed an increase in tree stability research between 2019 and 2022, with the USA, France, and Italy leading in research output, while Scotland and England notably demonstrated high research influence despite fewer publications. A keyword analysis showed that tree stability can be divided into four themes: tree species, architecture, anchorage, and environmental factors. Prominent studies on tree stability have focused on root anchorage. However, more recently, there has been a growing emphasis on urban forestry and disease-induced tree damage, underscoring a shift towards climate change and diversity research. It was concluded that considerable knowledge gaps still exist; that greater geographic diversification of research is needed and should include tropical and sub-tropical regions; that research relating to a wider range of soil types (and textures) should be conducted; and that a greater emphasis on large-scale physical modelling is required. Data and knowledge produced from these areas will improve our collective understanding of tree stability and therefore help decision makers and practitioners manage forestry resources in a more sustainable way into the future.

Performance evaluation of decision trees and multilayer perceptrons in seabed classification

Accurate knowledge of the oceanic propagation medium, and, thus, seabed properties, is of paramount importance in source localization, especially of weak targets. In this work, we employ machine learning techniques to perform seabed identification and classification using impulse responses of different media; supervised learning is employed. We first design a decision tree architecture that relies on features selected from the impulse responses corresponding to the different media. Examples of these features are kurtosis, skewness, and strength of the signal. Training sets are created by identifying features from noisy signals and testing follows after extracting corresponding features from a different data set. Performance is evaluated as a function of Signal-to-Noise Ratio. A principal component analysis is also implemented for the investigation of the potential for dimensionality reduction. Subsequently, multilayer perceptrons are employed using identical data both for training and testing and the two machine learning techniques are compared; advantages and disadvantages of each are identified and discussed in this work. [Work supported by ONR.]