Tree crown layer effective coverage estimation method and system based on three-dimensional point cloud data

Abstract

The present application relates to a kind of based on three-dimensional point cloud data's tree crown layer effective coverage estimation method and system.The based on three-dimensional point cloud data's tree crown layer effective coverage estimation method and system described in the application include: obtaining the three-dimensional point cloud data of target area;The three-dimensional point cloud data is preprocessed;Pulse is set in the three-dimensional point cloud data;The three-dimensional point cloud data is input into tree crown layer effective coverage estimation model, obtains crown layer data and the weight of point on pulse, according to the three-dimensional point cloud data, crown layer data and the weight of point on pulse, estimates tree crown layer effective coverage.The based on three-dimensional point cloud data's tree crown layer effective coverage estimation method and system described in the application have the advantages that data acquisition mode is simple, fast and accurate in calculating tree crown layer effective coverage.

Description

Technical Field

[0001] This invention relates to the field of tree canopy effective coverage monitoring technology, and in particular to a method and system for estimating tree canopy effective coverage based on three-dimensional point cloud data. Background Technology

[0002] Mangroves are woody plant communities that grow in the intertidal zone of tropical and subtropical coasts and are periodically submerged by seawater. They possess immense ecological and social value, especially their powerful carbon sequestration capacity. To determine the carbon sequestration capacity of mangrove ecosystems, it is necessary to estimate mangrove biomass. Mangrove biomass is strongly correlated with tree canopy cover, and the effective canopy coverage is also an essential parameter.

[0003] Existing methods for extracting effective tree cover are primarily image-based, analyzing the proportion of vegetation and soil pixels to determine the effective tree cover. However, since images are two-dimensional models, it is difficult to extract the effective canopy cover. While there are methods based on canopy height models to estimate effective tree canopy cover, the raster size and height threshold are difficult to determine. Summary of the Invention

[0004] Based on this, the purpose of the present invention is to provide a method and system for estimating the effective coverage of the tree canopy layer based on three-dimensional point cloud data, which has the advantages of fast, simple and accurate estimation of the effective coverage of the tree canopy layer.

[0005] A method for estimating the effective canopy cover based on 3D point cloud data includes the following steps:

[0006] Acquire 3D point cloud data of the target area;

[0007] Preprocess the three-dimensional point cloud data;

[0008] Pulses are set in the three-dimensional point cloud data;

[0009] The three-dimensional point cloud data is input into the tree canopy effective coverage estimation model to obtain the canopy data and the weights of the points on the pulse. Based on the three-dimensional point cloud data, canopy data and the weights of the points on the pulse, the effective coverage of the tree canopy is estimated.

[0010] A system for estimating the effective canopy cover based on 3D point cloud data includes:

[0011] Data acquisition device, used to acquire three-dimensional point cloud data of the target area;

[0012] A data preprocessing device is used to preprocess the three-dimensional point cloud data;

[0013] A pulse setting device for setting pulses in the three-dimensional point cloud data;

[0014] An effective canopy coverage estimation device is used to input the three-dimensional point cloud data into the tree canopy effective coverage estimation model, obtain the canopy data and the weights of the pulse points, and estimate the effective canopy coverage based on the three-dimensional point cloud data, canopy data and the weights of the pulse points.

[0015] A computer device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for estimating the effective canopy coverage based on three-dimensional point cloud data.

[0016] A computer-readable storage medium storing a computer program, characterized in that: when the computer program is executed by a processor, it implements the steps of the above-described method for estimating the effective canopy coverage based on three-dimensional point cloud data.

[0017] The tree canopy effective coverage estimation method based on 3D point cloud data in this application involves acquiring and preprocessing 3D point cloud data from a target area; then, setting pulses in the 3D point cloud data; finally, inputting the 3D point cloud data into a tree canopy effective coverage estimation model to obtain the weights of the canopy data and the points on the pulses; and estimating the effective canopy coverage based on the weights of the 3D point cloud data, canopy data, and the points on the pulses. This method for estimating tree canopy effective coverage based on 3D point cloud data is simple to acquire and does not damage the original environment; the use of weighted estimation in the tree canopy effective coverage estimation further improves the accuracy of coverage estimation.

[0018] To better understand and implement this invention, the following detailed description is provided in conjunction with the accompanying drawings. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the steps of the tree canopy layer effective coverage estimation method based on 3D point cloud data in the embodiments of this application;

[0020] Figure 2 This is a flowchart illustrating the steps for obtaining the three-dimensional point cloud data of the target region in an embodiment of this application.

[0021] Figure 3 This is a flowchart illustrating the steps of normalizing the three-dimensional point cloud data in this embodiment of the application.

[0022] Figure 4 This is a flowchart illustrating the steps for denoising the 3D point cloud data in an embodiment of this application.

[0023] Figure 5 This is a flowchart illustrating the steps involved in training the effective canopy coverage estimation model in an embodiment of this application.

[0024] Figure 6 This is a flowchart illustrating the steps for labeling the small sample point cloud data in this embodiment of the application.

[0025] Figure 7 This is a structural diagram of the tree canopy layer effective coverage estimation system based on 3D point cloud data in the embodiments of this application;

[0026] Figure 8 This is a schematic diagram of a computer device used for estimating the effective canopy coverage based on three-dimensional point cloud data in an embodiment of this application. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0028] Please see Figure 1 , Figure 1 This is a flowchart illustrating the steps of the tree canopy layer effective coverage estimation method based on 3D point cloud data in the embodiments of this application.

[0029] A method for estimating the effective canopy cover based on 3D point cloud data includes the following steps:

[0030] S101, acquire the 3D point cloud data of the target area;

[0031] S102, Preprocess the three-dimensional point cloud data;

[0032] S103, Set a pulse in the three-dimensional point cloud data;

[0033] S104, input the three-dimensional point cloud data into the tree canopy effective coverage estimation model, obtain the canopy data and the weights of the pulse points, and estimate the tree canopy effective coverage based on the three-dimensional point cloud data, canopy data and the weights of the pulse points.

[0034] The tree canopy effective coverage estimation method based on 3D point cloud data in this application involves acquiring and preprocessing 3D point cloud data from a target area; then, setting pulses in the 3D point cloud data; finally, inputting the 3D point cloud data into a tree canopy effective coverage estimation model to obtain the weights of the canopy data and the points on the pulses; and estimating the effective canopy coverage based on the weights of the 3D point cloud data, canopy data, and the points on the pulses. This method for estimating tree canopy effective coverage based on 3D point cloud data is simple to acquire and does not damage the original environment; the use of weighted estimation in the tree canopy effective coverage estimation further improves the accuracy of coverage estimation.

[0035] For step S101, acquire the three-dimensional point cloud data of the target area;

[0036] The target area is a pre-defined area where the effective coverage of the canopy layer needs to be estimated, and the three-dimensional point cloud data includes latitude, longitude, elevation, and echo information.

[0037] Please see Figure 2 , Figure 2 This is a flowchart illustrating the steps for obtaining the 3D point cloud data of the target region in an embodiment of this application. In one embodiment, obtaining the 3D point cloud data of the target region includes the following steps:

[0038] S201, Based on the target area, obtain the UAV flight path using flight path planning software;

[0039] S202, according to the drone flight path, control the drone to fly in the target area and acquire three-dimensional point cloud data of the target area.

[0040] The flight path planning software is used to plan the flight path of the UAV. In this embodiment, the flight path planning software is not limited; any software capable of planning the UAV flight path based on the target area can be used for UAV flight path acquisition in this application.

[0041] In this embodiment, the flight path planning software is used to plan a flight path for the target area, thereby obtaining the drone's flight path within the target area. Then, based on the drone's flight path, the drone is controlled to fly and acquire 3D point cloud data of the target area. In one embodiment, the 3D point cloud data of the target area is acquired using a lidar camera mounted on the drone.

[0042] In one embodiment, after planning the drone's flight path using flight path planning software, the DJI Matrice M300RTK drone equipped with a LiDAR lens is controlled to fly in the target area according to the drone's flight path and acquire three-dimensional point cloud data of the target area.

[0043] In this application, three-dimensional point cloud data is acquired by using a drone equipped with a lidar lens. The data acquisition method is simple and quick, and the method used is non-contact, which reduces a lot of manpower, material resources and financial resources without damaging the environment.

[0044] For step S102, the three-dimensional point cloud data is preprocessed;

[0045] Preprocessing the 3D point cloud data primarily aims to reduce data interference and improve the accuracy of estimating the effective canopy coverage. In one embodiment, the preprocessing includes denoising and normalization.

[0046] Please see Figure 3 , Figure 3 This is a flowchart illustrating the steps of normalizing the 3D point cloud data in an embodiment of this application. In one embodiment, the preprocessing of the 3D point cloud data includes the following steps:

[0047] S301, Fit the elevation of the data points in the three-dimensional point cloud data to a preprocessed elevation curve, obtain the abrupt change in the preprocessed slope in the preprocessed elevation curve, and take the protruding part of the preprocessed slope as the preprocessed segmentation point.

[0048] S302, based on the preprocessed segmentation points, obtain preliminary data of ground data points and non-ground data points, and select a specified area based on the preliminary data;

[0049] S303, In the specified area, calculate the elevation judgment value for each data point;

[0050] S304, if the elevation judgment value is higher than the preset threshold, then the data point is a non-ground point; if the elevation judgment value is lower than the preset threshold, then the data point is a ground point.

[0051] S305, subtract the ground points from the data points in the three-dimensional point cloud data to obtain normalized three-dimensional point cloud data.

[0052] For step S301, the elevation of the data points in the three-dimensional point cloud data is fitted to a preprocessed elevation curve, the abrupt change in the preprocessed slope in the preprocessed elevation curve is obtained, and the protruding part of the preprocessed slope is taken as the preprocessed segmentation point.

[0053] The points where the preprocessing slope changes abruptly include those points in the preprocessing elevation curve where the slope changes significantly. In this embodiment, since the number of ground point clouds is significantly higher than the number of point clouds in the main trunk of the tree, there is a relatively large slope change in the preprocessing elevation curve. Therefore, the points in the preprocessing elevation curve where the slope changes abruptly are used as preprocessing segmentation points.

[0054] In this embodiment of the application, when normalizing the three-dimensional point cloud data, the elevation of the data points in the three-dimensional point cloud data is first fitted to a preprocessed elevation curve, the abrupt change in the preprocessed slope in the preprocessed elevation curve is obtained, and the abrupt change in the preprocessed slope is used as the preprocessed segmentation point.

[0055] For step S302, preliminary data of ground data points and non-ground data points are obtained based on the preprocessed segmentation points, and a specified area is selected based on the preliminary data;

[0056] The preliminary data includes points that are initially filtered to separate ground and non-ground data points. This preliminary data is used for fine-tuning the separation of ground and non-ground data points. The designated area includes a small region centered on a point in the preliminary data. The selected designated area is the region formed by the ground and non-ground points selected in the preliminary data.

[0057] Preliminary data of ground data points and non-ground data points are obtained based on the preprocessed segmentation points, and then a specified region is selected from the preliminary data. In one embodiment, the specified region can be selected automatically by the system or manually.

[0058] For step S303, in the specified field, calculate the elevation judgment value for each data point;

[0059] The elevation judgment value is used to determine whether the data point is a ground point or a non-ground point.

[0060] The calculation of the elevation judgment value for each data point includes:

[0061] Elevation judgment value = (elevation value of the data point - elevation value of the lowest point in the three-dimensional point cloud data) / distance between the data point and the lowest point in the three-dimensional point cloud data.

[0062] For steps S304 to S305, the preset threshold is a pre-set threshold used to distinguish between ground data points and non-ground data points. After calculating and obtaining the elevation judgment value, the elevation judgment value is compared with the preset threshold. If the elevation judgment value is higher than the preset threshold, the data point is a non-ground point; if the elevation judgment value is lower than the preset threshold, the data point is a ground point.

[0063] After all data points in the specified area have been evaluated, the ground data points are subtracted from the data points in the 3D point cloud data to obtain the normalized data.

[0064] In this embodiment of the application, by normalizing the three-dimensional point cloud data, the stability of the three-dimensional point cloud data is further ensured, and the accuracy of the calculation of the effective coverage of the canopy layer is improved.

[0065] Please see Figure 4 , Figure 4 This is a flowchart illustrating the steps for denoising the 3D point cloud data in an embodiment of this application. In one embodiment, the preprocessing of the 3D point cloud data further includes the following steps:

[0066] S401, Set a sphere radius for each data point in the three-dimensional point cloud data, and count the number of neighboring points within the sphere radius;

[0067] S402, if the number of neighboring points is less than a set threshold, then the data point is defined as an isolated point and the isolated point is deleted.

[0068] The neighboring points include the data points contained within the sphere centered on the data point, and the radius of the sphere is the radius of the sphere used to determine the neighboring points. The sphere radius can be automatically selected by the system or set by the user.

[0069] In this embodiment, the number of neighboring points near the data point is obtained by setting the radius of the sphere. When the number of neighboring points is lower than a set threshold, the data point is defined as an isolated point and deleted, thereby reducing the impact of the isolated points on the effective coverage estimation of the canopy layer.

[0070] For step S103, a pulse is set in the three-dimensional point cloud data;

[0071] The pulse is a signal that occurs for a short time within the entire signal period, relative to a continuous signal. In this embodiment, multiple pulses are set in the three-dimensional point cloud data to accurately estimate the effective canopy coverage.

[0072] In one embodiment, by acquiring the two data points at the top left or top right of the three-dimensional point cloud data, the width of the two data points is set as the pulse width, and the pulses are set sequentially from left to right or from right to left.

[0073] S104, input the three-dimensional point cloud data into the tree canopy effective coverage estimation model, obtain the canopy data and the weights of the pulse points, and estimate the tree canopy effective coverage based on the three-dimensional point cloud data, canopy data and the weights of the pulse points;

[0074] The tree canopy effective coverage estimation model is based on the LSTM (Long Short-Term Memory) model. The LSTM model is built upon the RNN model by adding a threshold, thus solving the short-term memory problem of RNNs and enabling recurrent neural networks to effectively utilize long-distance temporal information.

[0075] In one embodiment, the canopy coverage estimation model further adds a fully connected layer to the LSTM long short-term memory model, using the sigmoid activation function. In this fully connected layer, each neuron receives all the outputs from the previous layer, performs a weighted summation based on corresponding weights, and then applies an activation function to the summation result to produce the final output. The sigmoid activation function is a commonly used non-linear function that maps the input real value to a range between 0 and 1. In this embodiment, the fully connected layer in the canopy coverage estimation model is used to classify canopy data and non-canopy data based on the 3D point cloud data, facilitating the calculation of the canopy coverage.

[0076] The weight of each point on a pulse is 1 / the number of points on each pulse. For example, if there are 5 points on a pulse, the weight of each point on the pulse is 0.2.

[0077] In this embodiment, after receiving the 3D point cloud data with pre-set pulses, the canopy coverage estimation model first obtains the corresponding canopy data and the weights of the points on the pulses. Based on the 3D point cloud data, canopy data, and the weights of the points on the pulses, it estimates the effective canopy coverage. In one embodiment, the estimated effective canopy coverage is the ratio of the sum of the weights of the canopy data points on the pulses to the total weights of the data points in the point cloud density.

[0078] Please see Figure 5 , Figure 5 This is a flowchart illustrating the steps of training the effective canopy cover estimation model in an embodiment of this application. In one embodiment, the method for estimating the effective canopy cover based on 3D point cloud data further includes training the effective canopy cover estimation model, which includes the following steps:

[0079] S501, acquire 3D point cloud training data, and preprocess the 3D point cloud training data;

[0080] S502, the three-dimensional point cloud data is divided into small sample point cloud data according to the latitude and longitude points;

[0081] S503, Set a pulse for the small sample point cloud data;

[0082] S504, In the small sample point cloud data, label the weights of canopy data, non-canopy data, and points on the pulse;

[0083] S505, the weights of the small sample point cloud data and the pulse points are used as inputs to the tree canopy effective coverage estimation model, and the ratio of the weights of the pulse points in the canopy data to the weights of the pulse points in the small sample point cloud data is used as the output of the tree canopy effective coverage estimation model.

[0084] S506, The accuracy of the tree canopy effective coverage estimation model is evaluated through error analysis.

[0085] For step S501, acquire 3D point cloud training data and preprocess the 3D point cloud training data;

[0086] In this embodiment, when training the tree canopy effective coverage estimation model, three-dimensional point cloud training data is first acquired by a drone equipped with a lidar, and the three-dimensional point cloud data is preprocessed. The preprocessing method is the same as that in step S102, and will not be described again here.

[0087] For step S502, the three-dimensional point cloud data is divided into small sample point cloud data according to the latitude and longitude points;

[0088] The small sample point cloud data refers to the small sample data obtained by dividing the three-dimensional point cloud data.

[0089] In this embodiment, the three-dimensional point cloud data is divided into small sample point cloud data based on the latitude and longitude information in the three-dimensional point cloud data, which facilitates the annotation of the three-dimensional point cloud data and the training of the tree canopy layer effective coverage estimation model.

[0090] For step S503, a pulse is set for the small sample point cloud data;

[0091] In one embodiment, setting a pulse for the sample point cloud data includes the following steps:

[0092] Extract the two data points at the top left of the small sample point cloud data and obtain the width of the two data points;

[0093] The pulses are set from left to right according to the width.

[0094] In this embodiment, the width between the two data points at the upper left end of the small sample point cloud data is obtained, and a pulse is set in the small sample point cloud data according to the width.

[0095] For step S504, the weights of canopy data, non-canopy data, and pulse points are labeled in the small sample point cloud data;

[0096] Please see Figure 6 , Figure 6 This is a flowchart illustrating the steps for labeling the small sample plot point cloud data in an embodiment of this application. In one embodiment, labeling the canopy data, non-canopy data, and the weights of points on impulses in the small sample plot point cloud data includes the following steps:

[0097] S601, Fit the elevation values ​​of the data points in the small sample point cloud data into an elevation curve;

[0098] S602, Obtain the point where the slope of the elevation curve changes abruptly, and use the point where the slope changes abruptly as a segmentation point;

[0099] S603, based on the segmentation points, label the canopy data and non-canopy data of the small sample point cloud data;

[0100] S604 extracts the points on each pulse, calculates and labels the weight of the points on each pulse.

[0101] Steps S601 to S603 are the methods for annotating canopy data and non-canopy data in this embodiment.

[0102] In this embodiment, when labeling canopy data and non-canopy data, since the number of points in the canopy data is significantly higher than that in the point cloud data of the main stem of the tree, an elevation curve is generated by fitting the elevation data of the data points in the small quadrat point cloud data, and the abrupt change in slope in the elevation curve is used as the dividing point, thereby dividing the small quadrat point cloud data into canopy data and non-canopy data according to the dividing point.

[0103] For step S604, extract the points on each pulse, calculate and label the weight of the points on each pulse;

[0104] In this embodiment, the points on the pulse are manually calculated and labeled, wherein the weight of each point on each pulse is 1 / the number of points on each pulse.

[0105] For steps S505 to S506, in this embodiment, after the weights of the canopy data, non-canopy data, and pulse points are labeled in the small quadrat point cloud data, the small quadrat point cloud data is input into the canopy effective coverage estimation model. The canopy effective coverage estimation model takes the small quadrat point cloud data and the weights of the pulse points as input, and outputs the ratio of the weights of the points in the pulse canopy data to the total weights of the pulse points in the small quadrat point cloud data. Finally, the output of the canopy effective coverage estimation model is evaluated by error analysis, and the canopy effective coverage estimation model is trained.

[0106] The tree canopy effective coverage estimation method based on 3D point cloud data in this application involves acquiring 3D point cloud data by flying a drone equipped with a lidar over a target area. After preprocessing the 3D point cloud data through denoising and normalization and setting pulses, the 3D point cloud data is input into a tree canopy effective coverage estimation model. Based on the fully connected layers in the tree canopy effective coverage estimation model, canopy data and non-canopy data are obtained. Then, the tree canopy effective coverage estimation model estimates the tree canopy effective coverage based on the canopy data, 3D point cloud data, and the weights of points on the pulses.

[0107] The tree canopy effective coverage estimation method based on 3D point cloud data in this application is simple and convenient in data acquisition; moreover, it adopts a non-contact estimation method, which can reduce a lot of manpower, material resources and financial resources without damaging the environment. This application also uses a fully connected layer for classification and weight estimation, making the tree canopy effective coverage estimation model more robust and the estimation of the effective coverage of the tree canopy more accurate.

[0108] Please see Figure 7 , Figure 7 This is a structural diagram of the tree canopy effective coverage estimation system based on 3D point cloud data in an embodiment of this application. This application also discloses a tree canopy effective coverage estimation system based on 3D point cloud data, comprising:

[0109] Data acquisition device 11 is used to acquire three-dimensional point cloud data of the target area;

[0110] Data preprocessing device 12 is used to preprocess the three-dimensional point cloud data;

[0111] Pulse setting device 13 is used to set pulses in the three-dimensional point cloud data;

[0112] The effective coverage estimation device 14 is used to input the three-dimensional point cloud data into the tree canopy effective coverage estimation model, obtain the canopy data and the weights of the pulse points, and estimate the effective coverage of the tree canopy based on the three-dimensional point cloud data, canopy data and the weights of the pulse points.

[0113] It should be noted that the tree canopy effective coverage estimation system based on 3D point cloud data provided in the above embodiments is only illustrated by the division of the above functional modules when executing the tree canopy effective coverage estimation method based on 3D point cloud data. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the tree canopy effective coverage estimation system based on 3D point cloud data and the tree canopy effective coverage estimation method based on 3D point cloud data provided in the above embodiments belong to the same concept, and its implementation process is detailed in the above embodiments, and will not be repeated here.

[0114] Please see Figure 8 , Figure 8 This is a schematic diagram of a computer device used for estimating the effective canopy coverage based on 3D point cloud data, as described in an embodiment of this application. Figure 8 As shown, the computer device 21 includes: a processor 211, a memory 212, and a computer program 213 stored in the memory 212 and executable on the processor 211, such as a tree canopy effective coverage estimation program based on three-dimensional point cloud data; the processor 211 executes the computer program 213 to implement the tree canopy effective coverage estimation method based on three-dimensional point cloud data described in the above embodiments.

[0115] The processor 211 may include one or more processing cores. The processor 211 connects to various parts within the computer device 21 using various interfaces and lines. It executes various functions of the computer device 21 and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 212, and by accessing data in memory 212. Optionally, the processor 211 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 211 may integrate one or more of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for the touch screen; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip, without being integrated into the processor 211.

[0116] The memory 212 may include random access memory (RAM) or read-only memory. Optionally, the memory 212 may include non-transitory computer-readable storage medium. The memory 212 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 212 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch instructions), instructions for implementing the various method embodiments described above, etc.; the data storage area may store data involved in the various method embodiments described above, etc. Optionally, the memory 212 may also be at least one storage device located remotely from the aforementioned processor 211.

[0117] This application also provides a computer-readable storage medium that can store multiple instructions. These instructions are applicable to being loaded by a processor and executed by the method steps of the above embodiments. For details of the execution process, please refer to the specific description of the above embodiments, which will not be repeated here.

[0118] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A method for estimating the effective canopy cover of a tree based on three-dimensional point cloud data, characterized in that, Includes the following steps: Acquire 3D point cloud data of the target area; Preprocess the three-dimensional point cloud data; Pulses are set in the three-dimensional point cloud data; Training a model for estimating effective canopy cover includes: Acquire 3D point cloud training data of a preset area, and preprocess the 3D point cloud training data; The three-dimensional point cloud data is divided into small sample square point cloud data according to the latitude and longitude points; Extract the two data points at the top left of the small sample point cloud data and obtain the width of the two data points; Based on the width, pulses are set from left to right; In the small sample point cloud data, the weights of canopy data, non-canopy data, and points on the pulses are labeled; The weights of the small sample point cloud data and the pulse points are used as inputs to the tree canopy effective coverage estimation model, and the ratio of the weights of the pulse points in the canopy data to the weights of the pulse points in the small sample point cloud data is used as the output of the tree canopy effective coverage estimation model. The accuracy of the tree canopy effective coverage estimation model was evaluated through error analysis; The three-dimensional point cloud data is input into the tree canopy effective coverage estimation model to obtain the canopy data and the weights of the points on the pulse. Based on the three-dimensional point cloud data, canopy data and the weights of the points on the pulse, the effective coverage of the tree canopy is estimated.

2. The method for estimating the effective canopy coverage based on three-dimensional point cloud data according to claim 1, characterized in that, The acquisition of the 3D point cloud data of the target area includes the following steps: Based on the target area, the drone flight path is obtained using flight path planning software; According to the drone flight path, control the drone to fly in the target area and acquire three-dimensional point cloud data of the target area.

3. The method for estimating the effective canopy coverage based on three-dimensional point cloud data according to claim 1, characterized in that, The preprocessing of the 3D point cloud data includes the following steps: The elevation of the data points in the three-dimensional point cloud data is fitted to a preprocessed elevation curve, and the abrupt change in the preprocessed slope in the preprocessed elevation curve is obtained. The abrupt change in the preprocessed slope is used as the preprocessed segmentation point. Based on the preprocessed segmentation points, preliminary data of ground data points and non-ground data points are obtained, and a specified area is selected based on the preliminary data; Within the specified area, calculate the elevation judgment value for each data point; If the elevation judgment value is higher than the preset threshold, the data point is a non-ground point; if the elevation judgment value is lower than the preset threshold, the data point is a ground point. Subtract the ground points from the data points in the 3D point cloud data to obtain the normalized 3D point cloud data.

4. The method for estimating the effective canopy coverage based on three-dimensional point cloud data according to claim 3, characterized in that, The preprocessing of the three-dimensional point cloud data further includes the following steps: Set a sphere radius for each data point in the point cloud data, and count the number of neighboring points within the sphere radius; If the number of neighboring points is less than a set threshold, the data point is defined as an isolated point and deleted.

5. The method for estimating the effective canopy coverage based on three-dimensional point cloud data according to claim 1, characterized in that, The step of labeling the weights of canopy data, non-canopy data, and pulse points in the small sample plot point cloud data includes the following steps: The elevation values ​​of the data points in the small sample point cloud data are fitted into an elevation curve; Identify the points where the slope of the elevation curve changes abruptly, and use these points as dividing points; Based on the segmentation points, label the canopy data and non-canopy data of the small sample point cloud data; Extract the points on each pulse, calculate and label the weight of each point on each pulse.

6. A system for estimating the effective canopy cover based on 3D point cloud data, characterized in that, include: Data acquisition device, used to acquire three-dimensional point cloud data of the target area; A data preprocessing device is used to preprocess the three-dimensional point cloud data; A pulse setting device for setting pulses in the three-dimensional point cloud data; An effective canopy cover estimation device, used to train a canopy layer effective canopy cover estimation model, includes: Acquire 3D point cloud training data of a preset area, and preprocess the 3D point cloud training data; The three-dimensional point cloud data is divided into small sample square point cloud data according to the latitude and longitude points; Extract the two data points at the top left of the small sample point cloud data and obtain the width of the two data points; Based on the width, pulses are set from left to right; In the small sample point cloud data, the weights of canopy data, non-canopy data, and points on the pulses are labeled; The weights of the small sample point cloud data and the pulse points are used as inputs to the tree canopy effective coverage estimation model, and the ratio of the weights of the pulse points in the canopy data to the weights of the pulse points in the small sample point cloud data is used as the output of the tree canopy effective coverage estimation model. The accuracy of the tree canopy effective coverage estimation model was evaluated through error analysis; The three-dimensional point cloud data is input into the tree canopy effective coverage estimation model to obtain the canopy data and the weights of the points on the pulse. Based on the three-dimensional point cloud data, canopy data and the weights of the points on the pulse, the effective coverage of the tree canopy is estimated.

7. A computer device, comprising: A processor, a memory, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the computer program, it implements the steps of the tree canopy layer effective coverage estimation method based on three-dimensional point cloud data as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, it implements the steps of the tree canopy layer effective coverage estimation method based on three-dimensional point cloud data as described in any one of claims 1 to 5.

Images