A method for setting forest monitoring standard plots based on airborne lidar data

Abstract

The application discloses a method for setting forest monitoring standard plots based on airborne laser radar data, comprising the following steps: S1: airborne laser radar surveying and mapping is carried out on the survey monitoring target forest stand; S2: surveying and mapping laser radar point cloud achievement data are acquired; S3: point cloud data are preprocessed; S4: a grid effective layout range is identified; S5: grid layout and effective grid screening are carried out; S6: forest structure parameter mean value extraction and high density index calculation are carried out; S7: effective grid layering is carried out; S8: HDI distribution is calculated layer by layer, and index concentrated distribution range grid is extracted; S9: typical grid screening is carried out; and S10: standard compliance verification and final determination are carried out. The method provided by the application can accurately extract forest stand factor information, and can provide standard plot settable plots according to standard plot setting principles, and can extract terrain factors of the plots, so that the typicality of the plots can be ensured, and the uncertainty brought by complex terrain to forest survey work can be avoided.

Description

Technical Field

[0001] This invention relates to the fields of forestry, tree surveying, and forest management, and in particular to a method for setting up forest monitoring standard plots based on airborne lidar data. Background Technology

[0002] [Forest Resource Survey] To understand the status and changing patterns of forest resources and meet the needs of forest resource management, forest surveys and monitoring should be conducted. However, since forest land accounts for a large proportion of land resources, in practice, it is generally not necessary to conduct field measurements of the entire forest stand. Instead, small-area, localized field surveys are carried out according to certain requirements and methods, and the status of the entire forest stand is inferred from the survey results. This survey method can meet the needs of forestry production while saving resources.

[0003] [Methods for Selecting Measured Areas] In local forest stand surveys, there are two methods for selecting measured areas: random sampling and typical sampling. Random sampling involves setting up a certain number of measured plots within the forest stand according to random principles; typical sampling involves setting up a certain number of measured plots based on the average condition of the forest stand.

[0004] [Concept of Standard Plot] In a forest stand, plots set up according to the principle of random sampling are called sample plots, or simply plots; plots set up according to the requirements of average condition that can fully represent the overall characteristic level of the forest stand are called typical plots, or simply standard plots.

[0005] [Purpose of Standard Plots] By establishing standard plots and conducting field surveys, the quantity and quality index values ​​of various survey factors can be obtained. The method of extrapolating the results of the entire forest stand based on the area ratio of the standard plot survey results is called the standard plot survey method. In the process of forest management, standard plot surveys provide a reliable basis for the rational organization of various management measures and the conduct of scientific research activities. In this process, standard plot surveys are an effective method, and for some specialized surveys, they are the only applicable method.

[0006] [Basic Requirements for Standard Plot Establishment] The establishment of standard plots must comply with the following requirements: ① The standard plot must be sufficiently representative of the predetermined requirements; ② The standard plot must be located within the same forest stand and cannot cross forest stands; ③ The standard plot cannot cross streams, roads, or cleared survey lines, and should be located away from the forest edge (at least one time the average height of the forest stand); ④ When the standard plot is located in a mixed forest, its tree species and tree density distribution should be uniform.

[0007] [Existing Standard Plot Setting Techniques] Currently, the following three technical methods are used to set up standard plots in forestry survey and monitoring work: ① Survey personnel go to the target forest stand and select the survey area on-site according to the basic requirements for standard plot setting; ② Set up survey standard plots by referring to recent forest management archives, combined with on-site surveys and the basic requirements for standard plot setting; ③ Based on recent optical remote sensing data and in accordance with the standard plot setting principles, standard plots are set up in the office through methods such as human-computer interactive interpretation, coordinate positions are recorded, and satellite navigation and positioning technology is used to go to the target area to conduct surveys.

[0008]

Problems and Shortcomings of Existing Technologies

[0009] To address the problems existing in the aforementioned background technology, the purpose of this invention is to provide a method for setting up forest monitoring standard plots based on airborne lidar data. This invention can ensure the typicality of the plots while avoiding the uncertainties brought about by complex terrain to forest survey work.

[0010] To achieve the above objectives, the present invention adopts the following technical solution:

[0011] A method for setting up forest monitoring standard plots based on airborne lidar data includes the following steps:

[0012] S1: Conduct airborne lidar aerial surveys of the target forest stands for investigation and monitoring;

[0013] S2: Post-processing settlement obtains aerial survey lidar point cloud data;

[0014] S3: Perform preprocessing on point cloud data;

[0015] S4: Identify the effective deployment range of the grid based on CHM;

[0016] S5: Grid layout and effective grid selection;

[0017] S6: Extraction of mean values ​​of effective grid forest structural parameters and calculation of high density index;

[0018] S7: Effective grid layering based on high-density index;

[0019] S8: Calculate the HDI distribution layer by layer and extract the grid of the index concentration distribution range;

[0020] S9: Selecting typical grids based on terrain factors;

[0021] S10: Standards compliance verification and final determination.

[0022] Furthermore, step S1 involves conducting airborne lidar aerial surveys of the target forest stands as follows: acquiring three-dimensional point cloud data of the target area, with a point density ≥ 100 pts / m². 2 Elevation accuracy ≤ 0.1m, horizontal accuracy ≤ 0.2m.

[0023] Furthermore, step S3, the method for preprocessing the point cloud data, includes the following steps:

[0024] S3-1: Noise Reduction and Ground Point Classification

[0025] Remove noise from power lines, birds, and isolated points to improve data quality; traverse and search for ground points in the point cloud data;

[0026] S3-2: Extraction and cropping of DEM, DSM, and CHM images

[0027] DSM and DEM are generated by interpolation of surface points and ground points; CHM is obtained by subtracting DEM values ​​from DSM values ​​based on the corresponding spatial locations; and mask analysis is performed on CHM based on the survey range vector to retain the pixel DN values ​​within the survey area.

[0028] S3-3: Point Cloud Normalization

[0029] The point cloud is normalized based on ground points or DEM;

[0030] S3-4: Forest Structure Parameter Raster Calculation and Pruning

[0031] Based on a high spatial resolution raster, all points above the corresponding two-dimensional plane are calculated pixel by pixel. The forest structure parameters of the corresponding pixel are calculated using this point set. The forest structure parameters include vegetation cover, leaf area index, and canopy height.

[0032] Furthermore, in step S4, the algorithm identifies areas within the survey scope where the canopy height is less than 6m and the continuous area is greater than 5m². 2 The area designated as invalid for grid deployment is identified, while the remaining area is considered a valid deployment range.

[0033] Furthermore, the method for grid layout and effective grid selection in step S5 includes the following steps:

[0034] S5-1: Based on the survey area and the preset grid side length, lay out a standard grid that can completely cover the survey area. The side length of the standard grid should be an integer multiple of the grid side length.

[0035] S5-2: When the grid is completely within the valid range identified in step S4, the grid is considered a valid grid; otherwise, it is an invalid grid. Valid grids are selected according to this rule.

[0036] Furthermore, the method for extracting the mean of effective grid forest structure parameters and calculating the high density index in step S6 includes the following steps:

[0037] S6-1: The forest structure parameters are normalized to maximum and minimum according to equation (1);

[0038] Equation (1)

[0039] S6-2: Overlay the grid with the forest structure parameter raster and calculate the mean value of the forest structure parameter in the region grid by grid.

[0040] S6-3: Based on formulas (2) and (3):

[0041] Equation (2)

[0042] Equation (3)

[0043] The stand height density index (HDI) is calculated grid by grid, where CHM is the average canopy height of the grid, CC is the average vegetation cover of the grid, LAI is the average leaf area index of the grid, and W... i This represents the weight of the corresponding factor; the default value is 1 / 3.

[0044] Furthermore, in step S7, the idea of ​​stratified sampling is used. Based on the high-density index calculated in step S6, Jenks' natural discontinuity method is used to divide the high-density index into N groups to ensure that the variance between groups is maximized and the variance within groups is minimized.

[0045] Furthermore, step S8, the method for calculating the HDI distribution layer by layer and extracting the grid of the index concentration distribution range, includes the following steps:

[0046] S8-1: Extract all high-density indices in the Nth layer and use the kernel density estimation method to find the distribution density function of the high-density indices in the Nth layer;

[0047] S8-2: Find the HDI corresponding to the maximum distribution density, and use the mean of HDI within N layers ± n times the standard deviation as the bandwidth to extract the grids whose HDI is contained within the bandwidth as candidate grids.

[0048] Furthermore, step S9, the method for selecting typical grids based on terrain factors, includes the following steps:

[0049] S9-1: Based on the DEM generated by point cloud data preprocessing, calculate the slope raster, and extract the terrain factors of mean elevation, elevation span, mean slope, and slope span for each candidate grid.

[0050] S9-2: Calculate the kernel density function for each factor and remove grids that are outside the mean ± n standard deviations of the terrain factors.

[0051] Furthermore, the method for standard compliance verification and final determination in step S10 includes the following steps:

[0052] S10-1: Verify the standard land setting principle according to the standard land setting requirements, check for the existence of typical grids that do not meet the standard land setting requirements, and manually remove them if they exist.

[0053] S10-2: Users select the corresponding area to conduct field surveys based on the typical grid spatial location provided by this invention according to work requirements.

[0054] The advantages of this invention compared to the prior art are as follows:

[0055] Laser point cloud data can accurately and completely replicate the three-dimensional structural information of forests. This invention utilizes airborne lidar point cloud data to extract forest structural parameters and topographic factors, and uses mathematical statistics to quantify and evaluate the distribution of each parameter and factor, thereby realizing a method for determining the spatial location of forest stand standard plots based on objective indicators. This method effectively overcomes the uncertainty of setting forest stand standard plots by relying on subjective judgment based on work experience and historical data; at the same time, lidar technology effectively compensates for many shortcomings of optical remote sensing data, and can more accurately and comprehensively reflect the actual condition of forest stands. Using this method to set forest monitoring standard plots can ensure the typicality of forest stands while avoiding the uncertainty brought by complex terrain to forest surveys. In addition, users can freely set the shape and area size of the grid according to the standard plot specifications. Attached Figure Description

[0056] Figure 1 This is a schematic flowchart of the method for setting forest stand standard plots based on airborne lidar data according to the present invention.

[0057] Figure 2 Diagram of ALS equipment;

[0058] Figure 3 A map for collecting data along flight routes;

[0059] Figure 4 Solve the graph for DJI Terra;

[0060] Figure 5 This is a data graph of the point cloud results;

[0061] Figure 6 This is a partial view of the original point cloud data;

[0062] Figure 7 This is the image after resampling and denoising;

[0063] Figure 8 A bottom view of the ground point classification results;

[0064] Figure 9 DEM image;

[0065] Figure 10 DSM diagram;

[0066] Figure 11 For CHM diagrams;

[0067] Figure 12 This is a cropped image of the CHM file.

[0068] Figure 13 This is a local point cloud data map before normalization;

[0069] Figure 14 This is a normalized local point cloud data map;

[0070] Figure 15 A vegetation cover map of the aerial survey area;

[0071] Figure 16 Leaf area index map of the aerial survey area;

[0072] Figure 17 Vegetation cover map of the target area;

[0073] Figure 18 Leaf area index map of the target area;

[0074] Figure 19 Image showing the results of forest window identification;

[0075] Figure 20 To effectively deploy the area map;

[0076] Figure 21 Grid layout diagram;

[0077] Figure 22 For an effective grid diagram;

[0078] Figure 23 A trend diagram of forest structure parameters within the grid;

[0079] Figure 24 HDI distribution statistics and stratified threshold plot;

[0080] Figure 25 Spatial distribution map of each grid layer;

[0081] Figure 26 This is a diagram showing the HDI distribution and extraction range of layer 0.

[0082] Figure 27 This is a diagram showing the distribution and extraction range of HDI in layer 1.

[0083] Figure 28 Typical grid diagrams for each layer;

[0084] Figure 29 Map showing the average altitude distribution and selected range;

[0085] Figure 30 This is a statistical map showing the distribution of altitude ranges and the selection of areas.

[0086] Figure 31 This is a map showing the average slope distribution and the selected range.

[0087] Figure 32 A statistical distribution map of slope span and the selection range;

[0088] Figure 33 This is a map showing the results of the terrain factor screening.

[0089] Figure 34 The final selection diagram. Detailed Implementation

[0090] Before further describing specific embodiments of the present invention, it should be understood that the scope of protection of the present invention is not limited to the specific embodiments described below; it should also be understood that the terminology used in the embodiments of the present invention is for describing specific embodiments and not for limiting the scope of protection of the present invention.

[0091] like Figure 1 As shown, a method for setting forest stand standard plots based on airborne lidar data includes the following steps:

[0092] S1: Conduct airborne lidar aerial surveys of the target forest stands for investigation and monitoring: acquire 3D point cloud data of the target area, data requirement: point density ≥ 100 pts / m² 2 Elevation accuracy ≤ 0.1m, horizontal accuracy ≤ 0.2m.

[0093] S2: Post-processing settlement to obtain aerial survey lidar point cloud data: Import the data into the post-processing settlement software that is compatible with the hardware to obtain point cloud data results in standard LAS format.

[0094] S3: Preprocessing point cloud data: Preprocessing point cloud data using open-source algorithms, plugins, or commercial software, including the following steps:

[0095] S3-1: Noise Reduction and Ground Point Classification

[0096] Remove data noise such as power lines, birds, and isolated points to improve data quality; traverse and search for ground points in the point cloud data;

[0097] S3-2: Extraction and cropping of DEM, DSM, and CHM images

[0098] DSM and DEM are generated by interpolation of surface points and ground points; CHM is obtained by subtracting DEM values ​​from DSM values ​​based on the corresponding spatial locations; and mask analysis is performed on CHM based on the survey range vector to retain the pixel DN values ​​within the survey area.

[0099] S3-3: Point Cloud Normalization

[0100] The point cloud is normalized based on ground points or DEM;

[0101] S3-4: Forest Structure Parameter Raster Calculation and Pruning

[0102] Based on a high spatial resolution raster, all points above the corresponding two-dimensional plane are calculated pixel by pixel. The forest structure parameters of the corresponding pixel are calculated using this point set. The forest structure parameters include vegetation cover, leaf area index, and canopy height.

[0103] S4: Identifying the Effective Grid Layout Area Based on CHM: In some areas of the forest stand, canopy gaps are formed due to sparse tree density. This step uses an algorithm to identify areas within the survey range where the canopy height is less than 6m and the continuous area is greater than 5m². 2 The area designated as invalid for grid deployment is identified, while the remaining area is considered a valid deployment range.

[0104] S5: Grid layout and effective grid selection:

[0105] S5-1: Based on the survey area and the preset grid side length, lay out a standard grid that can completely cover the survey area. The side length of the standard grid should be an integer multiple of the grid side length.

[0106] S5-2: When the grid is completely within the valid range identified in step S4, the grid is considered a valid grid; otherwise, it is an invalid grid. Valid grids are selected according to this rule.

[0107] S6: Extraction of mean values ​​of effective grid forest structure parameters and calculation of high density index:

[0108] S6-1: The forest structure parameters are normalized to maximum and minimum according to equation (1);

[0109] Equation (1)

[0110] S6-2: Overlay the grid with the forest structure parameter raster and calculate the mean value of the forest structure parameter in the region grid by grid.

[0111] S6-3: Based on formulas (2) and (3):

[0112] Equation (2)

[0113] Equation (3)

[0114] The stand height density index (HDI) is calculated grid by grid, where CHM is the average canopy height of the grid, CC is the average vegetation cover of the grid, LAI is the average leaf area index of the grid, and W... i This represents the weight of the corresponding factor; the default value is 1 / 3.

[0115] S7: Effective grid stratification based on high density index: Forest stands with large survey areas often require multiple standard plots. Therefore, this step utilizes the concept of stratified sampling. Based on the high density index calculated in step S6, the Jenks natural discontinuity method is used to divide the high density index into N groups, ensuring maximum between-group variance and minimum within-group variance.

[0116] S8: Calculate the HDI distribution layer by layer and extract the grid of the index concentration distribution range:

[0117] S8-1: Extract all high-density indices in the Nth layer and use the kernel density estimation method to find the distribution density function of the high-density indices in the Nth layer;

[0118] S8-2: Find the HDI corresponding to the maximum distribution density, and use the mean of HDI within N layers ± n times the standard deviation as the bandwidth to extract the grids whose HDI is contained within the bandwidth as candidate grids.

[0119] S9: Selecting typical grids based on terrain factors:

[0120] S9-1: Based on the DEM generated by point cloud data preprocessing, calculate the slope raster, and extract terrain factors such as mean elevation, elevation span, mean slope, and slope span for each candidate grid.

[0121] S9-2: Calculate the kernel density function for each factor and remove grids that are outside the mean ± n standard deviations of the terrain factors.

[0122] S10: Standards Compliance Verification and Finalization:

[0123] S10-1: Verify the standard land setting principle according to the standard land setting requirements, check for the existence of typical grids that do not meet the standard land setting requirements, and manually remove them if they exist.

[0124] S10-2: Users select the corresponding area to conduct field surveys based on the typical grid spatial location provided by this invention according to work requirements.

[0125] The following description uses more specific examples.

[0126] Example 1

[0127] A method for setting forest stand standard plots based on airborne lidar data includes the following steps:

[0128] 1. Obtain raw point cloud data of the survey area.

[0129] In this embodiment, a DJI M300 RTK drone is used, equipped with an L1 lidar (such as...). Figure 2 As shown), use aerial survey mission planning software to plan flight routes for the target area (e.g. Figure 3 (as shown), and carried out aerial surveying tasks.

[0130] 2. Post-processing and settlement to obtain point cloud result data

[0131] Import the raw aerial survey data into the accompanying post-processing software—DJI Terra (e.g., DJI Terra). Figure 4 As shown), perform data processing to obtain the point cloud results data of the target area (such as...). Figure 5 (As shown).

[0132] 3. Data Preprocessing

[0133] In this embodiment, LiDAR 360 is used for LiDAR point cloud data preprocessing. The preprocessing process is as follows:

[0134] (1) Data resampling and denoising

[0135] Because the original point cloud data is large and contains noisy point clouds such as birds and power lines, it is necessary to resample and denoise the original point cloud data.

[0136] Original point cloud partial data such as Figure 6 As shown, a noticeable noise point cloud caused by the high-voltage power lines can be seen in the upper left corner. After resampling and denoising, the noise points are effectively removed, as shown below. Figure 7 As shown. The total number of files decreased from 115.5 million to 57.5 million; the data size decreased from 3.55GB to 1.76GB.

[0137] (2) Ground point classification

[0138] Accurate and comprehensive extraction of ground points from point clouds can effectively improve the accuracy of derived data such as digital elevation models. Therefore, the preprocessing step involves classifying ground points in the resampled and denoised data, as shown in the following figure. Figure 8 As shown: the white area in the figure represents point clouds of surface cover, and the gray area represents point clouds of the ground.

[0139] (3) Extraction of DEM / DSM / CHM

[0140] Subsequent steps require extracting multiple factors from the point cloud-derived data. This process extracts the digital elevation model (DEM) of the area based on the ground point classification results, such as... Figure 9 As shown; a digital surface model (DSM) is extracted based on the point cloud of ground cover, such as... Figure 10 As shown; through spatial calculations, the DSM is subtracted from the DEM to obtain the canopy height model (CHM) of the aerial survey area, as shown. Figure 11 As shown; the CHM of the target area can be obtained by raster cropping using vector editing of the target area, such as... Figure 12 As shown.

[0141] (4) Point cloud normalization processing

[0142] To remove the influence of terrain undulations on the elevation values ​​of point cloud data and ensure the authenticity and validity of the extracted forest stand parameters, the data must be normalized based on ground points. The local point cloud data before normalization is as follows: Figure 13 As shown, it can realistically reflect the changes in terrain slope and undulation; the normalized local point cloud data is as follows: Figure 14 As shown, the topographic relief is eliminated, which can truly reflect the forest stand condition.

[0143] (5) Extraction of forest structure parameters

[0144] Based on the set pixel size, point clouds are extracted pixel by pixel from its XY projection plane, and vegetation cover and leaf area index are calculated, such as... Figure 15 , Figure 16 As shown; by using the target area vector data for raster cropping, the vegetation cover and leaf area index of the target area can be obtained, such as... Figure 17 , Figure 18 As shown.

[0145] 4. Effective deployment range of the grid based on CHM identification

[0146] In reality, a forest stand often experiences gaps due to tree death caused by various factors such as environmental conditions, its own internal factors, or human intervention. When establishing forest stand standard plots, it is necessary to avoid forest gap areas to prevent errors in survey and monitoring results. Furthermore, complete and accurate identification of forest gaps can accurately reflect the actual area of ​​the forest, thereby reducing errors in the point-to-area estimation process.

[0147] This invention, based on the CHM extracted during data preprocessing, identifies areas with canopy height less than a specified threshold and continuous area greater than a specified threshold as forest gaps, such as... Figure 19 As shown in the gray area; by inverting the forest gap identification results, the standard effective deployment range can be obtained, such as... Figure 20 As shown in the gray area.

[0148] 5. Grid layout and effective grid selection

[0149] Based on the target area vector data and the specified grid side length, a grid is laid out to completely cover its range, such as... Figure 21 As shown; select cells that are entirely within the effective layout area as effective grid cells, such as... Figure 22 As shown.

[0150] 6. Extraction of mean values ​​of effective grid forest structure parameters and calculation of high density index

[0151] Vegetation cover (CC), canopy height (CHM), and leaf area index (LAI) were normalized according to equation (1). Using the selected effective grid, the normalized mean values ​​of the parameters within the grid area were extracted, and the High Density Index (HDI) was calculated according to equations (2) and (3). The changing trends of effective grid forest structure parameters from low to high are as follows: Figure 23 As shown.

[0152] 7. Conduct effective grid layering based on high density index

[0153] Based on the HDI values ​​statistically analyzed within the effective grid, the Jenks natural discontinuity method was used, with a threshold of 0.5630, to divide the high-density index into two groups, as follows: Figure 24 As shown; the spatial distribution of each grid layer is as follows Figure 25 As shown.

[0154] 8. Calculate the HDI distribution layer by layer and extract the grid of the index concentration distribution range.

[0155] The kernel density estimation method is used to determine the two-layer HDI distribution density function, such as... Figure 26 , Figure 27 As shown. The median HDI corresponding to the maximum distribution density is used. Using one standard deviation as the bandwidth, grids whose HDI falls within this bandwidth are selected as candidate grids, such as... Figure 28 As shown.

[0156] 9. Selecting typical grids layer by layer based on terrain factors

[0157] Based on the DEM generated from preprocessed point cloud data, a slope raster is calculated. The mean elevation, elevation span, mean slope, and slope span are extracted as terrain factors by selecting grid cells one by one. The kernel density function of each factor is then calculated, such as... Figures 29-32 As shown, grid cells within the mean ± n standard deviations of terrain factors are retained, and the rest are removed.

[0158] 10: Standards compliance verification and final determination

[0159] The grid was manually reviewed in conjunction with data such as imagery, topography, and forest structure parameters. Further checks and screenings were conducted to select the areas for the final field survey. Figure 34 As shown, a field investigation was conducted.

[0160] The airborne lidar point cloud data of this invention can completely and accurately replicate the three-dimensional information of the real world. Based on the method provided by this invention, forest stand factor information can be accurately extracted, and standard plots can be provided according to the standard plot setting principle. At the same time, the topographic factors of the plots are extracted, which can ensure the typicality of the plots while avoiding the uncertainty brought by complex terrain to forest survey work.

[0161] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for setting a forest monitoring plot based on airborne lidar data, characterized by, The method comprises the following steps: S1: airborne laser radar surveying and mapping is carried out on the survey monitoring target forest stand; S2: post-processing settlement is carried out to obtain surveying and mapping laser radar point cloud result data; S3: point cloud data is preprocessed; S4: a grid effective layout range is identified based on CHM; S5: grid layout and effective grid screening; S6: effective grid forest structure parameter mean value extraction and high density index calculation; S7: effective grid layering is carried out based on the high density index; S8: HDI distribution is calculated layer by layer, and index concentrated distribution range grid is extracted; S9: typical grid is screened based on terrain factors; S10: standard compliance verification and final determination; In step S4, the algorithm identifies the area in the survey range with canopy height less than 6m and continuous area greater than 5m 2 , and determines it as a grid layout invalid area, and the rest is the effective layout range. The method of step S5, grid layout and effective grid screening, comprises the following steps: S5-1: based on the survey range and the preset grid length, a standard grid that can completely cover the survey range is laid out, and the standard grid length should be an integer multiple of the grid length; S5-2: when the grid is completely located in the effective range identified in step S4, the grid is determined as an effective grid, otherwise it is an invalid grid, and the effective grid is screened according to the rule; The method of step S6, effective grid forest structure parameter mean value extraction and high density index calculation, comprises the following steps: S6-1: the forest structure parameters are normalized according to formula (1); Formula (1) S6-2: the grid is overlaid with the forest structure parameter grid, and the mean value of the forest structure parameters in the region is calculated grid by grid; S6-3: according to formula (2) and formula (3): Equation (2) Equation (3) The stand height-density index (HDI) is calculated for each grid, where CHM is the average canopy height of the grid, CC is the average vegetation cover of the grid, LAI is the average leaf area index of the grid, W i is the weight of the corresponding factor, and the default value is 1 / 3; In step S7, the high density index calculated in step S6 is divided into N groups by using the Jenks natural breakpoint method, so as to ensure that the variance between groups is maximum and the variance within groups is minimum; The method of step S8, layer-by-layer calculation of HDI distribution and extraction of index concentrated distribution range grid, comprises the following steps: S8-1: all high density indexes in the Nth layer are extracted, and the kernel density estimation method is used to obtain the N-layer high density index distribution density function; S8-2: the maximum value of the distribution density is obtained, and the bandwidth is ±n times the standard deviation of the mean value of the N-layer HDI, and the grid within the bandwidth range in which the HDI is contained is selected as the candidate grid; The method of step S9, screening of typical grid based on terrain factors, comprises the following steps: S9-1: based on the DEM generated by preprocessing the point cloud data, the slope grid is calculated, and the terrain factors of the elevation mean value, elevation span, slope mean value and slope span are extracted for each candidate grid; S9-2: the kernel density function of each factor is calculated, and the grid outside the ±n times standard deviation of the mean value of the terrain factor is removed.

2. The method for setting a forest monitoring plot based on airborne LiDAR data according to claim 1, wherein, Step S1 involves conducting airborne lidar aerial surveys of the target forest stands: acquiring three-dimensional point cloud data of the target area. Data requirements: point density ≥ 100 pts / m². 2 Elevation accuracy ≤ 0.1m, horizontal accuracy ≤ 0.2m. 3.The method of claim 1, wherein, The method of step S3, preprocessing of point cloud data, comprises the following steps: S3-1: denoising and ground point classification Remove data noise such as wires, birds and isolated points to improve data quality; search for ground points in the point cloud data; S3-2: DEM, DSM, CHM extraction and cropping DSM and DEM are generated by interpolating the surface points and ground points; CHM is obtained by subtracting the DEM value from the DSM value based on the corresponding spatial position; the CHM is subjected to mask analysis according to the survey range vector to retain the DN value of the pixels in the survey area; S3-3: point cloud normalization According to the ground point or DEM, the point cloud is normalized; S3-4: Forest structure parameter grid calculation and cutting Based on high spatial resolution grid, all points above the corresponding position two-dimensional plane are calculated for each pixel, and the forest structure parameter value of the corresponding pixel is calculated by using the point set, and the forest structure parameters include vegetation coverage, leaf area index, and canopy height.

4. The method for setting a forest monitoring plot based on airborne LiDAR data according to claim 1, wherein, The method for standard compliance verification and final determination in step S10 includes the following steps: S10-1: According to the standard site setting requirements, the standard site setting principle is verified, and whether there are typical grids that do not meet the standard site setting requirements is checked, and if there are, they are manually removed; S10-2: According to the work requirements, the user selects the corresponding area to carry out field investigation according to the spatial position of the typical grid selected in step S9 based on the terrain factor.

Images