Forest canopy three-dimensional height evaluation method and device based on canopy structure connectivity

Abstract

The invention provides a forest canopy three-dimensional height evaluation method and device based on canopy structure connectivity, and relates to the technical field of monitoring three-floor canopy height by laser radar. The method comprises the following steps: acquiring forest point cloud three-dimensional data of a target area acquired in real time through an unmanned aerial vehicle laser radar; preprocessing is carried out, and a digital elevation model (DEM) and a digital earth surface model (DSM) of the target area are generated; generating a canopy height model CHM by calculating a difference value between the DSM and the DEM; calculating an optimal path of each grid of the CHM from a starting point to a target point based on a Dijkstra algorithm according to a cost curved surface generated by the CHM; and according to the optimal path, obtaining a canopy structure connectivity value, namely a CSC value, of each canopy unit by calculating the connectivity between each node of the CHM and the height of the similar canopy unit, thereby obtaining a forest canopy three-dimensional height model based on CSC value spatial distribution for further analysis and evaluation. According to the method, by calculating the CSC, the spatial heterogeneity of the forest canopy structure and the influence of the spatial heterogeneity on the ecological process are comprehensively analyzed.

Description

Technical Field

[0001] The present invention relates to the technical field of lidar monitoring of forest canopy height. Specifically, it relates to a method and device for evaluating the three-dimensional height of a forest canopy based on canopy structure connectivity. Background Art

[0002] The ecosystem services of forests largely depend on forest structure, especially the vertical distribution of the forest canopy. The vertical structure of forests not only determines the flow and distribution of light, matter, and energy among different layers, but also directly affects the unity and adaptability of forest ecosystems.

[0003] Although existing research has revealed the important role of forest canopy structure in light energy utilization and ecological functions, most research still focuses on the canopy height distribution at the macro scale or the analysis of single indicators, and fails to systematically quantify the relationships between canopy vertical structures and their ecological regulation mechanisms. In addition, limited by data resolution and spatial scale, traditional observation methods are difficult to comprehensively analyze the spatial heterogeneity of canopy structure and its specific impact on ecological processes. These deficiencies affect the accuracy of constructing the three-dimensional height model of the forest canopy, limit our in-depth evaluation and understanding of the functions of forest vertical structures, and also affect the scientific decision-making of forest ecosystem management.

[0004] In view of this, the applicant has specifically proposed this application after studying the existing technologies. Summary of the Invention

[0005] The present invention aims to provide a method and device for evaluating the three-dimensional height of a forest canopy based on canopy structure connectivity, so as to solve the shortcomings that existing methods are difficult to comprehensively analyze the spatial heterogeneity of forest canopy structure and its impact on ecological processes, and the evaluation and understanding of the functions of forest vertical structures are limited.

[0006] To solve the above technical problems, the present invention is achieved through the following technical solutions:

[0007] A method for evaluating the three-dimensional height of a forest canopy based on canopy structure connectivity, comprising:

[0008] Obtaining three-dimensional forest point cloud data of a target area collected in real time by an unmanned aerial vehicle lidar;

[0009] Preprocessing the three-dimensional forest point cloud data to generate a digital elevation model DEM and a digital surface model DSM of the target area;

[0010] Generating a canopy height model CHM by calculating the difference between DSM and DEM;

[0011] Based on the cost surface generated from the CHM, and based on the Dijkstra algorithm, calculate the optimal path from the starting point to the target point for each grid of the CHM to optimize the connectivity between canopy units;

[0012] According to the optimal path, create and initialize a node for each grid of the CHM. By calculating the connection degree between each node and the height of similar canopy units, obtain the canopy structure connection degree value of each canopy unit, that is, the CSC value, so as to obtain a three-dimensional height model of the forest canopy based on the spatial distribution of the CSC value;

[0013] According to the three-dimensional height model of the forest canopy, analyze and evaluate the distribution density and forest carbon flow function between different canopy units.

[0014] Furthermore, the preprocessing of the three-dimensional forest point cloud data includes:

[0015] Adopt the neighborhood point search algorithm to remove the high and low outliers of the three-dimensional forest point cloud data to improve the quality of the point cloud data;

[0016] According to the three-dimensional forest point cloud data with improved quality, use the progressive densification triangular mesh filtering algorithm to extract the ground points and use the interpolation method for interpolation to construct a digital elevation model DEM;

[0017] According to the three-dimensional forest point cloud data with improved quality, construct a digital surface model DSM including surface vegetation, where the grid resolution of the DSM is the same as that of the DEM.

[0018] Furthermore, the interpolation method includes: inverse distance weighted interpolation method, Kriging interpolation method or triangulated irregular network interpolation method.

[0019] Furthermore, before performing the denoising and filtering operation, correct and calculate the roll angle, pitch angle, and heading angle errors of the unmanned aerial vehicle to ensure the high precision of the three-dimensional forest point cloud data;

[0020] The calculation formulas for the roll angle ΔRoll, pitch angle ΔPitch, and heading angle ΔHeading errors are respectively:

[0021]

[0022] Among them, Δh represents the height difference of approximately homologous ground objects measured when opening the profile editing in the vertical flight direction, and r represents the horizontal distance between the approximately homologous ground objects measured in the 2D view and the center lines of the two flight strips;

[0023] D represents the distance difference along the flight direction of the center position of the same ground object measured when opening the profile editing parallel to the flight direction, and H represents the average flight height;

[0024] S represents the distance between the average center positions of the ground laser foot points measured twice, and D represents the distance between two flight strips.

[0025] Further, the CSC value of each canopy unit is obtained by segmenting the path, calculating the path suitability value between nodes, and accumulating the path connectivity. The specific formula is as follows:

[0026]

[0027] Among them, CSC i represents the structural connectivity of the i-th canopy unit, and C ji represents the path suitability value from canopy unit j to canopy unit i; N represents the total number of canopy units;

[0028] C ji The calculation formula of is:

[0029] Among them, C ji,p represents the suitability value from canopy unit j to canopy unit i along path p;

[0030] C ji,p is determined by cumulatively calculating the similarity value of the canopy height. The formula is:

[0031]

[0032] Among them, z k represents the canopy height of the k-th point in path p; L represents the number of nodes of the canopy height in path p; z j represents the canopy height of the j-th point in path p; σ is the standard deviation of the canopy height, which is used to control the influence amplitude of the height difference.

[0033] Further, when analyzing and evaluating according to the three-dimensional height model of the forest canopy, a data visualization graph is generated through the calculated CSC value, which is used to display the distribution density of different canopy heights in the target area.

[0034] The present invention also provides a three-dimensional height evaluation device for a forest canopy based on canopy structure connectivity, including:

[0035] An acquisition unit for acquiring the three-dimensional data of the forest point cloud in the target area collected in real time by the UAV lidar;

[0036] A preprocessing unit for preprocessing the three-dimensional data of the forest point cloud to generate a digital elevation model DEM and a digital surface model DSM of the target area;

[0037] A CHM model generation unit for generating a canopy height model CHM by calculating the difference between DSM and DEM;

[0038] The optimal path calculation unit calculates the optimal path from the starting point to the target point for each grid of the CHM based on the cost surface generated by the CHM and the Dijkstra algorithm to optimize the connectivity between canopy units;

[0039] The CSC calculation unit is used to create and initialize a node for each grid of the CHM according to the optimal path, and obtain the canopy structure connectivity value, i.e., the CSC value, of each canopy unit by calculating the connectivity degree between each node and the height of similar canopy units, so as to obtain a three-dimensional height model of the forest canopy based on the spatial distribution of the CSC value;

[0040] The evaluation unit is used to analyze and evaluate the distribution density and forest carbon flow function between different canopy units according to the three-dimensional height model of the forest canopy.

[0041] The present invention also provides a three-dimensional height evaluation device for a forest canopy based on canopy structure connectivity, including a processor and a memory. The memory stores a computer program, and the computer program can be executed by the processor to implement a three-dimensional height evaluation method for a forest canopy based on canopy structure connectivity as described above.

[0042] The present invention also provides a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by the processor of the device where the computer-readable storage medium is located, a three-dimensional height evaluation method for a forest canopy based on canopy structure connectivity as described above is implemented.

[0043] In summary, compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention proposes the concept of canopy structure connectivity (CSC) to evaluate the spatial connectivity between a certain height point and other similar height points in the forest canopy. The CSC value is obtained by calculating the sum of the optimal paths from the cells in the canopy height model CHM grid to all other cells, and the path selection is based on the ecological suitability of the canopy height. The higher the height suitability, the more likely the ecological process is to choose the path passing through this point, thereby reducing the energy consumption of the ecosystem.

[0045] The present invention takes the canopy height as the core index for path selection, combines the Dijkstra algorithm to calculate the optimal path suitability between grid cells, thereby reducing the energy consumption in the forest ecological process. At the same time, CSC can reflect the transfer efficiency of matter and energy in the vertical direction of the canopy. A higher CSC value indicates that the exchange of photosynthesis products, water, and nutrients inside the forest is smoother, which helps to improve the stability and productivity of the ecosystem. The three-dimensional forest canopy height model obtained by the present invention not only highlights the ecological empirical role of the connectivity of the forest vertical structure, but also provides scientific support and technical support for the assessment of forest carbon flow function and the realization of the "carbon neutral" goal, and can comprehensively analyze the spatial heterogeneity of the canopy structure and its specific impact on ecological processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0047] Figure 1 Schematic diagram of a method for evaluating the three-dimensional height of a forest canopy based on canopy structure connectivity provided for Example 1.

[0048] Figure 2 Schematic diagram for calculating the canopy height model provided for Example 1.

[0049] Figure 3 For a certain area provided in Example 2 using DJI Figure 3 LiDAR point cloud data map for three-dimensional reconstruction.

[0050] Figure 4 Point cloud classification result map for a certain area provided for Example 2.

[0051] Figure 5 Canopy height model structure diagram for a certain area provided for Example 2.

[0052] Figure 6 Canopy height and CSC result comparison map for a certain area provided for Example 2.

[0053] Figure 7 Canopy height frequency distribution and CSC frequency distribution group map for a certain area provided for Example 2.

[0054] Figure 8 Three-dimensional display map of CSC results for a certain area provided for Example 2.

[0055] Figure 9Schematic diagram of a three-dimensional forest canopy height evaluation device based on canopy structure connectivity provided for Embodiment 3.

[0056] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Specific Embodiments

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention. Therefore, the detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the present invention to be protected, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0058] Embodiment 1

[0059] Embodiment 1 of the present invention provides a method for evaluating the three-dimensional height of a forest canopy based on canopy structure connectivity, which can be implemented by a three-dimensional forest canopy height evaluation device based on canopy structure connectivity (hereinafter referred to as the evaluation device), and in particular, is executed by one or more processors in the evaluation device.

[0060] In this embodiment, the evaluation device may be an electronic device equipped with a processor, and the processor has a computer program for the method for evaluating the three-dimensional height of a forest canopy based on canopy structure connectivity and the computer program can be executed, such as a computer, a smart phone, a smart tablet, a workstation, etc., which is not limited herein.

[0061] As Figure 1 shown, a method for evaluating the three-dimensional height of a forest canopy based on canopy structure connectivity includes steps S1 to S6.

[0062] S1. Obtain the three-dimensional forest point cloud data of the target area collected in real time by an unmanned aerial vehicle lidar.

[0063] S2. Preprocess the three-dimensional forest point cloud data to generate a digital elevation model DEM and a digital surface model DSM of the target area.

[0064] Further, preprocessing the three-dimensional forest point cloud data includes:

[0065] S21. Use the neighborhood point search algorithm to remove the gross errors at the high and low levels of the three-dimensional forest point cloud data, improving the quality of the point cloud data. Specifically: For each point, define a search radius or neighborhood size, and search for other points within this neighborhood. Calculate the average height or median height of the current point and the points within the neighborhood; if the difference between the height of the current point and the average height or median height exceeds a certain set threshold, mark it as a gross error point; remove all the points marked as gross errors.

[0066] S22. Based on the three-dimensional forest point cloud data with improved quality, use the improved progressive densification triangulation filtering algorithm IPTD to extract ground points, providing accurate data for constructing the DEM.

[0067] In this embodiment, the IPTD algorithm is an effective method for extracting ground points from point cloud data. It combines the ideas of triangulation filtering and progressive densification, and can more accurately identify ground points and non-ground points.

[0068] Specifically:

[0069] Construct an initial triangulation network, which only contains the lowest points (assuming these points are most likely to be ground points); gradually add other points, and judge whether these points belong to the ground according to certain criteria (such as height difference, angle, etc.); if the newly added points meet the conditions of ground points with some edges or faces of the triangulation network, add them to the triangulation network until all points have been processed. The finally obtained triangulation network is the representation of the ground points.

[0070] In this embodiment, the inverse distance weighted interpolation method (IDW), Kriging interpolation method (Kriging), or triangulated irregular network interpolation method (TIN) is used for DEM interpolation.

[0071] Among them, the inverse distance weighted interpolation method (IDW): It is a method for interpolation based on the principle that "points closer have a greater influence", and is suitable for the case where the point cloud data is relatively evenly distributed.

[0072] Kriging interpolation method (Kriging): It takes into account the spatial autocorrelation of the point cloud data, can generate a smoother DEM surface, but has a large amount of calculation.

[0073] Triangulated irregular network interpolation method (TIN): It is a method for interpolation based on the triangulation network, and can preserve terrain features such as steep slopes and canyons.

[0074] S23. Based on the three-dimensional forest point cloud data with improved quality, construct a digital surface model DSM including surface vegetation, and the raster resolution of the DSM is the same as that of the DEM.

[0075] S3. Generate a canopy height model CHM by calculating the difference between the DSM and the DEM.

[0076] As shown Figure 2 in the figure, a canopy height model (CHM) is generated by the difference between DSM and DEM. In the figure, blue represents the DSM value, red represents the DEM value, and purple represents the CHM value. The CHM can visually display the height and spatial distribution of the vegetation canopy, providing basic data for forest structure, carbon storage assessment, and ecological process simulation.

[0077] S4. Based on the cost surface generated from the canopy height model CHM and using the Dijkstra algorithm, calculate the optimal path from each grid of the CHM from the starting point to the target point to optimize the connectivity between canopy units.

[0078] In this step, read the CHM data to obtain the height values of the grids. According to the CHM data, convert the height value of each grid into a movement cost. This can be done by simply taking the height value as the cost (assuming that the higher the height, the more difficult the movement), or by converting the height into a cost through a certain function (such as a linear exponent, etc.). Create a two-dimensional array of the same size as the CHM data to store the cost values. Consider each grid of the CHM as a node of the graph. Create edges between the nodes according to the adjacent relationship of the grids. Assign the cost value to each edge as the cost of movement.

[0079] For each starting grid (or select a set of representative starting grids to reduce the calculation amount), use the Dijkstra algorithm to calculate the shortest path to the target grid. Record the shortest path from each starting grid to the target grid and its total cost.

[0080] In this embodiment, the Dijkstra algorithm is a classic graph algorithm used to find the shortest path from a single source point to all other nodes in a weighted graph. This algorithm was proposed by the Dutch computer scientist Edsger Dijkstra in 1956.

[0081] Analyze the connectivity of the canopy units according to the calculated shortest path.

[0082] S5. According to the optimal path, create and initialize a node for each grid of the CHM. By calculating the connection degree of each node with the height of similar canopy units, obtain the canopy structure connection degree value of each canopy unit, that is, the CSC value, so as to obtain a three-dimensional height model of the forest canopy based on the spatial distribution of the CSC value.

[0083] In this embodiment, the Canopy Structure Connectivity (CSC) is an important indicator for quantifying the vertical structural connectivity of the forest canopy. It is calculated based on the Canopy Height Model (CHM) and can reveal the spatial connectivity of the canopy's vertical structure and its impact on material cycling and energy flow. CSC evaluates the height differences between different canopy units and their regulatory effects on ecological processes through the Dijkstra algorithm, providing insights into key functions such as energy transfer, carbon storage, and water cycling in forest ecosystems. The higher the CSC value, the more coherent the canopy structure and the more efficient the flow of ecological processes. The calculation formula for CSC is:

[0084]

[0085] where CSC i represents the structural connectivity of the i-th canopy unit, and C ji represents the path suitability value from canopy unit j to canopy unit i; N represents the total number of canopy units;

[0086] C ji The calculation formula for is:

[0087] where C ji,p represents the suitability value of the path from canopy unit j to canopy unit i along path p;

[0088] C ji,p is determined by cumulatively calculating the similarity values of the canopy heights, and the formula is:

[0089]

[0090] where z k represents the canopy height of the k-th point in path p; L represents the number of nodes of the canopy height in path p; z j represents the canopy height of the j-th point in path p; σ is the standard deviation of the canopy height, which is used to control the influence amplitude of the height difference. A smaller canopy height difference can enhance the path suitability value, thereby improving the overall connectivity of CSC.

[0091] S6. According to the three-dimensional forest canopy height model, analyze and evaluate the distribution density between different canopy units and the forest carbon flow function.

[0092] In this step, the CSC calculation results are written to the disk and stored in CSV format. This facilitates subsequent further analysis and data visualization, providing efficient data support for later decision-making. When analyzing and evaluating according to the three-dimensional forest canopy height model, through the calculated CSC value, a data visualization graph is generated to display the distribution density of different canopy heights in the target area.

[0093] For example, by constructing the function VisResult, two visualization windows are created. One shows the distribution of canopy height in the study area, and the other shows the distribution of CSC in the corresponding area. This function helps researchers visually observe the spatial distribution and connectivity of the canopy structure.

[0094] For example, by constructing the function VisHeightFrequency, the canopy height data is read from the CSV result file, a histogram is generated and the title and labels are set. At the same time, a density plot of the canopy height data is shown to display the distribution density of different canopy heights in the study area.

[0095] For example, by constructing the function VisCSCFrequency, the CSC data is read from the CSV result file, and a scatter plot of the CSC data is generated. This plot helps researchers understand the spatial distribution of CSC values and their sensitivity to canopy height.

[0096] For example, by constructing the function VisCSCZbar, it is used to visualize the relationship between CSC and canopy height. By reading the data from the output CSV file, a histogram of the canopy data is generated, and the weighted histogram and squared weight histogram are calculated in combination with the CSC data. The VisCSCZbar function also calculates the average CSC value and its standard deviation for each bin, and generates the corresponding line plot, scatter plot, and error bars. Finally, a new y-axis sharing the x-axis is added to plot the density plot of the canopy data, and the title and labels are set. This function helps to reveal the quantitative relationship between canopy height and CSC.

[0097] The distribution of canopy structure connectivity (CSC) in space is crucial for forest ecosystem analysis. In an application embodiment, to visually display the distribution of CSC, the Mayavi library is used to achieve the three-dimensional visualization effect of canopy structure connectivity. Mayavi is a powerful scientific data visualization tool that can effectively present high-dimensional data in three dimensions, especially suitable for displaying point cloud data, structured grids, and raster data, etc. The steps are as follows:

[0098] Step 1: Read the data and prepare the visualization data. Read the CSV result file containing CSC values. This file contains the xyz spatial coordinate information of each point and the corresponding CSC value. Separate the xyz coordinates and CSC values for storage to form four columns of data, where the first three columns represent the x, y, and z spatial coordinates respectively, and the fourth column is the corresponding CSC value. These data will serve as the basis for three-dimensional visualization using the Mayavi library.

[0099] Step 2: Use the mlab.points3d(x, y, z, CSC, colormap, scale_factor, opacity) function in Mayavi to draw a three-dimensional scatter plot. This function takes the xyz coordinates as input and maps the CSC value of each point to a color gradient. By choosing a suitable color scheme, the strength distribution of the data can be clearly shown. To optimize the visualization effect of the graph, the size and shape of the points can be adjusted according to needs, so that dense areas can be shown more clearly while reducing the computational burden.

[0100] Step 3: Display the three-dimensional visualization image. After completing the setting of visualization elements, use the mlab.show function to display the final three-dimensional scatter plot. At this time, users can rotate, zoom, and explore the spatial distribution of the canopy structure connectivity through an interactive view.

[0101] In summary, compared with the prior art, the present invention has the following beneficial effects:

[0102] The present invention proposes the concept of Canopy Structure Connectivity (CSC) to evaluate the spatial connectivity between a certain height point and other similar height points in the forest canopy. The CSC value is obtained by calculating the sum of the optimal paths from the cells within the grid to all other cells, and the path selection is based on the ecological suitability of the canopy height. The higher the height suitability, the more likely the ecological process is to choose the path passing through this point, thereby reducing the energy consumption of the ecosystem. Generally speaking, the higher the CSC value, the lower the ecological energy consumption, the stronger the connectivity of the canopy structure, reflecting the high coordination and efficiency of the forest canopy structure.

[0103] The present invention takes the canopy height as the core index for path selection, combines the Dijkstra algorithm to calculate the optimal path suitability between grid cells, thereby reducing the energy consumption in the forest ecological process. The three-dimensional height model of the forest canopy obtained by the present invention not only focuses on revealing the ecological empirical role of the vertical structure connectivity of the forest, but also provides scientific support and technical support for the evaluation of the forest carbon flow function and the realization of the "carbon neutral" goal, and can comprehensively analyze the spatial heterogeneity of the canopy structure and its specific impact on ecological processes.

[0104] Example Two

[0105] In the second embodiment of the present invention, a certain area is used as a research example to apply the evaluation method proposed by the present invention.

[0106] The east-west span of this area is about 126 kilometers, the north-south width is about 90 kilometers, and the total area reaches 5,824 square kilometers. The landform types are mainly low mountains and hills, lake-accumulated plains and alluvial plains. Generally, the terrain shows a slow inclination from west to east and from south to north. The climate characteristics belong to the subtropical monsoon humid climate, with abundant precipitation, rain and heat in the same period. The annual average temperature is between 12.2℃ and 17.3℃, the total annual precipitation is distributed between 761 and 1,780 millimeters, and the annual average relative humidity exceeds 80%. Due to the small undulation of the terrain in this area and the long-term interference of vegetation by human activities, the vertical distribution pattern and zonal differences are very significant, and the altitude difference causes the vegetation to be in the form of low shrubs. The vegetation community in this area belongs to the transitional type between the mid-subtropical evergreen broad-leaved forest and the northern transition zone. Among them, Cyclobalanopsis glauca and Castanopsis sclerophylla are the main representative species. The vegetation types are rich, including coniferous forests, broad-leaved forests, shrubs, meadows and wetlands and other diverse ecosystems.

[0107] In this embodiment, the DJI Terra lidar processing module is used to preprocess the original data collected by the Zenmuse L1 to generate three-dimensional point cloud data in LAS format, laying a foundation for the generation of the canopy height model (CHM) and ecosystem analysis.

[0108] Step 1: Relying on the DJI M300 RTK platform, equipped with the DJI L1 lidar sensor to collect lidar point cloud data, preheat the inertial navigation of the radar lens, set the flight height to 100m for terrain-following flight, the laser side overlap rate to 70%, and the three-echo laser scanning mode to collect data and check the data integrity.

[0109] The lidar accurately obtains the terrain undulation and vegetation height of the target area by actively emitting laser pulses and receiving their echo signals, measuring the time delay and intensity of the signals. This technology is especially suitable for areas with complex terrain and dense vegetation coverage, and can provide detailed three-dimensional spatial information.

[0110] First, accurately calibrate the inertial measurement unit (IMU) and compass to ensure the high precision of data collection. According to the scale, shape and complexity of the research area, formulate a flight plan, including the flight area, flight height, route planning and the selection of laser scanning mode.

[0111] Lidar data collection. During the flight, the lidar real-time collects point cloud data and combines it with the RTK positioning information of the UAV to ensure that the geographical coordinate accuracy of each point reaches the centimeter level. According to different research needs, select the single-echo or multi-echo mode to optimize the detection ability, especially in capturing the vegetation structure.

[0112] Data storage and inspection. The collected lidar point cloud data is stored in a file in the format of "DJI_YearMonthDay_Zenmuse-L1-Custom" and saved on the SD card of the L1 lens. The file should include camera calibration data, IMU calibration data, lidar point cloud data, RTK base station data, etc. to ensure the integrity of the file. If a missing file is found, the data should be collected again.

[0113] Step 2: Import DJI Terra to generate LAS data. Select the point cloud processing scenario according to the computer memory situation and set the point cloud density to 100%. Set the projection coordinate system to the WGS 84_UTM_zone_50N coordinate system and perform 3D reconstruction of the point cloud.

[0114] Create a new lidar point cloud task in the DJI Terra software and import the lidar data file of the Zenmuse L1. If the data file contains JPG images, true-color point clouds can be generated to enhance data visualization and help clearly identify vegetation types and surface features.

[0115] Set the point cloud density according to research needs and computer memory resources. The point cloud density can be resampled by percentage or the average distance between point clouds. The size of the point cloud data and the computer memory (RAM) requirements are shown in Table 1.

[0116] Table 1. Table of the size of point cloud data and computer memory (RAM) requirements

[0117] Memory (RAM) Original point cloud data size (GB) 16GB 4 32GB 8 64GB 16 128GB 32

[0118] Select an appropriate point cloud processing scenario and adjust the parameters according to the research area. In the advanced settings, convert the point cloud coordinate system to the UTM projection coordinate system to ensure data compatibility and accuracy.

[0119] After the settings are completed, perform 3D reconstruction of the point cloud to generate LAS format point cloud files that meet the research requirements. These point cloud files contain high-precision 3D spatial information and provide accurate data support for subsequent model construction.

[0120] Step 3: Import the generated LAS data into LiDAR360 for data processing. Import the SBET binary track file to ensure the correspondence between the UAV GPS time and the point cloud data; crop the point cloud to remove the low-density data caused by the "edge effect" to weaken the "edge effect" and optimize the data accuracy; calculate and correct the roll angle, pitch angle, and heading angle errors of the UAV to ensure the high precision of the point cloud data.

[0121] Among them, the calculation formulas for the roll angle ΔRoll, pitch angle ΔPitch, and heading angle ΔHeading errors are:

[0122]

[0123] Among them, Δh represents the height difference of approximately homologous ground objects measured when opening the profile editing in the vertical flight direction, and r represents the horizontal distance between the approximately homologous ground objects measured in the 2D view and the center lines of two flight strips.

[0124] D represents the distance difference along the flight direction of the center positions of the same ground object measured when opening the profile editing parallel to the flight direction, and H represents the average flight height.

[0125] S represents the distance between the average center positions of the laser foot points of the ground object measured twice, and D represents the distance between two flight strips.

[0126] Through the above steps, the three-dimensional data of the forest point cloud in the target area is collected. As Figure 3 shown, the lidar point cloud data reconstructed by DJI Smart Figure 3 Dimension.

[0127] Then, preprocess the three-dimensional data of the forest point cloud to generate the digital surface model DSM and the digital elevation model DEM of the target area. The preprocessing includes denoising filtering, ground point classification, and the construction of the digital elevation model (DEM) and the digital surface model (DSM).

[0128] Step 4: Use the neighborhood point search algorithm to remove the high and low gross errors of the three-dimensional data of the forest point cloud, set the number of neighborhood points to 10, the standard deviation multiple to 5, and perform point cloud denoising.

[0129] Step 5: Extract ground points based on the progressive densification triangular network algorithm, measure the maximum building size in the study area, set it to 20m, the maximum terrain slope to 88°, the iteration angle to 10°, the iteration distance to 1.6m, and stop constructing the triangular network when the side length is less than 1m to complete the extraction of ground points.

[0130] Step 6: As Figure 4 shown in the point cloud classification result map, observe the distribution of ground points. Since the terrain in the study area has large undulations and spatial autocorrelation needs to be considered, use the Kriging interpolation method to construct the DEM model, set the resolution to 1m, the buffer size to 2 grids, set the search radius to 5 pixels, and the number of search points to 12. Finally, obtain the DEM.tif file. Of course, other interpolation methods can also be used to construct the DEM model, such as the inverse distance weighting interpolation method or the triangulated irregular network interpolation method TIN, etc., which are not limited here.

[0131] Step 7: Use the vegetation points of the forest point cloud three-dimensional data to construct a DSM model. Similarly, set the resolution to 1m, the buffer size to 2 grids, and use the Kriging interpolation method for model construction. Set the search radius to 5 pixels and the number of search points to 12. Finally, obtain the DSM.tif file.

[0132] Step 8: Subtract the DSM.tif raster from the DEM.tif raster to obtain the Canopy Height Model CHM.tif file. As Figure 5 shown is the structural diagram of the canopy height model of this area.

[0133] Step 9: Create a python3.6 environment, import the canopy height data CHM.tif, and calculate the optimal path from the starting point to the target point for each grid based on the Dijkstra algorithm to obtain the canopy height grid file.

[0134] Step 10: Convert the canopy height grid file to a CSV file named CanopyHeigh.csv. Based on the optimal path from the starting point to the target point of each grid obtained previously, calculate the connectivity of each node to the similar canopy height and generate the spatial distribution of the CSC value.

[0135] Step 11: Output the CSC result to a folder and name it CSCResult.csv.

[0136] Step 12: Visualize the CSC result.

[0137] Generate images of the canopy height and CSC distribution in the study area (as Figure 6 shown), and save them to the specified folder named Plot.png. Generate the density map and histogram of the canopy height data in the study area (as Figure 7 shown in -a), and save them to the specified folder named HeightDF.png. Generate the scatter plot of the CSC changing with the canopy height in the study area (as Figure 7 shown in -b), and save them to the specified folder named CSCFre.png; generate the error bar and density map corresponding to the CSC and the canopy height (as Figure 7 shown in -c), and save them to the specified folder named CSCZbar.png. All the above result images are set to a display resolution of 300 dpi to ensure the display effect.

[0138] Step 13: 3D visualization of CSC results. Create a Python 3.8 environment, load the Mayavi library, import the CSCResult.csv file as the input file, and call the function mlab.figure(size=(800, 600)) to create a 3D scatter plot, where size is used to set the figure size. Call mlab.points3d(x, y, z, CSC, scale_mode='none', scale_factor=5, colormap) to visualize the CSC results. The scale_mode parameter specifies the size scaling mode of the points, and it can be set so that the points increase as CSC increases. The scale_factor controls the scaling factor of the points, and colormap specifies the color mapping scheme. In the example, a display scheme from blue to red for low values to high values is used (as Figure 8 shown).

[0139] Example 3

[0140] As Figure 9 shown, the third embodiment of the present invention also provides a three-dimensional height evaluation device for forest canopy based on canopy structure connectivity, including:

[0141] An acquisition unit for acquiring the three-dimensional forest point cloud data of the target area collected in real time by an unmanned aerial vehicle lidar;

[0142] A preprocessing unit for preprocessing the three-dimensional forest point cloud data to generate a digital elevation model DEM and a digital surface model DSM of the target area;

[0143] A CHM model generation unit for generating a canopy height model CHM by calculating the difference between DSM and DEM;

[0144] An optimal path calculation unit for calculating the optimal path from the starting point to the target point for each grid of CHM based on the cost surface generated by CHM and the Dijkstra algorithm to optimize the connectivity between canopy units;

[0145] A CSC calculation unit for creating and initializing a node for each grid of CHM according to the optimal path, and obtaining the canopy structure connectivity value, i.e., the CSC value, of each canopy unit by calculating the connectivity between each node and the height of similar canopy units, thereby obtaining a three-dimensional height model of the forest canopy based on the spatial distribution of CSC values;

[0146] An evaluation unit for analyzing and evaluating the distribution density and forest carbon flow function between different canopy units according to the three-dimensional height model of the forest canopy.

[0147] Example 4

[0148] The fourth embodiment of the present invention further provides a three-dimensional height evaluation device for a forest canopy based on the connectivity of the canopy structure, which includes a memory and a processor. A computer program is stored in the memory, and the computer program can be executed by the processor to implement the three-dimensional height evaluation method for a forest canopy based on the connectivity of the canopy structure as described above.

[0149] Embodiment Five

[0150] The fifth embodiment of the present invention further provides a computer-readable storage medium. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by the processor of the device where the computer-readable storage medium is located, the three-dimensional height evaluation method for a forest canopy based on the connectivity of the canopy structure as described above is implemented.

[0151] In several embodiments provided by the embodiments of the present invention, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device and method embodiments described above are merely illustrative. For example, the flowcharts in the drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to multiple embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0152] In addition, the functional modules in each embodiment of the present invention may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0153] When the above-mentioned functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, an electronic device, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs. It should be noted that in this article, the terms "including", "comprising", or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such a process, method, article, or device. Without further limitations, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article, or device including the said element.

[0154] The terms used in the embodiments of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The singular forms "a", "the", and "said" used in the embodiments of the present invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0155] It should be understood that the term "and / or" used herein is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0156] Depending on the context, the word "if" as used herein can be interpreted as "when", "while", "in response to determining", or "in response to detecting". Similarly, depending on the context, the phrase "if determined" or "if detecting (stated condition or event)" can be interpreted as "when determined", "in response to determining", "when detecting (stated condition or event)", or "in response to detecting (stated condition or event)".

[0157] The "first / second" mentioned in the embodiments is merely to distinguish similar objects and does not represent a specific order for the objects. It can be understood that the "first / second" can be interchanged in a specific order or sequence when permitted. It should be understood that the objects distinguished by the "first / second" can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order other than those illustrated or described herein.

[0158] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A three-dimensional height evaluation method for forest canopy based on the connectivity of canopy structure, characterized in that Including: Obtaining the three-dimensional forest point cloud data of the target area collected in real time by the UAV lidar; Preprocessing the three-dimensional forest point cloud data to generate a digital elevation model (DEM) and a digital surface model (DSM) of the target area; Generating a canopy height model (CHM) by calculating the difference between the DSM and the DEM; Based on the cost surface generated by the CHM, calculating the optimal path from the starting point to the target point for each grid of the CHM based on the Dijkstra algorithm to optimize the connectivity between canopy units; According to the optimal path, creating and initializing a node for each grid of the CHM, and obtaining the canopy structure connectivity value (i.e., CSC value) of each canopy unit by calculating the connectivity degree between each node and the height of similar canopy units, thereby obtaining a three-dimensional height model of the forest canopy based on the spatial distribution of the CSC value; Analyzing and evaluating the distribution density and forest carbon flow function between different canopy units according to the three-dimensional height model of the forest canopy; 2. The three-dimensional height evaluation method of a forest canopy based on the connectivity of the canopy structure according to claim 1, wherein , Preprocessing the three-dimensional forest point cloud data includes: Adopting a neighborhood point search algorithm to remove the high and low outliers of the three-dimensional forest point cloud data to denoise and filter the data and improve the quality of the point cloud data; According to the three-dimensional forest point cloud data with improved quality, extracting ground points using an incremental encrypted triangulation filtering algorithm and performing interpolation using an interpolation method to construct a digital elevation model (DEM); According to the three-dimensional forest point cloud data with improved quality, constructing a digital surface model (DSM) including surface vegetation, where the grid resolution of the DSM is the same as that of the DEM; 3. The three-dimensional height evaluation method of a forest canopy based on the connectivity of the canopy structure according to claim 2, wherein , The interpolation method includes: inverse distance weighted interpolation method, Kriging interpolation method or irregular triangulation interpolation method; 4. The three-dimensional height evaluation method of a forest canopy based on the connectivity of the canopy structure according to claim 2, wherein , Before performing the denoising and filtering operation, calibrating and calculating the roll angle, pitch angle, and heading angle errors of the UAV to ensure the high precision of the three-dimensional forest point cloud data; The calculation formulas for the roll angle ΔRoll, pitch angle ΔPitch, and heading angle ΔHeading errors are respectively: Where, Δh represents the height difference of approximately homologous ground objects measured when opening a profile edit in the vertical flight direction, and r represents the horizontal distance between the approximately homologous ground objects measured in the 2D view and the center lines of two flight strips; D represents the distance difference along the flight direction of the center position of the same ground object measured when opening a profile edit parallel to the flight direction, and H represents the average flight height; S represents the distance between the average center positions of the laser foot points of the ground objects measured twice, and D represents the distance between two flight strips; 5. The three-dimensional height evaluation method of a forest canopy based on the connectivity of the canopy structure according to claim 1, wherein , The CSC value of each canopy unit is obtained by splitting the path, calculating the path suitability value between nodes, and accumulating the path connectivity. The specific formula is as follows: Among them, CSC i represents the structural connectivity of the i-th canopy unit, and C ji represents the path suitability value from canopy unit j to canopy unit i; N represents the total number of canopy units; C ji The calculation formula of Among them, C ji,p represents the suitability value from canopy unit j to canopy unit i along path p; C ji,p Determined by cumulatively calculating the similarity value of the canopy height, and the formula is: where z k represents the canopy height of the k-th point in path p; L represents the number of nodes with canopy height in path p; z j represents the canopy height of the j-th point in path p; σ represents the standard deviation of the canopy height, which is used to control the influence amplitude of the height difference.

6. The three-dimensional height evaluation method of a forest canopy based on the connectivity of the canopy structure according to claim 1, characterized in that , When analyzing and evaluating according to the three-dimensional height model of the forest canopy, a data visualization graph is generated through the calculated CSC value to display the distribution density of different canopy heights in the target area; 7. A three-dimensional height evaluation device for forest canopy based on the connectivity of canopy structure, characterized in that, Including: An acquisition unit for obtaining the three-dimensional forest point cloud data of the target area collected in real time by the UAV lidar; A preprocessing unit for preprocessing the three-dimensional forest point cloud data to generate a digital elevation model (DEM) and a digital surface model (DSM) of the target area; The CHM model generation unit is used to generate the Canopy Height Model (CHM) by calculating the difference between the DSM and the DEM; The optimal path calculation unit calculates the optimal path from the starting point to the target point for each grid of the CHM based on the Dijkstra algorithm according to the cost surface generated by the CHM to optimize the connectivity between canopy units; The CSC calculation unit is used to create and initialize a node for each grid of the CHM according to the optimal path, and obtain the canopy structure connectivity value (i.e., CSC value) of each canopy unit by calculating the connectivity between each node and the height of similar canopy units, so as to obtain a three-dimensional height model of the forest canopy based on the spatial distribution of the CSC value; The evaluation unit is used to analyze and evaluate the distribution density and forest carbon flow function between different canopy units according to the three-dimensional height model of the forest canopy; 8. The three-dimensional height evaluation device of a forest canopy based on the connectivity of the canopy structure according to claim 7, wherein In the preprocessing unit, it specifically includes: Adopting the neighborhood point search algorithm to remove the high and low outliers of the three-dimensional forest point cloud data to improve the quality of the point cloud data; According to the three-dimensional forest point cloud data with improved quality, using the progressive densification triangular mesh filtering algorithm to extract the ground points and performing interpolation using the interpolation method to construct the Digital Elevation Model (DEM); According to the three-dimensional forest point cloud data with improved quality, construct the Digital Surface Model (DSM) including surface vegetation, where the grid resolution of the DSM is the same as that of the DEM; 9. The three-dimensional height evaluation device of a forest canopy based on the connectivity of the canopy structure according to claim 7, wherein In the CSC calculation unit, the CSC value of each canopy unit is obtained by splitting the path, calculating the path suitability value between nodes and accumulating the path connectivity. The specific formula is as follows: Among them, CSC i represents the structural connectivity of the i-th canopy unit, and C ji represents the path suitability value from canopy unit j to canopy unit i; C ji The calculation formula is as follows: Among them, C ji,p represents the suitability value from canopy unit j to canopy unit i along path p; C ji,p Determined by cumulatively calculating the similarity value of the canopy height, and the formula is: where z k represents the canopy height of the k-th point in the path p; L represents the number of nodes with canopy height in the path p; z j represents the canopy height of the j-th point in the path p; σ represents the standard deviation of the canopy height, which is used to control the influence range of the height difference.

Images