Branching Model Research Articles

Resource‐adaptive and OOD‐robust inference of deep neural networks on IoT devices

AbstractEfficiently executing inference tasks of deep neural networks on devices with limited resources poses a significant load in IoT systems. To alleviate the load, one innovative method is branching that adds extra layers with classification exits to a pre‐trained model, enabling inputs with high‐confidence predictions to exit early, thus reducing inference cost. However, branching networks, not originally tailored for IoT environments, are susceptible to noisy and out‐of‐distribution (OOD) data, and they demand additional training for optimal performance. The authors introduce BrevisNet, a novel branching methodology designed for creating on‐device branching models that are both resource‐adaptive and noise‐robust for IoT applications. The method leverages the refined uncertainty estimation capabilities of Dirichlet distributions for classification predictions, combined with the superior OOD detection of energy‐based models. The authors propose a unique training approach and thresholding technique that enhances the precision of branch predictions, offering robustness against noise and OOD inputs. The findings demonstrate that BrevisNet surpasses existing branching techniques in training efficiency, accuracy, overall performance, and robustness.

Structural and hydrodynamic modelling of the probability of breakage of branching and plate coral colonies

Climate change is amplifying the intensity of severe weather events, with coastal regions such as coral reefs facing heightened vulnerability to cyclonic wave forces. Structural models to predict bending stress and breakage of corals have been developed for coral colonies to enhance comprehension and prediction of the effects of hydrodynamic disturbances on coral reefs. However, there is scope for improving these predictions by evolving the methodology for quantifying complicated and variable coral morphologies. This study aims to predict breakage thresholds for two of the most prevalent coral morphologies: branching and plate corals (using Acropora muricata and Acropora hyacinthus as study species). Laboratory and field measurements were taken to assess coral morphologies and material characteristics. Morphological features of 47 branching colonies and 100 plate colonies were surveyed at the study site (Heron Reef, southern GBR) and the tensile strength of 80 coral samples was obtained by in situ and laboratory testing. Three-dimensional structural models of branching and plate coral colonies were developed, encompassing multiple coral colonies with varying morphological patterns from relatively shallow (5-7 m) to deep (9-12 m) zones. Model results were calibrated and verified with existing data, revealing that velocity thresholds of 1.7 m/s and 5.0 m/s would destroy 90% of the simulated branching coral structures growing in the deep and shallow parts of the forereef zone, respectively. In contrast, the plate corals have sufficient margins of safety even in extreme flow conditions (7 m/s). Additionally, skeletal strength and structural performance were adjusted based on varying degrees of bioerosion inside the coral skeleton. A higher probability of breakage was observed as the extent of bioerosion increased. The laboratory experiments of hydrodynamic loads on coral colony show that the sheltering effect due to one or two neighbouring colonies in the upwave direction is negligible. These models can be easily adjusted to provide predictions for other coral species, shapes, levels of bioerosion, and locations (e.g., sheltered or exposed areas). Comprehensive predictions about the level of expected damage and rubble generation in different areas can be used in reef management planning and restoration prioritization.

Mathematical modeling of sound propagation through the respiratory tract in a modified gas environment

The aim of the work was to develop a mathematical model of sound propagation through the human respiratory tract when breathing certain respiratory gas mixtures. We based the model on an assembler model of branching of the respiratory tract. For calculations, we presented the human tracheobronchial tree in the form of an equivalent T-shaped cut section diagram, which includes parameters of the walls of the respiratory tract. We carried the simulation out in two main stages. At the first stage, we calculated the impedance of the flow from the small airways to the larger airways. At the second stage, we calculated the sound pressure caused by breathing. As a result, we obtained the curves of the impedance dependence on the frequency spectrum, as well as the sound power from the respiratory flow. We noted that when breathing air, the impedance value remains constant with increasing frequency. Moreover, we also noted that when exposed to a heavier gas than air, the impedance value constantly increases, and when exposed to lighter gases, after reaching a maximum, the impedance value decreases. We found that the highest sound pressure is typical for the oxygen-krypton mixture, and the lowest for the oxygen–helium mixture. We also found that in the case of modeling the laminar flow regime (with a critical Reynolds number equal to 1800), sound affects how many generations of bronchi more than in the case of modeling the transient flow regime (with a critical Reynolds number equal to 2700). Thus, we drew conclusions about the influence of the physical properties of gaseous media on the amount of sound pressure and the propagation of sound through the human respiratory tract.

Benchmarking the effective temperature scale of red giant branch stellar models: The case of the metal-poor halo giant HD 122563

Context. There is plenty of evidence in the literature of significant discrepancies between the observations and models of metal-poor red giant branch stars, in particular regarding the effective temperature Teff scale. Aims. We revisit the benchmark star HD 122563 using the most recent observations from Gaia Data Release 3, to investigate if these new constraints may help in resolving this discrepancy. Methods. We review the most recent spectroscopic determinations of the metallicity [Fe/H] of HD 122563, and provide a new assessment of its fundamental parameters, specifically, bolometric luminosity, Teff, surface gravity, plus a photometric determination of its metal content. Using these constraints, we compare the position of the star in the Hertzsprung-Russell (H–R) diagram with various recent sets of stellar evolution tracks. Results. The H-R diagram analysis reveals a significant disagreement between observed and theoretical Teff values, when adopting the most recent spectroscopic estimate of [Fe/H]. On the other hand, by using the photometric determination of [Fe/H], some of the selected sets of stellar tracks appear in fair agreement with observations. The sets with discrepant Teff can be made to agree with observations either by modifying the prescription adopted to calculate the models’ outer boundary conditions, and/or by reducing the adopted value of the mixing length parameter with respect to the solar-calibration. Conclusions. A definitive assessment of whether the Teff scale of metal-poor stellar red giant branch models is consistent with observations requires a more robust determination of the fundamental parameters of HD 122563 and also a larger sample of calibrators. From the theoretical side, it is crucial to minimise the current uncertainties in the treatment (boundary conditions, temperature gradient) of the outer layers of stellar models with convective envelopes.

Developing Machine Vision in Tree-Fruit Applications-Fruit Count, Fruit Size and Branch Avoidance in Automated Harvesting.

Recent developments in affordable depth imaging hardware and the use of 2D Convolutional Neural Networks (CNN) in object detection and segmentation have accelerated the adoption of machine vision in a range of applications, with mainstream models often out-performing previous application-specific architectures. The need for the release of training and test datasets with any work reporting model development is emphasized to enable the re-evaluation of published work. An additional reporting need is the documentation of the performance of the re-training of a given model, quantifying the impact of stochastic processes in training. Three mango orchard applications were considered: the (i) fruit count, (ii) fruit size and (iii) branch avoidance in automated harvesting. All training and test datasets used in this work are available publicly. The mAP 'coefficient of variation' (Standard Deviation, SD, divided by mean of predictions using models of repeated trainings × 100) was approximately 0.2% for the fruit detection model and 1 and 2% for the fruit and branch segmentation models, respectively. A YOLOv8m model achieved a mAP50 of 99.3%, outperforming the previous benchmark, the purpose-designed 'MangoYOLO', for the application of the real-time detection of mango fruit on images of tree canopies using an edge computing device as a viable use case. YOLOv8 and v9 models outperformed the benchmark MaskR-CNN model in terms of their accuracy and inference time, achieving up to a 98.8% mAP50 on fruit predictions and 66.2% on branches in a leafy canopy. For fruit sizing, the accuracy of YOLOv8m-seg was like that achieved using Mask R-CNN, but the inference time was much shorter, again an enabler for the field adoption of this technology. A branch avoidance algorithm was proposed, where the implementation of this algorithm in real-time on an edge computing device was enabled by the short inference time of a YOLOv8-seg model for branches and fruit. This capability contributes to the development of automated fruit harvesting.

URINet: Unsupervised point cloud rotation invariant representation learning via semantic and structural reasoning

In recent years, many rotation-invariant networks have been proposed to alleviate the interference caused by point cloud arbitrary rotations. These networks have demonstrated powerful representation learning capabilities. However, most of those methods rely on costly manually annotated supervision for model training. Moreover, they fail to reason the structural relations and lose global information. To address these issues, we present an unsupervised method for achieving comprehensive rotation invariant representations without human annotation. Specifically, we propose a novel encoder–decoder architecture named URINet, which learns a point cloud representation by combining local semantic and global structural information, and then reconstructs the input without rotation perturbation. In detail, the encoder is a two-branch network where the graph convolution based structural branch models the relationships among local regions to learn global structural knowledge and the semantic branch learns rotation invariant local semantic features. The two branches derive complementary information and explore the point clouds comprehensively. Furthermore, to avoid the self-reconstruction ambiguity brought by uncertain poses, a bidirectional alignment is proposed to measure the quality of reconstruction results without orientation knowledge. Extensive experiments on downstream tasks show that the proposed method significantly surpasses existing state-of-the-art methods on both synthetic and real-world datasets.

Photo-induced changes in tissue stiffness alter epithelial budding morphogenesis in the embryonic lung.

Extracellular matrix (ECM) stiffness has been shown to influence the differentiation of progenitor cells in culture, but a lack of tools to perturb the mechanical properties within intact embryonic organs has made it difficult to determine how changes in tissue stiffness influence organ patterning and morphogenesis. Photocrosslinking of the ECM has been successfully used to stiffen soft tissues, such as the cornea and skin, which are optically accessible, but this technique has not yet been applied to developing embryos. Here, we use photocrosslinking with Rose Bengal (RB) to locally and ectopically stiffen the pulmonary mesenchyme of explanted embryonic lungs cultured ex vivo . This change in mechanical properties was sufficient to suppress FGF-10-mediated budding morphogenesis along the embryonic airway, without negatively impacting patterns of cell proliferation or apoptosis. A computational model of airway branching was used to determine that FGF-10-induced buds form via a growth-induced buckling mechanism and that increased mesenchymal stiffness is sufficient to inhibit epithelial buckling. Taken together, our data demonstrate that photocrosslinking can be used to create regional differences in mechanical properties within intact embryonic organs and that these differences influence epithelial morphogenesis and patterning. Further, this photocrosslinking assay can be readily adapted to other developing tissues and model systems.

Duality and the well-posedness of a martingale problem

For two Polish state spaces EX and EY, and an operator GX, we obtain existence and uniqueness of a GX-martingale problem provided there is a bounded continuous duality function H on EX×EY together with a dual process Y on EY which is the unique solution of a GY-martingale problem. For the corresponding solutions (Xt)t≥0 and (Yt)t≥0, duality with respect to a function H in its simplest form means that the relation Ex[H(Xt,y)]=Ey[H(x,Yt)] holds for all (x,y)∈EX×EY and t≥0. While duality is well-known to imply uniqueness of the GX-martingale problem, we give here a set of conditions under which duality also implies existence without using approximating sequences of processes of a different kind (e.g. jump processes to approximate diffusions) which is a widespread strategy for proving existence of solutions of martingale problems. Given the process (Yt)t≥0 and a duality function H, to prove existence of (Xt)t≥0 one has to show that the r.h.s. of the duality relation defines for each y a measure on EX, i.e. there are transition kernels (μt)t≥0 from EX to EX such that Ey[H(x,Yt)]=∫μt(x,dx′)H(x′,y) for all (x,y)∈EX×EY and all t≥0.As examples, we treat resampling and branching models, such as the Fleming–Viot measure-valued diffusion and its spatial counterparts (with both, discrete and continuum space), as well as branching systems, such as Feller’s branching diffusion. While our main result as well as all examples come with (locally) compact state spaces, we discuss the strategy to lift our results to genealogy-valued processes or historical processes, leading to non-compact (discrete and continuum) state spaces. Such applications will be tackled in forthcoming work based on the present article.

A Deep Learning Quantile Regression Photovoltaic Power-Forecasting Method under a Priori Knowledge Injection

Accurate and reliable PV power probabilistic-forecasting results can help grid operators and market participants better understand and cope with PV energy volatility and uncertainty and improve the efficiency of energy dispatch and operation, which plays an important role in application scenarios such as power market trading, risk management, and grid scheduling. In this paper, an innovative deep learning quantile regression ultra-short-term PV power-forecasting method is proposed. This method employs a two-branch deep learning architecture to forecast the conditional quantile of PV power; one branch is a QR-based stacked conventional convolutional neural network (QR_CNN), and the other is a QR-based temporal convolutional network (QR_TCN). The stacked CNN is used to focus on learning short-term local dependencies in PV power sequences, and the TCN is used to learn long-term temporal constraints between multi-feature data. These two branches extract different features from input data with different prior knowledge. By jointly training the two branches, the model is able to learn the probability distribution of PV power and obtain discrete conditional quantile forecasts of PV power in the ultra-short term. Then, based on these conditional quantile forecasts, a kernel density estimation method is used to estimate the PV power probability density function. The proposed method innovatively employs two ways of a priori knowledge injection: constructing a differential sequence of historical power as an input feature to provide more information about the ultrashort-term dynamics of the PV power and, at the same time, dividing it, together with all the other features, into two sets of inputs that contain different a priori features according to the demand of the forecasting task; and the dual-branching model architecture is designed to deeply match the data of the two sets of input features to the corresponding branching model computational mechanisms. The two a priori knowledge injection methods provide more effective features for the model and improve the forecasting performance and understandability of the model. The performance of the proposed model in point forecasting, interval forecasting, and probabilistic forecasting is comprehensively evaluated through the case of a real PV plant. The experimental results show that the proposed model performs well on the task of ultra-short-term PV power probabilistic forecasting and outperforms other state-of-the-art deep learning models in the field combined with QR. The proposed method in this paper can provide technical support for application scenarios such as energy scheduling, market trading, and risk management on the ultra-short-term time scale of the power system.

Three-Dimensional Obstacle Avoidance Harvesting Path Planning Method for Apple-Harvesting Robot Based on Improved Ant Colony Algorithm

The cultivation model for spindle-shaped apple trees is widely used in modern standard apple orchards worldwide and represents the direction of modern apple industry development. However, without an effective obstacle avoidance path, the robotic arm is prone to collision with obstacles such as fruit tree branches during the picking process, which may damage fruits and branches and even affect the healthy growth of fruit trees. To address the above issues, a three-dimensional path -planning algorithm for full-field fruit obstacle avoidance harvesting for spindle-shaped fruit trees, which are widely planted in modern apple orchards, is proposed in this study. Firstly, based on three typical tree structures of spindle-shaped apple trees (free spindle, high spindle, and slender spindle), a three-dimensional spatial model of fruit tree branches was established. Secondly, based on the grid environment representation method, an obstacle map of the apple tree model was established. Then, the initial pheromones were improved by non-uniform distribution on the basis of the original ant colony algorithm. Furthermore, the updating rules of pheromones were improved, and a biomimetic optimization mechanism was integrated with the beetle antenna algorithm to improve the speed and stability of path searching. Finally, the planned path was smoothed using a cubic B-spline curve to make the path smoother and avoid unnecessary pauses or turns during the harvesting process of the robotic arm. Based on the proposed improved ACO algorithm (ant colony optimization algorithm), obstacle avoidance 3D path planning simulation experiments were conducted for three types of spindle-shaped apple trees. The results showed that the success rates of obstacle avoidance path planning were higher than 96%, 86%, and 92% for free-spindle-shaped, high-spindle-shaped, and slender-spindle-shaped trees, respectively. Compared with traditional ant colony algorithms, the average planning time was decreased by 49.38%, 46.33%, and 51.03%, respectively. The proposed improved algorithm can effectively achieve three-dimensional path planning for obstacle avoidance picking, thereby providing technical support for the development of intelligent apple picking robots.

Face super resolution with a high frequency highway

AbstractFace shape priors such as landmarks, heatmaps, and parsing maps are widely used to improve face super resolution (SR). It is observed that face priors provide locations of high‐frequency details in key facial areas such as the eyes and mouth. However, existing methods fail to effectively exploit the high‐frequency information by using the priors as either constraints or inputs. This paper proposes a novel high frequency highway () framework to better utilize prior information for face SR, which dynamically decomposes the final SR face into a coarse SR face and a high frequency (HF) face. The coarse SR face is reconstructed from a low‐resolution face via a texture branch, using only pixel‐wise reconstruction loss. Meanwhile, the HF face is directly generated from face priors via an HF branch that employs the proposed inception–hourglass model. As a result, allows the face priors to have a direct impact on the SR face by adding the outputs of both branches as the final result and provides an extra face editing function. Extensive experiments show that significantly outperforms state‐of‐the‐art face SR methods, is general for different texture branch models and face priors, and is robust to dataset mismatch and pose variations.

Использование продвинутых функций Git при разработке программного обеспечения

Abstract. The key aspects of using the advanced features of the Git version control system in software development are considered: branching models, Git configuration settings and error correction methods useful for developers at all levels. The purpose of the study is to demonstrate the capabilities of Git to increase the efficiency and flexibility of software development. The issues of integrating advanced Git functions into the software development process, which are of interest to IT professionals, are considered. Practical significance: due to the fact that useful configuration parameters are considered that help maximize the performance and usability of Git. Examples of commands and settings of the Git system are given, which increase the accessibility and effectiveness of its application in practice. Discussion: through the analysis of technical and practical aspects, the advantages of advanced Git functions and their contribution to improving development processes are shown. It is shown that the use of advanced Git functions makes it possible to optimize development processes, increase the efficiency of team interactions and ensure high code quality

Two intuitions about free will—Some afterthoughts

AbstractIn 2014, Christian List and I published a paper that delineated our view regarding what it takes for an agent to act freely. We suggested that this requires the action to be endorsed by the agent and caused by this endorsement and yet not be necessitated. Free action requires indeterminism at the agential level—the kind of indeterminism that is compatible with physical determinism. I still think that our proposal was on the right track, but I believe it needs elaboration. As we already noted in the paper, our formal modelling—a standard branching model—was too extensional, and therefore in need of revision. Also, on a substantive side, what we say about the causal component of our proposal does not quite take care of the danger of ‘flukishness’ of an undetermined action. It was this threat to agential control that the requirement of causation by endorsement was meant to disarm in the first place. But the threat still remains and needs to be confronted. Revising the formal model and finding a solution to the problem of flukishness are my two objectives in this paper.

Contrary to phylogenetic predictions, crown coloration of the Eared Dove is not regulated by sex hormones

The mechanisms regulating plumage coloration appear to be phylogenetically conserved in avian species: the plesiomorphic character state is estrogen-dependent regulation, while testosterone, luteinizing hormone, and non-hormonal control are derived states. Limited data exist on the underlying regulatory processes of sexual dichromatism in the Eared Dove ( Zenaida auriculata (Des Murs, 1847)). Since the Columbiformes order is close to basal branches, we hypothesized that estrogen and testosterone play a central role in the regulation of crown plumage coloration in the Eared Dove. To test this, we subjected adult males to a forced molt accompanied by an exogenous increase of estradiol and testosterone to determine whether the presence of these hormones during molting modified the spectrophotometric characteristics of the plumage. No significant differences were found between treatments and controls in the colorimetric variables hue, ultraviolet saturation, and brightness. Similarly, the avian visual model showed no perceptible changes in chromatic and achromatic signals in the individual male adults. We also analyzed the effects of estrogen and testosterone on the growth speed of the new feathers. The hormone-treated groups increased in the speed of molting compared to the control group. These results suggest that the Eared Dove is an exception to basal branch models: rather than regulating coloration, sex hormones speed up the growth of the new feathers.

Medical-Surgical Implications of Branching Variation of Human Aortic Arch Known as Bovine Aortic Arch (BAA).

(1) Background: The aortic arch (AA) branching model is challenging, considering the multiple anatomical variations documented in existing research. The bovine aortic arch (BAA) is the most prevalent anatomical variation among these. This variant of AA branching has long been considered a nonsymptomatic malformation, having been discovered incidentally during imaging investigations for other causes. However, more recent studies have demonstrated that BAA shows a frequent association with coarctation of the aorta (CoA), thoracic aortic disease (TAD), and stroke. At the same time, given the current context of increasing activity in the fields of interventional and surgical procedures in the aorta and its branches, it is very important to know the medical-surgical implications of this anatomical variant. (2) Methods: We conducted a comprehensive review using PubMed and Embase, focusing specifically on randomized trials and cohort analyses that examined the medical-surgical implications of BAA. We assessed information related to studied groups, medical procedures, and study outcomes. Initially, we identified 8454 studies, and after rigorous evaluation, we narrowed down our review to 25 articles. (3) Discussions: The intervention consisted of assessing the risks associated with BAA through different imaging investigation methods such as computer tomographic angiography (CTA), magnetic resonance imaging (MRI), or ultrasonography (US). The following results were evaluated: the prevalence of the BAA, the importance of imaging investigations in establishing the diagnosis and the therapeutic management and monitoring the evolution of patients with the BAA, the association of the BAA with CoA, TAD, and stroke, and the potential risks of interventional treatment in patients with the BAA. (4) Conclusions: The prevalence of the BAA differs both between different ethnic groups and between genders. Advanced imaging methods such as CTA and 4D flow MRI allow detailed descriptions of supra-aortic vascular anatomy and information about blood flow velocities, direction, and turbulence in the AA. US remains an easy and valuable imaging investigation, with the potential to detect and correctly diagnose the BAA and its hemodynamic implications. Anatomical variations in the AA are associated with increased rates of TAD, CoA, and stroke, necessitating early diagnosis and increased supervision of patients with such incidentally observed abnormalities. In addition, there is a need to further develop and refine the surgical techniques used and personalize them to the individual characteristics of patients with the BAA.

MCNet: Multivariate long-term time series forecasting with local and global context modeling

Time series data typically exhibit various intra-sequence and inter-sequence correlations, resulting in intricate, intertwined dependencies, which pose challenges for accurately predicting future long-term trends. Previous studies have not fully considered the two correlations, and they also still face challenges of excessive time and memory complexity when dealing with long-term predictions. To address these challenges and establish high-precision prediction models, we propose MCNet that consists of a local branch and a global branch. The local branch aims at capturing short-term variations of intra-sequences, as well as capturing inter-sequence correlations. The global branch models long-term dependencies within sequences. Specifically, the local branch consists only of MLP module, which effectively captures short-term variations of intra-sequences by independently modeling the temporal information within and between patches of the most recent time series. Subsequently, inter-sequence dependencies are captured through the channel interaction module, which further explores more key information to improve the performance of MCNet. Meanwhile, global branch models long-term dependencies within the time series through structured global convolution. Experimental results on multiple popular long-term time series forecasting benchmarks demonstrate that MCNet outperforms state-of-the-art methods, yielding a relative improvement of 12% for multivariate time series while also being more efficient.

A novel plant type, leaf disease and severity identification framework using CNN and transformer with multi-label method.

The growth of plants is threatened by numerous diseases. Accurate and timely identification of these diseases is crucial to prevent disease spreading. Many deep learning-based methods have been proposed for identifying leaf diseases. However, these methods often combine plant, leaf disease, and severity into one category or treat them separately, resulting in a large number of categories or complex network structures. Given this, this paper proposes a novel leaf disease identification network (LDI-NET) using a multi-label method. It is quite special because it can identify plant type, leaf disease and severity simultaneously using a single straightforward branch model without increasing the number of categories and avoiding extra branches. It consists of three modules, i.e., a feature tokenizer module, a token encoder module and a multi-label decoder module. The LDI-NET works as follows: Firstly, the feature tokenizer module is designed to enhance the capability of extracting local and long-range global contextual features by leveraging the strengths of convolutional neural networks and transformers. Secondly, the token encoder module is utilized to obtain context-rich tokens that can establish relationships among the plant, leaf disease and severity. Thirdly, the multi-label decoder module combined with a residual structure is utilized to fuse shallow and deep contextual features for better utilization of different-level features. This allows the identification of plant type, leaf disease, and severity simultaneously. Experiments show that the proposed LDI-NET outperforms the prevalent methods using the publicly available AI challenger 2018 dataset.

Modeling branch attributes and biomass for Catalpa bungei plantations under various fertilization regimes

The development and morphology of branches, a crucial step in producing high-quality large-diameter lumber, may be influenced by fertilization. The response of branch attributes to different fertilization regimes, however, is still poorly understood. The Catalpa bungei plantations, which had been growing for 6 years in northern China, were chosen to study how various fertilization measures affected branch attributes. The two fertilization techniques used were hole fertilization (HF) and water and fertilizer integration (WF), with no fertilization (CK) as a control. The quantity, density, morphology (e.g., diameter, length, and angle), and position (e.g., height and orientation) of branches, and organ biomass of 18 standard trees (total of 516 branches) were investigated. The results demonstrated a considerable increase in tree height, diameter at breast height (DBH), canopy ratio, branch quantity, and organ biomass following the addition of fertilizer. Both the maximum branch diameter and the number of branches rose with fertilization. Following fertilization, the number of branches rose by 16% (HF) and 28% (WF) compared to non-fertilized trees, while the maximum branch diameter increased by 3.5% (HF) and 17.3% (WF), respectively. WF led to an increase in the number of branches and largest branch diameter in comparison to CK and HF. The length, angle, and diameter of branches, however, were not affected significantly by different fertilization treatments. There were roughly equal amounts of branches in four orientations. The mixed-model analysis revealed that the number of branches was positively correlated with branch density and tree height. The branch diameter increased with the increase of branch length and angle. The branch length was negatively correlated with branch height and angle. The branch angle showed a larger angle at the bottom of the canopy. Tree height plus diameter at breast height combined, or just the diameter at breast height indicator alone, can both reliably predict the total biomass of trees. The branch models created in this research may offer some theoretical backing for understanding the crown dynamics of valuable tree species in northern China.

Branch architecture quantification of large-scale coniferous forest plots using UAV-LiDAR data

Branch architecture plays an important role in forest physical and ecological processes, greatly affecting forest spatial structure and the accumulation, production and distribution of organic carbon. Very few studies have quantified the branch architecture of large-scale forest plots, due to the lack of high-resolution large-scale three-dimensional data. The emergence of unmanned aerial vehicle-light detecting and ranging (UAV-LiDAR) provides great potential to overcome this limitation. However, due to insufficient quality of UAV-LiDAR data for branch reconstruction, existing algorithms remain tremendously challenging. We propose an effective branch reconstruction algorithm for UAV-LiDAR data to quantify branch architecture. First, individual branches are extracted using a region growth method that is guided by the branch growth direction and local smoothness constraint. Second, incorrect branches are eliminated based on three pieces of branch features, i.e., specific angles between branches at the same whorl, specific height differences between branches at different whorls, and the growth pattern of branches from the stem outward, reaching maximum distances at tips. Finally, branches are reconstructed using polynomial fitting, and branch architecture parameters are extracted based on branch models. The proposed algorithm simplifies branch reconstruction by skipping the separation of photosynthetic and non-photosynthetic components. It also adapts well to UAV-LiDAR point clouds, generating more realistic reconstructions. The algorithm successfully identified non-photosynthetic components in experiments involving 240 trees, including Scots pine, Norway spruce, and Radiata pine. The average overall accuracy for these three species was 0.88, 0.76, and 0.74, respectively. The proposed algorithm was tested for branch reconstruction using UAV-LiDAR data from a two-hectare Larix principis-rupprechtii plot. The accuracy of branch identification and parameter extraction accuracy were evaluated branch by branch based on the individual branches manually identified in terrestrial laser scanning data. Results showed the F-score of branch and stem identification was 0.58 and 1. The relative RMSE of branch length and angle was 36.87% and 18.3%, and ones of stem length and diameter were 1.46% and 6.16%. The proposed algorithm outperformed the well-known reconstruction algorithms, including TreeQSM, AdTree and Laplacian-based algorithms.

Fully connected-convolutional (FC-CNN) neural network based on hyperspectral images for rapid identification of P. ginseng growth years

P. ginseng is a precious traditional Chinese functional food, which is used for both medicinal and food purposes, and has various effects such as immunomodulation, anti-tumor and anti-oxidation. The growth year of P. ginseng has an important impact on its medicinal and economic values. Fast and nondestructive identification of the growth year of P. ginseng is crucial for its quality evaluation. In this paper, we propose a FC-CNN network that incorporates spectral and spatial features of hyperspectral images to characterize P. ginseng from different growth years. The importance ranking of the spectra was obtained using the random forest method for optimal band selection. Based on the hyperspectral reflectance data of P. ginseng after radiometric calibration and the images of the best five VNIR bands and five SWIR bands selected, the year-by-year identification of P. ginseng age and its identification experiments for food and medicinal purposes were conducted, and the FC-CNN network and its FCNN and CNN branch networks were tested and compared in terms of their effectiveness in the identification of P. ginseng growth years. It has been experimentally verified that the best year-by-year recognition was achieved by utilizing images from five visible and near-infrared important bands and all spectral curves, and the recognition accuracy of food and medicinal use reached 100%. The FC-CNN network is significantly better than its branching model in the effect of edible and medicinal identification. The results show that for P. ginseng growth year identification, VNIR images have much more useful information than SWIR images. Meanwhile, the FC-CNN network utilizing the spectral and spatial features of hyperspectral images is an effective method for the identification of P. ginseng growth year.