Method and apparatus for analyzing three dimensional point cloud data

Abstract

To provide an analysis method of three dimensional point group data capable of performing shape analysis of a workpiece with high accuracy by using three dimensional point group data of an arbitrary measurement coordinate system.SOLUTION: The analysis method of the three-dimensional point group data D includes a data analysis step of analyzing the position and the direction of the central axis line Lm of the three-dimensional point group data D representing the shapes of the workpieces W1 and W2 by the processor 51, and a coordinate conversion step of converting the position and the direction of the central axis line Lm of the three-dimensional point group data D into the position and the direction of the reference axis line (Zm-axis) constituting the measurement coordinate system (Xm, Ym, Zm). A data conversion step of generating post-conversion three-dimensional point cloud data DC by performing an equal-magnification affine transformation on three-dimensional point cloud data D in a measurement coordinate system (Xm, Ym, Zm) and a feature amount analysis step of analyzing shape feature amounts F11, F12, F21, and F22 of the post-conversion three-dimensional point cloud data DC using the post-conversion three-dimensional point cloud data DC are executed.SELECTED DRAWING: Figure 13

Description

[Technical Field]

[0001] The present invention relates to a method and apparatus for analyzing three-dimensional point cloud data. [Background technology]

[0002] Known methods for determining the shape of a workpiece include a method using a contact-type measuring instrument, as described in Patent Document 1, and a method using image data acquired by an imaging device, as described in Patent Documents 2 and 3.

[0003] Whether a contact or non-contact method is used, the workpiece is fixed to the reference position of the measuring instrument. For example, the workpiece is fixed to the reference position of the measuring instrument so that the reference axis (e.g., Z-axis) that constitutes the measurement coordinate system coincides with the central axis of the workpiece. By measuring in this state, it is possible to obtain measurement values ​​in the desired measurement coordinate system. By comparing the obtained measurement values ​​with the design values, the shape of the workpiece can be determined. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-159397 [Patent Document 2] Patent Publication No. 2021-189822 [Patent Document 3] Patent No. 4873485 Summary of the Invention [Problem to be solved by the invention]

[0005] However, it is not easy to fix the workpiece at a reference position with high precision so that the reference axis (e.g., Z-axis) constituting the measurement coordinate system coincides with the central axis of the workpiece, and installation errors may occur. In such cases, the measured values ​​will contain errors. If the measured values ​​obtained in this way are used, the shape of the workpiece cannot be grasped with high precision. Therefore, there is a need to be able to grasp the shape of the workpiece with high precision without fixing the workpiece at a reference position with high precision.

[0006] The present invention has been made in consideration of such problems, and aims to provide a method and apparatus for analyzing three-dimensional point cloud data that can perform highly accurate shape analysis of a workpiece using three-dimensional point cloud data in an arbitrary measurement coordinate system. [Means for solving the problem]

[0007] A first aspect of the present invention is a method for executing a program by a processor, comprising: a data acquisition process for acquiring three-dimensional point cloud data representing the shape of a workpiece having a central axis, the data being data in a measurement coordinate system obtained by a measuring device; a data analysis step of analyzing the position and direction of a central axis of the three-dimensional point cloud data in the measurement coordinate system using an analysis algorithm; a data conversion step of generating converted three-dimensional point cloud data by performing an equal-magnification affine transformation on the three-dimensional point cloud data in the measurement coordinate system so that the position and direction of a central axis of the three-dimensional point cloud data coincide with the position and direction of a reference axis constituting the measurement coordinate system; and a feature amount analysis step of analyzing shape feature amounts of the converted three-dimensional point cloud data using the converted three-dimensional point cloud data.

[0008] A second aspect of the present invention is a data acquisition unit that acquires three-dimensional point cloud data representing the shape of a workpiece having a central axis, the data being data in a measurement coordinate system obtained by a measuring instrument; a data analysis unit that analyzes the position and direction of a central axis of the three-dimensional point cloud data in the measurement coordinate system using an analysis algorithm; a data conversion unit that generates transformed three-dimensional point cloud data by performing an equal-magnification affine transformation on the three-dimensional point cloud data in the measurement coordinate system so that the position and direction of a central axis of the three-dimensional point cloud data coincide with the position and direction of a reference axis that constitutes the measurement coordinate system; and a feature amount analysis unit that uses the transformed three-dimensional point cloud data to analyze shape feature amounts of the transformed three-dimensional point cloud data. [Effects of the Invention]

[0009] According to the above aspect, the acquired 3D point cloud data of the workpiece is data in an arbitrary measurement coordinate system. Therefore, there is no need to align the central axis of the workpiece with the reference axis that constitutes the measurement coordinate system in the measuring device. This increases the degree of freedom in measurement, and furthermore, there is no need to consider workpiece mounting errors during measurement. Therefore, measurement becomes easier.

[0010] However, because the reference axis that constitutes the measurement coordinate system for the acquired 3D point cloud data does not coincide with the central axis of the workpiece, it is not easy to analyze the shape features using the 3D point cloud data itself, and the information does not provide information that can be compared with the design features of the workpiece.As a result, it is not possible to grasp the shape of the workpiece as is.

[0011] Therefore, an equal-magnification affine transformation is performed on the three-dimensional point cloud data in the measurement coordinate system so that the position and direction of the central axis of the three-dimensional point cloud data coincide with the position and direction of the reference axis that constitutes the measurement coordinate system.The transformed three-dimensional point cloud data is then used to analyze the shape feature quantities of the transformed three-dimensional point cloud data.In other words, the shape feature quantities of the transformed three-dimensional point cloud data are information that can be compared with the design feature quantities of the workpiece.This makes it possible to perform shape analysis of the workpiece with high accuracy.

[0012] As described above, according to the above aspects, it is possible to provide a method and an apparatus for analyzing three-dimensional point cloud data that can perform highly accurate shape analysis of a workpiece using three-dimensional point cloud data in an arbitrary measurement coordinate system. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. [Figure 2] A diagram showing workpiece W2. [Figure 3] FIG. 1 is a diagram showing the configuration of an analysis system. [Figure 4] FIG. 2 is a diagram showing three-dimensional point cloud data D for a workpiece W1. [Figure 5] FIG. 2 is a diagram showing the hardware configuration of the analysis device. [Figure 6] FIG. 2 is a functional block diagram of the analysis device. [Figure 7] FIG. 2 is a diagram showing a plurality of algorithms stored in an algorithm storage unit in the analysis device. [Figure 8] FIG. 10 is a diagram illustrating an algorithm using the least squares method. [Figure 9] 10A and 10B are diagrams illustrating a method for determining a direction vector of a central axis in an algorithm using principal component analysis. [Figure 10] FIG. 4 is a diagram illustrating a first method for determining a central axis line from a direction vector. [Figure 11] FIG. 10 is a diagram illustrating a second method for determining a central axis line from a direction vector. [Figure 12] 10A and 10B are diagrams illustrating a method for determining a direction vector of a central axis in an algorithm that uses a normal vector. [Figure 13] FIG. 10 is a diagram showing an equal-magnification affine transformation performed by a data conversion unit in the analysis device. [Figure 14] FIG. 10 is a diagram for explaining first feature amounts F11 and F12. [Figure 15] 10 is a graph showing a first feature amount F11. [Figure 16] FIG. 10 is a diagram for explaining a first feature amount F12. [Figure 17] 10 is a graph showing a first feature amount F12. [Figure 18] FIG. 10 is a diagram for explaining a second feature amount F21. [Figure 19] 10 is a graph showing a second feature amount F21. [Figure 20] FIG. 10 is a diagram for explaining a second feature amount F22. [Figure 21] 10 is a graph showing a second feature amount F22. DETAILED DESCRIPTION OF THE INVENTION

[0014] (Embodiment) 1. Target work W1, W2 The workpieces W1 and W2 that are the targets of shape analysis will be described with reference to Figures 1 and 2. The workpieces W1 and W2 are rotating bodies such as bearings and gears. In other words, the workpieces W1 and W2 are rotating bodies having a central axis Ld.

[0015] The workpiece W1 shown in FIG. 1 is the outer ring of a bearing. However, the workpiece W1 may also be the inner ring of a bearing. The workpiece W1 has a central axis Ld. In FIG. 1, the inner peripheral surface of the outer ring, which is the workpiece W1, has a raceway groove 11, a first cylindrical surface 12, and a second cylindrical surface 13. In FIG. 1, the raceway groove 11 has a cross-sectional shape that approximates a circular arc in an axial cross section.

[0016] The workpiece W2 shown in Fig. 2 is a gear having external teeth. However, the workpiece W2 may also be a gear having internal teeth. The workpiece W2 has a central axis Ld. In Fig. 2, the workpiece W2 has a gear portion 21, a first circular cross-sectional portion 22, and a second circular cross-sectional portion 23. The first circular cross-sectional portion 22 and the second circular cross-sectional portion 23 form, for example, cylindrical surfaces.

[0017] The shapes of the workpieces W1 and W2 are defined by values ​​expressed in a design coordinate system (Xd, Yd, Zd). In this embodiment, the central axis Ld of the workpieces W1 and W2 is defined to coincide with the Zd axis, which is the reference axis of the design coordinate system (Xd, Yd, Zd).

[0018] 2. Configuration of analysis system 30 The analysis system 30 analyzes the shape feature quantities of the workpieces W1 and W2 using three-dimensional point cloud data D representing the shapes of the workpieces W1 and W2 obtained by the measuring instrument 31. Furthermore, the analysis system 30 may evaluate the shape feature quantities, which are the measurement results, by comparing the shape feature quantities obtained from the three-dimensional point cloud data D with the design feature quantities of the workpieces W1 and W2.

[0019] The evaluation portion of the workpiece W1, which is the target of the analysis system 30, is a raceway groove of a bearing, etc. The evaluation portion of the workpiece W2 is a tooth profile forming portion of a gear, etc. The following description will be given focusing on the workpiece W1, which is a bearing.

[0020] The analysis system 30 will be described with reference to Fig. 3. The analysis system 30 includes a measuring instrument 31 and an analysis device 32.

[0021] The measuring instrument 31 measures the shape of a measurement target portion of the target workpiece W1, and generates three-dimensional point cloud data D representing the measurement target portion. The generated three-dimensional point cloud data D does not need to be data in a specific measurement coordinate system (Xm, Ym, Zm), and may be data in any measurement coordinate system (Xm, Ym, Zm).

[0022] 3, the measuring instrument 31 includes a stage 41. The measuring instrument 31 generates three-dimensional point cloud data D of inner circumferential surfaces 11, 12, and 13, which are measurement target portions of the workpiece W1 placed on the stage 41, for example.

[0023] Here, it is not necessary to limit the mounting position of the workpiece W1 on the stage 41. In other words, it is not necessary to place the workpiece W1 on the stage 41 so that the central axis L of the workpiece W1 coincides with the Zm axis, which is the reference axis that constitutes the measurement coordinate system (Xm, Ym, Zm). Furthermore, it is not necessary for the normal to the mounting surface of the stage 41 to coincide with the Zm axis.

[0024] The measuring instrument 31 can acquire the three-dimensional point cloud data D of the inner circumferential surfaces 11, 12, and 13 of the bearings by, for example, a grid projection method. The measuring instrument 31 may acquire the three-dimensional point cloud data D of the inner circumferential surfaces 11, 12, and 13 of the bearings by using a contact probe or a non-contact laser displacement meter. The measuring instrument 31 may also acquire one or more image data of a measurement target portion of the workpiece W1 while irradiating the workpiece W1 with white light or light of a predetermined wavelength, and acquire the three-dimensional point cloud data D of the inner circumferential surfaces 11, 12, and 13 of the workpiece W1 based on the image data. Note that the measuring instrument 31 may acquire the three-dimensional point cloud data D of the inner circumferential surfaces 11, 12, and 13 of the bearings by any method other than the above method.

[0025] When the grid projection method is applied, the measuring instrument 31 further includes a projection device 42, an imaging device 43, and a three-dimensional point cloud data generation device 44. The projection device 42 projects a grid pattern onto the workpiece W1. For example, the projection device 42 projects the grid pattern onto the inner circumferential surfaces 11, 12, and 13 of the bearing, which is the measurement target portion. The imaging device 43 captures the diffuse reflection light of the grid pattern projected onto the inner circumferential surfaces 11, 12, and 13 of the bearing, which is the measurement target portion, to acquire image data. The three-dimensional point cloud data generation device 44 generates three-dimensional point cloud data D of the inner circumferential surfaces 11, 12, and 13 of the bearing, which is the measurement target portion, based on the image data acquired by the imaging device 43.

[0026] The analysis device 32 uses the three-dimensional point cloud data D obtained by the measuring instrument 31 to analyze the shape feature quantities of the workpiece W1 represented by the three-dimensional point cloud data D. Furthermore, the analysis device 32 compares the shape feature quantities obtained from the three-dimensional point cloud data D with the design feature quantities of the workpiece W1 to evaluate the shape feature quantities, which are the measurement results.

[0027] In addition, when the target workpiece W2 is a gear, the same process is essentially performed as when it is a bearing. In this case, too, the shape feature values ​​of the gear tooth profile forming portion can be analyzed and compared with the design feature values.

[0028] 3. 3D point cloud data D The three-dimensional point cloud data D of the workpiece W1 obtained by the measuring device 31 will be described with reference to Fig. 4. Fig. 4 shows a portion of the inner peripheral surfaces 11, 12, and 13 of the bearing, which is the workpiece W1. In reality, the three-dimensional point cloud data D includes data covering the entire circumferential direction.

[0029] As shown in FIG. 4, the three-dimensional point cloud data D has a portion corresponding to the raceway groove 11, a portion corresponding to the first cylindrical surface 12, and a portion corresponding to the second cylindrical surface 13.

[0030] As described above, the workpiece W1 has a central axis Ld. Therefore, the three-dimensional point cloud data D has a central axis Lm. However, a method for acquiring the central axis Lm of the three-dimensional point cloud data D will be described later. In FIG. 4, the two-dot chain line is a virtual line representing the contour of the shape.

[0031] Furthermore, the three-dimensional point cloud data D is defined by values ​​expressed in an arbitrary measurement coordinate system (Xm, Ym, Zm). The measurement coordinate system (Xm, Ym, Zm) does not need to be specified in terms of its positional relationship with the workpiece W1. For example, the reference axis (e.g., Zm axis) constituting the measurement coordinate system (Xm, Ym, Zm) does not need to coincide with the central axis Lm of the three-dimensional point cloud data D.

[0032] 4. Hardware configuration of the analysis device 32 The hardware configuration of the analysis device 32 will be described with reference to Fig. 5. The analysis device 32 is configured by a known computer. For example, the analysis device 32 is configured by a processor 51, a memory 52, a storage device 53, an input / output device 54, a communication interface 55, etc.

[0033] 5. Functional configuration of the analysis device 32 The functional configuration of the analysis device 32 and the processing method performed by the analysis device 32 will be described with reference to Fig. 6. As shown in Fig. 6, the analysis device 32 includes a point cloud data acquisition unit 61, a data analysis unit 62, an algorithm storage unit 63, a data conversion unit 64, a feature amount analysis unit 65, a design feature amount acquisition unit 66, and a comparison unit 67. The elements 61 to 62 and 64 to 67 are configured by the processor 51 in Fig. 5. The element 63 is configured by the storage device 53 in Fig. 5.

[0034] 6, steps in parentheses in each block indicate steps executed by the processor 51 in each of the elements 61-62 and 64-67. In other words, the analysis method by the analysis device 32 corresponds to the execution of each step in each of the elements 61-62 and 64-67.

[0035] 5-1. Point cloud data acquisition unit 61 A point cloud data acquisition unit 61 shown in Fig. 6 executes a process (data acquisition process) of acquiring three-dimensional point cloud data D of the inner peripheral surfaces 11, 12, and 13 of the bearing, which is the workpiece W1, using the processor 51 shown in Fig. 5. The point cloud data acquisition unit 61 acquires the three-dimensional point cloud data D generated by the three-dimensional point cloud data generation device 44 shown in Fig. 3. The three-dimensional point cloud data D is data of an arbitrary measurement coordinate system (Xm, Ym, Zm), as shown in Fig. 4. In other words, the central axis Lm of the three-dimensional point cloud data D does not coincide with the Zm axis, which is the reference axis that constitutes the measurement coordinate system (Xm, Ym, Zm).

[0036] 5-2. Data Analysis Section 62 The data analysis unit 62 shown in FIG. 6 executes a step (data analysis step) of analyzing the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm) acquired by the point cloud data acquisition unit 61 using an analysis algorithm via the processor 51 shown in FIG. 5. As shown in FIG. 4, the data analysis unit 62 analyzes the position and direction of the central axis Lm of the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm). The analysis of the position and direction of the central axis Lm is performed according to the analysis algorithm. A plurality of analysis algorithms are stored in the algorithm storage unit 63, and the data analysis unit 62 can select any one of them. The data analysis unit 62 can also select the optimal analysis algorithm depending on the type of data.

[0037] 5-3. Algorithm storage unit 63 The algorithm storage unit 63 shown in FIG. 6 stores a plurality of algorithms 71 to 75 for analyzing the central axis line Lm in the three-dimensional point cloud data D, as shown in FIG.

[0038] The algorithm storage unit 63 stores, for example, an algorithm 71 using the least squares method, a first algorithm 72 using principal component analysis, a second algorithm 73 using principal component analysis, a first algorithm 74 using normal vectors, and a second algorithm 75 using normal vectors. However, it is sufficient that at least one selected from the algorithms 71 to 75 is stored in the algorithm storage unit 63.

[0039] The algorithm 71 using the least squares method will be described with reference to Fig. 8. The algorithm 71 using the least squares method generates an approximate cylindrical shape by three-dimensional approximation for, for example, a cylindrical portion (first cylindrical surface 12 or second cylindrical surface 13) using a portion of the three-dimensional point cloud data D. Next, the algorithm 71 using the least squares method determines the central axis of the generated approximate cylindrical shape as the central axis line Lm of the three-dimensional point cloud data D.

[0040] The first algorithm 72 using principal component analysis will be described with reference to Fig. 9 and Fig. 10. As shown in Fig. 9, the first algorithm 72 using principal component analysis performs principal component analysis on the three-dimensional point cloud data D to determine one of the first principal component L1, the second principal component L2, and the third principal component L3 as the direction vector VL of the central axis Lm. In Fig. 9, the first principal component L1 is illustrated as the direction vector VL of the central axis Lm.

[0041] 10, a first algorithm 72 using principal component analysis determines, from among the straight lines having the direction vector VL, a straight line whose distance from each point of the three-dimensional point cloud data D is close to a certain value as the central axis Lm. For example, an optimization calculation can be used. For example, the point (position) through which the central axis Lm passes is used as a variable, and the standard deviation of the distance between each point and the central axis Lm is used as an objective function, and the position of the central axis Lm when the standard deviation, which is the objective function, is minimized is determined.

[0042] The second algorithm 73 using principal component analysis will be described with reference to Fig. 9 and Fig. 11. As shown in Fig. 9, the second algorithm 73 using principal component analysis determines the direction vector VL of the central axis Lm, similar to the first algorithm 72 using principal component analysis.

[0043] 11, the central axis of the approximate arc shape of the point group D1 located on the plane D2 obtained by projecting each point of the three-dimensional point group data D in the direction of the direction vector VL is determined as the central axis line Lm. In other words, the axis that passes through the center point of the approximate arc shape and has the direction vector VL is determined as the central axis line Lm.

[0044] The first algorithm 74 using normal vectors will be described with reference to FIGS. 10 and 12. As shown in FIG. 12, the first algorithm 74 using normal vectors uses a portion of the three-dimensional point cloud data D to generate, for example, a mesh region formed by multiple neighboring points in the cylindrical portion (first cylindrical surface 12 or second cylindrical surface 13) and a normal vector D3 at the points included in the mesh region for each point in the cylindrical portion (12, 13) of the three-dimensional point cloud data D. Here, the mesh region can be, for example, a region formed by three or more points positioned so as to surround a point of interest. The mesh region is assumed to be a plane. The normal vector D3 is assumed to be a vector that has a normal direction to the mesh region and passes through the point of interest.

[0045] Next, a first algorithm 74 using normal vectors determines a direction vector orthogonal to the normal vector D3 at each point of the cylindrical portion (12, 13) of the three-dimensional point cloud data D as the direction vector VL of the central axis.

[0046] 10, a first algorithm 74 using a normal vector determines, as the central axis line Lm, a straight line having a direction vector VL whose distance from each point of the cylindrical portion (12, 13) of the three-dimensional point cloud data D is close to a certain value, from among the straight lines. This is the same processing as the first algorithm 72 using the principal component analysis described above.

[0047] The second algorithm 75 using normal vectors will be described with reference to Fig. 11 and Fig. 12. As shown in Fig. 12, the second algorithm 75 using normal vectors determines the direction vector VL of the central axis Lm, similar to the first algorithm 74 using normal vectors. Subsequently, as shown in Fig. 11, the second algorithm 75 using normal vectors determines the central axis Lm, similar to the second algorithm 73 using principal component analysis.

[0048] 5-4. Data conversion unit 64 The data conversion unit 64 shown in Figure 6 executes a process (data conversion process) of generating transformed three-dimensional point cloud data DC by performing an equal-magnification affine transformation on the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm) using the processor 51 shown in Figure 5.

[0049] As shown in Fig. 4, the central axis Lm of the three-dimensional point cloud data D does not coincide with the Zm axis, which is the reference axis that constitutes the measurement coordinate system (Xm, Ym, Zm). Therefore, as shown in Fig. 13, the data conversion unit 64 performs an equal-magnification affine transformation so that the position and direction of the central axis Lm of the three-dimensional point cloud data D coincide with the position and direction of the Zm axis, which is the reference axis that constitutes the measurement coordinate system (Xm, Ym, Zm).

[0050] The equal-magnification affine transformation is performed using equation (1). That is, the equal-magnification affine transformation performs rotational coordinate transformation and translation while maintaining a constant magnification. Each point (x, y, z) in the three-dimensional point cloud data D is transformed into each point (x', y', z') in the transformed three-dimensional point cloud data DC.

[0051]

number

[0052] In this embodiment, the data conversion unit 64 generates the transformed three-dimensional point cloud data DC by performing an equal-magnification affine transformation on all of the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm). In other words, the number of data points in the three-dimensional point cloud data D and the transformed three-dimensional point cloud data DC are the same. However, depending on the shape of the workpiece W1, the transformed three-dimensional point cloud data DC may be generated by extracting only a portion of the three-dimensional point cloud data D and performing an equal-magnification affine transformation on only that portion.

[0053] Although the reference axis constituting the measurement coordinate system (Xm, Ym, Zm) is the Zm axis in the above example, it may be the Xm axis or the Ym axis. Also, the reference axis may be a specific axis different from the Xm axis, Ym axis, and Zm axis, for example, an axis parallel to the Zm axis.

[0054] 5-5. Feature Analysis Unit 65 The feature analysis unit 65 shown in Figure 6 executes a process (feature analysis process) of analyzing the shape features of the converted three-dimensional point cloud data DC using the converted three-dimensional point cloud data DC by the processor 51 shown in Figure 5.

[0055] The shape feature amount is a dimension for evaluating the shape of the workpiece W1, and different indices may be set for each workpiece W1. The shape feature amount may be first feature amounts F11, F12 generated using each of a plurality of first point cloud data groups DC1 obtained by dividing the converted three-dimensional point cloud data DC in the direction of the central axis Lm.

[0056] As an example of the first feature value F11, as shown in FIG. 14, when circle fitting is performed on each of a plurality of first point cloud data groups DC1 located at each position on the Zm axis, the position P1 of the center point of the fitting circle is taken as the position. In other words, the first feature value F11 is the position P1 of the center point of the raceway groove 11 of the bearing, which is the workpiece W1, at each position on the Zm axis. The position P1 is expressed by the Xm coordinate and the Ym coordinate. As shown in FIG. 15, the first feature value F11 is expressed by the Zm position and the position (Xm coordinate, Ym coordinate) of the center point P1. Here, the design feature value of the workpiece W1 corresponding to the first feature value F11 is the position P0. In other words, the Xm coordinate and the Ym coordinate of the design feature value Fa11 are (0, 0).

[0057] Another example of the first feature value F12 is the average value Rave of the radii R of multiple data points in each of multiple first point cloud data groups DC1 located at each position on the Zm axis, as shown in Figures 14 and 16. That is, the first feature value F12 is the average radius Rave at each position on the Zm axis. As shown in Figure 17, the first feature value F12 is expressed by the Zm position and the average radius Rave. Here, the design feature value of the workpiece W1 corresponding to the first feature value F12 is the dimension of the axial cross-sectional shape of the design shape.

[0058] The shape feature may be second feature F21, F22 generated using each of a plurality of second point cloud data groups DC2 obtained by dividing the converted three-dimensional point cloud data DC at predetermined angles in the circumferential direction.

[0059] As an example of the second feature quantity F21, as shown in FIG. 18, the maximum diameter Dmax of the raceway groove 11 is taken for each of a plurality of second point cloud data groups DC2 divided at predetermined angles in the circumferential direction. As shown in FIG. 18, the arc center P3 of the raceway groove 11 at a circumferential position, for example, 30°, is calculated, and a radius R3 of the raceway groove 11 centered on P3 is calculated. Furthermore, the arc center P4 of the raceway groove 11 at a circumferential position symmetrical by 180°, for example, 210°, is calculated, and a radius R4 of the raceway groove 11 centered on P4 is calculated. The distance ΔP3-P4 between positions P3 and P4 is calculated. Then, at the circumferential positions between 30° and 210°, the maximum diameter Dmax of the raceway groove 11 is taken as the sum of the radii R3, R4, and the distance ΔP3-P4. Similarly, the maximum diameter Dmax of the raceway groove 11 at other circumferential positions is calculated in the same manner.

[0060] The second feature amount F21 is represented by the circumferential position and the maximum diameter Dmax, as shown in Fig. 19. Here, the design feature amount of the workpiece W1 corresponding to the second feature amount F21 is a constant value Dst, as shown in Fig. 19.

[0061] Another example of the second feature amount F22 is the axial position Zm1 of the maximum radius portion DCmax of the raceway groove 11 in each of a plurality of second point cloud data groups DC2 divided at predetermined angles in the circumferential direction, as shown in Fig. 20. As shown in Fig. 20, the maximum radius portion DCmax of the raceway groove 11 is the deepest position of the raceway groove 11. In other words, the second feature amount F22 is the Zm coordinate value Zm1 of the point P5 at the maximum radius portion DCmax, which is the deepest position of the raceway groove 11.

[0062] The second feature amount F22 is expressed by the circumferential position and the Zm position, as shown in Fig. 21. Here, the design feature amount of the workpiece W1 corresponding to the second feature amount F22 is a constant value Zmst, as shown in Fig. 21.

[0063] 5-6. Design feature acquisition unit 66 The design feature acquisition unit 66 shown in Fig. 6 executes a step of acquiring design feature values ​​of the workpiece W1 (design feature acquisition step) using the processor 51 shown in Fig. 5. The design feature values ​​correspond to the design values ​​of the workpiece W1. As shown in Figs. 14 to 21, the design feature values ​​are feature values ​​corresponding to the shape feature values ​​F11, F12, F21, and F22.

[0064] The design feature quantity acquisition unit 66 may acquire design feature quantities stored in advance in the storage device 53 (shown in FIG. 5). In addition, the design feature quantity acquisition unit 66 can also apply the following two means.

[0065] First, the design feature quantity acquisition unit 66 may acquire the design feature quantities by acquiring CAD data including the design shape and design feature quantities of the workpiece W1. Second, the design feature quantity acquisition unit 66 may acquire the design feature quantities by acquiring shape measurement values ​​of a master workpiece for the workpiece W1. In this case, for example, it is preferable to use shape measurement values ​​of the master workpiece obtained using a high-precision measuring instrument.

[0066] 5-7. Comparison section 67 The comparison unit 67 shown in Fig. 6 executes a step (comparison step) of comparing the shape features F11, F12, F21, and F22 of the converted three-dimensional point cloud data DC with the design features using the processor 51 shown in Fig. 5. For example, the comparison unit 67 calculates an index indicating the difference between the shape features F11, F12, F21, and F22 and the design features, and outputs the index. Furthermore, if a tolerance for the difference is set, the comparison unit 67 may output information indicating whether the difference is within the tolerance range or outside the tolerance range.

[0067] 6.Effects The analysis system 30 of this embodiment includes a data acquisition process (a process using element 61) in which a processor 51 acquires three-dimensional point cloud data D representing the shape of a workpiece W1 having a central axis L, the data being data in a measurement coordinate system (Xm, Ym, Zm) obtained by a measuring instrument 31; a data analysis process (a process using element 62) in which analysis algorithms 71 to 75 are used to analyze the position and direction of the central axis Lm of the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm); The method executes a data conversion process (a process using element 64) for generating converted three-dimensional point cloud data DC by performing an equal-magnification affine transformation on the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm) so that the position and direction of the three-dimensional point cloud data D coincide with the position and direction of the reference axis (Zm axis) that constitutes the measurement coordinate system (Xm, Ym, Zm), and a feature analysis process (a process using element 65) for analyzing the shape features F11, F12, F21, and F22 of the converted three-dimensional point cloud data DC using the converted three-dimensional point cloud data DC.

[0068] As described above, the acquired three-dimensional point cloud data D of the workpiece W1 is data in an arbitrary measurement coordinate system (Xm, Ym, Zm). Therefore, it is not necessary to align the central axis L of the workpiece W1 with the reference axis (Zm axis) that constitutes the measurement coordinate system (Xm, Ym, Zm) in the measuring device 31. This increases the degree of freedom in measurement, and furthermore, there is no need to consider installation errors of the workpiece W1 during measurement. This makes measurement easier.

[0069] However, the reference axis (Zm axis) that constitutes the measurement coordinate system (Xm, Ym, Zm) for the acquired three-dimensional point cloud data D does not coincide with the central axis L of the workpiece W1, so it is not easy to analyze the shape features F11, F12, F21, and F22 using the three-dimensional point cloud data D itself, and the information does not become information that can be compared with the design features of the workpiece W1. Therefore, as it is, it is not possible to grasp the shape of the workpiece W1.

[0070] Therefore, an equal-magnification affine transformation is performed on the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm) so that the position and direction of the central axis Lm of the three-dimensional point cloud data D coincide with the position and direction of the reference axis (Zm axis) constituting the measurement coordinate system (Xm, Ym, Zm).The transformed three-dimensional point cloud data DC is then used to analyze the shape feature quantities F11, F12, F21, and F22 of the transformed three-dimensional point cloud data DC.In other words, the shape feature quantities F11, F12, F21, and F22 of the transformed three-dimensional point cloud data DC are information that can be compared with the design feature quantities of the workpiece W1.This allows for highly accurate shape analysis of the workpiece W1.

[0071] Therefore, according to the analysis system 30, it is possible to perform a shape analysis of the workpiece W1 with high accuracy using the three-dimensional point cloud data D in an arbitrary measurement coordinate system.

[0072] In addition, the analysis system 30 uses the processor 51 to execute a design feature acquisition process (a process using element 66) for acquiring the design features of the work W1, and a process (a process using element 67) for comparing the shape features F11, F12, F21, and F22 of the transformed three-dimensional point cloud data DC with the design features.

[0073] As a result, by comparing the shape feature values ​​obtained by measurement with the design feature values, the shape of the workpiece W1 obtained by measurement can be properly grasped.

[0074] Furthermore, in the design feature acquisition process (process by element 66), the design feature can be acquired by acquiring CAD data including the design shape and design feature of the workpiece W1. This makes it possible to easily acquire the design feature.

[0075] Furthermore, in the design feature acquisition process (process by element 66), design feature values ​​can be acquired by acquiring shape measurement values ​​of a master work for the workpiece W1. This allows comparison to be made with the master work as a reference.

[0076] The analysis algorithm can be at least one selected from the group consisting of an algorithm 71 using the least squares method, algorithms 72 and 73 using principal component analysis, and algorithms 74 and 75 using normal vectors. By using these analysis algorithms, the central axis Lm of the three-dimensional point cloud data D can be obtained with high accuracy.

[0077] Furthermore, in the data conversion process (process by element 64), converted three-dimensional point cloud data DC can be generated by performing equal-magnification affine transformation on all of the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm). This makes it possible to obtain data that appropriately represents all shapes acquired by the measuring instrument 31 as the converted three-dimensional point cloud data DC.

[0078] In this case, in the data analysis step (step by element 62), analysis algorithms 71 to 75 may be used to analyze the position and direction of the central axis Lm of the three-dimensional point cloud data D using a portion of the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm). That is, in the data analysis step, a portion of the three-dimensional point cloud data D is used, whereas in the data conversion step, all of the three-dimensional point cloud data D is converted. In the data analysis step, all of the three-dimensional point cloud data D may be used, and in the data conversion step, all of the three-dimensional point cloud data D may also be converted. In the analysis of the central axis Lm, it is advisable to use either all of the three-dimensional point cloud data D or a portion of the three-dimensional point cloud data D, depending on the target workpiece W1.

[0079] In the data conversion process (process by element 64), the converted three-dimensional point cloud data DC may be generated by performing equal-magnification affine transformation on only a portion of the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm). By extracting only the necessary parts, the data processing time can be shortened and the data volume can be reduced.

[0080] The shape feature amounts can be first feature amounts F11, F12 generated using each of a plurality of first point cloud data groups DC1 obtained by dividing the converted three-dimensional point cloud data DC in the direction of the central axis Lm. For example, the workpiece W1 is a bearing having a raceway groove 11, and the first feature amounts F11, F12 can be at least one of the position of the center point P1 of the fitting circle of the raceway groove 11 and the average radius Rave of the plurality of data points that make up the raceway groove 11. This makes it possible to grasp the shape of the bearing with high accuracy.

[0081] The shape feature values ​​can be second feature values ​​F21, F22 generated using each of a plurality of second point cloud data groups DC2 obtained by dividing the converted three-dimensional point cloud data DC at predetermined angles in the circumferential direction. For example, the workpiece W1 is a bearing having a raceway groove 11, and the second feature values ​​F21, F22 can be at least one of the maximum diameter Dmax of the raceway groove 11 and the axial position Zm1 of the maximum radius portion DCmax of the raceway groove 11. This makes it possible to grasp the shape of the bearing with high accuracy.

[0082] 7. Work W2 In the case of a gear as the workpiece W2, as shown in FIG. 2, the workpiece W2 has circular cross-sectional portions 22 and 23 whose cross-sectional shape perpendicular to the axis is circular. In this case, the data analysis step (step using element 62) may analyze the position and direction of the central axis Lm of the circular cross-sectional portions 22 and 23 in the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm) using analysis algorithms 71 to 75. Furthermore, in the data conversion step (step using element 64), the transformed three-dimensional point cloud data DC may be generated by performing an equal-magnification affine transformation on all of the three-dimensional point cloud data D in the measurement coordinate system (Xm, Ym, Zm) so that the position and direction of the central axis Lm of the circular cross-sectional portions 22 and 23 in the three-dimensional point cloud data D coincides with the position and direction of the reference axis (Zm axis) constituting the measurement coordinate system (Xm, Ym, Zm).

[0083] In particular, the workpiece W2 is a gear having circular cross-sectional portions 22 and 23 and a gear portion 21, and the shape feature quantity is preferably at least one of the gear specifications. This allows the shape of the gear portion 21 of the gear to be grasped with high accuracy. [Explanation of symbols]

[0084] 30 Analysis System 31 Measuring instruments 32 Analyzer 51 processors 61 Data Acquisition Department (Data Acquisition Process) 62 Data Analysis Department (Data Analysis Process) 64 Data Conversion Department (Data Conversion Process) 65 Feature analysis unit (feature analysis process) 71-75 Analysis Algorithm W1, W2 work D 3D point cloud data DC converted 3D point cloud data L, Ld, Lm center axis line Zm Reference axis F11, F12, F21, F22 shape features (Xm,Ym,Zm) Measurement coordinate system

Claims

1. The processor a data acquisition process for acquiring three-dimensional point cloud data representing the shape of a workpiece having a central axis, the data being data in a measurement coordinate system obtained by a measuring device; a data analysis step of analyzing the position and direction of a central axis of the three-dimensional point cloud data in the measurement coordinate system using an analysis algorithm; a data conversion step of generating converted three-dimensional point cloud data by performing an equal-magnification affine transformation on the three-dimensional point cloud data in the measurement coordinate system so that the position and direction of a central axis of the three-dimensional point cloud data coincide with the position and direction of a reference axis constituting the measurement coordinate system; a feature amount analysis step of analyzing shape feature amounts of the converted three-dimensional point cloud data using the converted three-dimensional point cloud data.

2. Further, by the processor: a design feature acquisition step of acquiring a design feature of the workpiece; The three-dimensional point cloud data analysis method according to claim 1 , further comprising: a comparison step of comparing the shape feature quantities of the converted three-dimensional point cloud data with the design feature quantities.

3. The three-dimensional point cloud data analysis method according to claim 2 , wherein the design feature quantity acquisition step acquires the design feature quantities by acquiring CAD data including the design shape of the workpiece and the design feature quantities.

4. The three-dimensional point cloud data analysis method according to claim 2 , wherein the design feature quantity acquisition step acquires the design feature quantities by acquiring shape measurement values ​​of a master work for the workpiece.

5. The method for analyzing three-dimensional point cloud data according to any one of claims 1 to 4, wherein the analysis algorithm is at least one selected from an algorithm using the least squares method, an algorithm using principal component analysis, and an algorithm using normal vectors.

6. The method for analyzing three-dimensional point cloud data according to any one of claims 1 to 4, wherein the data conversion step generates the transformed three-dimensional point cloud data by performing an equal-magnification affine transformation on all of the three-dimensional point cloud data in the measurement coordinate system.

7. The workpiece has a circular cross-sectional portion whose cross-sectional shape perpendicular to the axis is circular, In the data analysis step, the position and direction of a central axis of the circular cross-sectional portion in the three-dimensional point cloud data in the measurement coordinate system are analyzed using the analysis algorithm; 7. The method for analyzing three-dimensional point cloud data according to claim 6, wherein the data conversion step generates the transformed three-dimensional point cloud data by performing an equal-magnification affine transformation on all of the three-dimensional point cloud data in the measurement coordinate system so that the position and direction of a central axis of the circular cross-sectional portion of the three-dimensional point cloud data coincides with the position and direction of the reference axis that constitutes the measurement coordinate system.

8. A method for analyzing three-dimensional point cloud data according to any one of claims 1 to 4, wherein the shape feature is a first feature generated using each of a plurality of first point cloud data groups obtained by dividing the transformed three-dimensional point cloud data in the central axis direction.

9. the workpiece is a bearing having a raceway groove, 9. The method for analyzing three-dimensional point cloud data according to claim 8, wherein the first feature amount is at least one of a position of a center point of a fitting circle of the raceway groove and an average radius of a plurality of data points that form the raceway groove.

10. The method for analyzing three-dimensional point cloud data according to any one of claims 1 to 4, wherein the shape feature is a second feature generated using each of a plurality of second point cloud data groups obtained by dividing the transformed three-dimensional point cloud data at predetermined angles in the circumferential direction.

11. the workpiece is a bearing having a raceway groove, 11. The method for analyzing three-dimensional point cloud data according to claim 10, wherein the second feature amount is at least one of a maximum diameter of the raceway groove and an axial position of a maximum radius portion of the raceway groove.

12. the workpiece is a gear having the circular cross-sectional portion and a gear portion, The method for analyzing three-dimensional point cloud data according to claim 7 , wherein the shape feature amount is at least one of specifications of the gear.

13. a data acquisition unit that acquires three-dimensional point cloud data representing the shape of a workpiece having a central axis, the data being data in a measurement coordinate system obtained by a measuring instrument; a data analysis unit that analyzes the position and direction of a central axis of the three-dimensional point cloud data in the measurement coordinate system using an analysis algorithm; a data conversion unit that generates transformed three-dimensional point cloud data by performing an equal-magnification affine transformation on the three-dimensional point cloud data in the measurement coordinate system so that the position and direction of a central axis of the three-dimensional point cloud data coincide with the position and direction of a reference axis that constitutes the measurement coordinate system; a feature amount analysis unit that uses the transformed three-dimensional point cloud data to analyze shape feature amounts of the transformed three-dimensional point cloud data.

Images