Target ball point cloud extraction method, device, equipment and storage medium

Abstract

The application discloses a target ball point cloud extraction method and device, equipment and a storage medium, and relates to the technical field of engineering surveying. The method comprises the following steps: obtaining a first total reflection material point cloud set by scanning a target ball by a scanner, wherein the surface of a target ball supporting rod of the target ball is covered with total reflection material; performing noise point filtering on the first total reflection material point cloud set to obtain a second total reflection material point cloud set; performing clustering on the second total reflection material point cloud set to obtain a clustering center set; and extracting a target ball point cloud through the clustering center set. By covering the surface of the target ball supporting rod of the target ball with total reflection material, the total reflection characteristics of the total reflection material are utilized to realize fast extraction of the target ball point cloud and improve the efficiency and accuracy of automatic search of the target ball.

Description

Technical Field

[0001] This application relates to the field of engineering measurement technology, and in particular to a method, apparatus, equipment and storage medium for extracting point clouds from a target sphere. Background Technology

[0002] With the rapid development of engineering surveying, in practical engineering, people often use targets (such as spherical targets, planar targets, etc.) to assist in the registration of point cloud data. Among them, spherical targets have been widely used due to their isotropic properties, high registration accuracy, and flexible configuration.

[0003] Spherical target detection methods mainly include curvature-based methods and point cloud data segmentation methods based on multiple geometric primitives. However, curvature-based methods are easily affected by noise, which may result in low accuracy; point cloud data segmentation methods based on multiple geometric primitives require segmenting the entire scene to obtain a finite number of spherical targets, which leads to high time consumption and low efficiency. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, device, and storage medium for extracting point clouds from a target sphere, aiming to solve the technical problem of low efficiency in the current point cloud data acquisition process.

[0005] To achieve the above objectives, this application proposes a target sphere point cloud extraction method, which includes: A first total reflection material point cloud set is obtained by scanning the target sphere with a scanner, wherein the surface of the target sphere support rod is covered with total reflection material; Noise filtering is performed on the first total internal reflection material point cloud set to obtain the second total internal reflection material point cloud set; Cluster the point cloud set of the second total internal reflection material to obtain a set of cluster centers; The target sphere point cloud is extracted using the cluster center set.

[0006] In one embodiment, the step of performing noise filtering on the first total internal reflection material point cloud set to obtain the second total internal reflection material point cloud set includes: Using each sampling point in the first total reflection material point cloud set as the center, fit a sphere corresponding to each sampling point according to a preset radius; The number of neighboring points in the sphere and the point cloud intensity of the neighboring points are statistically analyzed. When the number of neighboring points is less than the number threshold and the point cloud intensity of the neighboring points is greater than the preset intensity threshold, the sampling point corresponding to the sphere is determined to be a noise point; The noise points in the first total internal reflection material point cloud set are filtered out to obtain the second total internal reflection material point cloud set.

[0007] In one embodiment, after the step of counting the number of neighboring points in the sphere and the point cloud intensity of the neighboring points, the method further includes: The resolution of the scanner, the distance between the target sphere and the scanner, and the area of ​​the total reflective material are obtained. The linear point spacing is determined based on the resolution and the distance; The number of points per unit area is obtained based on the linear point spacing. The quantity threshold is determined based on the number of points per unit area and the area.

[0008] In one embodiment, the step of acquiring the first total internal reflection material point cloud set obtained by scanning the target sphere with a scanner includes: The point cloud intensity of each sampling point on the surface of the target sphere is obtained by scanning the target sphere with a scanner; When the point cloud intensity of the sampling point is less than a preset intensity threshold, the sampling point is filtered to obtain the filtered sampling point. The first total internal reflection material point cloud set is determined based on the filtered sampling points.

[0009] In one embodiment, the step of clustering the second total internal reflection material point cloud set to obtain a cluster center set includes: Select a preset number of sampling points from the second total internal reflection material point cloud set as the initial cluster centers; Calculate the Euclidean distance from each sampling point in the second total internal reflection material point cloud set to the initial cluster center; The preset number of clusters is obtained based on the Euclidean distance; Calculate the centroid of the cluster, and determine the updated cluster center based on the centroid; The updated cluster center is used as the initial cluster center, and the step of calculating the Euclidean distance from each sampling point in the second total internal reflection material point cloud set to the initial cluster center is returned to obtain the offset of the target cluster center and the target cluster; When the offset of the target cluster centers is less than the preset convergence threshold, a set of cluster centers is obtained based on the target cluster.

[0010] In one embodiment, the step of extracting the target sphere point cloud through the cluster center set includes: Obtain the distance between the total reflection material and the center of the target sphere; The spatial search interval is determined based on the distance between the total reflection material and the center of the target sphere. Extract the target sphere point cloud based on the cluster center set and the spatial search interval.

[0011] In one embodiment, the step of extracting the target sphere point cloud based on the cluster center set and the spatial search interval includes: Candidate point clouds are extracted based on the cluster center set and the spatial search interval; The curvature range is determined based on the surface curvature of the target sphere; Based on the candidate point cloud, the candidate curvature corresponding to each candidate point cloud is calculated. When the candidate curvature is not within the curvature range, the candidate point cloud corresponding to the candidate curvature is filtered to obtain the target sphere point cloud.

[0012] Furthermore, to achieve the above objectives, this application also proposes a target sphere point cloud extraction device, which includes: The acquisition module is used to acquire a first total reflection material point cloud set obtained by scanning the target sphere by the scanner, wherein the surface of the target sphere support rod is covered with total reflection material; The processing module is used to perform noise filtering on the first total reflection material point cloud set to obtain the second total reflection material point cloud set; The processing module is also used to cluster the second total internal reflection material point cloud set to obtain a cluster center set; The extraction module is used to extract the target sphere point cloud through the cluster center set.

[0013] In addition, to achieve the above objectives, this application also proposes a target sphere point cloud extraction device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the target sphere point cloud extraction method as described above.

[0014] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the target sphere point cloud extraction method described above.

[0015] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the target sphere point cloud extraction method described above.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: 1) By covering the surface of the target ball support rod with a total reflection material, the total reflection properties of the material are utilized to quickly extract the target ball point cloud, thereby improving the efficiency and accuracy of automatic target ball search.

[0017] 2) By filtering out the noise in the first total internal reflection material point cloud set, a second total internal reflection material point cloud set after noise filtering is obtained, which lays the foundation for subsequent accurate search of target sphere point cloud.

[0018] 3) By clustering the point cloud set of the second total reflection material, a set of cluster centers can be obtained, thereby enabling the fine extraction of the target sphere point cloud. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating an embodiment of the target point cloud extraction method of this application. Figure 2 A schematic diagram of the point cloud data registration process provided in an embodiment of the target ball point cloud extraction method of this application; Figure 3 A schematic diagram of the target sphere provided in an embodiment of the target sphere point cloud extraction method of this application; Figure 4 A schematic diagram of the target sphere fitting result provided in an embodiment of the target sphere point cloud extraction method of this application; Figure 5 This is a flowchart illustrating Embodiment 2 of the target sphere point cloud extraction method of this application; Figure 6 This is a schematic diagram comparing the point clouds before and after noise filtering, provided in an embodiment of the target ball point cloud extraction method of this application. Figure 7 This is a flowchart illustrating Embodiment 3 of the target sphere point cloud extraction method of this application; Figure 8 This is a schematic diagram of the coarse extraction result of the target sphere point cloud provided in an embodiment of the target sphere point cloud extraction method of this application; Figure 9 This is a schematic diagram showing the result of fine extraction of target sphere point cloud provided in an embodiment of the target sphere point cloud extraction method of this application; Figure 10 This is a schematic diagram of the module structure of the target ball point cloud extraction device according to an embodiment of this application; Figure 11 This is a schematic diagram of the hardware operating environment involved in the target ball point cloud extraction method in this application embodiment.

[0022] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0023] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0024] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0025] The main solution of this application embodiment is: to obtain a first total reflection material point cloud set obtained by scanning the target sphere with a scanner, wherein the surface of the target sphere support rod is covered with total reflection material; to filter noise in the first total reflection material point cloud set to obtain a second total reflection material point cloud set; to cluster the second total reflection material point cloud set to obtain a cluster center set; and to extract the target sphere point cloud through the cluster center set.

[0026] Existing technologies use curvature-based methods to obtain curvature and normal, and then determine the possible locations of spherical targets based on these parameters. Theoretically, this method can identify spherical targets with various radii. However, calculations based on curvature and normal are easily affected by noise. Alternatively, point cloud data segmentation methods based on multiple geometric primitives can be used, but these methods require segmenting the entire scene to obtain a finite number of spherical targets, resulting in high time consumption and low efficiency.

[0027] This application provides a solution that involves covering the surface of a target ball support rod with a total reflective material. Using this special target ball, a first total reflective material point cloud set is obtained by scanning the target ball with a scanner. The total reflective properties of the material guide a search algorithm to quickly and automatically locate the target ball. Noise filtering is performed on the first total reflective material point cloud set to remove noise, resulting in a second total reflective material point cloud set, ensuring the reliability of the point cloud data. The second total reflective material point cloud set is then clustered to obtain a set of cluster centers. The target ball point cloud is extracted from these cluster centers, forming an efficient and accurate target ball point cloud extraction process.

[0028] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device capable of performing the above functions, such as a target sphere point cloud extraction system. The following description uses a target sphere point cloud extraction system as an example to illustrate this embodiment and the subsequent embodiments. All actions involving the acquisition of signals, information, or data in this application are performed in accordance with the relevant data protection regulations of the country where the application is located and with authorization from the owner of the corresponding device.

[0029] Based on this, embodiments of this application provide a method for extracting target sphere point clouds, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the target point cloud extraction method of this application.

[0030] In this embodiment, the target sphere point cloud extraction method includes steps S10 to S40: Step S10: Obtain the first total reflection material point cloud set obtained by scanning the target sphere with a scanner, wherein the surface of the target sphere support rod is covered with total reflection material.

[0031] It should be noted that the scanner is typically a 3D laser scanner, a device that rapidly acquires a large number of 3D coordinates (point clouds) and optical properties (such as reflectance intensity) of an object or environment surface through non-contact measurement. Total internal reflection materials generally refer to materials with extremely high reflectivity and ideal diffuse reflection characteristics to the laser wavelength emitted by the scanner (usually near-infrared). The first total internal reflection material point cloud set is a subset of point cloud data, sharing the common characteristic that the point cloud intensity corresponding to each point in the first total internal reflection material point cloud set exceeds a preset intensity threshold.

[0032] Furthermore, in the field of engineering surveying, in order to obtain complete surface information of ground features, terrestrial laser scanners (at least terrestrial laser scanners are included in the category of 3D laser scanners) typically need to be set up at multiple stations. The point cloud data obtained from different stations have independent coordinate systems, so they need to be transformed to the same coordinate system before they can be used for subsequent processing. This transformation process is called point cloud data registration, such as... Figure 2 As shown, Figure 2 This is a schematic diagram of the point cloud data registration process, which typically includes coordinate rotation and coordinate translation.

[0033] Furthermore, in practical engineering, targets (such as spherical targets and planar targets) are frequently used to assist in the registration process of point cloud data. Among them, spherical targets have gained widespread application due to their isotropy, high registration accuracy, and flexible configuration. They use the center point of a corresponding spherical target (also called a target sphere) as the corresponding point, overcoming the difficulty in forming corresponding points due to factors such as scanner resolution and occlusion. This application uses a target sphere with a total reflection material covering the surface of the target sphere support rod as the target sphere, such as... Figure 3 As shown, Figure 3 This is a schematic diagram of the target ball. Figure 3 The sphere is the target ball (made of ordinary material), and the cylinder is the target ball support rod (made of ordinary material). A ring of total reflection material is pasted on the cylindrical grid area. When the laser scans the target ball from different directions, it can scan the total reflection material. The intensity value of the total reflection material is very stable and in a saturated state at different incident angles and scanning distances. It has a high degree of distinction from most ground features, which is conducive to guiding the target ball to be extracted quickly and automatically.

[0034] In addition, the layout of the total reflection material can be changed to other ways to facilitate scanning other shapes at different angles, such as arranging the total reflection material on a polygonal prism.

[0035] In one embodiment, the step of obtaining the first total internal reflection material point cloud set obtained by scanning the target sphere with a scanner includes: obtaining the point cloud intensity of each sampling point on the surface of the target sphere obtained by scanning the target sphere with a scanner; when the point cloud intensity of the sampling point is less than a preset intensity threshold, filtering the sampling point to obtain filtered sampling points; and determining the first total internal reflection material point cloud set based on the filtered sampling points.

[0036] Specifically, the point cloud intensity range obtained by the scanner is usually 0~n. When the scanner scans a totally reflective material, the point cloud intensity corresponding to the totally reflective material is usually very high (close to or equal to n). Based on this characteristic of totally reflective materials, the preset intensity threshold I is determined by subtracting a preset value from n. First, the target sphere is scanned by the scanner to determine the point cloud intensity of each sampling point on the surface of the target sphere, and the point cloud intensity data of each sampling point on the surface are obtained. When the point cloud intensity of a certain sampling point is less than the preset intensity threshold I, this sampling point is filtered out. The remaining sampling points after filtering are combined into a set, and this set is used as the first totally reflective material point cloud set.

[0037] By taking advantage of the characteristic that the point cloud intensity corresponding to the total reflection material is usually close to or equal to the upper limit of the scanning point cloud intensity of the scanner, the first total reflection material point cloud set is obtained according to the point cloud intensity of each sampling point on the surface of the target sphere, thereby improving the efficiency of subsequent extraction of the target sphere point cloud.

[0038] Step S20: Perform noise filtering on the first total reflection material point cloud set to obtain the second total reflection material point cloud set.

[0039] It should be noted that the second total internal reflection material point cloud set refers to the point cloud set after filtering out noise from the first total internal reflection material point cloud set. Since there may be objects in the scene with reflectivity similar to that of the total internal reflection material, or some sporadic noise points with intensities close to or equal to n, the first total internal reflection material point cloud set will usually contain some noise (noise typically refers to at least one of the following: discrete points in the point cloud data that do not belong to the surface of the target object, useless discrete points, or erroneous discrete points). To address this potential "noise situation," noise filtering can be performed on the first total internal reflection material point cloud set to obtain the second total internal reflection material point cloud set.

[0040] Step S30: Cluster the second total internal reflection material point cloud set to obtain a cluster center set.

[0041] It's important to note that clustering is typically an unsupervised machine learning method that aims to divide a set of data objects into several distinct groups or subsets based on their similarity. The cluster centroid set is usually a direct product of the clustering algorithm and is a set of points consisting of the centroid coordinates of all clusters.

[0042] In one embodiment, the step of clustering the second total internal reflection material point cloud set to obtain a set of cluster centers includes: selecting a preset number of sampling points from the second total internal reflection material point cloud set as initial cluster centers; calculating the Euclidean distance from each sampling point in the second total internal reflection material point cloud set to the initial cluster centers; obtaining the preset number of clusters based on the Euclidean distances; calculating the centroids of the clusters and determining updated cluster centers based on the centroids; using the updated cluster centers as initial cluster centers and returning to execute the step of calculating the Euclidean distance from each sampling point in the second total internal reflection material point cloud set to the initial cluster centers to obtain the offset of the target cluster centers and the target clusters; and obtaining a set of cluster centers based on the target clusters when the offsets of the target cluster centers are all less than a preset convergence threshold.

[0043] Specifically, the preset number of clusters is a pre-defined hyperparameter, typically determined by prior knowledge (such as the known number of objects) or by methods like the elbow rule and contour coefficients. It represents the desired number of cluster centers. The preset number of sampling points are randomly selected from the second total internal reflection material point cloud set as initial cluster centers (or sampling points that maximize the distance between initial centers are selected to improve convergence speed and clustering quality). For each sampling point in the second total internal reflection material point cloud set, its Euclidean distance to all initial cluster centers is calculated. Each sampling point is then assigned to the cluster corresponding to its nearest initial cluster center, resulting in the preset number of clusters. At this point, the second total internal reflection material point cloud set is divided into the preset number of disjoint subsets (i.e., clusters). For each cluster, its centroid is recalculated as the updated cluster center: that is, the arithmetic mean of the coordinates of all points within the cluster is calculated. If a cluster is empty (no point assignment), the center is usually re-initialized randomly. The step of calculating the Euclidean distance from each sampling point in the second total internal reflection material point cloud set to the initial cluster center is repeated until the iteration termination condition is met, and the final target cluster center and target cluster are obtained. The iteration termination condition is usually when the offset of each target cluster center (the offset of the target cluster center refers to the Euclidean distance between the final target cluster center and the cluster center obtained after the previous iteration update) is less than a preset convergence threshold (or the number of iterations reaches the maximum limit). When the offset of each target cluster center is less than the preset convergence threshold, the final target cluster center is determined according to the target cluster, and the cluster center set is obtained according to the final target cluster center.

[0044] It should be emphasized that the clustering method in this application is usually K-means clustering. Other implementations may also use density-based spatial clustering with noise (DBSCAN), agglomerative hierarchical clustering, spectral clustering, and other clustering methods to obtain the set of cluster centers.

[0045] This embodiment obtains a cluster center set by clustering the point cloud set of the second total reflection material, which provides a data foundation for subsequent fine extraction of the target sphere point cloud.

[0046] Step S40: Extract the target sphere point cloud through the cluster center set.

[0047] It should be noted that after obtaining the set of cluster centers, each cluster center in the set provides an "anchor point" for the possible position of the target sphere. Therefore, subsequent fine extraction and spherical fitting can be performed within a small space around the set of cluster centers, reducing the search range and thus significantly reducing the amount of computation and improving the processing efficiency.

[0048] In addition, after extracting the target sphere point cloud, the target sphere center can be fitted using the Random Sample Consensus (Ransac) method based on the target sphere point cloud and the known target sphere radius.

[0049] Specifically, based on the quality of the target sphere point cloud and computational resources, the obtained target sphere point clouds are combined into a target sphere point cloud set, and three key parameters are preset: (1) The first is the interior point determination distance threshold. It is used to determine whether a point lies on the surface of a sphere. If the distance from a point to the candidate sphere is less than this threshold, then the point is considered an interior point; (2) The second is the number of random sampling points. Since the radius of the target sphere is known, fitting a sphere requires determining the center position at least, and the center can be uniquely calculated through four non-coplanar points on the sphere. Therefore, the number of sampling points is set to four. (3) The third is the maximum number of iterations. It is set according to the estimated proportion of the outer points in the target sphere point cloud and is used to control the upper limit of the loop.

[0050] While presetting three key parameters, initialize the optimal number of interior points to zero and set the optimal target ball center coordinates to empty. Then, repeatedly execute the following four steps until the maximum number of iterations is reached: (1) Randomly select four sampling points from the target sphere point cloud set; (2) Using the spatial coordinates of these four sampling points and the known radius of the target ball, calculate a sphere that meets the conditions and obtain the center coordinates of the candidate target ball; (3) Traverse each point in the target sphere point cloud set and calculate the Euclidean distance from that point to the center coordinates of the candidate target sphere. If the absolute value of the difference between this distance and the known target sphere radius is less than the inlier determination distance threshold, then the point is counted as an inlier of the current model; (4) Compare the number of interior points in the current model with the number of the best interior points recorded in the past. If the current model has more interior points, it means that a better spherical model has been found. In this case, the current model is taken as the best model, and the center coordinates of the candidate target sphere corresponding to the current model are taken as the center coordinates of the best target sphere.

[0051] After the loop ends, the optimal target sphere center coordinates are the result of the coarse fitting. All interior points corresponding to the optimal model are extracted to form a pure set of interior points on the sphere; these points are considered to best represent the data of the real target sphere surface. On this set of interior points, the least squares method is used to refit the sphere, obtaining the optimized target sphere center coordinates. Finally, the optimized target sphere center coordinates are used as the final fitted target sphere center output.

[0052] Understandably, the purpose of extracting the target sphere point cloud is to obtain the optimized target sphere center coordinates, and thus obtain the target sphere fitting result, such as... Figure 4 As shown, Figure 4 This is a schematic diagram of the target sphere fitting results, which are ultimately used for point cloud data registration.

[0053] This embodiment provides a target sphere point cloud extraction method. It involves acquiring a first total internal reflection material point cloud set obtained by scanning a target sphere with a scanner, wherein the surface of the target sphere's support rod is covered with total internal reflection material. Noise filtering is performed on the first total internal reflection material point cloud set to obtain a second total internal reflection material point cloud set. Clustering is then performed on the second total internal reflection material point cloud set to obtain a cluster center set. The target sphere point cloud is extracted using the cluster center set. By covering the surface of the target sphere's support rod with total internal reflection material, and utilizing the total internal reflection properties of the material, rapid extraction of the target sphere point cloud is achieved, improving the efficiency and accuracy of automatic target sphere search.

[0054] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 5 Step S20 includes steps S201 to S204: Step S201: Using each sampling point in the first total reflection material point cloud set as the center, fit the sphere corresponding to each sampling point according to the preset radius.

[0055] It should be noted that step S20 is a step of noise removal using the radius filtering method. The core logic of the radius filtering method is: if a sampling point does not have enough "neighbors" within a sphere of a preset radius around it, then it is likely an isolated noise point and should be filtered out.

[0056] It is understandable that, with each sampling point in the first total internal reflection material point cloud set as the center, fitting the sphere corresponding to each sampling point according to the preset radius is to search for information about the "neighbors" in this sphere.

[0057] Step S202: Count the number of neighborhood points in the sphere and the point cloud intensity of the neighborhood points.

[0058] It should be noted that neighborhood points are usually the "neighbors" of the sampling points corresponding to the sphere. They typically refer to all other points contained within a spherical space of a specified radius centered on any sampling point selected as a reference in the point cloud set.

[0059] In one embodiment, after the step of counting the number of neighboring points in the sphere and the point cloud intensity of the neighboring points, the method further includes: obtaining the resolution of the scanner, the distance between the target sphere and the scanner, and the area of ​​the total reflection material; determining the linear point spacing based on the resolution and the distance; obtaining the number of points per unit area based on the linear point spacing; and determining a quantity threshold based on the number of points per unit area and the area.

[0060] Specifically, based on the scanner's resolution and the distance between the target sphere and the scanner, the calculation process for determining the linear dot spacing at that distance is as follows:

[0061] In the above formula, The linear dot pitch is L, where L is the distance between the target sphere and the scanner. Let be the scanner's resolution. Based on the linear dot pitch obtained in the previous step, the calculation process for the number of dots per unit area (i.e., dot density) is as follows:

[0062] In the above formula, Points per unit area The linear dot spacing is used. After obtaining the number of dots per unit area, the calculation process for the theoretically required number of dots within an area, based on the number of dots per unit area and the area of ​​the total reflective material, is as follows:

[0063] In the above formula, The number of points per unit area is given, where A is the area of ​​the total reflective material. This represents the theoretical number of points that should be included within the area. After obtaining the theoretical number of points that should be included within the area, a threshold is set to a certain percentage (usually 50% to 80%) of the theoretically required number of points to determine whether a sampling point is noise.

[0064] Step S203: When the number of neighboring points is less than the number threshold and the point cloud intensity of the neighboring points is greater than the preset intensity threshold, the sampling point corresponding to the sphere is determined to be a noise point.

[0065] In practice, when the number of neighboring points in a sphere is less than the number threshold, it means that the number of "neighbors" of the sampling point corresponding to the sphere is small, and when the point cloud intensity of the neighboring points is greater than the preset intensity threshold (i.e., the neighboring point is identified as a point in the total reflection material region), the sampling point corresponding to the sphere is determined to be a noise point.

[0066] Step S204: Filter out the noise points in the first total reflection material point cloud set to obtain the second total reflection material point cloud set.

[0067] For easier understanding, please refer to Figure 6 , Figure 6 This is a diagram comparing the point clouds before and after noise filtering, as shown below. Figure 6 As shown, before filtering, there were some scattered points (i.e., noise); after filtering, almost all of the scattered points were removed.

[0068] In addition, this application can use the radius filtering method described above for noise filtering, and can also use statistical filtering, Gaussian filtering and bilateral filtering methods for noise filtering.

[0069] This embodiment fits a sphere corresponding to each sampling point in the first total internal reflection material point cloud set to a preset radius; counts the number of neighboring points and the point cloud intensity of the neighboring points in the sphere; when the number of neighboring points is less than a quantity threshold and the point cloud intensity of the neighboring points is greater than a preset intensity threshold, the sampling point corresponding to the sphere is determined to be a noise point; the noise points in the first total internal reflection material point cloud set are filtered to obtain a second total internal reflection material point cloud set. Noise filtering can systematically remove these erroneous data (i.e., noise points), so that the output second total internal reflection material point cloud set can more accurately reflect the real surface of the scanned object, improving the reliability and usability of the point cloud data.

[0070] Based on the first embodiment of this application, in the third embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 7 Step S40 includes steps S401 to S403: Step S401: Obtain the distance between the total reflection material and the center of the target sphere.

[0071] It should be noted that the distance between the total reflection material and the center of the target sphere is the straight-line distance from the total reflection material to the geometric center of the target sphere, which is a known fixed value determined by the design specifications of the target sphere.

[0072] Step S402: Determine the spatial search interval based on the distance between the total reflection material and the center of the target sphere.

[0073] It should be noted that the spatial search range usually refers to a three-dimensional spatial range set in the coarse extraction step of point cloud in order to filter out points that may belong to the surface of the target sphere.

[0074] Understandably, since the scanned point cloud has thickness, the actual point position may deviate. Therefore, a first tolerance value can be preset. The distance between the reflective material and the center of the target sphere minus the preset first tolerance value is used as the lower limit of the spatial search interval. The distance between the reflective material and the center of the target sphere plus the first tolerance value is used as the upper limit of the spatial search interval. Finally, the spatial search interval can be determined based on the lower limit of the spatial search interval and the first tolerance value of the target sphere.

[0075] Step S403: Extract the target sphere point cloud based on the cluster center set and the spatial search interval.

[0076] In one embodiment, the step of extracting the target sphere point cloud based on the cluster center set and the spatial search interval includes: extracting candidate point clouds based on the cluster center set and the spatial search interval; determining the curvature range based on the surface curvature of the target sphere; calculating the candidate curvature corresponding to each candidate point cloud based on the candidate point clouds; and filtering the candidate point clouds corresponding to the candidate curvature when the candidate curvature is not within the curvature range to obtain the target sphere point cloud.

[0077] Specifically, points in the cluster center set that satisfy the spatial search interval can be retained, and these points can be used as candidate point clouds. The candidate point clouds extracted through the cluster center set and the spatial search interval are the coarse extraction results of the target sphere point cloud. Figure 8 As shown, Figure 8 This is a schematic diagram of the coarse extraction results of the target sphere point cloud.

[0078] After extracting candidate point clouds, the curvature range is determined using the known, constant curvature of the target sphere surface and a preset second tolerance value. The candidate curvature corresponding to each candidate point cloud is calculated. When a candidate curvature is outside the curvature range, the candidate point cloud corresponding to that curvature is filtered out. The remaining candidate point clouds within the curvature range are used as the target sphere point cloud. The resulting target sphere point cloud is the refined extraction result of the target sphere point cloud. Figure 9 As shown, Figure 9 This is a schematic diagram of the refined extraction results of the target sphere point cloud. The curvature mentioned above can be at least one of the following: mean curvature, Gaussian curvature, discrete curvature, and surface variation derived from the structure tensor.

[0079] This embodiment obtains the distance between the total internal reflection material and the center of the target sphere; determines the spatial search interval based on the distance between the total internal reflection material and the center of the target sphere; and extracts the target sphere point cloud based on the cluster center set and the spatial search interval. Through the above-described coarse and fine extraction methods, the points on the target sphere surface can be gradually and accurately separated from massive noise and interference, thereby improving the accuracy of the target sphere point cloud extraction.

[0080] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the target sphere point cloud extraction method of this application. Any simple transformations based on this technical concept are within the protection scope of this application.

[0081] This application also provides a target sphere point cloud extraction device, please refer to... Figure 10 The target sphere point cloud extraction device includes: The acquisition module 10 is used to acquire a first total reflection material point cloud set obtained by scanning the target ball with a scanner, wherein the surface of the target ball support rod of the target ball is covered with total reflection material.

[0082] Processing module 20 is used to perform noise filtering on the first total reflection material point cloud set to obtain a second total reflection material point cloud set.

[0083] The processing module 20 is further configured to cluster the second total reflection material point cloud set to obtain a cluster center set.

[0084] Extraction module 30 is used to extract the target sphere point cloud through the cluster center set.

[0085] The target sphere point cloud extraction device provided in this application, employing the target sphere point cloud extraction method described in the above embodiments, can solve the technical problem of low efficiency in the current point cloud data acquisition process. Compared with the prior art, the beneficial effects of the target sphere point cloud extraction device provided in this application are the same as those of the target sphere point cloud extraction method provided in the above embodiments, and other technical features in the target sphere point cloud extraction device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0086] In one embodiment, the processing module 20 is further configured to: fit a sphere corresponding to each sampling point in the first total internal reflection material point cloud set as the center, according to a preset radius; count the number of neighboring points and the point cloud intensity of the neighboring points in the sphere; when the number of neighboring points is less than a number threshold and the point cloud intensity of the neighboring points is greater than a preset intensity threshold, determine the sampling point corresponding to the sphere as a noise point; filter the noise points in the first total internal reflection material point cloud set to obtain a second total internal reflection material point cloud set.

[0087] In one embodiment, the processing module 20 is further configured to acquire the resolution of the scanner, the distance between the target sphere and the scanner, and the area of ​​the total reflection material; determine the linear dot pitch based on the resolution and the distance; and obtain the number of dots per unit area based on the linear dot pitch. The quantity threshold is determined based on the number of points per unit area and the area.

[0088] In one embodiment, the acquisition module 10 is further configured to acquire the point cloud intensity of each sampling point on the surface of the target target sphere obtained by scanning the target target sphere with a scanner; when the point cloud intensity of the sampling point is less than a preset intensity threshold, the sampling point is filtered to obtain filtered sampling points; and a first total internal reflection material point cloud set is determined based on the filtered sampling points.

[0089] In one embodiment, the processing module 20 is further configured to: select a preset number of sampling points from the second total internal reflection material point cloud set as initial cluster centers; calculate the Euclidean distance from each sampling point in the second total internal reflection material point cloud set to the initial cluster centers; obtain the preset number of clusters based on the Euclidean distances; calculate the centroid of the clusters; determine the updated cluster centers based on the centroids; use the updated cluster centers as initial cluster centers, and return to execute the step of calculating the Euclidean distance from each sampling point in the second total internal reflection material point cloud set to the initial cluster centers, thereby obtaining the offset of the target cluster centers and the target clusters; and when the offsets of the target cluster centers are all less than a preset convergence threshold, obtain a set of cluster centers based on the target clusters.

[0090] In one embodiment, the extraction module 30 is further configured to obtain the distance between the total reflection material and the center of the target sphere; determine the spatial search interval based on the distance between the total reflection material and the center of the target sphere; and extract the target sphere point cloud based on the cluster center set and the spatial search interval.

[0091] In one embodiment, the extraction module 30 is further configured to extract candidate point clouds based on the cluster center set and the spatial search interval; determine the curvature range based on the surface curvature of the target sphere; calculate the candidate curvature corresponding to each candidate point cloud based on the candidate point clouds; and filter the candidate point clouds corresponding to the candidate curvature when the candidate curvature is not within the curvature range to obtain the target sphere point cloud.

[0092] This application provides a target sphere point cloud extraction device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the target sphere point cloud extraction method in the above embodiment 1.

[0093] The following is for reference. Figure 11 The diagram illustrates a structural schematic of a target sphere point cloud extraction device suitable for implementing embodiments of this application. The target sphere point cloud extraction device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), tablets, PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 11 The target sphere point cloud extraction device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0094] like Figure 11 As shown, the target ball point cloud extraction device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM (Read Only Memory) 1002 or a program loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the target ball point cloud extraction device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, LCDs (Liquid Crystal Displays), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the target ball point cloud extraction device to communicate wirelessly or wiredly with other devices to exchange data. Although a target ball point cloud extraction device with various systems is shown in the figure, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0095] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0096] The target sphere point cloud extraction device provided in this application, employing the target sphere point cloud extraction method described in the above embodiments, can solve the technical problem of low efficiency in the current point cloud data acquisition process. Compared with the prior art, the beneficial effects of the target sphere point cloud extraction device provided in this application are the same as those of the target sphere point cloud extraction method provided in the above embodiments, and other technical features in this target sphere point cloud extraction device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0097] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0098] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0099] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the target sphere point cloud extraction method in the above embodiments.

[0100] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory or Flash Memory), optical fibers, CD-ROM (CD-Read Only Memory), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0101] The aforementioned computer-readable storage medium may be included in the target sphere point cloud extraction device; or it may exist independently and not assembled into the target sphere point cloud extraction device.

[0102] The aforementioned computer-readable storage medium carries one or more programs. When the one or more programs are executed by the target ball point cloud extraction device, the target ball point cloud extraction device: acquires a first total reflection material point cloud set obtained by scanning the target ball with a scanner, wherein the surface of the target ball support rod of the target ball is covered with total reflection material; performs noise filtering on the first total reflection material point cloud set to obtain a second total reflection material point cloud set; clusters the second total reflection material point cloud set to obtain a cluster center set; and extracts the target ball point cloud through the cluster center set.

[0103] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including LAN (Local Area Network) or WAN (Wide Area Network)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0104] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0105] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0106] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described target sphere point cloud extraction method, which can solve the technical problem of low efficiency in the current point cloud data acquisition process. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the target sphere point cloud extraction method provided in the above embodiments, and will not be repeated here.

[0107] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the target sphere point cloud extraction method described above.

[0108] The computer program product provided in this application can solve the technical problem of low efficiency in the current point cloud data acquisition process. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the target sphere point cloud extraction method provided in the above embodiments, and will not be repeated here.

[0109] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for extracting point clouds of a target sphere, characterized in that, The target sphere point cloud extraction method includes: A first total reflection material point cloud set is obtained by scanning the target sphere with a scanner, wherein the surface of the target sphere support rod is covered with total reflection material; Noise filtering is performed on the first total internal reflection material point cloud set to obtain the second total internal reflection material point cloud set; Cluster the point cloud set of the second total internal reflection material to obtain a set of cluster centers; The target sphere point cloud is extracted using the cluster center set.

2. The method as described in claim 1, characterized in that, The step of filtering noise from the first total internal reflection material point cloud set to obtain the second total internal reflection material point cloud set includes: Using each sampling point in the first total reflection material point cloud set as the center, fit a sphere corresponding to each sampling point according to a preset radius; The number of neighboring points in the sphere and the point cloud intensity of the neighboring points are statistically analyzed. When the number of neighboring points is less than the number threshold and the point cloud intensity of the neighboring points is greater than the preset intensity threshold, the sampling point corresponding to the sphere is determined to be a noise point; The noise points in the first total internal reflection material point cloud set are filtered out to obtain the second total internal reflection material point cloud set.

3. The method as described in claim 2, characterized in that, After the steps of counting the number of neighborhood points in the sphere and the point cloud intensity of the neighborhood points, the method further includes: The resolution of the scanner, the distance between the target sphere and the scanner, and the area of ​​the total reflective material are obtained. The linear point spacing is determined based on the resolution and the distance; The number of points per unit area is obtained based on the linear point spacing. The quantity threshold is determined based on the number of points per unit area and the area.

4. The method as described in claim 1, characterized in that, The step of acquiring the first total internal reflection material point cloud set obtained by scanning the target sphere with a scanner includes: The point cloud intensity of each sampling point on the surface of the target sphere is obtained by scanning the target sphere with a scanner; When the point cloud intensity of the sampling point is less than a preset intensity threshold, the sampling point is filtered to obtain the filtered sampling point. The first total internal reflection material point cloud set is determined based on the filtered sampling points.

5. The method as described in claim 1, characterized in that, The step of clustering the second total internal reflection material point cloud set to obtain the cluster center set includes: Select a preset number of sampling points from the second total internal reflection material point cloud set as the initial cluster centers; Calculate the Euclidean distance from each sampling point in the second total internal reflection material point cloud set to the initial cluster center; The preset number of clusters is obtained based on the Euclidean distance; Calculate the centroid of the cluster, and determine the updated cluster center based on the centroid; The updated cluster center is used as the initial cluster center, and the step of calculating the Euclidean distance from each sampling point in the second total internal reflection material point cloud set to the initial cluster center is returned to obtain the offset of the target cluster center and the target cluster; When the offset of the target cluster centers is less than the preset convergence threshold, a set of cluster centers is obtained based on the target cluster.

6. The method as described in claim 1, characterized in that, The step of extracting the target sphere point cloud through the cluster center set includes: Obtain the distance between the total reflection material and the center of the target sphere; The spatial search interval is determined based on the distance between the total reflection material and the center of the target sphere. Extract the target sphere point cloud based on the cluster center set and the spatial search interval.

7. The method as described in claim 6, characterized in that, The step of extracting the target sphere point cloud based on the cluster center set and the spatial search interval includes: Candidate point clouds are extracted based on the cluster center set and the spatial search interval; The curvature range is determined based on the surface curvature of the target sphere; Based on the candidate point cloud, the candidate curvature corresponding to each candidate point cloud is calculated. When the candidate curvature is not within the curvature range, the candidate point cloud corresponding to the candidate curvature is filtered to obtain the target sphere point cloud.

8. A target sphere point cloud extraction device, characterized in that, The target sphere point cloud extraction device includes: The scanning module is used to acquire a first total reflection material point cloud set obtained by scanning the target sphere with a scanner, wherein the surface of the target sphere support rod is covered with total reflection material; The processing module is used to perform noise filtering on the first total reflection material point cloud set to obtain the second total reflection material point cloud set; The processing module is also used to cluster the second total internal reflection material point cloud set to obtain a cluster center set; The extraction module is used to extract the target sphere point cloud through the cluster center set.

9. A target sphere point cloud extraction device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the target sphere point cloud extraction method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the target sphere point cloud extraction method as described in any one of claims 1 to 7.

Images