A method for calculating single leaf biomass from ground-based lidar point clouds

Abstract

The present invention discloses a method for calculating the biomass of individual tree leaves using ground-based laser radar point clouds, comprising collecting point cloud data of individual trees through a ground-based laser radar, randomly harvesting leaves from different street trees and at different canopy heights, matching the point cloud data of each site with corresponding panoramic photos, performing RGB rendering on the point cloud to color the point cloud, and obtaining a preliminary leaf point cloud through site splicing, denoising, normalization, cropping, thinning, individual tree point cloud segmentation, and leaf point cloud extraction, and manually removing the remaining branch point clouds to obtain an accurate leaf point cloud, and processing the harvested leaves to obtain the volume and mass of each leaf. The present invention can extract leaf point clouds and calculate leaf volume using a more efficient and accurate processing method through the point cloud data scanned by the ground-based laser radar and the actual harvested leaves, thereby improving the accuracy and efficiency of leaf biomass estimation.

Description

Technical Field

[0001] The present invention relates to the technical field of leaf biomass statistics, and in particular to a method for calculating the leaf biomass of a single tree using a ground-based laser radar point cloud. Background Art

[0002] Biomass is a fundamental quantitative characteristic of ecosystems. Accurate estimation of aboveground biomass provides a scientific basis for evaluating vegetation carbon sequestration. Aboveground biomass includes the biomass of stems, stumps, branches, bark, seeds, and leaves. Leaf biomass, as a component of aboveground biomass, is the most important factor determining tree productivity. Its spatial distribution effectively reflects vegetation growth trends and energy flows.

[0003] Traditional leaf biomass measurements rely primarily on manual field measurements, which require harvesting and weighing leaves in the field. This method is inherently destructive to standing trees and is irreversible. Furthermore, it is labor-intensive and time-consuming. Because leaves fluctuate significantly with the seasons, leaf biomass measurements are often limited to annual averages, making it difficult to obtain real-time values.

[0004] In recent years, with the continuous improvement of radar technology, lidar detection has replaced traditional manual sampling methods. Lidar detection can quickly obtain detailed structural information of trees and has great advantages in vegetation visualization. As an active remote sensing tool, lidar can quickly collect high-density point cloud data, and the stable acquisition platform ensures the accuracy of data acquisition. By selecting appropriate lidar scanning tools, detailed coverage information of the entire vegetation surface can be obtained without loss, and real, detailed three-dimensional tree data can be obtained. At present, domestic and foreign scholars often use tools such as terrestrial laser scanning (TLS) and mobile laser scanning (MLS) to efficiently obtain three-dimensional point cloud data of trees.

[0005] Currently, most calculation methods based on ground-based lidar data obtain leaf biomass density through field measurements and indirectly calculate individual tree leaf biomass by combining it with canopy volume. Specifically, leaf biomass per unit volume is obtained and then extracted from the canopy volume or leaf volume. However, leaf biomass density calculations directly obtain the ratio of leaf volume to biomass in a crude manner, without considering the regression relationship between the two, resulting in weak robustness. Furthermore, the multitude of methods for calculating canopy volume makes the extraction accuracy difficult to assess, resulting in significant variation in the accuracy of canopy-based leaf biomass estimates. Summary of the Invention

[0006] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of this application to avoid blurring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.

[0007] In view of the above and / or existing problems in the existing leaf biomass estimation, the present invention is proposed.

[0008] Therefore, the purpose of the present invention is to provide a method for calculating the leaf biomass of a single tree from a ground-based lidar point cloud to improve the estimation accuracy and efficiency of leaf biomass.

[0009] To solve the above technical problems, according to one aspect of the present invention, the present invention provides the following technical solutions:

[0010] A method for calculating the biomass of individual leaves from ground-based laser radar point clouds, comprising:

[0011] S1. Collect point cloud data of individual trees using ground-based LiDAR, and randomly harvest leaves from different street trees at different canopy heights.

[0012] S2. Match the point cloud data of each station to the corresponding panoramic photo, and perform RGB rendering on the point cloud to color the point cloud. Then, preliminary leaf point clouds are obtained through station stitching, denoising, normalization, cropping, thinning, single tree point cloud segmentation, and leaf point cloud extraction. The remaining branch point clouds are manually removed to obtain accurate leaf point clouds.

[0013] S3. Processing the harvested leaves to obtain the volume and mass of each leaf;

[0014] S4. Using leaf mass as the dependent variable and leaf volume as the independent variable, a regression relationship between leaf mass and leaf volume was constructed to obtain the leaf biomass-eLVV equation;

[0015] S5, divide the leaf point cloud into several layers, according to Expand the points in each layer, count the area occupied by the expanded point cloud of each layer, integrate and obtain the effective three-dimensional green volume (eLVV) of each tree, traverse all the point clouds of individual trees, and count the effective three-dimensional green volume (eLVV) of all individual trees;

[0016] Where P(i)dilated refers to the expanded size of the point cloud of the i-th tree, LA refers to the average leaf area of the collected leaves, plength(i) and pwidth(i) refer to the canvas length and width set for the i-th tree, and vlength and vwidth refer to the actual crown width of the i-th tree in two directions, which is obtained by counting the extracted single tree point cloud.

[0017] S6. Substitute the calculated eLVV into the leaf biomass-eLVV equation to obtain the leaf biomass of a single tree.

[0018] As a preferred solution of the method for calculating the biomass of individual tree leaves using a ground-based lidar point cloud described in the present invention, in step S1, the steps for collecting point cloud data of individual trees using a ground-based lidar are as follows: the ground-based lidar is set up in the center of the road, stations are arranged at equal intervals along the road direction, and the roadside trees are scanned.

[0019] As a preferred solution of the method for calculating the biomass of individual tree leaves from a ground-based lidar point cloud described in the present invention, in step S2, site stitching, denoising, normalization and cropping are completed in the ground-based radar supporting software Riscan Pro 64bit v2.6.1, the thinning process is performed by downsampling according to the establishment of a 0.05×0.05×0.05m filter, the individual tree point cloud is segmented according to manual segmentation to obtain a complete individual tree crown and trunk point cloud, and the leaf point cloud is extracted according to the LeWoS method to classify the branch and leaf point clouds.

[0020] As a preferred solution of the method for calculating the biomass of individual leaves from a ground-based lidar point cloud described in the present invention, in step S3, the specific steps for obtaining the volume of each leaf after processing the harvested leaves are as follows: flatten the leaves with a transparent acrylic plate and take orthographic photos. When taking photos, a one-yuan coin is used as a reference and recorded at the same time, the number of leaf pixels and coin pixels is extracted, and the leaf length and area are obtained in combination with the actual area of the coin, and the leaf thickness is set to 0.2 cm to obtain the volume of each leaf.

[0021] As a preferred embodiment of the method for calculating the biomass of individual leaves from a ground-based lidar point cloud described in the present invention, in step S3, the specific steps for obtaining the mass of each leaf after processing the harvested leaves are as follows: number the harvested leaves in the order of extraction, place them in envelopes with corresponding numbers, dry them in a 65°C oven to a constant weight, take them out, and place a blank envelope for white balance. Then, weigh each leaf with a precision balance to obtain the weight of a single leaf.

[0022] As a preferred solution of the method for calculating the biomass of individual leaves from a ground-based lidar point cloud described in the present invention, in step S5, the leaf point cloud is divided into several layers at 5 cm intervals.

[0023] Compared with the existing technology, the present invention has the following beneficial effects: the present invention uses point cloud data scanned by ground-based lidar and actual harvested leaves. The ground-based lidar can display the vertical structure of vegetation and detailed information inside trees in more detail, which is used to obtain leaf point clouds. By calculating the three-dimensional green volume, a green space landscape indicator, the leaf volume is obtained. The actual harvested leaves are used to obtain leaf quality. The leaf point cloud can be extracted and the leaf volume can be calculated with a more efficient and accurate processing method, thereby improving the estimation accuracy and efficiency of leaf biomass. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and detailed embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be derived from these drawings without inventive effort. Among them:

[0025] Figure 1 This is a flow chart of a method for calculating the biomass of a single tree leaf from a ground-based laser radar point cloud according to the present invention;

[0026] Figure 2 This is a schematic diagram of a point cloud processing flow of a method for calculating the biomass of individual leaves based on a ground-based laser radar point cloud according to the present invention;

[0027] Figure 3 A schematic diagram of leaf area extraction in a method for calculating single tree leaf biomass using a ground-based lidar point cloud according to the present invention;

[0028] Figure 4 This is a leaf biomass-eLVV regression relationship diagram obtained by fitting a method for calculating leaf biomass of a single tree using ground-based lidar point cloud.

[0029] Figure 5 This is a flowchart of the layered area integration method for calculating the biomass of individual leaves from ground-based laser radar point clouds of the present invention;

[0030] Figure 6 This is a Bland-Altman plot of the single leaf biomass calculated based on eLVV and the allometric model value of the ground-based lidar point cloud single leaf biomass calculation method of the present invention. DETAILED DESCRIPTION

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0032] Next, the present invention is described in detail with reference to schematic diagrams. For ease of illustration, cross-sectional views of device structures may be partially enlarged and not to scale when describing the embodiments of the present invention. Furthermore, the schematic diagrams are merely illustrative and should not limit the scope of protection of the present invention. Furthermore, in actual production, three-dimensional dimensions, including length, width, and depth, should be included.

[0033] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0034] The present invention provides a method for calculating the leaf biomass of a single tree based on a ground-based laser radar point cloud, which improves the estimation accuracy and efficiency of leaf biomass.

[0035] like Figure 1 As shown in the figure, the specific steps of the method for calculating the biomass of individual leaves from ground-based lidar point clouds are as follows:

[0036] S1. Collect point cloud data of individual trees using a ground-based LiDAR. Randomly harvest leaves from different street trees at different canopy heights. The steps for collecting point cloud data of individual trees using a ground-based LiDAR are as follows: Set up the ground-based LiDAR in the middle of the road, arrange stations at equal intervals along the road, and scan the street trees.

[0037] S2. Match the point cloud data of each site to the corresponding panoramic photo, and perform RGB rendering on the point cloud to colorize the point cloud. Then, a preliminary leaf point cloud is obtained through site stitching, denoising, normalization, cropping, thinning, single tree point cloud segmentation, and leaf point cloud extraction. The remaining branch point cloud is manually removed to obtain an accurate leaf point cloud. Among them, site stitching, denoising, normalization and cropping are completed in Riscan Pro 64bit v2.6.1, the software supporting the ground-based radar. The thinning process is performed by downsampling according to the establishment of a 0.05×0.05×0.05m filter. The single tree point cloud segmentation is based on manual segmentation to obtain a complete single tree crown and trunk point cloud. The leaf point cloud extraction is based on the LeWoS method to classify the branch and leaf point cloud.

[0038] S3. After processing the harvested leaves, the volume and mass of each leaf are obtained. The specific steps for obtaining the volume of each leaf after processing the harvested leaves are as follows: flatten the leaves with a transparent acrylic plate and take orthophotographs. When taking photos, a one-yuan coin is used as a reference and recorded at the same time. The number of leaf pixels and coin pixels is extracted. Combined with the actual area of the coin, the length and area of the leaf are obtained, and the leaf thickness is set to 0.2 cm to obtain the volume of each leaf. The specific steps for obtaining the mass of each leaf after processing the harvested leaves are as follows: number the harvested leaves in the order of extraction, and put them in envelopes with corresponding numbers. After drying in a 65°C oven to a constant weight, take them out and put them in a blank envelope for white balance. Then, each leaf is weighed with a precision balance to obtain the weight of a single leaf.

[0039] S4. Using leaf mass as the dependent variable and leaf volume as the independent variable, a regression relationship between leaf mass and leaf volume was constructed to obtain the leaf biomass-eLVV equation;

[0040] S5, divide the leaf point cloud into several layers, according to Expand the points in each layer, count the area occupied by the expanded point cloud of each layer, integrate and obtain the effective three-dimensional green volume (eLVV) of each tree, traverse all the point clouds of individual trees, and count the effective three-dimensional green volume (eLVV) of all individual trees;

[0041] Where P(i)dilated refers to the expanded size of the point cloud of the i-th tree, LA refers to the average leaf area of the collected leaves, plength(i) and pwidth(i) refer to the canvas length and width set for the i-th tree, and vlength and vwidth refer to the actual crown width of the i-th tree in two directions. The crown width is obtained by counting the extracted single tree point cloud. The leaf point cloud is divided into several layers with each layer being 5 cm.

[0042] S6. Substitute the calculated eLVV into the leaf biomass-eLVV equation to obtain the leaf biomass of a single tree.

[0043] The following takes the street trees on Yuanding Road in Nanjing City, Jiangsu Province (32°33′~32°57′N, 120°07′~120°53′E) as an example. The main street tree species is hybrid tulip tree.

[0044] S1. A RIEGL VZ-400i terrestrial 3D laser scanner was set up in the middle of Gardener Road, with 10 stations set up at equal intervals. 30 street trees were scanned and approximately 30 healthy hybrid tulipwood leaves were randomly harvested and brought back to the laboratory for processing.

[0045] S2, such as Figure 2As shown in the figure, the point cloud data of each station is matched with the corresponding panoramic photos, and the point cloud is rendered in RGB to color the point cloud. Point stitching, denoising, normalization and cropping are completed in the ground-based radar supporting software Riscan Pro 64bitv2.6.1. The thinning process is performed by downsampling according to the establishment of a 0.05×0.05×0.05m filter. The single tree point cloud segmentation is mainly based on manual segmentation to obtain complete single tree crown, trunk and other point clouds. The leaf point cloud extraction is based on the LeWoS method to classify the branch and leaf point clouds to obtain a preliminary leaf point cloud. After that, the remaining branch point clouds are manually removed to obtain an accurate leaf point cloud.

[0046] S3, such as Figure 3 As shown, the leaves were flattened with a transparent acrylic plate and photographed orthogonally. During the photography, a one-yuan coin was also recorded as a reference. The number of leaf pixels and coin pixels was extracted, and the leaf length and area were obtained by combining the actual area of the coin. Based on experience, the leaf thickness was assumed to be 0.2 cm. The volume of each leaf was obtained, and the leaves were numbered in the order of extraction and placed in envelopes with corresponding numbers. After drying in a 65°C oven to a constant weight, they were taken out and placed in a blank envelope for white balance. Each leaf was weighed with a precision balance to obtain the weight of a single leaf.

[0047] S4. Using the mass of the leaf as the dependent variable and the volume of the leaf as the independent variable, the regression relationship between the mass of the leaf and the volume of the leaf is constructed, such as Figure 4 As shown, the leaf biomass-eLVV equation is LB = 24.54 × eLVV - 7 × 10 -5 , where R 2 =0.8857, the equation has extremely high fitting accuracy;

[0048] S5, such as Figure 5 As shown, the leaf point cloud is divided into several layers according to 5cm. The points in each layer are expanded, and the area occupied by the expanded point cloud of each layer is counted. The effective three-dimensional green volume (eLVV) of each tree is obtained by integration. All the point clouds of individual trees are traversed, and the effective three-dimensional green volume (eLVV) of all individual trees is counted. Among them, eLVV refers to the spatial volume occupied by the leaves, that is, the leaf volume.

[0049] S6. Substitute the calculated eLVV into the leaf biomass-eLVV equation to obtain the leaf biomass of a single tree.

[0050] In order to verify the accuracy of biomass calculation, the leaf biomass obtained by the traditional leaf biomass allometric growth model was used as a reference value for accuracy comparison. Since the biomass model of hybrid tulipwood in Jiangsu Province has not yet been published, in order to ensure the accuracy of the extracted leaf biomass, we chose the hard-broad leaf biomass allometric growth equation to obtain the leaf biomass of hybrid tulipwood under the traditional model and used it as the true value. Formula:

[0051] V = 6.01228 × 10 -5 ×DBH 1.87550 ×H 0.98496

[0052] LB=0.22526×DBH -0.38874 ×H -0.21925 ×V

[0053] Where V refers to the volume of a single tree trunk (m 3 ), LB refers to the biomass of single wood leaf (kg).

[0054] like Figure 6 As shown in the figure, the correlation coefficient of the two sets of results is 0.635 and p = 0.074. Among them, the p value is greater than 0.05, which means that the two sets of data are not significant and have strong consistency, which means that the leaf biomass value calculated based on eLVV is reliable.

[0055] Although the present invention has been described above with reference to embodiments, various modifications may be made thereto and equivalent components may be substituted without departing from the scope of the present invention. In particular, as long as there are no structural conflicts, the various features of the embodiments disclosed herein may be combined with each other in any manner, and the omission of an exhaustive description of such combinations in this specification is solely for the sake of space and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for calculating the biomass of individual leaves from ground-based laser radar point clouds, characterized in that: include: S1. Collect point cloud data of individual trees using ground-based LiDAR, and randomly harvest leaves from different street trees at different canopy heights. S2. Match the point cloud data of each station to the corresponding panoramic photo, and perform RGB rendering on the point cloud to color the point cloud. Then, preliminary leaf point clouds are obtained through station stitching, denoising, normalization, cropping, thinning, single tree point cloud segmentation, and leaf point cloud extraction. The remaining branch point clouds are manually removed to obtain accurate leaf point clouds. S3. Processing the harvested leaves to obtain the volume and mass of each leaf; S4. Using leaf mass as the dependent variable and leaf volume as the independent variable, a regression relationship between leaf mass and leaf volume was constructed to obtain the leaf biomass-eLVV equation; S5, divide the leaf point cloud into several layers, according to Expand the points in each layer, count the area occupied by the expanded point cloud of each layer, integrate and obtain the effective three-dimensional green volume (eLVV) of each tree, traverse all the point clouds of individual trees, and count the effective three-dimensional green volume (eLVV) of all individual trees; Where P(i)dilated refers to the expanded size of the point cloud of the i-th tree, LA refers to the average leaf area of the collected leaves, plength(i) and pwidth(i) refer to the canvas length and width set for the i-th tree, and vlength and vwidth refer to the actual crown width of the i-th tree in two directions, which is obtained by counting the extracted single tree point cloud. S6. Substitute the calculated eLVV into the leaf biomass-eLVV equation to obtain the leaf biomass of a single tree.

2. The method for calculating the biomass of individual leaves from ground-based laser radar point clouds according to claim 1, characterized in that: In step S1, the steps of collecting point cloud data of single trees by ground-based laser radar are as follows: the ground-based laser radar is set up in the middle of the road, stations are arranged at equal intervals along the road direction, and the roadside trees are scanned.

3. The method for calculating the biomass of individual leaves from ground-based laser radar point clouds according to claim 1, characterized in that: In step S2, site stitching, denoising, normalization, and cropping are completed in Riscan Pro 64bitv2.6.1, the software supporting the ground-based radar. The thinning process is performed by downsampling according to the establishment of a 0.05×0.05×0.05m filter. The single tree point cloud is segmented according to manual segmentation to obtain a complete single tree crown and trunk point cloud. The leaf point cloud is extracted according to the LeWoS method to classify the branch and leaf point cloud.

4. The method for calculating the biomass of individual leaves from ground-based laser radar point clouds according to claim 1, characterized in that: In step S3, the specific steps for obtaining the volume of each leaf after processing the harvested leaves are as follows: flatten the leaves with a transparent acrylic plate and take orthographic photos. During the photography, a one-yuan coin is used as a reference and recorded at the same time. The number of leaf pixels and coin pixels is extracted, and the leaf length and area are obtained by combining the actual area of the coin. The leaf thickness is set to 0.2 cm to obtain the volume of each leaf.

5. The method for calculating the biomass of individual leaves from ground-based laser radar point clouds according to claim 1, characterized in that: In step S3, the specific steps for obtaining the mass of each leaf after processing the harvested leaves are as follows: number the harvested leaves according to the extraction order, place them in envelopes with corresponding numbers, dry them in a 65°C oven to a constant weight, take them out, and place a blank envelope for white balance. Then, weigh each leaf using a precision balance to obtain the weight of a single leaf.

6. The method for calculating the biomass of individual leaves from ground-based laser radar point clouds according to claim 1, characterized in that: In step S5, the leaf point cloud is divided into several layers at 5 cm intervals.

Images