A method for establishing a shelterbelt biomass estimation model based on laser radar point cloud data

Abstract

This invention proposes a method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data, comprising the following steps: Step 1) Using lidar to scan the farmland shelterbelt to obtain three-dimensional point cloud data of the farmland shelterbelt, and constructing a single-tree voxel space from the preprocessed point cloud data; Step 2) Excavating the underground parts of the trees, scanning them with lidar to obtain point cloud data of the underground parts of individual trees; Step 3) Stitching together the scanned data of the aboveground and underground parts of individual trees to construct a complete single-tree model; Step 4) Establishing a single-tree biomass estimation model, and then calculating the biomass of the shelterbelt. Compared with traditional acquisition methods, this invention is convenient, non-contact, and accurate. It can completely extract the parameters of the aboveground and underground parts of individual trees, and can quickly obtain the three-dimensional structural information of the aboveground and underground parts of individual trees. It also develops a method for extracting and calculating the parameters of the aboveground and underground parts of individual trees from lidar point cloud data to estimate the biomass of shelterbelts.

Description

Technical Field

[0001] This invention belongs to the field of forestry technology, and specifically relates to a method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data. Background Technology

[0002] Calculating the biomass of shelterbelts is of great significance for understanding regional biomass. Currently, estimates of shelterbelt biomass primarily use parameters such as sapwood area, diameter at breast height (DBH), tree height, and crown width. Existing technologies for studying tree volume often employ field measurements, which are destructive, time-consuming, and labor-intensive.

[0003] Light Detection and Ranging (LiDAR), as a novel and popular remote sensing technology, has attracted widespread attention from scholars since its inception. It uses laser ranging principles (including pulsed laser and phased laser) to instantly measure three-dimensional spatial coordinates. Simultaneously, the spatial point cloud data acquired using this technology can be used to quickly build three-dimensional visualization models of complex and irregular scenes, saving both time and effort. In forestry, foreign countries were the first to utilize this technology for extracting individual tree parameters (tree height, diameter at breast height, crown, trunk diameter), estimating forest biomass and volume, and creating digital terrain models, achieving significant results. Domestic application of this technology started later, with related research mainly focusing on forestry resource surveys and three-dimensional reconstruction of individual trees. Compared to the two-dimensional planar information provided by traditional optics, LiDAR can acquire tree point cloud data with accurate spatial geometric information, reconstructing three-dimensional tree models and laying the foundation for the extraction and inversion of forest structure parameters. Summary of the Invention

[0004] To address the problems of time-consuming and labor-intensive calculations of shelterbelt biomass in existing technologies, which often neglect underground components, this invention proposes a method for estimating the water consumption of shelterbelts based on lidar point cloud data. This invention employs the following technical solution:

[0005] A method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data includes the following steps:

[0006] Step 1: Use lidar to acquire point cloud data of the shelterbelt, and use the point cloud data of the shelterbelt to separate the parameters of each individual tree in the shelterbelt, including tree height, diameter at breast height, etc.

[0007] Step 2: Excavate the underground part of the tree, use LiDAR to scan and obtain point cloud data, and use the point cloud data of the underground part to extract the parameters of the underground part of a single tree.

[0008] Step 3: Combine the point cloud data of the above-ground and underground parts of a single log to establish a single log volume calculation model;

[0009] Step 4: Based on the single timber volume calculation model, establish a biomass estimation model for the shelterbelt.

[0010] Optionally, step one specifically includes:

[0011] 1) Obtain point cloud data of shelterbelts;

[0012] 2) Perform individual tree segmentation on the point cloud data of the shelterbelt to obtain individual tree point cloud data;

[0013] 3) Calculate the volume of the aboveground portion of a single tree using point cloud data.

[0014] Optionally, step two specifically includes:

[0015] 1) Excavate the underground part of the single log;

[0016] 2) Obtain point cloud data of the underground portion of a single log;

[0017] 3) Calculate the volume of the underground portion of a single log using point cloud data.

[0018] Optionally, the acquisition of shelterbelt point cloud data specifically includes: scanning the farmland shelterbelt using a ground-based lidar with an S-shaped scanning path; stitching and exporting the acquired shelterbelt point cloud data in the scanning instrument's software, removing noise, and finally exporting the point cloud data in txt format.

[0019] Optionally, the single-log splitting process includes manual single-log splitting and automatic single-log splitting:

[0020] Manual single-tree segmentation involves importing the point cloud data of the shelterbelt into CloudCompare software, and then using the Segment function in the software to crop the data to obtain single-tree point cloud data that does not include ground points.

[0021] The automatic single-tree segmentation process involves importing the point cloud data of the protective forest belt into CloudCompare software for ground point separation. Using the CSFFilter function, corresponding parameters are selected based on the terrain conditions of the data to separate ground points from non-ground points. Point clouds that are not filtered out are manually removed, and the point cloud data is normalized. The forest belt point cloud that does not contain ground points is cropped using the Segment function. Finally, the single-tree point cloud data is exported one by one in txt format.

[0022] Optionally, the volume of a single timber can be calculated using point cloud data as follows:

[0023] Using Matlab software, the area of ​​each leaf was calculated individually using the aef.m and PointAverageSpacing.m algorithms.

[0024] Optionally, the biomass of a single tree can be calculated based on the parameters.

[0025] Optionally, the biomass estimation model for farmland shelterbelts established in step four is a function model with individual tree biomass as the dependent variable and individual tree height and diameter at breast height as independent variables.

[0026] Optionally, the trees in the farmland shelterbelt are Xinjiang poplar.

[0027] A method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data includes the following steps:

[0028] Point cloud data of the shelterbelt was acquired using lidar, and the volume of each individual tree in the shelterbelt was separated using the point cloud data. The volume of each individual tree was then substituted into the established farmland shelterbelt biomass estimation model to calculate the biomass of each individual tree. The biomass of each individual tree was then summed to obtain the biomass of the shelterbelt.

[0029] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0030] This invention proposes a method to directly acquire three-dimensional point cloud data of shelterbelts using lidar, combined with computer algorithms, to conveniently and accurately calculate the volume of shelterbelt timber. Furthermore, a model for estimating the biomass of shelterbelts is established based on the timber volume. This invention can quickly, conveniently, efficiently, and accurately extract and calculate the above-ground and below-ground volume of trees in shelterbelts, and estimate the biomass of the shelterbelts. Attached Figure Description

[0031] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0032] Figure 1 This is a LiDAR scanning path diagram;

[0033] Figure 2 This is a schematic diagram of the single-tree scanning layout;

[0034] Figure 3 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0035] The following embodiments are provided to better understand the present invention and are not limited to the preferred embodiments described. They do not constitute a limitation on the content and scope of protection of the present invention. Any product that is the same as or similar to the present invention, derived by any person under the guidance of the present invention or by combining the features of the present invention with other prior art, falls within the protection scope of the present invention.

[0036] For experiments not specifically described in the examples, the procedures or conditions should be followed according to the conventional experimental procedures described in the literature in this field. Reagents or instruments whose manufacturers are not specified are all commercially available conventional reagent products.

[0037] This embodiment uses a two-row Xinjiang poplar farmland shelterbelt as the research object. A ground-based LiDAR scanner, FARO Focus S 150 (parameter information shown in Table 1), was used to acquire the three-dimensional point cloud data of the research object. The data was stitched together using SCENE software, and denoised and segmented into individual trees using Compare Cloud software to obtain the shelterbelt point cloud data.

[0038] Table 1 FARO Parameters for Ground-Based Laser Scanner

[0039] Scan size 20480*8533Pt Horizontal and vertical field of view 0°-360° and -60°-90° Laser scanning distance 4m laser wavelength 1550nm resolution 1280×720

[0040] A method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data, the method comprising the following steps:

[0041] (1) Acquisition and preprocessing of point cloud data of shelterbelt. Ground-based lidar was used to scan the farmland shelterbelt. The scanning path is shown in Figure 1. The path follows the order of numbers 1 to 7 in the figure and uses an S-shaped path to circumnavigate the shelterbelt. The purpose is to ensure that information on all trees in the shelterbelt is obtained.

[0042] The forest belt point cloud data obtained by scanning is stitched together and exported in the software accompanying the scanning instrument. Noise is then removed by importing the data into CloudCompare software in las / txt / xyz format, using the SOR function in the software to remove noise from the point cloud data, manually deleting any remaining noisy point clouds, and finally exporting the point cloud data in txt format.

[0043] (2) Acquisition and preprocessing of point cloud data of the underground part of a single tree. The underground part of the tree is completely excavated, and the underground part of the single tree is scanned using a ground-based lidar, such as... Figure 2 As shown.

[0044] The forest belt point cloud data obtained by scanning is stitched together and exported in the software accompanying the scanning instrument. Noise is then removed by importing the data into CloudCompare software in las / txt / xyz format, using the SOR function in the software to remove noise from the point cloud data, manually deleting any remaining noisy point clouds, and finally exporting the point cloud data in txt format.

[0045] (3) Individual Tree Segmentation in Shelterbelts. There are two methods for individual tree segmentation: manual and automatic. The manual method involves importing the denoised TXT format point cloud data into CloudCompare software, using the Segment function to trim the data and obtain individual tree point cloud data that does not contain ground points. This data is then exported in TXT format as the data source for branch and leaf separation. The automatic method involves importing the denoised TXT format point cloud data into CloudCompare software for ground point separation. In this software, the CSF Filter function under the plugins option is used to select the corresponding parameters based on the terrain conditions to separate ground points from non-ground points. Point clouds that are not properly filtered are manually removed, and the point cloud data is normalized. The forest belt point cloud that does not contain ground points is trimmed using the Segment function in the software, and finally, the individual tree point cloud data is exported one by one in TXT format as the data source for branch and leaf separation. Both methods can be selected according to the amount of data and can achieve individual tree segmentation in shelterbelts.

[0046] (4) Calculation of single timber volume. The AdQSM algorithm is used to realize the automatic reconstruction of the fine structure of single timber branches and trunks. Its principle is based on the tree skeleton and cylinder fitting algorithm to describe the geometric structure and spatial topological relationship of branches and trunks.

[0047] Based on the generated 3D structural model, a circular ring of the tree trunk section at a distance of 1.3m from the ground is extracted. The circumference of the ring (in meters) is calculated, and the diameter at breast height (DBH) (in centimeters) is calculated using the circumference formula. The tree height (in meters) is calculated using the maximum-minimum method. Using a volume estimation method based on convex polyhedra, the volume of the 3D tree model can be calculated, thereby estimating the volume of the branches and trunk of the sample tree.

[0048]

[0049] In the formula:

[0050] D1.3 represents the diameter at breast height (DBH) at a height of 1.3m above the ground;

[0051] k represents the total number of point clouds on the cross-sectional annulus;

[0052] xi,yi,zi represent the three-dimensional coordinates of the i-th point cloud on the cross-sectional annulus;

[0053] xi−1,yi−1,zi−1 represent the three-dimensional coordinates of the i−1th point cloud on the cross-sectional annulus;

[0054] H represents the tree height;

[0055] Zmax represents the highest Z-axis coordinate value in the point cloud model of the target tree;

[0056] Zmin represents the lowest Z-axis coordinate value in the point cloud model of the target tree;

[0057] x0, y0, z0 represent the coordinates of the reference point inside the convex polyhedron;

[0058] x1, y1, z1, x2, y2, z2, x3, y3, z3 represent the three-dimensional coordinates of the three vertices of the triangular facets that make up the surface of the convex polyhedron.

[0059] a, b, c represent three spatial vectors formed by the reference point and the three vertices of the triangular facet;

[0060] Vtree represents the volume of a 3D tree model;

[0061] u and w represent the number of faces and polyhedra when constructing a 3D model, respectively.

[0062] (5) Biomass estimation

[0063] Based on the diameter at breast height (DBH) and tree height extracted using TLS, a corresponding binary biomass model was selected. Since this embodiment focuses on *Populus simonii*, an empirical binary biomass model (allometric growth equation) pre-constructed for *Populus simonii* in existing technology was used to calculate the biomass model value of the target tree. Then, based on the corresponding carbon content coefficient, the biomass of different organs was calculated. The relevant formulas are as follows:

[0064] W1=0.0219D 1.3 ³-1.5473D 1.3 ²+3.744D 1.3 +121.79

[0065] W2=0.0121(D 1.3 ²H) 0 · 8378

[0066] W3=0.0268D 1.3 2.1717

[0067] In the formula: W1 is the trunk biomass, W2 is the branch biomass, W3 is the root biomass; D1.3 is the diameter at breast height; H is the tree height.

[0068] The total biomass W of the target tree is the sum of the biomass of the trunk, branches, and roots, indicating a relationship:

[0069] W = W1 + W2 + W3.

[0070] To establish a more direct, diameter-of-thorax (D) based 1.3 The total biomass prediction model for tree height H was developed using 50 sets of sample data, with 35 sets used as the training set and 15 sets as the test set. During the training phase, the total biomass W of the samples was used as the dependent variable, and the diameter at breast height (DBH) D was used as the dependent variable. 1.3 Using tree height H as the independent variable, a multiple linear regression method was used for model training and parameter fitting to obtain a biomass formula containing only parameters D1.3 and H:

[0071] W = 118.9863 + 33.859418D 1.3 -1.552810D 1.3 ²+0.021916D 1.3 ³+2.719547×10 -2 (HD 1.3 ²) 0.8 +2.883582×10 -2 D 1.3 2.1717 The model performed well, with an R² of approximately 0.8, indicating high prediction accuracy.

[0072] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data, characterized in that, Includes the following steps: Step 1: Use lidar to acquire point cloud data of the shelterbelt, and use the point cloud data of the shelterbelt to separate the parameters of each individual tree in the shelterbelt, including tree height, diameter at breast height, etc. Step 2: Excavate the underground part of the tree, use LiDAR to scan and obtain point cloud data, and use the point cloud data of the underground part to extract the parameters of the underground part of a single tree. Step 3: Combine the point cloud data of the above-ground and underground parts of a single log to establish a single log volume calculation model; Step 4: Based on the single timber volume calculation model, establish a biomass estimation model for the shelterbelt.

2. The method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data according to claim 1, characterized in that, Step one specifically includes: 1) Obtain point cloud data of shelterbelts; 2) Perform individual tree segmentation on the point cloud data of the shelterbelt to obtain individual tree point cloud data; 3) Calculate the volume of the aboveground portion of a single tree using point cloud data.

3. The method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data according to claim 1, characterized in that, Step two specifically includes: 1) Excavate the underground part of the single log; 2) Obtain point cloud data of the underground portion of a single log; 3) Calculate the volume of the underground portion of a single log using point cloud data.

4. The method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data according to claim 2, characterized in that, The acquisition of point cloud data of protective forest belts specifically includes: scanning the farmland protective forest belts using a ground-based lidar with an S-shaped scanning path; stitching and exporting the scanned point cloud data in the software accompanying the scanning instrument, removing noise, and finally exporting the point cloud data in txt format.

5. The method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data according to claim 2, characterized in that, Single-log splitting includes manual single-log splitting and automatic single-log splitting: The manual single-tree segmentation process involves importing the point cloud data of the shelterbelt into CloudCompare software, and then using the Segment function in the software to crop the data to obtain single-tree point cloud data that does not include ground points. The automatic single-tree segmentation process involves importing the point cloud data of the protective forest belt into CloudCompare software for ground point separation. Using the CSFFilter function, corresponding parameters are selected based on the terrain conditions of the data to separate ground points and non-ground points. Point clouds that are not filtered out are manually removed, and the point cloud data is normalized. The forest belt point cloud that does not contain ground points is cropped using the Segment function. Finally, the single-tree point cloud data is exported one by one in txt format.

6. The method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data according to claim 2, characterized in that, The calculation of single-tree volume using point cloud data involves using Matlab software and employing the aef.m and PointAverageSpacing.m algorithms to calculate the leaf area of ​​each single tree individually.

7. The method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data according to claim 3, characterized in that, Calculate the biomass of a single tree based on the parameters.

8. The method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data according to claim 1, characterized in that, The biomass estimation model for farmland shelterbelts established in step four is a function model with individual tree biomass as the dependent variable and individual tree height and diameter at breast height as independent variables.

9. The method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data according to claim 1, characterized in that, The trees in the farmland shelterbelt are Xinjiang poplars.

10. A method for establishing a biomass estimation model for shelterbelts based on lidar point cloud data, characterized in that, Includes the following steps: The point cloud data of the shelterbelt is acquired using lidar, and the volume of each individual tree in the shelterbelt is separated using the point cloud data. The volume of each individual tree is substituted into the farmland shelterbelt biomass estimation model established according to any one of claims 1 to 9 to calculate the biomass of each individual tree. The biomass of each individual tree is summed to obtain the biomass of the shelterbelt.

Images