Tree wind load assessment method based on point cloud model

Abstract

The invention relates to a tree wind load assessment method based on a point cloud model. The tree wind load assessment method comprises the steps of performing semantic segmentation on an image collected by an unmanned aerial vehicle, extracting a tree crown part image for three-dimensional reconstruction, generating a point cloud model, extracting point cloud data of a tree crown and a tree trunk, and generating a three-dimensional point cloud model of a target tree. And calculating the wind load influence area of the tree crown through an Alpha Shape algorithm. In combination with wind speed, wind direction and tree geometric characteristics, ANSYS Fluent software is used for simulating wind load distribution under different wind direction and wind speed conditions, and a wind load result under the wind load effect is calculated. And finally, based on a finite element analysis method, performing tree structure stress analysis in combination with the wind load data, and evaluating the fatigue life and the damage risk of the tree under the wind action. The method can accurately evaluate the dynamic response of the tree under the action of wind power, has the characteristics of high efficiency, accuracy and automation, and is widely applied to the fields of wind disaster early warning, tree health monitoring, environmental protection and the like.

Description

Technical Field

[0001] The invention belongs to the technical field of tree assessment, and in particular relates to a tree wind load assessment method based on a point cloud model. Background Art

[0002] With climate change and the increase in extreme weather events, the threat posed by wind disasters to cities has gradually intensified. The behavior of trees under wind, including bending, swaying and even breaking, has become an important part of wind disaster prevention research. Especially in high-wind weather, the wind resistance of trees is directly related to environmental safety, public facilities, and the safety of people's lives and property. At present, the assessment of tree wind loads mainly relies on manual observations and traditional wind models, but these methods usually have problems such as insufficient assessment accuracy, poor real-time performance and limited scope of application.

[0003] Traditional methods for evaluating tree wind loads mainly use simple wind tunnel experiments or estimation methods based on past experience, which cannot accurately reflect the dynamic response of trees in complex environments. With the development of science and technology, images collected by drones and 3D reconstruction technology have provided new possibilities for tree wind load evaluation. However, in the existing technology, research on how to use images to generate accurate tree point cloud models and perform automatic evaluation and structural analysis of wind loads based on these models is still in its infancy, and the comprehensive analysis of multi-directional wind loads has not yet been effectively solved. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a tree wind load assessment method based on a point cloud model.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] In view of the shortcomings of the existing technology, the present invention proposes a tree wind load assessment method based on a point cloud model. This method uses drone image acquisition, semantic segmentation and point cloud reconstruction technology, combined with wind load calculation and tree structure stress analysis, to more accurately assess the wind resistance of trees under wind force and provide a scientific basis for wind disaster warning and emergency response.

[0007] The present invention provides a tree wind load assessment method based on a point cloud model, comprising the following steps:

[0008] Use drones to shoot target trees from multiple angles to obtain images of target trees;

[0009] Annotate the image using an image semantic segmentation algorithm to generate a label for each pixel in the image, wherein the label includes a tree crown, a tree trunk, and a background;

[0010] Constructing a preliminary three-dimensional point cloud model based on the annotated image, wherein each point cloud data in the preliminary three-dimensional point cloud model includes label information;

[0011] Generate a 3D point cloud model of the target tree based on the preliminary 3D point cloud model and its label information;

[0012] Calculating the wind load impact area of ​​the tree crown according to the three-dimensional point cloud model of the target tree;

[0013] Based on the wind load impact area of ​​the tree canopy, wind speed, and geometric characteristics of the target tree, multi-directional wind loads are simulated to obtain wind load simulation results, and the geometric characteristics of the target tree are obtained according to a three-dimensional point cloud model of the target tree;

[0014] Combined with the wind load simulation results, tree structure stress analysis is carried out to predict the risk of tree damage under wind action.

[0015] Furthermore, the image is annotated by using an image semantic segmentation algorithm to generate a label for each pixel in the image, including the following steps:

[0016] Perform preprocessing operations on the collected images, including denoising and color correction;

[0017] The preprocessed image is input into a pretrained convolutional neural network model for semantic segmentation, and a semantic segmentation map of each image is output, in which each pixel in the semantic segmentation map is assigned a label. The convolutional neural network model includes a MaskR-CNN model.

[0018] Furthermore, constructing a preliminary three-dimensional point cloud model based on the annotated image includes the following steps:

[0019] Use COLMAP 3D reconstruction software to perform multi-view 3D reconstruction on the acquired images and generate a point non-semantic 3D point cloud model;

[0020] Calculate the 3D coordinates of each pixel in the semantic segmentation map of each image in the non-semantic 3D point cloud model, determine the point cloud data corresponding to the pixel according to the 3D coordinates of each pixel, transfer the label of each pixel to the corresponding point cloud data, and generate a preliminary 3D point cloud model.

[0021] Furthermore, the method of using COLMAP 3D reconstruction software to perform multi-view 3D reconstruction on the collected images to generate a point non-semantic 3D point cloud model includes the following steps:

[0022] The SIFT algorithm is used to extract feature points and their descriptors from each image, the nearest neighbor matching algorithm is used to match feature points between multiple images based on the feature points and their descriptors of each image, and the RANSAC algorithm is used to eliminate mismatched feature points.

[0023] Based on the matched feature points, COLMAP is used to construct a non-semantic 3D point cloud model through an incremental reconstruction method.

[0024] Furthermore, generating a target tree three-dimensional point cloud model based on the preliminary three-dimensional point cloud model and its label information includes the following steps:

[0025] Through the semantic segmentation map of each image and the geometric features of each point cloud data in the preliminary 3D point cloud model, the geometric conditions based on color, position and normal vector are used to remove the debris not related to the tree;

[0026] Extracting point cloud data labeled as tree crowns and tree trunks from the preliminary 3D point cloud model;

[0027] Use voxel gridding algorithm to simplify the extracted point cloud data, grid the point cloud in space, and reduce the number of redundant points;

[0028] The SOR filtering algorithm is used to remove noise from the point cloud data and generate a three-dimensional point cloud model of the target tree.

[0029] Furthermore, the calculation of the wind load influence area of ​​the tree crown according to the three-dimensional point cloud model of the target tree includes the following steps:

[0030] Extracting point cloud data labeled as tree crown from the three-dimensional point cloud model of the target tree to generate a three-dimensional point cloud model of the tree crown;

[0031] Projecting the tree crown point cloud data in the tree crown three-dimensional point cloud model along a preset direction, and converting the tree crown three-dimensional point cloud model into two-dimensional plane data;

[0032] The three-dimensional coordinates in the three-dimensional point cloud model of the tree crown are converted into corresponding two-dimensional coordinates according to a preset direction to obtain a two-dimensional point set of the tree crown;

[0033] The Alpha Shape algorithm is used to calculate the boundary envelope of the two-dimensional point set of the tree crown to obtain the wind load influence area of ​​the tree crown.

[0034] Furthermore, the preset direction is dynamically adjusted according to changes in wind speed and wind direction.

[0035] Furthermore, the wind load influence area of ​​the tree canopy is calculated as follows:

[0036]

[0037] Where A is the wind load influence area of ​​the tree canopy, (x i ,y i ) is the two-dimensional coordinate of the i-th point cloud data in the crown two-dimensional point set, n is the total number of point clouds in the crown two-dimensional point set, (x n+1 ,yn+1 ) is considered as (x 1 ,y 1 ) forms a closed path.

[0038] Furthermore, simulating multi-directional wind loads based on the wind load impact area of ​​the tree crown, wind speed, and geometric characteristics of the target tree to obtain wind load simulation results includes the following steps:

[0039] The wind resistance coefficients at the crown and trunk of the tree are calculated according to the geometric characteristics of the target tree. The three-dimensional point cloud model of the crown is converted into a three-dimensional CAD model and imported into Fluent. The wind resistance coefficients at the crown and trunk are input into ANSYS Fluent software. Fluent is used to simulate the different wind speeds and wind pressure distributions generated in multiple directions. The flow field simulation results are analyzed to obtain the wind load simulation results. The formula is:

[0040]

[0041] Among them, F is the wind load, C is the d is the drag coefficient, A is the wind load influence area of ​​the tree crown, ρ is the air density, and v is the wind speed.

[0042] Furthermore, the tree structure stress analysis is performed in combination with the wind load simulation results to predict the damage risk of the tree under the action of wind, including the following steps:

[0043] Based on the three-dimensional point cloud model of the target tree, a finite element model of the tree is constructed, wherein the finite element model includes the geometric shape, material properties, mechanical properties and boundary conditions of the tree;

[0044] The wind load simulation results are input into the finite element model as external loads acting on each node or element of the finite element model to calculate the structural stress of the tree under the wind load. The formula is:

[0045]

[0046] Among them, σ is stress, F is wind load, A cross is the cross-sectional area of ​​the wind load action area;

[0047] Based on the calculated structural stress, combined with the fatigue strength and fatigue index of the target tree material, the fatigue life of the target tree under long-term wind load is predicted using the formula:

[0048]

[0049] Among them, N f is the fatigue life, Δσ is the stress amplitude, obtained according to the calculated structural stress, σ fis the fatigue strength of the target tree material, β is the fatigue index of the target tree material;

[0050] Based on the predicted fatigue life of the target trees under long-term wind loads, the damage risk of the target trees under wind loads is assessed.

[0051] Compared with the prior art, the present invention has the following advantages:

[0052] (1) The present invention solves the deficiencies of the prior art in wind load assessment and fatigue damage analysis of trees. Traditional methods often rely on static wind speed data and simplified mechanical models, which make it difficult to accurately simulate the dynamic response of trees in complex wind environments and cannot comprehensively assess the stress state of trees and their potential risks. The present invention uses advanced CFD simulation technology and tree geometric modeling methods to accurately simulate the dynamic response of trees under different wind speeds and wind directions, and through stress analysis and fatigue calculations, to assess the structural stability of trees in real time. Compared with the prior art, the present invention can overcome the defects of traditional methods that ignore the dynamic response of trees and the complex wind load effects, and provide a more accurate and comprehensive wind load assessment. Combined with urban emergency management systems, disaster prevention and mitigation systems, etc., this system can realize the prediction and analysis of tree damage in strong wind weather, provide a scientific basis for wind disaster warning and emergency response, and thus reduce the loss of public facilities and human life and property in wind disasters.

[0053] (2) The present invention uses drone image acquisition and image semantic segmentation algorithms to achieve high-precision three-dimensional point cloud model construction of target trees. This method avoids the high cost and low efficiency of traditional manual measurement and can accurately extract the structural information of trees (such as crowns, trunks, etc.), thereby providing more accurate geometric data for subsequent wind load analysis.

[0054] (3) The present invention constructs a point cloud model based on COLMAP 3D reconstruction software, combines the image semantic segmentation results, and extracts the 3D geometric information of trees through an automated algorithm, which greatly improves the efficiency and accuracy of point cloud model construction. This method can effectively remove background debris and noise, ensuring the accuracy and operability of the tree model.

[0055] (4) The present invention uses the Alpha Shape algorithm to calculate the wind load influence area of ​​the tree crown, which can dynamically adjust the wind load influence area of ​​the tree model under different wind direction conditions and accurately calculate the actual exposed area of ​​the tree crown, providing reliable data support for the subsequent simulation and evaluation of wind loads.

[0056] (5) The present invention combines Fluent software to simulate multi-directional wind loads. By inputting the three-dimensional CAD model of the tree and the wind resistance coefficient, the multi-directional wind speed and wind pressure distribution are simulated, thereby obtaining accurate wind load simulation results. This technical means can comprehensively analyze wind loads under different wind directions and wind speeds, providing a more scientific basis for the structural evaluation of trees under wind force.

[0057] (6) The present invention combines finite element models to perform tree structural stress analysis. By calculating the structural stress of trees under wind loads and combining the fatigue strength and fatigue index of tree materials, the fatigue life of trees under long-term wind loads can be accurately predicted. This technology can provide a scientific basis for tree health assessment and wind damage risk prediction, and help achieve early warning and management of tree wind damage risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 is a flow chart of the overall method of the present invention;

[0059] Figure 2 It is the data collection and semantic segmentation flow chart of the present invention;

[0060] Figure 3 A flow chart for point cloud model generation of the present invention;

[0061] Figure 4 A flow chart of wind load simulation calculation of the present invention;

[0062] Figure 5 The flowchart of the tree trunk stress analysis and fracture prediction of the present invention is shown in FIG. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0064] Embodiment 1:

[0065] This embodiment provides a tree wind load assessment method based on a point cloud model, which can efficiently and accurately assess the damage risk of trees under wind force through technical means such as drone image acquisition, image semantic segmentation, three-dimensional point cloud reconstruction, wind load simulation and structural stress analysis. Figure 1 As shown, the following steps are included:

[0066] Use drones to shoot target trees from multiple angles to obtain images of target trees;

[0067] Annotate the image using the image semantic segmentation algorithm to generate a label for each pixel in the image;

[0068] Construct a preliminary 3D point cloud model based on the annotated images;

[0069] Generate a 3D point cloud model of the target tree based on the preliminary 3D point cloud model and its label information;

[0070] Calculate the wind load impact area of ​​the tree canopy based on the three-dimensional point cloud model of the target tree;

[0071] Based on the wind load impact area of ​​the tree crown, wind speed, and geometric characteristics of the target trees, multi-directional wind loads are simulated to obtain wind load simulation results;

[0072] Combined with the wind load simulation results, tree structure stress analysis is carried out to predict the risk of tree damage under wind action.

[0073] Specifically include the following:

[0074] The target trees are photographed from multiple angles by drones to obtain image data of the trees. The images collected by drones have high resolution and accuracy, and can obtain comprehensive and detailed information about the appearance of the trees. The advantage of this step is that through the efficient shooting of drones, a large number of high-quality images can be obtained in a short time, avoiding the inefficiency and high cost of traditional manual measurement.

[0075] After acquiring the image, the image is annotated using an image semantic segmentation algorithm. Each pixel in the image is assigned label information, including the crown, trunk, and background. This process is performed through a pre-trained convolutional neural network (such as MaskR-CNN), which can achieve accurate recognition of different parts of the tree. Through this technology, the crown, trunk, and background can be effectively separated, and data in irrelevant areas can be removed, ensuring that the extraction of tree parts in subsequent processing is more accurate. The advantage of this technical feature is that it can efficiently process images and achieve high-precision area annotation, reducing manual intervention and improving processing speed and accuracy.

[0076] Next, a preliminary 3D point cloud model is constructed based on the annotated images. The COLMAP 3D reconstruction software is used to perform multi-view 3D reconstruction on the acquired images to generate a point non-semantic 3D point cloud model. This process extracts image feature points through the SIFT algorithm, and uses the RANSAC algorithm to remove mismatched points, ultimately achieving the fusion of multi-view point cloud data. Through this step, a more accurate 3D point cloud model of trees can be generated, providing basic data for subsequent wind load analysis. The advantage of this technical feature is that it utilizes the geometric relationship between images and automatically processes them through algorithms, greatly improving the efficiency and accuracy of model construction and avoiding the tediousness and errors of manual modeling.

[0077] The information of each part of the tree is extracted from the preliminary 3D point cloud model to generate a 3D point cloud model of the target tree. In this process, based on the geometric conditions of color, position and normal vector, the debris irrelevant to the tree is removed, and the point cloud data labeled as the crown and trunk are extracted. The point cloud is simplified by using a voxel gridding algorithm to further reduce the number of redundant points, and the SOR filtering algorithm is used to remove noise in the point cloud. The final generated 3D point cloud model of the target tree accurately reflects the geometric shape of the tree and provides a reliable data source for the calculation of the area affected by the wind load of the tree. The advantage of this technical feature is that by removing debris, simplifying data and denoising, the accuracy and effectiveness of the point cloud data are ensured, and the errors that may occur in subsequent analysis are reduced.

[0078] Based on the three-dimensional point cloud model of the target tree, the wind load influence area of ​​the tree crown is calculated. First, the point cloud data labeled as the tree crown is extracted from the three-dimensional point cloud model of the target tree to generate a three-dimensional point cloud model of the tree crown. Then, the three-dimensional point cloud model of the tree crown is projected along the preset direction to obtain the two-dimensional plane data of the tree crown. Next, the three-dimensional coordinates of the three-dimensional point cloud data of the tree crown are converted into two-dimensional coordinates to obtain the two-dimensional point set of the tree crown. The Alpha Shape algorithm is used to perform boundary envelope calculation on the two-dimensional point set of the tree crown to obtain the wind load influence area of ​​the tree crown. The specific calculation formula is:

[0079]

[0080] Where A is the wind load influence area of ​​the tree canopy, (x i ,y i ) is the two-dimensional coordinate of the i-th point cloud data in the crown two-dimensional point set, n is the total number of point clouds in the crown two-dimensional point set, (x n+1 ,y n+1 ) is considered as (x 1 ,y 1 ) forms a closed path. Through the calculation of two-dimensional projection and Alpha Shape algorithm, the effective wind load influence area of ​​the tree crown can be accurately obtained, thus providing an accurate basis for wind load simulation.

[0081] In the wind load simulation stage, the wind resistance coefficient of the tree crown and trunk is calculated according to the geometric characteristics of the target tree, and the three-dimensional point cloud model of the tree crown is converted into a three-dimensional CAD model and imported into the ANSYS Fluent software. Fluent is used to simulate the different wind speeds and wind pressure distributions generated in multiple directions, and finally the wind load simulation results are obtained. The wind load calculation formula is:

[0082]

[0083] Among them, F is the wind load, C is the dis the drag coefficient, A is the wind load influence area of ​​the tree canopy, ρ is the air density, and v is the wind speed. By accurately simulating the wind load effect on trees at different wind speeds and wind directions, reliable data is provided for structural stress analysis. Using Fluent to simulate wind field flow can take into account multiple wind directions and different wind speed conditions, providing comprehensive and accurate results for wind load analysis of trees.

[0084] Combined with the wind load simulation results, the tree structural stress analysis is performed. Based on the 3D point cloud model of the target tree, a finite element model of the tree is constructed. The model includes the tree's geometry, material properties, mechanical properties, and boundary conditions. The wind load simulation results are input into the finite element model and acted on each node or element as an external load to calculate the structural stress of the tree under the wind load. The formula for calculating the structural stress is:

[0085]

[0086] Among them, σ is stress, F is wind load, A cross is the cross-sectional area of ​​the wind load action area; based on the calculated structural stress, combined with the fatigue strength and fatigue index of the tree material, the fatigue life of the tree under long-term wind load is predicted. The fatigue life calculation formula is:

[0087]

[0088] Among them, N f is the fatigue life, Δσ is the stress amplitude, obtained according to the calculated structural stress, σ f is the fatigue strength of the target tree material, and β is the fatigue index of the target tree material; through this step, the fatigue damage and destruction risk of trees under wind can be accurately assessed. The advantage of its technical features is that through the calculation of structural stress and fatigue life, it can provide a scientific basis for the assessment of the health status of trees and provide early warning of the damage risk caused by wind disasters.

[0089] In summary, this embodiment systematically completes the damage risk assessment of trees under wind force through technical means such as drone image acquisition, image semantic segmentation, 3D point cloud reconstruction, wind load simulation and finite element analysis. This method is efficient and accurate, and can provide reliable technical support for wind disaster prevention and control, tree health monitoring and other fields.

[0090] Embodiment 2:

[0091] The parts not mentioned in this embodiment are the same as those in Embodiment 1.

[0092] This embodiment provides a tree wind load assessment system based on a tree point cloud model, including: a data acquisition and semantic segmentation module, a point cloud model generation and processing module, a wind load simulation calculation module, and a tree trunk stress analysis and fracture prediction module. Each module works together to provide a complete solution from data acquisition to final wind disaster prediction.

[0093] The data acquisition and semantic extraction module includes a data acquisition submodule and an image semantic segmentation submodule. It collects images through drones and uses neural networks to perform semantic segmentation of the images, outputting semantic segmentation maps of each image. Each pixel in the semantic segmentation map is assigned a label, including the crown, trunk, and background.

[0094] The data acquisition submodule uses a drone to take photos of the target trees from multiple angles using a high-resolution camera, ensuring that the images cover the entire tree area and are as high quality as possible. The flight path of the drone is precisely planned to ensure image overlap and coverage, facilitating subsequent 3D reconstruction.

[0095] The image semantic segmentation submodule performs semantic segmentation on the images taken by the drone and uses an open source neural network model to automatically identify the tree crown area. This step can effectively distinguish the tree crown from other non-target areas such as the trunk and environmental debris, avoiding noise interference in the data. The model can handle complex background and lighting conditions.

[0096] The point cloud model generation and processing module includes a point cloud model generation submodule and a noise reduction processing submodule. The preliminary data obtained by the drone image acquisition and semantic extraction modules are further processed to generate a complete point cloud model.

[0097] The point cloud model generation submodule uses open source 3D reconstruction software such as COLMAP to perform multi-view 3D reconstruction of the collected images and generate point cloud data. This process uses the SIFT algorithm to extract image feature points, combines the estimation of camera position and posture, and performs incremental 3D reconstruction through nearest neighbor matching, relative posture constraints, and five-point algorithm. The final output point cloud data is a 3D surface model of the tree, including all point cloud data of the crown, trunk, and background environment.

[0098] The denoising submodule uses the above semantic segmentation results and geometric features of the point cloud data to remove debris that is not related to the tree using geometric conditions based on features such as color, position, and normal vectors. The point cloud data is simplified using a voxel gridding algorithm, and the number of redundant points is reduced by gridding the point cloud in space, thereby reducing the amount of calculation. In addition, the SOR filtering algorithm is used to remove noise from the point cloud data, eliminating noise data caused by factors such as lighting changes and weather effects, in order to obtain uniform density and high-accuracy point cloud data for the crown area.

[0099] The wind load simulation calculation module includes the affected area calculation submodule and the wind load simulation calculation submodule. Based on the point cloud model obtained previously, this module combines parameters such as wind speed and tree structure to calculate the effect of wind load on trees. This module uses physical models and computational fluid dynamics simulation to calculate the wind load on trees under different wind speeds and wind directions, and evaluates the structural stability of trees:

[0100] The affected area calculation submodule projects the crown point cloud data along a given direction. The projection process converts the three-dimensional point cloud into two-dimensional plane data to form a point set. The point set represents the cross-sectional shape of the crown in that direction. The projection direction is dynamically adjusted according to the changes in wind speed and direction. In view of the irregular graphics of the crown part, this module uses the Alpha Shape algorithm to calculate the envelope of the point set. By performing Alpha Shape processing on the crown point cloud data, the shape of the crown can be accurately captured and its wind load affected area can be calculated.

[0101] The wind load simulation submodule measures the complexity or roughness of the surface by analyzing the density of the point cloud and calculates the drag coefficient of the tree under different projection forms. For example, dense areas (such as leaves) have a higher drag coefficient, while sparse areas (such as bare branches) have a lower drag coefficient.

[0102] The extracted tree crown point cloud data is used to generate a 3D CAD model and imported into Fluent. The function of the drag coefficient is input into ANSYS Fluent software, and Fluent is used to simulate the different wind speeds and wind pressure distributions generated in multiple directions. By analyzing the flow field simulation results, the pressure distribution and wind load results of each area can be obtained, and the structural stress and fracture risk of each tree part can be further calculated.

[0103] Due to the influence of factors such as the shape and layout of tree crowns, branches and leaves on airflow. When analyzing a tree patch, trees will create obstacles to airflow, change the velocity and pressure distribution, and this local wind flow change will affect the calculation of wind load. This module uses Fluent to simulate the wind channel effect, and provides visualization means to realize streamlines, velocity vector diagrams and wind pressure distribution diagrams, intuitively displaying the direction and velocity of wind flow, thereby showing the possible intensification or weakening of wind in local areas of the forest, to help identify areas that may be severely affected.

[0104] The tree trunk stress analysis and fracture prediction module uses the wind pressure and other results obtained from the above CFD simulation calculations, combined with ANSYS Mechanical structural analysis software, to realize the deformation and displacement analysis of trees under wind loads and the dynamic response analysis under wind loads, and then determine the risk of tree fracture when encountering strong winds.

[0105] The point cloud model and material properties of the trees were input into the software, and the wind load data information obtained from the CFD simulation was input to perform structural stress analysis.

[0106] Fluid-Structure Interaction (FSI) analysis is used to simulate the deformation and structural response of trees under wind loads. This method not only considers the static effect of wind loads, but also simulates the transient response of trees under the dynamic action of wind, such as vibration and swaying of trees.

[0107] Fatigue fracture analysis: Trees experience periodic wind loads in the wind, which can lead to fatigue damage. ANSYS Mechanical can perform fatigue analysis to evaluate the fatigue life of trees under long-term wind loads and identify areas where fatigue damage may occur. Wind loads can easily cause trees to break or be damaged in vulnerable parts of trees, such as where branches and trunks meet. Fracture mechanics analysis can be used to simulate the fracture caused by wind loads to help determine the risk of trees breaking when encountering strong winds.

[0108] Instability and collapse analysis: Taller trees may become unstable and collapse when encountering strong winds. The wind load simulation results can be used to assess the risk of instability and collapse of trees under strong winds.

[0109] Embodiment 3:

[0110] The parts not mentioned in this embodiment are the same as those in Embodiment 1.

[0111] This embodiment relates to a tree wind load assessment system based on a tree point cloud model, which mainly consists of the following parts:

[0112] The data acquisition and semantic analysis module includes a data acquisition submodule and an image semantic segmentation submodule.

[0113] The point cloud model generation and processing module includes a point cloud model generation submodule and a noise reduction processing submodule.

[0114] The wind load simulation calculation module includes an affected area calculation submodule and a wind load simulation calculation submodule.

[0115] The tree trunk stress analysis and fracture prediction module includes fluid-structure coupling and fatigue and instability analysis submodules.

[0116] The data collection and semantic segmentation process is as follows: Figure 2 As shown, the following steps are included:

[0117] The drone flies and captures images according to a predetermined trajectory. The flight trajectory should take into account the overlapping pixels of images in various directions and cover all angles of the tree for subsequent 3D reconstruction.

[0118] Pre-process the images extracted by the drone to remove the unclear images due to occlusion and light occlusion, and import the images into the neural network;

[0119] The neural network performs semantic segmentation on the processed images, accurately distinguishes trees from the background, and extracts the crown area as the source image for subsequent point cloud model generation.

[0120] The process of point cloud model generation is as follows: Figure 3 As shown, the following steps are included:

[0121] The results of semantic segmentation are imported into 3D reconstruction software, which converts the image into 3D point cloud data;

[0122] Use voxel gridding and SOR filtering algorithms to simplify point cloud data, remove redundancy and noise, and ensure the clarity and accuracy of point cloud data;

[0123] The software uses the SIFT algorithm to extract key feature points, and performs 3D reconstruction based on the extracted feature points to generate a 3D surface model of the tree. It uses geometric conditions based on features such as color, position, and normal vectors to remove debris that is not related to the tree, ensuring that the point cloud model has a high degree of 3D restoration and can accurately reflect the 3D information of the tree for subsequent wind load calculations.

[0124] The process of wind load simulation calculation is as follows Figure 4 As shown, the following steps are included:

[0125] Project the tree crown point cloud data along a given direction. The data obtained by two-dimensional plane projection forms a point set, which represents the cross-sectional shape of the tree crown in this direction. Use the Alpha Shape algorithm to select a suitable radius parameter to calculate the area of ​​the irregular image, which is the area affected by the wind load in this direction.

[0126] By analyzing the density of the point cloud to measure the complexity or roughness of the surface, the drag coefficient of the tree in different projection forms is calculated for CFD simulation;

[0127] The 3D point cloud data and drag coefficient function information are imported into ANSYS Fluent for multi-directional wind speed and wind pressure distribution simulation. The software can simulate wind channel effects and provide visualization to help identify areas susceptible to wind loads. CFD simulation obtains data such as pressure field and wind load distribution, and can display the streamlines and vortex structures formed by fluid around trees.

[0128] The process of tree trunk stress analysis and fracture prediction is as follows Figure 5 As shown, the following steps are included:

[0129] Import the wind pressure data obtained after wind load simulation, together with the point cloud model and the material properties of the tree, into ANSYS Mechanical structural analysis software for analysis;

[0130] Through fluid-structure interaction (FSI) analysis, the deformation and structural changes of trees under wind loads are simulated, and the displacement and vibration responses of trees under different wind effects can be analyzed;

[0131] Combined with the FSI analysis results, fatigue fracture risk analysis is conducted on parts of trees that are prone to fatigue damage (such as the connection between branches and trunks), and the risk of instability and collapse of taller trees under strong winds is assessed. The risk assessment results can be reflected in the upper-level management system to further realize the risk warning function.

[0132] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0133] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A tree wind load assessment method based on a point cloud model, characterized in that: The following steps are involved: Use drones to shoot target trees from multiple angles to obtain images of target trees; Annotate the image using an image semantic segmentation algorithm to generate a label for each pixel in the image, wherein the label includes a tree crown, a tree trunk, and a background; Constructing a preliminary three-dimensional point cloud model based on the annotated image, wherein each point cloud data in the preliminary three-dimensional point cloud model includes label information; Generate a 3D point cloud model of the target tree based on the preliminary 3D point cloud model and its label information; Calculating the wind load impact area of ​​the tree crown according to the three-dimensional point cloud model of the target tree; Based on the wind load impact area of ​​the tree canopy, wind speed, and geometric characteristics of the target tree, multi-directional wind loads are simulated to obtain wind load simulation results, and the geometric characteristics of the target tree are obtained according to a three-dimensional point cloud model of the target tree; Combined with the wind load simulation results, tree structure stress analysis is carried out to predict the risk of tree damage under wind action.

2. A tree wind load assessment method based on a point cloud model according to claim 1, characterized in that: The process of labeling the image by using an image semantic segmentation algorithm and generating a label for each pixel in the image includes the following steps: Perform preprocessing operations on the collected images, including denoising and color correction; The preprocessed image is input into a pretrained convolutional neural network model for semantic segmentation, and a semantic segmentation map of each image is output, in which each pixel in the semantic segmentation map is assigned a label. The convolutional neural network model includes a Mask R-CNN model.

3. A tree wind load assessment method based on a point cloud model according to claim 1 or 2, characterized in that: The method of constructing a preliminary three-dimensional point cloud model based on the annotated image includes the following steps: Use COLMAP 3D reconstruction software to perform multi-view 3D reconstruction on the acquired images and generate a point non-semantic 3D point cloud model; Calculate the 3D coordinates of each pixel in the semantic segmentation map of each image in the non-semantic 3D point cloud model, determine the point cloud data corresponding to the pixel according to the 3D coordinates of each pixel, transfer the label of each pixel to the corresponding point cloud data, and generate a preliminary 3D point cloud model.

4. The tree wind load assessment method based on point cloud model according to claim 3 is characterized in that: The method of using COLMAP 3D reconstruction software to perform multi-view 3D reconstruction on the collected images to generate a point non-semantic 3D point cloud model includes the following steps: The SIFT algorithm is used to extract feature points and their descriptors from each image, the nearest neighbor matching algorithm is used to match feature points between multiple images based on the feature points and their descriptors of each image, and the RANSAC algorithm is used to eliminate mismatched feature points. Based on the matched feature points, COLMAP is used to construct a non-semantic 3D point cloud model through an incremental reconstruction method.

5. A tree wind load assessment method based on a point cloud model according to claim 1 or 2, characterized in that: The method of generating a target tree 3D point cloud model based on the preliminary 3D point cloud model and its label information includes the following steps: Through the semantic segmentation map of each image and the geometric features of each point cloud data in the preliminary 3D point cloud model, the geometric conditions based on color, position and normal vector are used to remove the debris not related to the tree; Extracting point cloud data labeled as tree crowns and tree trunks from the preliminary 3D point cloud model; Use voxel gridding algorithm to simplify the extracted point cloud data, divide the point cloud into grids in space, and reduce the number of redundant points; The SOR filtering algorithm is used to remove noise from the point cloud data and generate a three-dimensional point cloud model of the target tree.

6. The tree wind load assessment method based on point cloud model according to claim 1, characterized in that: The calculation of the wind load influence area of ​​the tree crown according to the three-dimensional point cloud model of the target tree includes the following steps: Extracting point cloud data labeled as tree crown from the three-dimensional point cloud model of the target tree to generate a three-dimensional point cloud model of the tree crown; Projecting the tree crown point cloud data in the tree crown three-dimensional point cloud model along a preset direction, and converting the tree crown three-dimensional point cloud model into two-dimensional plane data; The three-dimensional coordinates in the three-dimensional point cloud model of the tree crown are converted into corresponding two-dimensional coordinates according to a preset direction to obtain a two-dimensional point set of the tree crown; The Alpha Shape algorithm is used to perform boundary envelope calculation on the two-dimensional point set of the tree crown to obtain the wind load influence area of ​​the tree crown.

7. The tree wind load assessment method based on point cloud model according to claim 6 is characterized in that: The preset direction is dynamically adjusted according to changes in wind speed and wind direction.

8. The tree wind load assessment method based on point cloud model according to claim 6 is characterized in that: The wind load influence area of ​​the tree canopy is calculated as follows: Where A is the wind load influence area of ​​the tree canopy, (x i ,y i ) is the two-dimensional coordinate of the i-th point cloud data in the crown two-dimensional point set, n is the total number of point clouds in the crown two-dimensional point set, (x n+1 ,y n+1 ) is considered as (x1,y1) forming a closed path.

9. A tree wind load assessment method based on a point cloud model according to claim 1 or 6, characterized in that: The method of simulating multi-directional wind loads based on the wind load impact area of ​​the tree crown, wind speed, and geometric characteristics of the target tree to obtain wind load simulation results includes the following steps: The wind resistance coefficients at the crown and trunk of the tree are calculated according to the geometric characteristics of the target tree. The three-dimensional point cloud model of the crown is converted into a three-dimensional CAD model and imported into Fluent. The wind resistance coefficients at the crown and trunk are input into ANSYS Fluent software. Fluent is used to simulate the different wind speeds and wind pressure distributions generated in multiple directions. The flow field simulation results are analyzed to obtain the wind load simulation results. The formula is: Among them, F is the wind load, C is the d is the drag coefficient, A is the wind load influence area of ​​the tree crown, ρ is the air density, and v is the wind speed.

10. The tree wind load assessment method based on point cloud model according to claim 1, characterized in that: The method of combining the wind load simulation results to perform tree structure stress analysis and predicting the damage risk of trees under wind force includes the following steps: Based on the three-dimensional point cloud model of the target tree, a finite element model of the tree is constructed, wherein the finite element model includes the geometric shape, material properties, mechanical properties and boundary conditions of the tree; The wind load simulation results are input into the finite element model as external loads acting on each node or element of the finite element model to calculate the structural stress of the tree under the wind load. The formula is: Among them, σ is stress, F is wind load, A cross is the cross-sectional area of ​​the wind load action area; Based on the calculated structural stress, combined with the fatigue strength and fatigue index of the target tree material, the fatigue life of the target tree under long-term wind load is predicted using the formula: Among them, N f is the fatigue life, Δσ is the stress amplitude, obtained according to the calculated structural stress, σ f is the fatigue strength of the target tree material, β is the fatigue index of the target tree material; Based on the predicted fatigue life of the target trees under long-term wind loads, the damage risk of the target trees under wind loads is assessed.

Images