Point cloud data processing method, device, equipment and storage medium

Abstract

The present application provides a point cloud data processing method, apparatus, device, and storage medium. First, surveying and mapping data of crown trees are obtained, the surveying and mapping data including to-be-processed point cloud data. Target segmentation points for segmenting the trunk and crown of each crown tree are then obtained based on the to-be-processed point cloud data and a preset processing strategy. Target point cloud data are then obtained based on the target segmentation points and an estimated radius of each crown tree. The target point cloud data is used to characterize the trunk area of each crown tree, thereby extracting the effective portion of the crown tree without manual operation, achieving automatic cleaning and segmentation of the point cloud data of the crown trees. This not only improves data cleaning efficiency but also makes the cleaning and segmentation processes more controllable.

Description

Technical Field

[0001] The present application relates to the field of data processing, and specifically to a point cloud data processing method, device, equipment and storage medium. Background Art

[0002] In the process of 3D surface modeling of point clouds scanned by lidar, although some methods have achieved 3D reconstruction of crown tree models under data cleaning conditions, it cannot be ignored that the data cleaning process cannot be automated. At present, it is still necessary to manually extract the effective parts of the crown trees and manually remove outliers. In other words, the current point cloud data cleaning relies on manual operations and has not yet been automated.

[0003] However, when point cloud cleaning relies primarily on manual operation of point cloud processing software, the operation relies primarily on visual judgment of point cloud positions, which not only consumes a lot of labor costs but also may lead to inaccurate cutting, resulting in missing data and failure to successfully build models. Furthermore, manual operation and cleaning may also lead to the loss of significant effective information, resulting in a certain degree of uncontrollability in the cleaning process. Summary of the Invention

[0004] The present application provides a point cloud data processing method, apparatus, device and storage medium, which provide an automatic cleaning and segmentation solution for point cloud data for three-dimensional reconstruction of crown trees.

[0005] In a first aspect, the present application provides a point cloud data processing method, comprising:

[0006] Acquiring surveying and mapping data of crown trees, wherein the surveying and mapping data includes point cloud data to be processed;

[0007] Obtaining target segmentation points according to the point cloud data to be processed and a preset processing strategy, wherein the target segmentation points are used to segment the trunk and crown of each crown tree;

[0008] Target point cloud data is acquired according to the target segmentation point and the estimated radius value of each crown tree. The target point cloud data is used to characterize the trunk area of each crown tree.

[0009] In one possible design, obtaining target segmentation points according to the point cloud data to be processed and a preset processing strategy includes:

[0010] Performing a first projection and a first clustering process on the point cloud data to be processed to obtain tree point cloud data, wherein the tree point cloud data is used to characterize each crown tree;

[0011] Performing a second projection and a second clustering process on the tree point cloud data to obtain the target segmentation points;

[0012] The preset processing strategy includes the first projection, the second projection, the first clustering processing, and the second clustering processing.

[0013] In a possible design, performing a first projection and a first clustering process on the to-be-processed point cloud data to obtain tree point cloud data includes:

[0014] Performing a preset vertical projection on the point cloud data to be processed to obtain a plane point set, wherein the first projection includes the preset vertical projection;

[0015] The first clustering process is performed on the plane point set according to a preset accuracy to eliminate non-tree point sets, thereby obtaining the tree point cloud data.

[0016] In a possible design, performing a second projection and a second clustering process on the tree point cloud data to obtain the target segmentation points includes:

[0017] performing a preset longitudinal projection and a preset transverse projection on the tree point cloud data in sequence to obtain a curve point set, wherein the second projection includes the preset longitudinal projection and the preset transverse projection;

[0018] The second clustering process is performed on the curve point set according to a preset convergence principle to obtain the target segmentation point, and the number of clusters in the second clustering process is 2.

[0019] In a possible design, obtaining target point cloud data according to the target segmentation point and the estimated radius of each crown tree includes:

[0020] Obtaining the trunk center of each crown tree according to the target segmentation point;

[0021] determining a trunk area of each crown tree based on the trunk center and the radius estimate;

[0022] The point set of the trunk area is extracted as the target point cloud data.

[0023] In a possible design, before acquiring the target point cloud data, the method further includes:

[0024] Performing a preset fitting process on the tree point cloud data to obtain an outer diameter fitting value and an inner diameter fitting value of each crown tree;

[0025] The radius estimation value of each crown tree is obtained according to the outer diameter fitting value and the inner diameter fitting value of each crown tree and a preset estimation model.

[0026] In a possible design, after acquiring the target point cloud data, the method further includes:

[0027] Normalizing the coordinates of each point cloud representing the target point cloud data with the center of the tree trunk as the origin;

[0028] A surface model is generated according to the normalized target point cloud data, and the surface model is used to realize three-dimensional reconstruction of crown trees.

[0029] In a second aspect, the present application provides a point cloud data processing device, comprising:

[0030] An acquisition module, configured to acquire surveying and mapping data of crown trees, wherein the surveying and mapping data includes point cloud data to be processed;

[0031] A first processing module is configured to obtain target segmentation points based on the point cloud data to be processed and a preset processing strategy, wherein the target segmentation points are used to segment the trunk and crown of each crown tree;

[0032] The second processing module is configured to obtain target point cloud data according to the target segmentation point and the estimated radius value of each crown tree, wherein the target point cloud data is used to characterize the trunk area of each crown tree.

[0033] In one possible design, the first processing module includes:

[0034] a first processing submodule, configured to perform a first projection and a first clustering process on the point cloud data to be processed to obtain tree point cloud data, wherein the tree point cloud data is used to characterize each crown tree;

[0035] A second processing submodule, configured to perform a second projection and a second clustering process on the tree point cloud data to obtain the target segmentation points;

[0036] The preset processing strategy includes the first projection, the second projection, the first clustering processing, and the second clustering processing.

[0037] In one possible design, the first processing submodule is specifically configured to:

[0038] Performing a preset vertical projection on the point cloud data to be processed to obtain a plane point set, wherein the first projection includes the preset vertical projection;

[0039] The first clustering process is performed on the plane point set according to a preset accuracy to eliminate non-tree point sets, thereby obtaining the tree point cloud data.

[0040] In one possible design, the second processing submodule is specifically configured to:

[0041] performing a preset longitudinal projection and a preset transverse projection on the tree point cloud data in sequence to obtain a curve point set, wherein the second projection includes the preset longitudinal projection and the preset transverse projection;

[0042] The second clustering process is performed on the curve point set according to a preset convergence principle to obtain the target segmentation point, and the number of clusters in the second clustering process is 2.

[0043] In one possible design, the second processing module is specifically configured to:

[0044] Obtaining the trunk center of each crown tree according to the target segmentation point;

[0045] determining a trunk area of each crown tree based on the trunk center and the radius estimate;

[0046] The point set of the trunk area is extracted as the target point cloud data.

[0047] In a possible design, the point cloud data processing device further includes: a third processing module; the third processing module is specifically configured to:

[0048] Performing a preset fitting process on the tree point cloud data to obtain an outer diameter fitting value and an inner diameter fitting value of each crown tree;

[0049] The radius estimation value of each crown tree is obtained according to the outer diameter fitting value and the inner diameter fitting value of each crown tree and a preset estimation model.

[0050] In one possible design, the point cloud data processing device further includes: a fourth processing module; the fourth processing module is specifically configured to:

[0051] Normalizing the coordinates of each point cloud representing the target point cloud data with the center of the tree trunk as the origin;

[0052] A surface model is generated according to the normalized target point cloud data, and the surface model is used to realize three-dimensional reconstruction of crown trees.

[0053] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0054] The memory stores computer-executable instructions;

[0055] The processor executes the computer-executable instructions stored in the memory to implement any possible point cloud data processing method provided in the first aspect.

[0056] In a fourth aspect, the present application provides a computer-readable storage medium having computer-executable instructions stored thereon, which, when executed by a processor, implement any possible point cloud data processing method provided in the first aspect.

[0057] In a fifth aspect, the present application also provides a computer program product, comprising computer instructions, which, when executed by a processor, implement any possible point cloud data processing method provided in the first aspect.

[0058] The present application provides a point cloud data processing method, apparatus, device, and storage medium. First, surveying and mapping data of crown trees are obtained, the surveying and mapping data including to-be-processed point cloud data. Target segmentation points for segmenting the trunk and crown of each crown tree are then obtained based on the to-be-processed point cloud data and a preset processing strategy. Target point cloud data are then obtained based on the target segmentation points and an estimated radius of each crown tree. The target point cloud data is used to characterize the trunk area of each crown tree, thereby extracting the effective portion of the crown tree without manual operation, achieving automatic cleaning and segmentation of the point cloud data of the crown trees. This not only improves data cleaning efficiency but also makes the cleaning and segmentation processes more controllable. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A schematic diagram of an application scenario provided in an embodiment of the present application;

[0060] Figure 2 A schematic diagram of a process for processing point cloud data provided in an embodiment of the present application;

[0061] Figure 3 A flowchart of another point cloud data processing method provided in an embodiment of the present application;

[0062] Figure 4 A flowchart of another point cloud data processing method provided in an embodiment of the present application;

[0063] Figure 5 A schematic diagram of a point cloud image provided in an embodiment of the present application;

[0064] Figure 6 A schematic structural diagram of a point cloud data processing device provided in an embodiment of the present application;

[0065] Figure 7 This is a schematic diagram of the structure of an electronic device provided in this application. DETAILED DESCRIPTION

[0066] To make the purpose, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work, including but not limited to combinations of multiple embodiments, are within the scope of protection of this application.

[0067] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the numbers used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0068] Currently, point cloud data cleaning relies on manual operations, and automated cleaning has yet to be achieved. However, when point cloud processing software is primarily used for manual operation, the operation relies primarily on visual judgment of point cloud positions, which not only consumes a lot of labor costs but also may lead to inaccurate cropping, resulting in missing data and unsuccessful modeling. Furthermore, manual operation and cleaning may lead to the loss of significant effective information, resulting in a certain degree of uncontrollability in the cleaning process.

[0069] To address the aforementioned issues in the prior art, the present application provides a point cloud data processing method, apparatus, device, and storage medium. The inventive concept of the point cloud data processing method provided herein is to automatically clean and segment the unprocessed point cloud data included in the surveying and mapping of crown trees using a preset processing strategy, thereby replacing manual operations and improving point cloud data processing efficiency. Furthermore, the processing is based on machine learning, and by controlling processing boundaries, the process is highly controllable, which helps reduce the cycle time for 3D reconstruction of crown trees using the target point cloud data.

[0070] The following describes exemplary application scenarios of the embodiments of the present application.

[0071] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of the present application. Figure 1As shown, the point cloud data processing method provided by the embodiment of the present application can be executed by the point cloud data processing device provided by the embodiment of the present application. The point cloud data processing device provided by the embodiment of the present application can be configured in the electronic device 100. The processor of the electronic device 100 can be configured to execute corresponding computer instructions to execute the point cloud data processing method provided by the embodiment of the present application, so as to realize automatic cleaning and segmentation of the surveying and mapping data of the crown trees, and extract the effective data required for the three-dimensional reconstruction of the crown trees, that is, the target point cloud data for characterizing the trunk area of each crown tree. Among them, the point cloud data for characterizing the trunk area of each crown tree is the target point cloud data for characterizing the trunk area of each crown tree. Figure 1 The point cloud data of the point cloud image 200 of the mid-crown tree is the point cloud data to be processed.

[0072] It is understood that the electronic device 100 can be any device that can be configured with a processor to execute a corresponding computer program to perform the point cloud data processing method provided in the embodiment of the present application. For example, the electronic device 100 can be a computer, a smart phone, a smart watch, etc. This embodiment does not limit the type of the electronic device 100. Figure 1 The electronic device 100 in the figure is shown as a computer as an example.

[0073] It should be noted that the above application scenarios are merely illustrative, and the point cloud data processing methods, devices, equipment, and storage media provided in the embodiments of the present application include but are not limited to the above application scenarios.

[0074] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0075] Figure 2 This is a flow chart of a point cloud data processing method provided in an embodiment of the present application. Figure 2 As shown, the embodiment of the present application includes:

[0076] S101: Acquire crown tree mapping data.

[0077] Among them, the surveying and mapping data includes point cloud data to be processed.

[0078] The crown tree survey data including the crown tree point cloud image is loaded, wherein the survey data includes the point cloud data to be processed.

[0079] For example, by loading .las or other point cloud files to read point cloud coordinates, the process of reading point cloud coordinates is the process of obtaining surveying and mapping data of crown data. The surveying and mapping data of crown trees refers to the data containing crown trees obtained through surveying and mapping means.

[0080] Assuming that each point cloud is represented by three-dimensional coordinates, that is, it can be represented as C(x, y, z), the data described by the point cloud data to be processed is the data represented by multiple C(x, y, z), which can be used to depict, for example Figure 5 The point cloud image data on the left is the cloud data to be processed.

[0081] S102: Obtain target segmentation points according to the point cloud data to be processed and a preset processing strategy.

[0082] Among them, the target segmentation point is used to segment the trunk and crown of each crown tree.

[0083] Since the point cloud data to be processed comes from real surveying and mapping data, the point cloud data to be processed also includes point clouds of non-crown trees. Therefore, it is necessary to perform data cleaning on the point cloud data to eliminate noise and obtain the effective parts of the trunks and crowns that represent the crown trees.

[0084] The preset processing strategy is a data processing strategy involved in the embodiment of this application for the technical problem to be solved by this application, and its purpose is to segment the point clouds representing the trunks and crowns of crown trees from the point cloud data to be processed.

[0085] In one possible implementation, the possible implementation of step S102 includes:

[0086] First, the processing point cloud data is subjected to a first projection and a first clustering process to obtain tree point cloud data. The obtained tree point cloud data is used to represent each crown tree. In other words, the processing point cloud data is processed using the first projection and first clustering processes to clean up a point cloud representing each crown tree. Specifically, the processing point cloud data is segmented into a point cloud representing crown trees and a point cloud representing non-crown trees. Thus, the point cloud representing the crown trees is cleaned, thereby obtaining the tree point cloud data.

[0087] Based on the tree point cloud data, the second projection and the second clustering processing are then applied to the tree point cloud data to distinguish the trunk and crown of each crown tree. Specifically, the target segmentation point is obtained, and the target segmentation point is used to segment the trunk and crown of each crown tree.

[0088] The preset processing strategies include the aforementioned first projection, second projection, and first and second clustering processes. The first and second projections are processing strategies configured based on the characteristics of the surveying and mapping data. For example, the first projection may be a vertical projection, such as projecting the vertical direction z in C(x, y, z) onto the geodetic plane where xy lies. The second projection may be projections in the x and y directions, respectively. Both the first and second clustering processes may employ a Gaussian clustering model, classifying the data to be processed by setting the number of clusters to segment the desired data.

[0089] S103: Obtain target point cloud data according to the target segmentation points and the estimated radius of each crown tree.

[0090] Among them, the target point cloud data is used to characterize the trunk area of each crown tree.

[0091] After obtaining the target segmentation point of each crown tree, target point cloud data is obtained according to the target segmentation point of each crown tree and the estimated radius value of each crown tree obtained through estimation.

[0092] Each crown tree consists of a trunk and a crown. The target segmentation point is the point cloud data that distinguishes the trunk and crown of the crown tree. Therefore, once the target segmentation point is obtained, the center of the trunk of each crown tree can be obtained based on the target segmentation point. The estimated radius of each crown tree is the estimated radius of each crown tree. Therefore, the target point cloud data can be obtained based on the center of each crown tree and the estimated radius of each crown tree.

[0093] In a possible design, step S103 may be implemented as follows: Figure 3 shown. Figure 3 This is a flow chart of another point cloud data processing method provided in an embodiment of the present application. Figure 3 As shown, the embodiment of the present application includes:

[0094] S1031: Obtain the trunk center of each crown tree according to the target segmentation point.

[0095] S1032: Determine the trunk area of each crown tree based on the trunk center and radius estimation value.

[0096] S1033: Extract the point set in the trunk area as target point cloud data.

[0097] The target segmentation point is used to distinguish the trunk and crown of a crown tree. If the target segmentation point is known, point cloud data representing the trunk and crown of each crown tree can be obtained based on the target segmentation point. If the point cloud data of each crown tree's crown is known, the point cloud data representing the center of each crown tree's trunk can also be determined, thus obtaining the trunk center of each crown tree. Therefore, if the trunk center of each crown tree and the estimated radius of each crown tree are known, the circular area formed by the crown center as the origin and the estimated radius as the radius is determined to be the trunk area of the crown tree. The point set of the trunk area is then extracted, and this extracted point set becomes the target point cloud data.

[0098] By setting a preprocessing strategy, the system extracts point cloud data representing the trunk area of crown trees, automatically segmenting the trunk and crown of the crown tree, without relying on manual visual selection of point cloud locations and the removal of outliers. This achieves the goal of automatic point cloud data cleaning and segmentation. Furthermore, the pre-set processing strategy, implemented through corresponding projection and distance algorithms, effectively controls the processing boundaries during the processing process, making it highly controllable.

[0099] The point cloud data processing method provided in the embodiment of the present application first obtains surveying data of crown trees, the surveying data including point cloud data to be processed, then obtains target segmentation points that can segment the trunk and crown of each crown tree based on the point cloud data to be processed and a preset processing strategy, and then obtains target point cloud data based on the target segmentation points and the estimated radius of each crown tree. The target point cloud data is used to characterize the trunk area of each crown tree, so that the effective part of the crown tree can be extracted without relying on manual operation, thereby realizing automatic cleaning and segmentation of the point cloud data of the crown trees, which not only improves the data cleaning efficiency, but also makes the cleaning and segmentation process more controllable.

[0100] Based on the above embodiments, Figure 4 A flow chart of another point cloud data processing method provided in the embodiment of the present application. Figure 4 As shown, the embodiment of the present application includes:

[0101] S201: Acquire surveying data of crown trees.

[0102] Among them, the surveying and mapping data includes point cloud data to be processed.

[0103] The implementation method, principle and technical effect of step S201 are similar to those of step S101 and will not be repeated here.

[0104] S202: Project the point cloud data to be processed in a preset vertical direction to obtain a plane point set.

[0105] The first projection includes a preset vertical projection.

[0106] Project the point cloud data to be processed in a preset vertical direction to obtain a planar point set. In other words, the data F(C(x,y,z)) is converted to P(x,y), converting the three-dimensional point cloud data to be processed into a two-dimensional planar point set. The preset vertical direction is the Z axis, specifically the direction perpendicular to the ground, which is the direction of the z data in C representing the height of the crown tree.

[0107] Through the first projection process, for example, Figure 5 The point cloud image of the crown tree on the left is converted into Figure 5 The image shown on the right, Figure 5 A schematic diagram of a point cloud image provided in an embodiment of the present application.

[0108] S203: performing a first clustering process on the plane point set according to a preset accuracy to eliminate non-tree point sets and obtain tree point cloud data.

[0109] The purpose of the first clustering process is to remove the point set of non-crown trees from the plane point set, that is, the non-tree point set, and obtain the point set representing each crown tree, that is, the tree point cloud data.

[0110] For example, the first clustering process may be a Gaussian clustering algorithm. The specific process is as follows:

[0111] First, set the number of clusters and randomly initialize the Gaussian distribution parameters of each cluster, such as setting the initial mean and initial variance;

[0112] Second, set the Gaussian distribution for each cluster and calculate the probability that each point in the plane point set belongs to each cluster. For example, the closer a point is to the center of the Gaussian distribution, the more likely it is to belong to that cluster.

[0113] Third, the Gaussian distribution parameters are adjusted according to the calculated probability to maximize the probability of the points in the plane point concentration. For example, the Gaussian distribution parameters can be adjusted based on the weighted probabilities of the points in the plane point concentration, and each weight is the probability that the point in the plane point concentration belongs to the cluster.

[0114] The above process is iterated a preset number of times, for example, 2-3 times, until the accuracy of the Gaussian clustering algorithm output during each iteration is less than a preset accuracy. This terminates the first clustering process using the Gaussian clustering algorithm, and non-tree point sets are removed from the plane point set based on the Gaussian clustering algorithm output that meets the preset accuracy. Optionally, the preset accuracy can be set to 0.0001.

[0115] By restoring the area represented by the plane point set after removing the non-tree point set to the point cloud control, a point cloud representing each crown tree can be obtained, that is, tree point cloud data.

[0116] It can be understood that through the first clustering process in this step, the planar point set P(x, y) can be processed into P1(x, y), P2(x, y), …, Pn(x, y). Here, n is the number of clusters. Assuming that the point cloud data to be processed contains multiple crown trees, the first clustering process can also obtain separate point cloud data for each crown tree, eliminating non-tree point sets. If the point cloud data to be processed contains a single crown tree, n is set to 2. If the point cloud data to be processed contains multiple crown trees, n can be set to the number of crown trees to segment each crown tree.

[0117] S204: Performing preset longitudinal projection and preset transverse projection on the tree point cloud data in sequence to obtain a curve point set.

[0118] The second projection includes a preset longitudinal projection and a preset transverse projection.

[0119] The tree point cloud data representing each crown tree is sequentially projected along the Y-axis and the X-axis to convert the planar point set into a curved point set. The Y-axis projection is the preset longitudinal projection, and the X-axis projection is the preset transverse projection.

[0120] S205: Performing a second clustering process on the curve point set according to a preset convergence principle to obtain target segmentation points.

[0121] The number of clusters in the second clustering process is 2.

[0122] The second clustering process can also be achieved through the Gaussian clustering algorithm. The clustering process of the curve point set using the Gaussian clustering algorithm in this step is similar to the process in step S203. The difference is that the number of clusters in this step is set to 2. Its purpose is to distinguish the trunk and crown of each crown tree.

[0123] The preset convergence principle can be implemented by setting a preset convergence threshold, the purpose of which is to obtain a stable target segmentation point. The stability of the obtained target segmentation point can be reflected by the convergence condition, that is, the feature data describing the convergence condition reaches the preset convergence threshold. The specific setting of the preset convergence threshold is not limited in the present embodiment.

[0124] It should be noted that in the actual processing process, if the noise is more complex, the second clustering process of this step can be repeated until a stable target segmentation point is obtained.

[0125] S206: Obtain target point cloud data according to the target segmentation point and the estimated radius of each crown tree.

[0126] Among them, the target point cloud data is used to characterize the trunk area of each crown tree.

[0127] The implementation method, principle and technical effect of step S206 are similar to those of step S103 and will not be repeated here.

[0128] The point cloud data processing method provided in an embodiment of the present application first obtains surveying data of crown trees, the surveying data including point cloud data to be processed. The point cloud data to be processed is then projected in a preset vertical direction to obtain a plane point set. The plane point set is then clustered according to a preset accuracy to eliminate non-tree point sets, thereby obtaining tree point cloud data. The tree point cloud data is then sequentially projected in a preset longitudinal direction and a preset transverse direction to obtain a curve point set, thereby obtaining target segmentation points capable of segmenting the trunk and crown of each crown tree. Finally, target point cloud data is obtained based on the target segmentation points and the estimated radius of each crown tree. The target point cloud data is used to characterize the trunk area of each crown tree, thereby extracting the effective portion of the crown tree without manual operation, thereby achieving automatic cleaning and segmentation of the point cloud data of the crown trees. This not only improves data cleaning efficiency but also makes the cleaning and segmentation processes more controllable.

[0129] In a possible design, before acquiring the target point cloud data, a step of estimating the radius of each crown tree is also included.

[0130] For example, after obtaining tree point cloud data, a preset fitting process is performed on the tree point cloud data to fit the outer diameter and inner diameter of each crown tree based on the tree point cloud data. Specifically, the preset fitting process is performed on the tree point cloud data to obtain a fitted outer diameter value and a fitted inner diameter value for each crown tree. The preset fitting process refers to any method for estimating the outer diameter and inner diameter, and the specific content of the preset fitting process is not limited in this embodiment of the present application. After obtaining the fitted outer diameter value and the fitted inner diameter value for each crown tree, the fitted outer diameter value and the fitted inner diameter value for each crown tree are input into a preset estimation model, and the output is the estimated radius value for each crown tree.

[0131] Optionally, the preset estimation model is shown in the following formula (1):

[0132] R′=2R*r / (R+r) (1)

[0133] Where R represents the outer diameter fitting value, r represents the inner diameter fitting value, and R′ represents the radius estimation value.

[0134] The process of acquiring target point cloud data described in the above embodiments can achieve the purpose of automatically cleaning and segmenting the point cloud data to be processed. Furthermore, based on the above embodiments, the acquired target point cloud data can be used to perform three-dimensional reconstruction of crown trees.

[0135] For example, the coordinates of each point cloud representing the target point cloud data are first normalized with the center of the tree trunk as the origin, and then a surface model is generated based on the normalized target point cloud data, and the surface model is used to perform three-dimensional reconstruction of the crown tree.

[0136] Specifically, the method of normalizing the coordinates of each point cloud representing the target point cloud data with the center of the tree trunk as the origin can be implemented by using the Direct Linear Transform algorithm. For example, solving the following equation (2):

[0137]

[0138] Among them, P i and Represent the target point cloud data before and after normalization, respectively. i represents each point cloud included in the target point cloud data and is a positive integer. α and A are preset parameters, and their specific values are set according to the actual working conditions.

[0139] The surface model refers to the surface reconstruction algorithm, the core of which is to map the normalized target point cloud data into a predefined three-dimensional space, and use the stage signed distance function to represent the area of the surface of the real crown tree to construct a surface model, that is, to use the normalized target point cloud data as input to realize the three-dimensional reconstruction of the crown tree.

[0140] The point cloud data processing method provided in the embodiment of the present application can extract target point cloud data that can represent the trunk area of crown trees without relying on manual operation, and then normalize the obtained target point cloud data. The normalized target point cloud data is used to perform three-dimensional reconstruction of the crown trees. This not only improves the accuracy of the three-dimensional reconstruction effect, but also reduces the three-dimensional reconstruction cycle of the crown trees because the process of obtaining the target point cloud data is based on machine learning algorithms such as clustering. For example, it can reduce the original manual operation time of tens of minutes to a few minutes of automatic operation time.

[0141] Figure 6 This is a schematic diagram of the structure of a point cloud data processing device provided in an embodiment of the present application. Figure 6 As shown, the point cloud data processing device 400 provided in the embodiment of the present application includes:

[0142] The acquisition module 401 is used to acquire the surveying and mapping data of crown trees.

[0143] Among them, the surveying and mapping data includes point cloud data to be processed.

[0144] The first processing module 402 is configured to obtain target segmentation points according to the point cloud data to be processed and a preset processing strategy.

[0145] Among them, the target segmentation point is used to segment the trunk and crown of each crown tree.

[0146] The second processing module 403 is configured to obtain target point cloud data according to the target segmentation points and the estimated radius of each crown tree.

[0147] Among them, the target point cloud data is used to characterize the trunk area of each crown tree.

[0148] In one possible design, the first processing module 402 includes:

[0149] A first processing submodule is configured to perform a first projection and a first clustering process on the point cloud data to be processed to obtain tree point cloud data, wherein the tree point cloud data is used to characterize each crown tree;

[0150] A second processing submodule is used to perform a second projection and a second clustering process on the tree point cloud data to obtain target segmentation points;

[0151] The preset processing strategy includes a first projection, a second projection, a first clustering process, and a second clustering process.

[0152] In one possible design, the first processing submodule is specifically configured to:

[0153] Performing a preset vertical projection on the point cloud data to be processed to obtain a plane point set, wherein the first projection includes a preset vertical projection;

[0154] The first clustering process is performed on the plane point set according to the preset accuracy to eliminate non-tree point sets and obtain tree point cloud data.

[0155] In one possible design, the second processing submodule is specifically configured to:

[0156] Performing a preset longitudinal projection and a preset transverse projection on the tree point cloud data in sequence to obtain a curve point set, wherein the second projection includes the preset longitudinal projection and the preset transverse projection;

[0157] The curve point set is subjected to a second clustering process according to a preset convergence principle to obtain target segmentation points. The number of clusters in the second clustering process is 2.

[0158] In one possible design, the second processing module 403 is specifically configured to:

[0159] Obtain the trunk center of each crown tree based on the target segmentation point;

[0160] The trunk area of each crown tree was determined based on the trunk center and radius estimates;

[0161] The point set in the trunk area is extracted as the target point cloud data.

[0162] In a possible design, the point cloud data processing device 400 further includes: a third processing module. The third processing module is specifically configured to:

[0163] Perform preset fitting processing based on the tree point cloud data to obtain the outer diameter fitting value and inner diameter fitting value of each crown tree;

[0164] The radius estimation value of each crown tree is obtained based on the outer diameter fitting value and inner diameter fitting value of each crown tree and the preset estimation model.

[0165] In a possible design, the point cloud data processing device 400 further includes: a fourth processing module. The fourth processing module is specifically configured to:

[0166] Normalize the coordinates of each point cloud representing the target point cloud data with the center of the tree trunk as the origin;

[0167] A surface model is generated based on the normalized target point cloud data, and the surface model is used to realize the three-dimensional reconstruction of the crown trees.

[0168] The point cloud data processing device provided in the embodiment of the present application can execute the corresponding steps of the point cloud data processing method in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here.

[0169] The above-mentioned device embodiments provided in this application are merely illustrative, and the module division therein is merely a logical functional division, and there may be other division methods in actual implementation. For example, multiple modules can be combined or integrated into another system. The coupling between the modules can be achieved through some interfaces, which are usually electrical communication interfaces, but it is not ruled out that they may be mechanical interfaces or other forms of interfaces. Therefore, the modules described as separate components may or may not be physically separated, and may be located in one place or distributed in different locations on the same or different devices.

[0170] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 7 As shown, the electronic device 500 may include: at least one processor 501 and a memory 502. Figure 7 An electronic device is shown using a processor as an example.

[0171] The memory 502 is used to store programs. Specifically, the programs may include program codes, and the program codes include computer-executable instructions.

[0172] The memory 502 may include a high-speed RAM memory, and may also include a non-volatile memory (MoM), such as at least one disk memory.

[0173] The processor 501 is configured to execute computer-executable instructions stored in the memory 502 to implement a point cloud data processing method.

[0174] The processor 501 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0175] Optionally, the memory 502 may be independent or integrated with the processor 501. When the memory 502 is a device independent of the processor 501, the electronic device 500 may further include:

[0176] Bus 503 is used to connect processor 501 and memory 502. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be divided into address buses, data buses, control buses, etc., but this does not mean that there is only one bus or only one type of bus.

[0177] Optionally, in a specific implementation, if the memory 502 and the processor 501 are integrated on a chip, the memory 502 and the processor 501 can communicate through an internal interface.

[0178] The present application also provides a computer-readable storage medium, which may include: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, and other media that can store program codes. Specifically, the computer-readable storage medium stores computer execution instructions, and the computer execution instructions are used in the point cloud data processing method in the above embodiment.

[0179] The present application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the point cloud data processing method in the above embodiment.

[0180] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered merely as exemplary, and the true scope and spirit of the present application are indicated by the claims.

[0181] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A point cloud data processing method, characterized in that: include: Acquiring surveying and mapping data of crown trees, wherein the surveying and mapping data includes point cloud data to be processed; Obtaining target segmentation points according to the point cloud data to be processed and a preset processing strategy, wherein the target segmentation points are used to segment the trunk and crown of each crown tree; Acquiring target point cloud data according to the target segmentation point and the estimated radius value of each crown tree, wherein the target point cloud data is used to characterize the trunk area of each crown tree; Obtaining target segmentation points according to the point cloud data to be processed and a preset processing strategy includes: Performing a first projection and a first clustering process on the point cloud data to be processed to obtain tree point cloud data, wherein the tree point cloud data is used to characterize each crown tree; Performing a second projection and a second clustering process on the tree point cloud data to obtain the target segmentation points; The preset processing strategy includes the first projection, the second projection, the first clustering processing, and the second clustering processing.

2. The point cloud data processing method according to claim 1, characterized in that: The performing a first projection and a first clustering process on the point cloud data to be processed to obtain tree point cloud data includes: Performing a preset vertical projection on the point cloud data to be processed to obtain a plane point set, wherein the first projection includes the preset vertical projection; The first clustering process is performed on the plane point set according to a preset accuracy to eliminate non-tree point sets, thereby obtaining the tree point cloud data.

3. The point cloud data processing method according to claim 2, characterized in that: The performing a second projection and a second clustering process on the tree point cloud data to obtain the target segmentation points includes: performing a preset longitudinal projection and a preset transverse projection on the tree point cloud data in sequence to obtain a curve point set, wherein the second projection includes the preset longitudinal projection and the preset transverse projection; The second clustering process is performed on the curve point set according to a preset convergence principle to obtain the target segmentation point, and the number of clusters in the second clustering process is 2.

4. The point cloud data processing method according to any one of claims 1 to 3, characterized in that: The step of obtaining target point cloud data according to the target segmentation point and the radius estimation value of each crown tree includes: Obtaining the trunk center of each crown tree according to the target segmentation point; determining a trunk area of each crown tree based on the trunk center and the radius estimate; The point set of the trunk area is extracted as the target point cloud data.

5. The point cloud data processing method according to claim 4, characterized in that: Before acquiring the target point cloud data, the method further includes: Performing a preset fitting process on the tree point cloud data to obtain an outer diameter fitting value and an inner diameter fitting value of each crown tree; The radius estimation value of each crown tree is obtained according to the outer diameter fitting value and the inner diameter fitting value of each crown tree and a preset estimation model.

6. The point cloud data processing method according to claim 5, characterized in that: After obtaining the target point cloud data, the method further includes: Normalizing the coordinates of each point cloud representing the target point cloud data with the center of the tree trunk as the origin; A surface model is generated according to the normalized target point cloud data, and the surface model is used to realize three-dimensional reconstruction of crown trees.

7. A point cloud data processing device, characterized in that: include: An acquisition module, configured to acquire surveying and mapping data of crown trees, wherein the surveying and mapping data includes point cloud data to be processed; A first processing module is configured to obtain target segmentation points based on the point cloud data to be processed and a preset processing strategy, wherein the target segmentation points are used to segment the trunk and crown of each crown tree; a second processing module, configured to obtain target point cloud data according to the target segmentation point and the estimated radius value of each crown tree, wherein the target point cloud data is used to represent the trunk area of each crown tree; The first processing module is specifically used to perform a first projection and a first clustering process on the point cloud data to be processed to obtain tree point cloud data, and the tree point cloud data is used to characterize each crown tree; and to perform a second projection and a second clustering process on the tree point cloud data to obtain the target segmentation point; wherein the preset processing strategy includes the first projection, the second projection, the first clustering process, and the second clustering process.

8. An electronic device, characterized in that: include: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the point cloud data processing method according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that: When the computer-executable instructions are executed by a processor, the point cloud data processing method according to any one of claims 1 to 6 is implemented.

Images