Method for quickly projecting large-scale point cloud to parametric curved surface, storage medium and equipment

Abstract

The invention discloses a method for rapidly projecting a large-scale point cloud to a parametric curved surface, a storage medium and equipment, and relates to the technical field of industrial measurement, and the method mainly comprises the steps: carrying out the point cloud extraction according to a CAD theoretical model and a corresponding point cloud grid model, and obtaining a surface parallelogram point cloud; parameterization division is carried out on the curved surface of the CAD theoretical model to obtain parametric curved surface point clouds, a projection initial value is obtained by combining the parametric surface point clouds, and projection points of the to-be-projected points to the parametric curved surface are obtained by using a least square method; by implementing the method, the storage medium and the equipment for rapidly projecting the large-scale point cloud to the parametric curved surface, the large-scale point cloud error analysis speed in industrial measurement can be increased.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of industrial measurement technology, in particular to a large-scale point cloud to parameter surface fast projection method, a storage medium and equipment. BACKGROUND

[0002] With the rapid development of laser three-dimensional scanning technology and optical measurement digital technology, the demand for surface digitization of free complex surface physical models is increasing. Through three-dimensional scanning technology, complex part surface discrete sampling point cloud data information can be quickly obtained, and the visual error field distribution can be compared with the CAD model. Since the unordered point cloud data has no topological structure, the measurement points have no corresponding relationship with the theoretical surface, and a large number of point cloud calculations are time-consuming and costly, which seriously affects the comparison efficiency of point cloud and CAD model.

[0003] How to quickly project large-scale point cloud to parameter surface is a technical problem to be solved. SUMMARY

[0004] The purpose of the present application is to provide a large-scale point cloud to parameter surface fast projection method, a storage medium and equipment, which can improve the speed of large-scale point cloud error analysis in industrial measurement.

[0005] The present application provides a large-scale point cloud to parameter surface fast projection method, comprising the following steps: S1: extracting point cloud according to the theoretical model and the corresponding point cloud grid model to obtain a surface bias body point cloud; S2: parameterizing and dividing the surface of the theoretical model to obtain a parameter surface point cloud; S3: obtaining a projection initial value according to the surface bias body point cloud and the parameter surface point cloud; S4: obtaining a projection point of the point to be projected to the parameter surface by using the least square method according to the projection initial value.

[0006] The present application also provides a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the steps of the large-scale point cloud to parameter surface fast projection method described above.

[0007] The present application also provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the large-scale point cloud to parameter surface fast projection method described above when executing the program.

[0008] The large-scale point cloud to parameter surface fast projection method, the storage medium and the equipment provided by the present application have the following advantages: The present application aims at the problem of long comparison calculation time and large cost of large-scale measurement point cloud and CAD theoretical model in three-dimensional scanning. Firstly, the corresponding relationship between the matching point cloud and the surface of the CAD theoretical model is obtained by intersecting the parametric surface bias body and the scanning point cloud grid model, and the corresponding point cloud data PC_Face is extracted. Secondly, the surface is divided into regions according to the equal parameter to obtain the parameter grid Grid_UV, and the center points of all sub-regions constitute the point cloud PC_UV. Then, the nearest point of the sampling point in PC_UV is queried to match the point and the corresponding relationship of Grid_UV region and obtain the parameter range start_uv. Finally, the start_uv is taken as the initial value of Newton projection, and the projection point of the measurement point to the parametric surface is solved based on the least square method. The present application adopts the idea of point cloud segmentation and block projection, which greatly improves the speed of large-scale point cloud error analysis in industrial measurement. BRIEF DESCRIPTION OF DRAWINGS

[0009] The present application will be further described below in combination with the drawings and embodiments, and the drawings are as follows: Figure 1 It is the flow chart of the large-scale point cloud fast projection method to the parametric surface provided by the present application; Figure 2 It is the implementation method schematic diagram of the large-scale point cloud fast projection method to the parametric surface provided by the present application; Figure 3 It is the CAD theoretical model and scanning point cloud model example diagram of the large-scale point cloud fast projection method to the parametric surface provided by the present application; Figure 4 It is the point cloud extraction principle schematic diagram based on the surface feature of the large-scale point cloud fast projection method to the parametric surface provided by the present application; Figure 5 It is the surface bias body schematic diagram of the large-scale point cloud fast projection method to the parametric surface provided by the present application; Figure 6 It is the parameter surface region division schematic diagram of the large-scale point cloud fast projection method to the parametric surface provided by the present application; Figure 7 It is the effect diagram of the large-scale point cloud fast projection method to the parametric surface provided by the present application; Figure 8 It is the structure block diagram of the computer equipment provided by the present application. DETAILED DESCRIPTION

[0010] In order to have a clearer understanding of the technical features, purposes and effects of the present application, the specific embodiments of the present application will be described in detail with reference to the drawings.

[0011] Figure 1A schematic diagram of the large-scale point cloud to parameter surface fast projection method of the embodiment is shown. In the embodiment, the large-scale point cloud to parameter surface fast projection method comprises the following steps: S1: extracting point cloud according to a theoretical model and a corresponding point cloud grid model to obtain a surface bias body point cloud; In an exemplary embodiment, step S1 specifically comprises: S11: obtaining a surface bias body according to a theoretical machining allowance and a surface bias value of the theoretical model, and obtaining a minimum bounding box according to the surface bias body; In an exemplary embodiment, the surface bias body comprises an upper bias surface and a lower bias surface, and the upper bias surface and the lower bias surface are biased by 3 times; S12: filtering out non-corresponding points by using the minimum bounding box according to the point cloud grid model to obtain a minimum bounding box point cloud; S13: obtaining a surface bias body point cloud by using a region growing algorithm according to the minimum bounding box point cloud; It should be noted that the theoretical model in the embodiment is a CAD theoretical model.

[0012] In an exemplary embodiment, step S13 specifically comprises: establishing a binary tree according to the minimum bounding box point cloud, taking a point on a surface of the theoretical model as a seed point, finding a nearest point by using a region growing algorithm, if the nearest point is inside the surface bias body, taking the nearest point as a growing point, and recursively growing from the growing point until a new growing point cannot be found, to obtain the surface bias body point cloud.

[0013] S2: parameterizing and dividing the surface of the theoretical model to obtain a parameter surface point cloud; In an exemplary embodiment, step S2 specifically comprises: dividing the surface of the theoretical model into a plurality of parameter regions according to equal parameters to obtain a parameter grid; The parameter region comprises a region center point and a parameter range, and the parameter surface point cloud is obtained according to the parameter region.

[0014] S3: obtaining a projection initial value according to the surface bias body point cloud and the parameter surface point cloud; In an exemplary embodiment, step S3 specifically comprises: establishing a binary tree according to the parameter surface point cloud, finding a nearest point of a distance between a sampling point in the surface bias body point cloud and a region center point of the parameter surface point cloud, and taking a parameter range corresponding to the nearest point as a projection initial value of the sampling point.

[0015] S4: obtaining a projection point of a to-be-projected point to a parameter surface by using a least square method according to the projection initial value; In an exemplary embodiment, the minimization objective function of the least square method is: , , wherein, represents the distance minimization objective function of the projection point to the surface; represents the parameter surface; and represent two parameters of the curve base function in the surface; represents the point to be projected; and D is the parameter domain.

[0016] In an exemplary embodiment, the large-scale point cloud fast projection method to the parameter surface further comprises: calculating an error value according to the projection point, picking up a color according to the error value to fill the point cloud grid model, and obtaining an error field color spot map.

[0017] In some embodiments, the large-scale point cloud fast projection method to the parameter surface can also be implemented in the following manner.

[0018] As Figure 2 shown is a large-scale point cloud fast projection method implementation method schematic diagram; in this embodiment, the large-scale point cloud fast projection method to the parameter surface comprises the following steps: Step 1: model import.

[0019] Import the CAD theoretical model (STEP / IGES), and import the scanned point cloud grid model (STL). As Figure 3 is the CAD theoretical model and the scanned point cloud model example diagram of the present application, wherein the gray model is the CAD theoretical model, and the yellow model is the scanned point cloud model, both of which are in the registration state.

[0020] Step 2: point cloud extraction.

[0021] The parallel technology is adopted to traverse all the faces of the CAD theoretical model to extract the corresponding data points, determine the corresponding relationship between the point cloud data and the faces of the CAD theoretical model, and extract the corresponding point cloud data PC_Face.

[0022] As Figure 4 is the point cloud extraction principle based on the face feature of the present application, wherein the face bias body is the thickened body of the face to be extracted, and the bounding box is the minimum bounding box of the face bias body.

[0023] Firstly, when a face in the CAD theoretical model is traversed, the face bias body and its minimum bounding box are calculated, wherein the face bias body is calculated according to the theoretical machining allowance and the profile deviation value, and a thickened body is formed between the upper deviation face and the lower deviation face.

[0024] As Figure 5As shown in the schematic diagram of the face bias body of the present application, the space between the upper and lower deviation surfaces forms the face bias body, like a profile thickening, wherein the thickness can be determined according to the actual working condition, and the upper and lower deviation surfaces are generally biased by 3 times the optimum.

[0025] Then, the minimum bounding box of the face bias body is used to filter most of the non-corresponding points to obtain the minimum bounding box point cloud PC_Box, thereby improving the point cloud extraction efficiency.

[0026] Finally, a binary tree kd-tree is established based on the point cloud PC_Box in the minimum bounding box of the face bias body, a seed point is taken as a point on the theoretical surface, and the nearest point is searched through a region growing algorithm, if the point is in the face bias body, the point is included in the growing point, and the growing point is recursively grown from the growing point until no growing point is found, thereby obtaining the face bias body point cloud PC_Face.

[0027] Step 3: parameter division of the curved surface.

[0028] The theoretical curved surface is divided according to the equal parameters The parameter region is obtained, wherein m is the number of u-directional uniform segmentation, and n is the number of v-directional uniform segmentation. Each region has a center point p, each center point has corresponding uv parameters, and all region center points constitute a parameter curved surface point cloud PC_UV.

[0029] As Figure 6 As shown in the schematic diagram of the parameter curved surface region division of the present application, the parameter curved surface is uniformly divided into 6 segments in the u-direction and 4 segments in the v-direction, and there are 24 regions and region center points. After the parameter curved surface is divided into regions, the data structure Cell is used to save the single region information, including the center point of the region and the parameter range uMin, uMax, vMin and vMax, wherein the region center point is shown as the black point in the figure.

[0030] Step 4: projection initial value query.

[0031] The point cloud extracted in step 2 is used to query the nearest point in the parameter curved surface point cloud divided in step 3.

[0032] Specifically, the parallel technology is used to classify the extracted point cloud according to the curved surface parameter grid Grid_UV, that is, the nearest point of the point cloud sampling point in PC_UV is queried to match the point and the corresponding relationship of the Grid_UV sub-grid, and the uv range start_uv is obtained.

[0033] As Figure 6For the parameter surface region division diagram of the present application, the red dots are sampling points of the point cloud, and the black dot set is the parameter surface point cloud PC_UV. The nearest point of the sampling point is found by establishing a kd-tree based on the parameter surface point cloud PC_UV, and the uv parameter (range start_uv) corresponding to the nearest point (corresponding to the Cell) is used as the initial value of the projection of the point.

[0034] Step 5: Newton projection.

[0035] The projection point of the point cloud to the parameter surface is solved based on the least square method, and the initial value start_uv is obtained from step 4.

[0036] Given the projection point , the parameter surface , the projection point of the projection point to the parameter surface is solved, that is, the nearest point of the projection point to the parameter surface is solved. By establishing a minimum objective function: , the Newton iteration method is used to solve the optimal parameter and the corresponding projection point.

[0037] Step 6: error field color spot map generation.

[0038] The error value is calculated according to the projection point, and the color is picked up to fill the cloud grid model to form an error field color spot map. For example Figure 7 The effect diagram of the present application for fast projection of large-scale point cloud to parameter surface is shown in the figure, in which the projection distance of the point is distinguished by color.

[0039] The embodiment provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the steps of the large-scale point cloud fast projection method to the parameter surface. Wherein, the storage medium can be a disc, an optical disc, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above types of memories.

[0040] The embodiment provides a computer device, which includes a memory, a processor and a computer program stored on the memory and executable on the processor, and the processor executes the program to realize the steps of the large-scale point cloud fast projection method to the parameter surface.

[0041] For example Figure 8As shown, the computer device 120 can include at least one processor 121, such as a central processing unit (CPU), at least one communication interface 123, a memory 124, and at least one communication bus 122. The communication bus 122 is configured to enable communication between these components. The communication interface 123 can include a display, a keyboard, and optionally the communication interface 123 can further include a standard wired interface, a wireless interface. The memory 124 can be a high-speed random access memory (RAM), and can also be a non-volatile memory, such as at least one disk memory. The memory 124 can also be at least one storage device located away from the aforementioned processor 121. The memory 124 stores an application program, and the processor 121 invokes the program code stored in the memory 124 to execute any of the above method steps. The communication bus 122 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The communication bus 122 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8The bus 123 is used to connect the above-mentioned elements in the system 120, and only one bus is represented, but it does not mean that there is only one bus or only one type of bus. Among them, the memory 124 can include volatile memory such as random-access memory (RAM); the memory can also include non-volatile memory such as flash memory, a hard disk drive (HDD) or a solid-state drive (SSD); the memory 124 can also include a combination of the above-mentioned types of memory. Among them, the processor 121 can be a central processing unit (CPU), a network processor (NP) or a combination of CPU and NP. The processor 121 can further include a hardware chip. The hardware chip can be an application-specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof. The PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof. Alternatively, the memory 124 is also used to store program instructions. The processor 121 can invoke the program instructions to implement the large-scale point cloud fast projection method to the parametric surface as in the embodiment.

[0042] The embodiments of the present application are described above in conjunction with the drawings, but the present application is not limited to the specific embodiments described above, and the specific embodiments described above are only illustrative, not restrictive, and those of ordinary skill in the art can make many forms under the inspiration of the present application without departing from the purpose of the present application and the scope protected by the claims, which are all within the protection of the present application.

Claims

1. A method for fast projection of large-scale point clouds onto parametric surfaces, characterized in that, Includes the following steps: S1: Extract point cloud data based on theoretical model and corresponding point cloud mesh model to obtain surface-biased volume point cloud; S2: Parametrically divide the surface of the theoretical model to obtain the parametric surface point cloud; S3: Obtain the initial projection value based on the surface-biased volume point cloud and the parametric surface point cloud; S4: Based on the initial projection values, the projection points from the point to be projected to the parametric surface are obtained using the least squares method.

2. The method for fast projection of large-scale point clouds onto parametric surfaces according to claim 1, characterized in that, The method for fast projection of large-scale point clouds onto parametric surfaces further includes: calculating error values ​​based on the projection points, picking colors to fill the point cloud mesh model based on the error values, and obtaining an error field color patch map.

3. The method for fast projection of large-scale point clouds onto parametric surfaces according to claim 1, characterized in that, Step S1 specifically includes: S11: Obtain the surface deviation body based on the theoretical machining allowance and surface deviation value of the theoretical model, and obtain the minimum bounding box based on the surface deviation body; S12: Based on the point cloud mesh model, use the minimum bounding box to filter out non-corresponding points to obtain the minimum bounding box point cloud; S13: Based on the minimum bounding box point cloud, obtain the surface-biased volume point cloud using the region growing algorithm.

4. The method for fast projection of large-scale point clouds onto parametric surfaces according to claim 3, characterized in that, The surface-biased body includes an upper biased surface and a lower biased surface, with the upper and lower biased surfaces biased by 3 times.

5. The method for fast projection of large-scale point clouds onto parametric surfaces according to claim 3, characterized in that, Step S13 specifically includes: establishing a binary tree based on the minimum bounding box point cloud, using points on the surface of the theoretical model as seed points, using a region growing algorithm to find the nearest point, and if the nearest point is inside the surface eccentricity, then using the nearest point as the growth point, and then recursively growing from the growth point until no new growth point can be found, thus obtaining the surface eccentricity point cloud.

6. The method for fast projection of large-scale point clouds onto parametric surfaces according to claim 1, characterized in that, Step S2 specifically includes: The surface of the theoretical model is divided into multiple parametric regions according to equal parameters to obtain a parametric mesh; The parameter region includes the region center point and the parameter range, and the parameter surface point cloud is obtained based on the parameter region.

7. The method for fast projection of large-scale point clouds onto parametric surfaces according to claim 1, characterized in that, Step S3 specifically includes: establishing a binary tree based on the parametric surface point cloud, finding the closest point between the sampling point in the surface biased volume point cloud and the regional center point of the parametric surface point cloud, and using the parameter range corresponding to the closest point as the initial projection value of the sampling point.

8. The method for fast projection of large-scale point clouds onto parametric surfaces according to claim 1, characterized in that, The objective function minimized by the least squares method is: ， ， in, This represents the objective function that minimizes the distance from the projection point to the surface. Represents a parametric surface; and These represent the two parameters of the basis function of the curve in the surface; D represents the point to be projected; D is the parameter domain.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method for fast projection of large-scale point clouds onto parametric surfaces as described in any one of claims 1-8.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for fast projection of large-scale point clouds onto parametric surfaces as described in any one of claims 1-8.

Images