A method for extracting the spatial distribution characteristics of leaf veins based on the point cloud of broad-leaved plant leaves

Abstract

The present invention belongs to the field of methods for extracting phenotypic characteristics of plant leaf physiological structures, and provides a method for extracting the spatial distribution characteristics of leaf veins based on the point cloud of broad-leaved plant leaves, including: leaf point cloud extraction, spatial clustering, search radius setting, local point cluster screening, average normal calculation, directional projection of the point cloud, digital surface model construction, local elevation iterative optimization, local inclination calculation, extraction of a leaf vein candidate pixel set based on morphological filtering, and patch skeleton analysis. The present invention realizes the extraction of in-situ non-destructive leaf vein distribution structures and morphological parameters by calculating the height difference change between the leaf veins and the neighboring leaf surfaces.

Description

Technical Field

[0001] The present invention relates to the field of methods for extracting phenotypic characteristics of plant leaf physiological structures, and particularly to a method for extracting the spatial distribution characteristics of leaf veins based on the point cloud of broad-leaved plant leaves. Background Art

[0002] The leaf veins of broad-leaved plants refer to the vascular bundle system within the leaf blade, which dominates the transportation of water, minerals, and organic substances in the leaf blade. The vessels in the leaf veins are responsible for transporting water and minerals from the plant roots to the leaf blade. This water is dissipated through the stomata of the leaf blade via transpiration, thereby maintaining the water balance within the plant body and driving the roots to absorb more water and nutrients. The sieve tubes in the leaf veins are responsible for transporting the organic substances produced through photosynthesis, such as glucose, sucrose, etc., from the leaf blade to other parts of the plant body, including the roots, stems, and fruits. These organic substances provide energy and raw materials for the growth, storage, and reproduction of the plant. By transporting water, the leaf veins maintain the water potential within the leaf blade and regulate the opening and closing of the stomata. This regulatory mechanism is achieved through transpiration, which helps the leaf blade maintain appropriate moisture while promoting the absorption of carbon dioxide and supporting photosynthesis. While transporting organic substances, the leaf veins are also responsible for the distribution of energy and nutrients among different organs of the plant. This is crucial for the growth and development of the plant, the formation of storage organs, and the reproductive process. In addition, the leaf veins provide support for the broad-leaved leaf blade to maintain a certain shape and stiffness, enabling the leaf blade to better capture light for photosynthesis and also enabling the leaf blade to remain intact and not easily damaged when facing environmental factors such as wind and rain. The leaf veins also participate in the disease resistance defense of the leaf blade. When the leaf blade is damaged, the vessels and sieve tubes in the leaf veins can redistribute resources, repair the damaged tissues, help the wound heal, and enhance the survival rate of the plant. According to the distribution form of the leaf veins, the leaf veins of broad-leaved plants are usually divided into types such as reticulate veins, parallel veins, palmate veins, and pinnate veins. Reticulate veins are the most common type of leaf veins, and their main characteristic is that the leaf veins form a reticulate structure within the leaf blade. The main veins and lateral veins intersect to form a complex reticulate system. Usually, there is a main vein in the center, that is, the midrib, and the lateral veins branching out from the midrib further branch to form a reticulate structure. Parallel leaf veins start from the base of the leaf blade and extend to the tip of the leaf blade. Parallel veins are common in monocotyledonous plants but also occur in some broad-leaved plants, such as certain gramineous plants. Palmate veins originate from the base of the petiole and radiate outward. There are multiple main leaf veins extending in various directions towards the edge of the leaf blade. Pinnate veins have an obvious main vein running through the entire leaf blade, and many feather-like arranged lateral veins branch out from both sides of the midrib. The above distribution patterns result in obvious differences in the leaf vein distribution density among different plant species. The leaf vein distribution density, that is, the number or length of leaf veins per unit area, can reveal the physiological characteristics, adaptation strategies, and ecological functions of plants. During the agricultural breeding process, the leaf vein density is an important selective trait, which is related to the drought resistance, heat resistance, and disease resistance of crops. By studying the leaf vein density of different plants, the environmental adaptation strategies of plant phenotypic characteristics can be inferred and used for crop growth modeling. Understanding the distribution of the leaf vein density of crops can help agricultural managers optimize water resource and nutrient application strategies, and can also help develop bionic materials with excellent strength and flexibility for use in fields such as architecture, textiles, and biomedicine.By using high-precision multi-angle LiDAR or structured light imaging and stitching, we can obtain high-precision point clouds of leaves and depict the millimeter-level fine structure of the vein distribution area on the leaf surface, including the shape, distribution pattern and direction of the veins. We can also analyze the three-dimensional shape of the veins, including bending, undulations, etc. With certain computer vision analysis algorithms and deep learning models, we can realize the recognition of vein distribution patterns based on the digital surface model of the leaf surface, which can be used for plant classification, health status assessment and ecological research.

[0003] Existing structural analysis based on leaf point clouds mainly focuses on the extraction of structural parameters related to leaf area and morphology, but there is still a lack of available solutions for the extraction of the structure and distribution patterns of fine-scale leaf organs such as veins. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a method for extracting the spatial distribution features of leaf veins based on point clouds of broad-leaved plants, and to realize in-situ non-destructive extraction of leaf vein distribution structure and morphological parameters by calculating the height difference between leaf veins and neighboring leaf surfaces.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for extracting spatial distribution features of veins of broad-leaved plants based on leaf point clouds, comprising:

[0007] Collecting a broad-leaved plant point cloud, and performing leaf separation on the broad-leaved plant point cloud to obtain a leaf point cloud extraction result;

[0008] Performing spatial clustering on the leaf point cloud extraction results to obtain a single broadleaf point cloud;

[0009] Setting a search radius according to the mean value of the distance data;

[0010] Using the search radius to filter the distance data to obtain a local point cluster;

[0011] Using the local point cluster to construct a local discrete point space coordinate matrix, and calculating and averaging the point cluster normals according to the local discrete point space coordinate matrix to obtain an average normal;

[0012] Projecting the single broadleaf point cloud onto a plane perpendicular to the average normal and parallel to the spatial growth direction of the target leaf to obtain a projection image;

[0013] Constructing a digital surface model after directional projection of the point cloud using a voxel method according to the projection image;

[0014] The digital surface model is optimized and iterated locally using a 3*3 square sliding window to obtain a local elevation difference lifting model;

[0015] Calculate the local inclination angles of each pixel in the local height difference elevation model to obtain leaf surface orientation data;

[0016] Extract the candidate vein pixel set in the local height difference elevation model according to the leaf surface orientation data using the Freeman algorithm;

[0017] Perform a dilation operation on the candidate vein pixel set using a disk-shaped filtering window with a radius of 1 pixel, and perform a morphological closing operation on the dilated candidate vein pixel set using a disk-shaped filtering window with a radius of 2 pixels to obtain linearly continuously distributed vein pixels;

[0018] Perform a patch skeleton analysis on the linearly continuously distributed vein pixels to obtain a vein patch skeleton;

[0019] Perform a vein distribution density calculation and multi-type multi-scale analysis on the vein patch skeleton to obtain vein distribution characteristics.

[0020] Preferably, the projection formula of the projection image is: P proj = P - (P × n)n; where P proj is the projection coordinate; P is the coordinate of any point in the single broad-leaf point cloud; n is the average normal.

[0021] Preferably, the calculation formula of the local height difference elevation model is:

[0022]

[0023] where DSM LS is the local height difference elevation model; DSM is the digital surface model; DSM Mean is the digital surface model processed by mean filtering; N is the number of iterative analyses; i represents the current calculation number.

[0024] Preferably, the calculation formula of the local inclination angle is: where θ is the local inclination angle; G x is the height component of the pixel in the X direction in the local height difference elevation model; G y is the height component of the pixel in the Y direction in the local height difference elevation model.

[0025] Preferably, constructing a digital surface model after point cloud directional projection using the voxel method according to the projection image includes:

[0026] Randomly extract 1% of the points in the single broad-leaf point cloud to obtain a point cloud subset;

[0027] Calculate the Euclidean distance of the nearest neighbor point of each point in the point cloud subset in three-dimensional space, and calculate the mean value of all the Euclidean distances of the nearest neighbor points to obtain the average Euclidean distance;

[0028] Set twice the average Euclidean distance as the spatial resolution;

[0029] Divide the projection image into several elevation statistical units according to the spatial resolution;

[0030] Calculate the mean elevation of the point clusters in each elevation statistical unit to obtain the digital surface model.

[0031] Preferably, use the Freeman algorithm to extract the candidate vein pixel set in the local height difference elevation model according to the leaf surface orientation data, including:

[0032] Calculate the slope direction and local height difference between each pixel in the local height difference elevation model and its 8 neighboring pixels to obtain slope change data;

[0033] Use the slope change data to judge the edge position, move to the next pixel according to the edge position, and record the moving direction to obtain vein growth direction data;

[0034] Use the vein growth direction data to calculate weights by a recursive method to obtain a vein candidate pixel weight grid;

[0035] Calculate the pixel mean value of the vein candidate pixel weight grid to obtain a pixel threshold;

[0036] Binarize the vein candidate pixel weight grid using the pixel threshold to obtain the candidate vein pixel set.

[0037] Preferably, the calculation formula for the vein distribution density is: ρ pixel =S YM / S P ; where ρ pixel is the vein distribution density; S YM is the number of vein pixels in the retrieval window in the vein patch skeleton; S P is the total number of pixels in the retrieval window in the vein patch skeleton.

[0038] The present invention discloses the following technical effects:

[0039] The present invention provides a method for extracting the spatial distribution characteristics of veins based on the point cloud of broad-leaved plant leaves. By calculating the height difference change between the veins and the neighboring leaf surfaces, it solves the defect that the traditional structure analysis method based on leaf point clouds lacks vein extraction, and realizes the in-situ non-destructive extraction of vein distribution structures and morphological parameters. Description of the Drawings

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the accompanying drawings required in the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0041] Figure 1 Schematic diagram of the extraction process of the vein spatial distribution characteristics based on the point cloud of broad-leaved plant leaves provided by the embodiment of the present invention;

[0042] Figure 2 Example reference diagram provided by the embodiment of the present invention;

[0043] Figure 3 Flow chart of the implementation provided by the embodiment of the present invention. Specific implementation manners

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0045] The purpose of the present invention is to provide a method for extracting the vein spatial distribution characteristics based on the point cloud of broad-leaved plant leaves, and by calculating the height difference change between the veins and the adjacent leaf surfaces, the extraction of the in-situ non-destructive vein distribution structure and morphological parameters is realized.

[0046] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will further describe the present invention in detail with reference to the accompanying drawings and specific implementation manners.

[0047] Figure 1 Schematic diagram of the extraction process of the vein spatial distribution characteristics based on the point cloud of broad-leaved plant leaves provided by the embodiment of the present invention, Figure 2 Example reference diagram provided by the embodiment of the present invention, as Figure 1 and Figure 2 shown, the present invention provides a method for extracting the vein spatial distribution characteristics based on the point cloud of broad-leaved plant leaves, including:

[0048] Collect the point cloud of broad-leaved plant plants, and separate the leaves from the point cloud of broad-leaved plant plants to obtain the leaf point cloud extraction result;

[0049] Perform spatial clustering on the leaf point cloud extraction result to obtain a single broad-leaved point cloud;

[0050] Set the search radius according to the mean of the distance data;

[0051] Use the search radius to filter the distance data and obtain local point clusters;

[0052] The local point cluster is used to construct a local discrete point space coordinate matrix, and the point cluster normal is calculated and averaged according to the local discrete point space coordinate matrix to obtain the average normal;

[0053] Project the single broadleaf point cloud onto a plane perpendicular to the average normal and parallel to the spatial growth direction of the target leaf to obtain a projection image;

[0054] According to the projection image, a digital surface model after directional projection of the point cloud is constructed by using the voxel method;

[0055] The local elevation optimization and iteration of the digital surface model are performed using a 3*3 square sliding window to obtain a local elevation difference lifting model;

[0056] Calculate the local inclination angle of each pixel in the local height difference lifting model to obtain the leaf orientation data;

[0057] According to the leaf orientation data, the Freeman algorithm is used to extract the candidate leaf vein pixel set in the local height difference lifting model;

[0058] A disc-shaped filter window with a radius of 1 pixel is used to dilate the candidate leaf vein pixel set, and a disc-shaped filter window with a radius of 2 pixels is used to perform a morphological closing operation on the dilated candidate leaf vein pixel set to obtain linearly continuously distributed leaf vein pixels.

[0059] Performing pattern skeleton analysis on linear continuously distributed leaf vein pixels to obtain leaf vein pattern skeleton;

[0060] The vein distribution density is calculated and multi-type and multi-scale analysis is performed on the vein pattern skeleton to obtain the vein distribution characteristics.

[0061] Specifically, the projection formula of the projected image is: proj =P-(P×n)n; where P proj are projection coordinates; P is the coordinate of any point in a single broadleaf point cloud; n is the average normal.

[0062] Furthermore, the calculation formula of the local height difference lifting model is:

[0063]

[0064] Among them, DSM LS is the local height difference lifting model; DSM is the digital surface model; DSM Mean is the digital surface model after mean filtering; N is the number of iterative analysis; i represents the current number of calculations.

[0065] Preferably, the calculation formula for the local dip angle is: where θ is the local dip angle; G x is the height component of the pixel in the X direction in the local height difference elevation model; G y is the height component of the pixel in the Y direction in the local height difference elevation model.

[0066] Furthermore, based on the projected image, a digital surface model after directional projection of the point cloud is constructed using the voxel method, including:

[0067] Randomly extract 1% of the points of a single broad-leaf point cloud to obtain a point cloud subset;

[0068] Calculate the Euclidean distance of the nearest neighbor point of each point in the point cloud subset in three-dimensional space, and calculate the mean value of all the Euclidean distances of the nearest neighbor points to obtain the average Euclidean distance;

[0069] Set twice the average Euclidean distance as the spatial resolution;

[0070] Divide the projected image into several elevation statistical units according to the spatial resolution;

[0071] Calculate the mean elevation of the point clusters of each elevation statistical unit to obtain the digital surface model.

[0072] Specifically, based on the leaf surface orientation data, a set of candidate vein pixels in the local height difference elevation model is extracted using the Freeman algorithm, including:

[0073] Calculate the slope direction and local height difference between each pixel in the local height difference elevation model and its 8 neighboring pixels to obtain slope change data;

[0074] Use the slope change data to determine the edge position, move to the next pixel according to the edge position, and record the moving direction to obtain vein growth direction data;

[0075] Perform weight calculation using the recursive method according to the vein growth direction data to obtain a vein candidate pixel weight grid;

[0076] Calculate the pixel mean value of the vein candidate pixel weight grid to obtain a pixel threshold;

[0077] Binarize the vein candidate pixel weight grid using the pixel threshold to obtain a set of candidate vein pixels.

[0078] Preferably, the calculation formula for the vein distribution density is: ρ pixel = S YM / S P ; where ρ pixel is the vein distribution density; S YMis the number of vein pixels in the retrieval window of the vein patch skeleton; S P is the total number of pixels in the retrieval window of the vein patch skeleton.

[0079] Specifically, preprocessing of the leaf point cloud. In the natural growth state, the overall orientation of broad-leaved leaves is variable and not necessarily parallel to the ground. To analyze the distribution pattern of veins, it is necessary to project the leaf point cloud onto a plane parallel to the growth orientation according to its overall orientation and construct a digital surface model (DSM) of the leaf. In the point cloud, each point defines its three-dimensional spatial position through its X, Y, and Z coordinates. To characterize the orientation of local point clusters, the normal vector of the plane fitted by the spatial coordinates of the point clusters can be extracted. The normal vector refers to the direction vector perpendicular to a certain point on the surface of the point cloud. Averaging the normal vectors of the planes locally fitted to the points in the point cloud one by one, the result obtained is called the average normal direction. The average normal direction reflects the overall distribution orientation of the leaf point cloud and is used to indicate the projection plane of the leaf.

[0080] Furthermore, the point cloud projection process converts the three-dimensional point cloud onto a two-dimensional plane. Projecting according to the average normal direction means projecting each point in the point cloud onto a plane perpendicular to the average normal direction. The projected data retains the direction information of the point cloud parallel to the projection plane and eliminates the influence of other non-projection directions. The projection formula is:

[0081] P proj = P - (P × n)n

[0082] Use a neighborhood search algorithm (such as K-nearest neighbor or radius neighborhood search) to calculate the local normal vector for each point in the point cloud. Project each point in the point cloud onto a plane perpendicular to the average normal direction, which will be used to construct the DSM of the leaf point cloud in the projection direction and provide a data source for vein extraction and distribution pattern analysis.

[0083] Specifically, construction of the leaf surface model. After completing the point cloud projection, determine the range interval of the outer bounding rectangle of the leaf point cloud in the projection plane according to the projection result. Using this interval as a mask, use the voxel method to construct a digital surface model (DSM) after projecting the leaf point cloud. The voxel method divides the projected point cloud on the projection plane into several elevation statistical units according to the resolution W, and takes the elevation mean of the point cluster in each unit as the elevation of the unit. To determine W, randomly extract 1% of the points in the point cloud, and statistically calculate the Euclidean distance D from each point to its nearest neighbor point in its three-dimensional space. Then, calculate twice the mean of the corresponding D of all the retrieved points as the spatial resolution W of the DSM. The generated DSM is used as a key data source for subsequent vein extraction.

[0084] Furthermore, considering that the height difference between the vein distribution area and the surrounding leaf surface may not be obvious, it is necessary to iteratively use grid arithmetic filtering to increase the height difference between the vein distribution area and the surrounding leaf surface to improve the accuracy of vein extraction. First, use mean filtering with a 3*3 window to process the DSM to obtain the DSM Mean . For the central pixel P of an analysis window in the DSM, the mean filtering result is the mean of the 8-direction neighborhood pixels of P. Then, calculate the difference between the DSM and the DSM Mean , and add it to the DSM, and iterate N times. If the elevation value of a pixel is greater than its DSM Mean , then the height difference with the vein pixels lower than the DSM Mean increases after addition. Then, divide the iterative cumulative result by N to obtain the DSM LS after local height difference elevation to avoid leaf edge extraction errors.

[0085]

[0086] Considering the operation efficiency and the rationality of height difference analysis, N is set to 3 here.

[0087] Specifically, vein pixel extraction. Use the DSM LS after projection of the high-precision leaf point cloud to extract the pixels covered by the veins. Since the local elevation of the vein distribution area is lower than the surrounding non-vein area, the set of low-elevation pixels showing a valley-like distribution can be extracted by using the local pixel elevation change, that is, the vein-covered pixels. First, use the DSM LS to analyze the leaf surface orientation within the local 3*3 window range. For each pixel, use the elevation values of the 8 neighboring pixels to calculate the local neighborhood inclination angle of the pixel. The leaf surface orientation is determined by the inclination components in two directions (such as the X and Y directions) obtained from the calculation of the local leaf surface inclination angle. The local inclination angle θ of the central pixel of the window is calculated using the following formula:

[0088]

[0089] The calculation result is the inclination radian of the central pixel. Based on this, the local leaf surface orientations of the neighborhoods of each pixel in the DSM LS are calculated.

[0090] Further, after completing the leaf surface orientation analysis, a set of candidate vein pixels is extracted using local orientation change information. During the extraction process, the Freeman algorithm is utilized. The growth of veins on the leaf surface exhibits the characteristics of spatial continuity and forking growth. The growth path of veins always heads towards the direction with the largest local slope change on the leaf surface, that is, growing from the leaf margin towards the petiole. The Freeman algorithm depicts the distribution of gullies on a spatially continuous surface by recording the direction of a pixel relative to its previous pixel. Similar to the process of extracting the slope runoff path based on the digital elevation model, in the process of extracting candidate vein pixels based on the Freeman algorithm, for each pixel in the DSM LS , the growth path of the vein from the leaf margin to the petiole is calculated according to its slope aspect and local height difference. Generally, the directions of the 8 neighboring pixels of the central pixel are represented by integers from 0 to 7 (including horizontal, vertical, and diagonal directions). The vein tracking extraction usually starts from a point on the edge of the DSM LS and moves step by step along the edge, recording the direction from the current point to the next point each time until it returns to the starting point or reaches the contour end point of the low elevation area of the DSM LS . If the contour is closed, it will eventually return to the starting point; otherwise, the contour will end at the open point of the low elevation area of the DSM LS . For broad-leaved blades, this low elevation open point is often at the petiole. For each pixel, according to the orientation encoding, the chain code of the possible extension path of the vein is recorded.

[0091] Furthermore, once the vein growth directions at each pixel of the DSM LS are determined, the vein growth path can be analyzed by tracing back and accumulating. Usually, a recursive method is used to calculate its weight pixel by pixel from downstream to upstream for extracting candidate pixels covered by the vein. The weight of a pixel near the petiole is equal to the sum of the weights of the upper pixels plus its own weight. The weight raster YM of candidate vein pixels based on the DSM LS is obtained. In YM, the possibility of the existence of veins is high in the area where high-value pixels are distributed. Further, the mean value of the pixels in YM is used as a threshold to binarize YM, obtaining a set of candidate vein pixels.

[0092] Specifically, the optimization of vein extraction. In the above extraction results, the extraction results of candidate vein pixels in the locally flat leaf surface area may show patchy distribution and breakage. Here, morphological opening and closing operations for binary grayscale images are used to further optimize the vein extraction results, improving the continuity of the extracted veins and the extraction accuracy of the subsequent vein distribution density. For the binarized YM (YM BL ), first, the morphological dilation operation is used to increase the edge width of independent foreground patches, connecting the spatially separated adjacent candidate vein patches that are locally broken. Here, a disk-shaped filtering window with a radius of 1 pixel is set for YM BLPerform the dilation operation. Then, use the morphological closing operation on the dilated YM BL Perform erosion to accelerate the peeling of YM BL The redundant edge pixels of the vein pixels with connection optimization in YM. Define the neighborhood range of the closing operation using a disk-shaped filtering window with a radius of 2 pixels, and the result is a linearly continuous distribution of vein pixels.

[0093] Furthermore, using the vein pixels after connectivity optimization, perform patch skeleton analysis to reduce the vein pixel patch to its centerline while preserving the topological structure of the vein pixels. The skeletonization analysis reduces the shape of an object by iteratively peeling off the edge pixels of the object, generating the thinnest connected path that preserves the topological structure of the object. As a result, the vein pixels are presented as a set of foreground pixels with a continuous distribution and a width of 1 pixel in YM BL In each iteration, the algorithm identifies and marks the edge pixels of the current object. For each marked edge pixel, further check whether it can be deleted without destroying the topological structure of the image. If deleting this pixel will cause the object to disconnect or the number of holes to change, then this pixel will not be deleted. The above process will continue iteratively, gradually thinning the vein patch from the outside to the inside until no more pixels can be deleted. Finally, the remaining pixel set is the vein patch skeleton, which is used as the final result of extracting vein pixels from DSM LS and is used as the data source for vein distribution pattern extraction and local density analysis.

[0094] Specifically, vein distribution density analysis. Count the number of pixels N LS in DSM after leaf projection P and the number of vein pixels N YM extracted from it, and calculate the overall distribution density of the veins using the following formula:

[0095] ρ = N YM / N P

[0096] Since the veins are not evenly distributed inside the leaf, further analyze the change of vein density inside the leaf through the local window statistical method. According to the coverage range of the leaf DSM LS , set the analysis window size to S. Use a sliding window to traverse and retrieve the number of vein pixels S LS inside the window starting from the upper left corner of DSM YM . For an analysis window, the vein distribution density ρ pixel inside it is:

[0097] ρ pixel = S YM / S P

[0098] Through the above analysis, the variation trend of the vein distribution density in the leaf projection plane can be obtained. Based on the DSM LS , a density distribution map is generated on the leaf surface and visualized, which can intuitively depict the distribution characteristics of the veins in each part of the leaf.

[0099] Furthermore, the leaf is divided into different radial segments (from the petiole to the leaf tip) and transverse segments (approximately parallel to the direction of the main vein). The vein density of each segment is calculated, and based on this, the variation trend of the vein density at different developmental positions of the leaf is analyzed. Then, the leaf is divided into several concentric elliptical regions along the vein extension direction (usually from the midrib to the leaf margin). The midpoint of the ellipse is the center point of the long axis of the outer rectangle of the leaf DSM LS . The distance between the circular rings refers to the distance between the intersection points of the lateral veins and the main vein. The vein distribution densities in the near-center ellipse and the multi-layer outer rings are analyzed respectively. On this basis, the regions with and without vein distribution are extracted to realize the multi-type and multi-scale analysis of the vein distribution density on the leaf surface, which is used for the interactive modeling of the leaf structure phenotypic characteristics and physiological processes.

[0100] Specifically, the specific implementation steps are as follows:

[0101] S1: Use an RGB-D camera or lidar to collect the point cloud of broad-leaved plant plants with millimeter-level high precision. Manually separate the leaf and branch point clouds in the plant in a desktop visualization software platform (such as: Cloudcompare or Leica Cyclone software). It is also possible to use a local point cloud distribution pattern analysis algorithm or use the additional attribute information of the point cloud (such as the surface reflectivity of each point, the RGB color of the target, and the echo intensity attribute) to assist in separating the branch and leaf point clouds.

[0102] S2: According to the leaf point cloud extraction result, further obtain a single broad-leaved point cloud by spatial clustering segmentation based on the change of the point cloud density between the leaves. Set the search radius based on the Euclidean distance between adjacent points to obtain the local point clusters in a single broad-leaved point cloud. Use the point clusters to construct the local discrete point space coordinate matrix and calculate the normal of the point clusters. Project each point in the point cloud along the average normal direction onto a plane parallel to the leaf space growth direction, set the spatial resolution W, and construct the DSM covering the range of the leaf point cloud. Then, use the non-0 pixel coverage area in the DSM as a mask to extract the DSM after the leaf point cloud is projected.

[0103] S3: Use a 3*3 square sliding window to traverse and analyze the entire DSM. Calculate the elevation change rate and mean value of the local leaf surface pixels in each analysis window. If the elevations in the analysis window are the same, abandon the extraction and continue the next analysis. Then, iteratively perform local elevation optimization of the DSM, increasing the difference between the high-value area and the low-value area in the local DSM of the leaf surface while maintaining the local height difference fluctuation pattern of the leaf surface to construct the DSM LSTo highlight the distribution of vein pixels, which is conducive to subsequent extraction of DSM LS The spatially continuous vein pixels in it.

[0104] S4: Extract DSM by local slope analysis and calculation of the weight of the existence of veins based on the Freeman algorithm LS The set of candidate vein pixels in it. On the premise of not destroying the connectivity of the veins, an iterative morphological dilation and thinning method is adopted, a disk-shaped sliding analysis unit is set, and the dilation and thinning analysis window sizes with radii of 1 and 2 are set to increase the spatial connectivity of the candidate vein patches, and the edge points of the block-aggregated distributed vein patches are stripped from them to obtain a skeletonized set of vein pixels.

[0105] S5: Count the number of pixels in DSM LS And the number of vein pixels extracted therefrom, and calculate the overall distribution density of the veins. Through the local window statistical method, the local vein density change within the leaf is further analyzed. Based on the local vein density distribution map of the leaf surface obtained by DSM LS Analyze the distribution characteristics and change trends of the veins in each part of the leaf quantitatively. The leaf is divided into multiple horizontal and vertical sections, and the vein distribution map extracted based on DSM LS is superimposed to further analyze the change trend of the vein density in each section. Then, along the vein extension direction of the leaf (usually from the midrib to the leaf margin), referring to the spacing between the intersections of the veins, DSM LS is divided into several concentric elliptical regions, and the vein distribution densities in the near-center ellipse and the multi-layer outer-ring are analyzed respectively. Through the above analysis, the total length of the veins inside the target leaf, the local vein density change in the leaf surface space, and the vein density change in sections and regions can be obtained, so as to quantitatively describe the relevant leaf structure phenotypic characteristics for the modeling related to the leaf physiological process and the three-dimensional digital display of the fine structure of the leaf.

[0106] The beneficial effects of the present invention are as follows:

[0107] (1) Using the solution proposed by the present invention, the complex three-dimensional spatial distribution pattern of the veins inside the leaf can be extracted in-situ and non-destructively based on the high-precision leaf point cloud. On this basis, the structure and distribution density information of the veins are quantitatively analyzed.

[0108] (2) Based on the vein extraction results of the leaf point cloud, combined with the measured values of the leaf physiological indexes, a special model can be constructed to simulate the growth and development process of the plant organs inside the leaf, providing reliable data for constructing a refined digital crop model.

[0109] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0110] In this article, specific examples are used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for helping to understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A method for extracting the spatial distribution characteristics of leaf veins based on the point cloud of broad-leaved plant leaves, characterized in that, include: Collecting a broad-leaved plant point cloud, and performing leaf separation on the broad-leaved plant point cloud to obtain a leaf point cloud extraction result; Performing spatial clustering on the leaf point cloud extraction results to obtain a single broadleaf point cloud; Set the search radius according to the mean of the distance data; Using the search radius to filter the distance data to obtain a local point cluster; Using the local point cluster to construct a local discrete point space coordinate matrix, and calculating and averaging the point cluster normals according to the local discrete point space coordinate matrix to obtain an average normal; Projecting the single broadleaf point cloud onto a plane perpendicular to the average normal and parallel to the spatial growth direction of the target leaf to obtain a projection image; Constructing a digital surface model after directional projection of the point cloud using a voxel method according to the projection image; The digital surface model is optimized and iterated locally using a 3*3 square sliding window to obtain a local elevation difference lifting model; Calculating the local inclination angle of each pixel in the local height difference lifting model to obtain leaf orientation data; Extracting a set of candidate leaf vein pixels in the local height difference lifting model using the Freeman algorithm according to the leaf orientation data; Using a disc-shaped filter window with a radius of 1 pixel to perform an expansion operation on the candidate leaf vein pixel set, and using a disc-shaped filter window with a radius of 2 pixels to perform a morphological closing operation on the expanded candidate leaf vein pixel set to obtain linearly continuously distributed leaf vein pixels; Performing a pattern skeleton analysis on the linear continuously distributed leaf vein pixels to obtain a leaf vein pattern skeleton; The vein pattern skeleton is subjected to vein distribution density calculation and multi-type and multi-scale analysis to obtain vein distribution characteristics.

2. The method for extracting the spatial distribution characteristics of leaf veins based on the point cloud of broad-leaved plant leaves according to claim 1, wherein, The projection formula of the projected image is: P proj = P - (P × n)n; where P proj is the projection coordinate; P is the coordinate of any point in the single broad-leaved point cloud; n is the average normal vector.

3. A method for extracting the spatial distribution characteristics of leaf veins based on the point cloud of broad-leaved plant leaves according to claim 1, characterized in that, The calculation formula of the local height difference lifting model is: Among them, DSM LS is the local elevation difference elevation model; DSM is the digital surface model; DSM Mean is the digital surface model processed by mean filtering; N is the number of iterative analyses; i represents the current calculation number.

4. The method for extracting the spatial distribution characteristics of leaf veins based on the point cloud of broad-leaved plant leaves according to claim 1, wherein The calculation formula for the local dip angle is as follows: where θ is the local dip angle; G x is the height component of the pixel in the X direction in the local height difference elevation model; G y is the height component of the pixel in the Y direction in the local height difference elevation model.

5. A method for extracting the spatial distribution characteristics of leaf veins based on the point cloud of broad-leaved plant leaves according to claim 1, characterized in that The digital surface model after the point cloud directional projection is constructed by using the voxel method according to the projection image, including: Randomly extract 1% of the points of the single broadleaf point cloud to obtain a point cloud subset; Calculating the Euclidean distance of the nearest neighbor point of each point in the point cloud subset in the three-dimensional space, and calculating the average of the Euclidean distances of all the nearest neighbor points to obtain an average Euclidean distance; setting twice the average Euclidean distance as the spatial resolution; Dividing the projection image into a plurality of elevation statistical units according to the spatial resolution; The point cluster elevation mean of each elevation statistical unit is calculated to obtain the digital surface model.

6. The method for extracting the spatial distribution characteristics of leaf veins based on the point cloud of broad-leaved plant leaves according to claim 1, wherein Extracting a candidate leaf vein pixel set in the local height difference lifting model using the Freeman algorithm according to the leaf orientation data includes: Calculate the slope aspect and local height difference between each pixel and 8 neighboring pixels in the local height difference lifting model to obtain slope change data; Using the slope change data to determine the edge position, moving to the next pixel according to the edge position, and recording the moving direction to obtain leaf vein growth direction data; A weight calculation is performed using a recursive method according to the leaf vein growth direction data to obtain a leaf vein candidate pixel weight grid; Calculating the pixel mean of the leaf vein candidate pixel weight grid to obtain a pixel threshold; The pixel threshold is used to binarize the leaf vein candidate pixel weight grid to obtain the candidate leaf vein pixel set.

7. A method for extracting the spatial distribution characteristics of leaf veins based on the point cloud of broad-leaved plant leaves according to claim 1, wherein, The calculation formula for the vein distribution density is: ρ pixel = S YM / S P ; where ρ pixel is the vein distribution density; S YM is the number of vein pixels within the retrieval window in the vein patch skeleton; S P is the total number of pixels within the retrieval window in the vein patch skeleton.

Images