Point cloud robust sampling method based on polymerization rate

Abstract

The invention relates to a point cloud robust sampling method based on an aggregation rate. The method comprises the steps that firstly, the Euclidean distance between each point in a point cloud set and a neighbor point of the point cloud set is calculated, and the local diameter of a neighbor point set is calculated based on the Euclidean distance; calculating the polymerization rate of each point based on the Euclidean distance and the local diameter; then, determining a sampling point candidate set based on the aggregation rate; and finally, determining a target sampling point set based on the sampling point candidate set and the Euclidean distance. That is to say, when point cloud sampling is carried out based on the farthest point sampling method, local density analysis is carried out on sampling points, the density weight and the distance threshold are introduced, noise points are not sampled, and therefore it is guaranteed that the robustness of the improved farthest point sampling strategy is higher. The accuracy of point cloud data sampling is improved, and the range and applicability of the 3D recognition technology in real life can be expanded.

Description

Technical Field

[0001] The present application relates to the technical field of three-dimensional point cloud processing, and in particular to a point cloud robust sampling method based on aggregation rate. Background Art

[0002] With the popularization of 3D scanning technology and the rapid progress of data storage technology, the scale of point cloud data has shown an unprecedented growth trend. In cutting-edge fields such as cloud computing and the Internet of Things, point cloud data, as a core information carrier, has become particularly important for efficient processing. In recent years, the development of deep learning technology has brought new breakthroughs in 3D point cloud recognition, making it a key application in computer vision, robotics, autonomous driving, industrial inspection, medical imaging and other fields. It can be used for multiple tasks such as object detection, target tracking, scene reconstruction, map construction, etc., providing important information for automation systems.

[0003] However, the real world poses challenges to point cloud recognition, including noise and distortion caused by sensor errors, occlusions, and missing data, which can significantly impair the accuracy and robustness of the recognition model. The original 3D point cloud data usually contains a large number of points. When processing large-scale 3D point cloud data, important feature points in the point cloud can be extracted through sampling, and the data can be simplified, which can reduce the amount of calculation and memory consumption and speed up the execution of the algorithm. This is helpful for subsequent object recognition, classification, registration and other tasks.

[0004] Traditional point cloud sampling methods are mostly based on dense sampling strategies, that is, uniform sampling from dense point cloud data. However, this method is inadequate for large-scale point cloud data, as it faces severe challenges in both computing and storage resources, and it is difficult to meet the needs of efficient processing of large-scale point cloud data.

[0005] In the prior art, Farthest Point sampling (FPS) is a simple, efficient and widely used point cloud sampling method that does not rely on specific point cloud features, can maintain the uniformity of the point cloud to a certain extent, and is applicable to various types of point cloud data. Although farthest point sampling is good at preserving the skeleton structure of a given point cloud, it ignores potential data corruption, making it susceptible to noise and distortion. Farthest point sampling may select noise points as representative points, which may cause sampling bias and even make the sampling results inconsistent with the real data distribution, resulting in inaccurate sampling results.

[0006] Therefore, in the related technology, there is an urgent need for a method that can improve the sampling accuracy of point cloud data. Summary of the invention

[0007] Based on this, it is necessary to provide a point cloud robust sampling method based on aggregation rate that can improve the sampling accuracy of point cloud data in response to the above technical problems.

[0008] In a first aspect, the present application provides a point cloud robust sampling method based on aggregation rate. The method comprises:

[0009] Calculate the Euclidean distance between each point in the point cloud set and its neighboring points, and calculate the local diameter of the neighboring point set based on the Euclidean distance;

[0010] Calculate the aggregation rate of each point based on the Euclidean distance and the local diameter;

[0011] Determine a candidate set of sampling points based on the aggregation rate;

[0012] A target sampling point set is determined based on the candidate sampling point set and the Euclidean distance.

[0013] Optionally, in one embodiment of the present application, the calculation formula of the local diameter is:

[0014]

[0015] Among them, LD is the local diameter, P is the point cloud set, and p i is a point in the point cloud set, d i is the average of the Euclidean distances between all points and their neighboring points, and |P| is the number of points in the point cloud set.

[0016] Optionally, in one embodiment of the present application, the calculation formula of the polymerization rate is:

[0017]

[0018] Among them, N K For point p i The set of K nearest neighbor points (K-Nearest Neighbored, KNN), ∥p i -p j ∥ represents point p i With p j The Euclidean distance between them, I[·] is an exponential function, defined as:

[0019]

[0020] Where LD is the local diameter.

[0021] Optionally, in one embodiment of the present application, determining a candidate set of sampling points based on the aggregation rate includes:

[0022] Adding the point with the highest aggregation rate in the point cloud set to the sampling point candidate set;

[0023] Points in the point cloud set whose aggregation rate is higher than a preset threshold are added to the sampling point candidate set.

[0024] Optionally, in one embodiment of the present application, determining a target sampling point set based on the sampling point candidate set and the Euclidean distance includes:

[0025] Adding the point with the highest aggregation rate in the point cloud set to the target sampling point set;

[0026] Calculate the Euclidean distance from each point in the candidate set of sampling points to the point with the highest aggregation rate;

[0027] The point with the largest Euclidean distance is added to the target sampling point set, and the Euclidean distances of the remaining points are updated.

[0028] Optionally, in one embodiment of the present application, the method further includes:

[0029] The polymerization rate was normalized.

[0030] In a second aspect, the present application also provides a point cloud robust sampling device based on aggregation rate. The device comprises:

[0031] The Euclidean distance and local diameter calculation module is used to calculate the Euclidean distance between each point in the point cloud set and its neighboring points, and calculate the local diameter of the neighboring point set based on the Euclidean distance;

[0032] An aggregation rate calculation module, used for calculating the aggregation rate of each point based on the Euclidean distance and the local diameter;

[0033] A sampling point candidate set determination module, configured to determine a sampling point candidate set based on the aggregation rate;

[0034] The target sampling point set determination module is used to determine the target sampling point set based on the sampling point candidate set and the Euclidean distance.

[0035] In a third aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the steps of the methods described in the above embodiments.

[0036] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method described in each of the above embodiments are implemented.

[0037] The above-mentioned robust sampling method of point cloud based on aggregation rate first calculates the Euclidean distance between each point in the point cloud set and its neighboring points, and calculates the local diameter of the neighboring point set based on the Euclidean distance; then, the aggregation rate of each point is calculated based on the Euclidean distance and the local diameter; then, the candidate set of sampling points is determined based on the aggregation rate; finally, the target sampling point set is determined based on the candidate set of sampling points and the Euclidean distance. In other words, when sampling point clouds based on the farthest point sampling method, local density analysis is performed on the sampling points, and density weights and distance thresholds are introduced so that noise points are not sampled, thereby ensuring that the improved farthest point sampling strategy is more robust. It not only improves the accuracy of point cloud data sampling, but also expands the scope and applicability of 3D recognition technology in real life. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 FIG. 1 is an application environment diagram of a point cloud robust sampling method based on aggregation rate in one embodiment;

[0039] Figure 2 Schematic diagram of a flow chart of a point cloud robust sampling method based on aggregation rate in one embodiment;

[0040] Figure 3 It is a flowchart of specific steps of a point cloud robust sampling method based on aggregation rate in one embodiment;

[0041] Figure 4 is a structural block diagram of a point cloud robust sampling device based on aggregation rate in one embodiment;

[0042] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0044] The point cloud robust sampling method based on aggregation rate provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Among them, the terminal 102 can be but is not limited to various personal computers, laptops, smart phones, tablets, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented with an independent server or a server cluster consisting of multiple servers.

[0045] In one embodiment, Figure 2 As shown in the figure, a robust sampling method of point cloud based on aggregation rate is provided. Figure 1 The server in the example is used to illustrate the following steps:

[0046] S201: Calculate the Euclidean distance between each point in the point cloud set and its neighboring points, and calculate the local diameter of the neighboring point set based on the Euclidean distance.

[0047] In the embodiment of the present application, first, for each point p in the point cloud set P i Calculate the Euclidean distance to the K nearest neighbor points and find the average to get the average distance d i , the specific calculation formula is as follows.

[0048]

[0049] Among them, N K For point p i The set of K nearest neighbor points (K-Nearest Neighbored, KNN), K is an input parameter, set by humans, ∥p i -p j ∥ represents point p i With p j The Euclidean distance between .

[0050] After that, calculate the KNN average distance d of all points in the point cloud set P i The average value of the local diameter LD (LocalDiameter, LD) is obtained.

[0051] Specifically, in one embodiment of the present application, the calculation formula of the local diameter is:

[0052]

[0053] Among them, LD is the local diameter, P is the point cloud set, and pi is a point in the point cloud set, d i is the average of the Euclidean distances between all points and their neighboring points, and |P| is the number of points in the point cloud set.

[0054] S203: Calculate the aggregation rate of each point based on the Euclidean distance and the local diameter.

[0055] In the embodiment of the present application, each point p in the point cloud set P is calculated based on the Euclidean distance and the local diameter. i The polymerization rate w i The aggregation rate is used to quantify the degree of clustering of points in space. It reveals whether points tend to condense into dense groups. If the aggregation rate is low, it indicates that some points may behave as outliers and are relatively isolated in space. Specifically, if the Euclidean distance between the point and the K nearest points is less than the local diameter, the aggregation rate value of the point is increased by one.

[0056] Specifically, in one embodiment of the present application, the calculation formula of the polymerization rate is:

[0057]

[0058] Among them, N K For point p i The set of K nearest neighbor points (K-Nearest Neighbored, KNN), ∥p i -p j ∥ represents point p i With p j The Euclidean distance between them, I[·] is an exponential function, defined as:

[0059]

[0060] Where LD is the local diameter.

[0061] In one embodiment of the present application, the method further includes:

[0062] The polymerization rate was normalized.

[0063] In one embodiment of the present application, in order to eliminate the influence of the scale of the data set itself, so that the aggregation rate can be used to compare the aggregation degree of data in different data sets and increase the generalization of the indicator, it is necessary to normalize the aggregation rate to a number between 0 and 1. The specific calculation formula is as follows.

[0064]

[0065] Among them, w i For point p i The polymerization rate, w j For point pj The polymerization rate.

[0066] S205: Determine a candidate set of sampling points based on the aggregation rate.

[0067] In the embodiment of the present application, a target sampling point set S and a sampling point candidate set P are created. * , are initialized to empty sets, that is, Based on the calculated aggregation rate of each point, the corresponding point is selected and added to the candidate set of sampling points.

[0068] Specifically, in one embodiment of the present application, determining a candidate set of sampling points based on the aggregation rate includes:

[0069] S301: adding the point with the highest aggregation rate in the point cloud set to a sampling point candidate set.

[0070] S303: adding points in the point cloud set whose aggregation rate is higher than a preset threshold to the sampling point candidate set.

[0071] In one embodiment of the present application, in order to avoid selecting outliers, the point with the highest aggregation rate is selected from the point cloud set P. Right now Point Add to the sampling point candidate set P * In At the same time, delete the point from the original point cloud set P Right now Afterwards, for each point p in the current point cloud set P i , if the polymerization rate w i If it is higher than the preset threshold Ω, it will be added to the sampling point candidate set P * In, that is, P * =P * ∪{p i |p i ∈P,w i ≥Ω}, the preset threshold Ω is an input parameter, which is set manually.

[0072] S207: Determine a target sampling point set based on the candidate sampling point set and the Euclidean distance.

[0073] In the embodiment of the present application, based on the initially determined candidate set of sampling points, the corresponding Euclidean distance is recalculated, and the corresponding points are screened based on the Euclidean distance, and added to the target sampling point set to complete the point cloud sampling.

[0074] Specifically, in one embodiment of the present application, determining the target sampling point set based on the sampling point candidate set and the Euclidean distance includes:

[0075] S401: adding the point with the highest aggregation rate in the point cloud set to the target sampling point set.

[0076] S403: Calculate the Euclidean distance from each point in the candidate set of sampling points to the point with the highest aggregation rate.

[0077] S403: Add the point with the largest Euclidean distance to the target sampling point set, and update the Euclidean distances of the remaining points.

[0078] In one embodiment of the present application, first, the point with the highest polymerization rate is Add to the target sampling point set S, that is After that, create an n-dimensional array Dist∈R n (n=|P * |), used to store the candidate set P of sampling points * Each point p corresponding to the aggregation rate higher than the preset threshold j To point The Euclidean distance of

[0079]

[0080] Where j = {1,…,n}, p j ∈P * .

[0081] After that, select the point with the largest Euclidean distance in the current array Dist Add it to the target sampling point set S. That is, for any 1≤j≤n, Dist[j * ]≥Dist[j] holds:

[0082]

[0083] At the same time, update the array Dist, that is, update the Euclidean distance. * Every point p in j If the point-to-point If it is less than Dist[j], then update Dist[j], that is, for any 1≤j≤n:

[0084]

[0085] In a specific application, the above selection and update steps are repeated until one of the sampling end conditions is met, and the sampling ends. The sampling end condition is:

[0086] 1) The number of points in the target sampling point set S reaches the required number of sampling points m, where m is the input parameter.

[0087] m≤|P|, set manually.

[0088] 2) All values ​​in the array Dist are 0.

[0089] like Figure 3 As shown, it is a flowchart of the specific implementation steps of the aggregation rate-based point cloud robust sampling method of the present application.

[0090] In the above-mentioned robust sampling method of point cloud based on aggregation rate, first, the Euclidean distance between each point in the point cloud set and its neighboring points is calculated, and the local diameter of the neighboring point set is calculated based on the Euclidean distance; then, the aggregation rate of each point is calculated based on the Euclidean distance and the local diameter; then, the candidate set of sampling points is determined based on the aggregation rate; finally, the target sampling point set is determined based on the candidate set of sampling points and the Euclidean distance. In other words, when performing point cloud sampling based on the farthest point sampling method, local density analysis is performed on the sampling points, and density weights and distance thresholds are introduced so that noise points are not sampled, thereby ensuring that the improved farthest point sampling strategy is more robust. It not only improves the accuracy of point cloud data sampling, but also expands the scope and applicability of 3D recognition technology in real life.

[0091] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0092] Based on the same inventive concept, the embodiment of the present application also provides a point cloud robust sampling device based on aggregation rate for implementing the above-mentioned point cloud robust sampling method based on aggregation rate. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodiments of the point cloud robust sampling device based on aggregation rate provided below can refer to the limitations of the point cloud robust sampling method based on aggregation rate above, and will not be repeated here.

[0093] In one embodiment, Figure 4 As shown, a point cloud robust sampling device 400 based on aggregation rate is provided, comprising: a Euclidean distance and local diameter calculation module 401, an aggregation rate calculation module 403, a sampling point candidate set determination module 405 and a target sampling point set determination module 407, wherein:

[0094] The Euclidean distance and local diameter calculation module 401 is used to calculate the Euclidean distance between each point in the point cloud set and its neighboring points, and calculate the local diameter of the neighboring point set based on the Euclidean distance.

[0095] The aggregation rate calculation module 403 is used to calculate the aggregation rate of each point based on the Euclidean distance and the local diameter.

[0096] The sampling point candidate set determination module 405 is configured to determine a sampling point candidate set based on the aggregation rate.

[0097] The target sampling point set determination module 407 is configured to determine a target sampling point set based on the sampling point candidate set and the Euclidean distance.

[0098] In one embodiment of the present application, the calculation formula of the local diameter is:

[0099]

[0100] Among them, LD is the local diameter, P is the point cloud set, and p i is a point in the point cloud set, d i is the average of the Euclidean distances between all points and their neighboring points, and |P| is the number of points in the point cloud set.

[0101] In one embodiment of the present application, the calculation formula of the polymerization rate is:

[0102]

[0103] Among them, N K For point p i The set of K nearest neighbor points (K-Nearest Neighbored, KNN), ∥p i -p j ∥ represents point p i With p j The Euclidean distance between them, I[·] is an exponential function, defined as:

[0104]

[0105] Where LD is the local diameter.

[0106] In one embodiment of the present application, the sampling point candidate set determination module is further configured to:

[0107] Adding the point with the highest aggregation rate in the point cloud set to the sampling point candidate set;

[0108] Points in the point cloud set whose aggregation rate is higher than a preset threshold are added to the sampling point candidate set.

[0109] In one embodiment of the present application, the target sampling point set determination module is further configured to:

[0110] Adding the point with the highest aggregation rate in the point cloud set to the target sampling point set;

[0111] Calculate the Euclidean distance from each point in the candidate set of sampling points to the point with the highest aggregation rate;

[0112] The point with the largest Euclidean distance is added to the target sampling point set, and the Euclidean distances of the remaining points are updated.

[0113] In one embodiment of the present application, the aggregation rate calculation module is further used to:

[0114] The polymerization rate was normalized.

[0115] Each module in the above-mentioned point cloud robust sampling device based on aggregation rate can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0116] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a point cloud robust sampling method based on aggregation rate is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0117] Those skilled in the art will understand that Figure 5The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0118] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0119] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0120] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0121] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0122] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0123] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0124] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A point cloud robust sampling method based on aggregation rate, characterized in that: The method comprises: Calculate the Euclidean distance between each point in the point cloud set and its neighboring points, and calculate the local diameter of the neighboring point set based on the Euclidean distance; Calculate the aggregation rate of each point based on the Euclidean distance and the local diameter; Determine a candidate set of sampling points based on the aggregation rate; A target sampling point set is determined based on the candidate sampling point set and the Euclidean distance.

2. The point cloud robust sampling method based on aggregation rate according to claim 1 is characterized in that: The calculation formula of the local diameter is: Among them, LD is the local diameter, P is the point cloud set, and p i is a point in the point cloud set, d i is the average of the Euclidean distances between all points and their neighboring points, and |P| is the number of points in the point cloud set.

3. The point cloud robust sampling method based on aggregation rate according to claim 1, characterized in that: The calculation formula of the polymerization rate is: Among them, N K For point p i The set of K nearest neighbor points (K-Nearest Neighbored, KNN), ||p i -p j || represents point p i With p j The Euclidean distance between them, I[·] is an exponential function, defined as: Where LD is the local diameter.

4. The point cloud robust sampling method based on aggregation rate according to claim 1, characterized in that: The determining of a candidate set of sampling points based on the aggregation rate comprises: Adding the point with the highest aggregation rate in the point cloud set to the sampling point candidate set; Points in the point cloud set whose aggregation rate is higher than a preset threshold are added to the sampling point candidate set.

5. The point cloud robust sampling method based on aggregation rate according to claim 4 is characterized in that: Determining a target sampling point set based on the sampling point candidate set and the Euclidean distance comprises: Adding the point with the highest aggregation rate in the point cloud set to the target sampling point set; Calculate the Euclidean distance from each point in the candidate set of sampling points to the point with the highest aggregation rate; The point with the largest Euclidean distance is added to the target sampling point set, and the Euclidean distances of the remaining points are updated.

6. The point cloud robust sampling method based on aggregation rate according to claim 1, characterized in that: The method further comprises: The polymerization rate was normalized.

7. A point cloud robust sampling device based on aggregation rate, characterized in that: The device comprises: The Euclidean distance and local diameter calculation module is used to calculate the Euclidean distance between each point in the point cloud set and its neighboring points, and calculate the local diameter of the neighboring point set based on the Euclidean distance; An aggregation rate calculation module, used for calculating the aggregation rate of each point based on the Euclidean distance and the local diameter; A sampling point candidate set determination module, configured to determine a sampling point candidate set based on the aggregation rate; The target sampling point set determination module is used to determine the target sampling point set based on the sampling point candidate set and the Euclidean distance.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

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

Images