JP2005208868A - Method and device for comparing point group data, device for acquiring point group data, and system for comparing point group data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily remove noises from point group data, concerning point group data comparison for comparing by collating point group data acquired for the shape of an object with design shape data being the origin of the object. <P>SOLUTION: Concerning a point group data comparison method which acquires point group data about the shape of an object, and compares by collating the point group data with design shape data being the origin of the object, polygon data is created from the point group data by a polygonization processing part 17, and noises are removed form the polygon data by a polygon noise removal processing part 18. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention acquires point cloud data about the shape of an object such as an industrial product produced based on a design shape made by a computer-aided design system such as CAD, and collates this point cloud data with the design shape data. It relates to the comparison of point cloud data to be compared.

  For example, in a cast product, the shape may differ from the mold due to deformation due to sink marks or the like. In such a case, point cloud data (three-dimensional point cloud data) is acquired from the actual cast product, and this point cloud data is compared with the design shape data (for example, three-dimensional CAD data) of the mold for comparison. Thus, an error relative to the design value of the actual product can be measured.

  In such point cloud data comparison, first, point cloud data is acquired for the shape (three-dimensional shape) of an object (such as a cast product). This point cloud data is acquired using a point cloud data acquisition device, or is obtained by obtaining point cloud data from an image of an object. Then, after removing noise from the acquired point cloud data, the point cloud data is synthesized. This synthesis of the point cloud data is performed when the point cloud data is acquired in units of planes obtained by dividing each plane in the shape of the object or each plane by a certain area. This is a process of combining unit faces. Then, after the point cloud data and the design shape data are aligned, the point cloud data is collated and compared with the design shape data. In this comparison / comparison, the process of visually comparing the point cloud data overlaid on the screen of the display device with the design shape data, and obtaining the dimension value of the required part from the point cloud data and comparing it with the design value. It is normal to perform processing to do.

  For alignment of point cloud data and design shape data in such point cloud data comparison, as seen in Patent Document 1, an inertia main axis that is one of the feature quantities of an object is used, and an inertia main axis in point cloud data is used. There is known a method of aligning point cloud data and CAD data by superimposing the principal axis of inertia in CAD data. A method disclosed in Patent Document 2 is also known. In this patent document 2, target surface designation means for associating a reference CAD data alignment surface with a target point group corresponding to a non-contact measurement point data alignment surface, arbitrarily extracted from the target point group Approximate alignment means for performing approximate alignment based on information on the point group to be extracted consisting of points and information on the point on the surface that is an arbitrary point on the alignment plane; There has been disclosed a non-contact measurement point data alignment apparatus provided with detailed alignment means for performing alignment based on information on alignment point groups that have been aligned and information on alignment surfaces. Regarding the point cloud data comparison, examples disclosed in, for example, Patent Documents 3 to 5 are known in addition to these.

JP-A-9-204532 JP 2001-82951 A JP 2001-243500 A JP 2002-7485 A JP 2003-123057 A

  As described above, in the point cloud data comparison, noise removal from the point cloud data acquired for the target object, synthesis of point cloud data in units of planes, alignment of point cloud data and design shape data, point cloud data and design shape A series of processing called data comparison is required. Each of these processes is an operation that places a heavy burden on the operator. For this reason, it is required to reduce the burden in each of these processes.

  As described above, many proposals have already been made on the alignment between the point cloud data and the design shape data, as in the examples known from Patent Document 1 and Patent Document 2, and improvements are progressing accordingly. However, noise removal, synthesis of point cloud data, and comparison between point cloud data and design shape data are still processing that places a heavy burden on the operator.

  Accordingly, an object of the present invention is to provide a point cloud data comparison method and apparatus capable of more easily removing noise from point cloud data, and to compare point cloud data with design shape data. Point cloud data comparison method and apparatus for enabling point cloud data acquisition that enables easy acquisition of point cloud data that can be combined more easily Another object is to provide a point cloud data comparison system that combines these devices.

  To achieve the above object, the present invention provides a point cloud data comparison method for acquiring point cloud data for the shape of an object and comparing the point cloud data against the design shape data based on the object. The method includes the step of creating polygon data from the point cloud data and removing noise from the polygon data.

  Further, in the present invention, for the purpose described above, a point cloud data comparison method for acquiring point cloud data for the shape of an object and comparing the point cloud data with the design shape data based on the object. In comparing the point cloud data and the design shape data with respect to the dimensions of an arbitrary location, the size comparison location can be designated on the design shape data or the point cloud data. Yes.

  Further, in the present invention, for the above purpose, in the point cloud data comparison device for comparing and comparing the point cloud data acquired for the shape of the object with the design shape data based on the object, the point cloud A polygonization processing unit for converting data into polygon data, and a polygon noise removal processing unit for performing noise removal processing on polygon data obtained by the processing in the polygonization processing unit are provided.

  Further, in the present invention, for the above purpose, in the point cloud data comparison device for comparing and comparing the point cloud data acquired for the shape of the object with the design shape data based on the object, the point cloud A comparison processing unit for comparing the data and the design shape data with respect to dimensions at an arbitrary position, and the comparison processing unit is configured to be able to designate a size comparison position on the design shape data or the point cloud data. It is characterized by having.

  Further, in the present invention, for the above purpose, point cloud data is acquired for the shape of the object, and the point cloud data is compared with the design shape data that is the basis of the object. The point cloud data acquisition apparatus for acquiring the point cloud data to be used includes a sensor for measuring the surface shape of the object as a collection of a large number of points, and an X-axis direction moving mechanism, a Y-axis direction moving mechanism, Z An axial movement mechanism, a θ-axis rotation mechanism, and a φ-axis rotation mechanism, and the sensor is configured such that X is orthogonal to each other by the X-axis direction movement mechanism, the Y-axis direction movement mechanism, and the Z-axis direction movement mechanism. It can be arbitrarily moved in three directions, Y and Z, and the measuring axis can be arbitrarily rotated in the vertical direction by the θ-axis rotating mechanism, and the object can be moved horizontally by the φ-axis rotating mechanism. The X-axis direction moving mechanism, the Y-axis direction moving mechanism, the Z-axis direction moving mechanism, the θ-axis rotating mechanism, and the φ-axis rotating mechanism are An operation amount detection means for detecting the operation amount is provided.

  In the present invention, for the purpose described above, the point cloud data comparison system is configured by combining the above point cloud data comparison device and the point cloud data acquisition device.

  In the point cloud data comparison method or apparatus of the present invention, polygon data is created from the point cloud data, and noise is removed from the polygon data. Therefore, according to the present invention, it is easy to set a standard for automatic noise removal, and in manual noise removal necessary for obtaining highly accurate point cloud data, the operation is performed by using polygon data. The noise can be more easily determined by the operator, and the operator can easily confirm the noise removal result.

  In the point cloud data comparison method or apparatus of the present invention, the point comparison data can be designated on the design shape data or the point cloud data in order to compare the point cloud data and the design shape data with respect to the dimensions of arbitrary locations. ing. For this reason, according to the present invention, it becomes possible to more appropriately compare the dimensions, and it becomes possible to more easily compare the point cloud data and the design shape data.

  In the point cloud data acquisition device of the present invention, a sensor for measuring the surface shape of an object can be moved by each moving mechanism in three directions of X, Y, and Z, and the measuring axis of the sensor is a rotating mechanism. Thus, the object can be arbitrarily rotated in the vertical direction, the object can be arbitrarily rotated in the horizontal direction by the rotating mechanism, and the operation amount of each mechanism is detected by the operation amount detecting means. For this reason, according to the present invention, it becomes possible to automate the synthesis of point cloud data (or polygon data) by using the operation amount information of each mechanism, compared with the conventional method in which the synthesis is performed manually. Therefore, workability can be greatly improved.

  Hereinafter, modes for carrying out the present invention will be described. FIG. 1 shows a configuration of a point cloud data acquisition apparatus according to an embodiment. This point cloud data acquisition device 1 includes an X-axis direction moving mechanism 2, a Y-axis direction moving mechanism 3, a Z-axis direction moving mechanism 4, a θ-axis rotating mechanism 5, a φ-axis rotating mechanism 6, a driver 7, a sensor 8, and a control. A device 9 is provided.

  The sensor 8 is a means for measuring the surface shape of the target object M from which point cloud data is acquired as a collection of a large number of points. For example, a laser measurement method or a moire interference method is used. The sensor 8 can be arbitrarily moved in three directions of X, Y, and Z orthogonal to each other by a combination of the X-axis direction moving mechanism 2, the Y-axis direction moving mechanism 3, and the Z-axis direction moving mechanism 4. The θ axis rotation mechanism 5 can arbitrarily rotate the measurement axis (line of sight) in the vertical direction. On the other hand, the object (object to be measured) M is supported so as to be arbitrarily rotatable in the horizontal direction by a φ-axis rotating mechanism 6 formed with, for example, a turntable structure. Therefore, the sensor 8 can measure the object M from an arbitrary position at an arbitrary measurement axis angle, that is, at an arbitrary “angle”.

  The X-axis direction moving mechanism 2, Y-axis direction moving mechanism 3, Z-axis direction moving mechanism 4, θ-axis rotating mechanism 5, and φ-axis rotating mechanism 6 are each operated by a control device 9 via a driver 7. Be controlled. The control device 9 can acquire the point cloud data of the object M with “angle” corresponding to the 3D design shape data (for example, 3D CAD data) for comparing and comparing the 3D shape of the object M. To control the operation. Such a function of the control device 9 is normally imposed on a point cloud data acquisition device to be described later, and in the present invention, a point cloud data comparison system is formed by doing so.

  The X-axis direction moving mechanism 2, the Y-axis direction moving mechanism 3, the Z-axis direction moving mechanism 4, the θ-axis rotating mechanism 5, and the φ-axis rotating mechanism 6 are encoders as means for detecting the respective operation amounts. 2e, 3e, 4e, 5e, 6e are provided. The operation amount information of each mechanism from these encoders 2e, 3e, 4e, 5e, 6e is stored in the control device 9 via the driver 7.

  The operation amount information of each mechanism stored in the control device 9 is used to obtain coordinate values necessary for the point cloud data of the object M. That is, the point cloud data of the object M is data obtained by measuring the surface of the object M as a number of points related to the point cloud data coordinates appropriately set for the object M. Since coordinate values relating to data coordinates are required, the movement amount information of each mechanism is used to obtain the coordinate values. Further, the operation amount information of each mechanism is also used in the synthesis of polygon data, as will be described later.

  FIG. 2 shows a configuration of a point cloud data comparison apparatus according to an embodiment. The point cloud data comparison device 10 includes a display device 11, an input device 12, an arithmetic processing device 13, an auxiliary storage device 14, and a main storage device 15. The arithmetic processing unit 13 accepts the processing conditions input from the input device 2 and reads the point cloud data / processing condition input unit 16 for reading the point cloud data stored in the auxiliary storage device 14 from the point cloud data. Polygon processing unit 17 that creates polygon data, polygon noise removal processing unit 18 that removes noise from the polygonized data, polygon synthesis processing unit 19 that synthesizes a plurality of polygon data into one polygon data, and points the polygon data again. A re-point grouping processing unit 20 that converts the data into group data, a comparison processing unit 21 that compares and compares the design shape data with the point group data, and a comparison verification result output instruction unit 22 that outputs the comparison verification result to the display device 11 are provided. Yes. Here, the data stored in the auxiliary storage device 14 may be data obtained by the above point cloud data acquisition device 1 or data obtained by other methods. As described above, when the point cloud data comparison device 10 is combined with the point cloud data acquisition device 1 to configure the point cloud data comparison system, the data obtained by the point cloud data acquisition device 1 is obtained.

  FIG. 3 shows the flow of collation / comparison processing performed by the point cloud data comparison apparatus 10 configured as described above. In the collation / comparison processing, first, the point cloud data / processing condition input unit 16 reads data from the auxiliary storage device 14 and stores it in the main storage device 15 (processing S1). The data read from the auxiliary storage device 14 includes the point cloud data obtained by measuring the object M illustrated in FIG. 4 with the point cloud data acquisition device 1 in FIG. It is operation amount data.

  An example of point cloud data for the object M in FIG. 4 is shown in FIG. The point cloud data obtained by measuring the object M with the point cloud data acquisition apparatus 1 contains noise N generated during the measurement, as seen in FIG. FIG. 6 shows the point cloud data viewed from the side.

  Next, various processing conditions are input from the input device 2 (processing S2). The processing conditions to be input are polygon processing conditions for converting point cloud data to polygon data, noise removal processing conditions for removing the above-described noise of polygon data obtained by polygon processing, Synthesis processing conditions for the synthesis processing for polygon data, re-point grouping processing conditions for converting the synthesized polygon data into point cloud data again, and comparison verification processing for comparing and collating design shape data with point cloud data And display processing conditions when displaying the comparison result.

  Next, polygon processing for creating polygon data from the point cloud data is performed by the polygon processing unit 7 (processing S3). The polygonization processing is performed based on the polygonization processing conditions input in step S2. For example, when a quadrilateral polygon is used, each point in the point cloud data is sequentially connected by adjacent four points to form a polygon. That ’s it. An example of polygon data obtained by converting the point cloud data of FIG. 5 into a polygon is shown in FIG.

  When polygon data is obtained, the polygon noise removing unit 8 performs noise removal processing on the polygon data (processing S4). As for noise removal, automatic processing by the polygon noise removal unit 8 and manual processing by the operator instructing the polygon noise removal unit 8 of noise to be removed can be performed. Automatic noise removal processing is performed based on the noise removal processing conditions input in step S2. The noise removal processing conditions include, for example, setting a reference value S for the distance between the points constituting the polygon, deleting a polygon whose distance between the constituent points is larger than the reference value S, or creating a virtual surface for the polygon data. It is set so that the polygon whose distance from the virtual surface exceeds the reference value S ′ is deleted. On the other hand, the manual noise removal processing is performed by the operator instructing the noise to be removed while viewing the polygon data displayed on the display device 11, and the polygon noise removal unit 8 deleting it in response to this. The Such noise removal processing is sequentially repeated for each noise location (processing S4 ′). An example of polygon data that has undergone the noise removal processing is shown in FIG.

  As described above, the point cloud data is converted into polygon data, and noise removal is performed on the polygon data, so that the noise removal operation can be performed more easily. That is, first, it becomes easy to set a reference for automatic noise removal by making a polygon. Secondly, there is a limit to automatic noise removal, and manual noise removal is necessary to obtain point cloud data with higher accuracy. However, in this manual noise removal, noise is caused by polygon data. It appears as a graphic that is easy to visually recognize, and the operator can more easily determine noise. Thirdly, it becomes easier for the operator to confirm the result of noise removal.

  When the noise removal is completed, polygon data is synthesized by the polygon synthesis processing unit 9 (processing S5). Polygon data is synthesized in accordance with the synthesis processing conditions input in step S2. This process is necessary because the point cloud data is acquired in units of planes. That is, since the point cloud data for the object M in FIG. 4 is acquired in units of surfaces obtained by dividing each surface F1 to F7 in the shape of the object M or each surface by a certain size, the design shape data. In order to collate and compare with each other, it is necessary to synthesize the point cloud data of each surface unit to a state corresponding to the design shape data, and for this reason, a synthesis process is required. By performing this synthesis on polygon data as in the present invention, it is easy for the operator to visually recognize the synthesis state, and the processing efficiency can be increased. This composition can be automated by using the operation amount information of each mechanism as described above obtained by the point cloud data acquisition apparatus 1 according to the present invention. That is, the line of sight (measurement axis) of the sensor 8 with respect to the object M can be determined based on the operation amount information of each mechanism. Then, it is possible to derive the relationship between the polygon data in units of planes from the line-of-sight information, thereby automatically synthesizing the polygon data. By enabling automatic synthesis in this way, it is possible to significantly improve workability compared to the conventional method in which synthesis is performed manually.

  When the synthesis is completed, the re-point grouping process is performed by the re-point group processing unit 9 to convert the polygon data back to the point group data (processing S6). An example of point cloud data obtained by the re-point cloud processing is shown in FIG. Next, the point cloud data and design shape data (three-dimensional CAD data, etc.) obtained by the re-point cloud processing are displayed on the display device 11 to align them (processing S7). FIG. 10 shows a state in which the point cloud data and the design shape data are aligned. In this state of alignment, the operator visually compares and compares both data to determine whether there is an error in the point cloud data with respect to the design shape data.

  Next, the comparison processing unit 10 compares and collates the point cloud data with the design shape data (processing S8). This comparison and collation is usually performed as a dimension comparison process for comparing the dimensions of the sides of each surface in the point cloud data with the corresponding parts in the design shape data. In this embodiment, the comparison is performed by the dimension comparison process. In this dimension comparison process, the comparison processing unit 10 is provided with a function capable of designating a point to be compared on either the point cloud data or the design shape data. Therefore, the operator can appropriately designate the required dimension comparison portion on the point cloud data or the design shape data based on the result of the visual comparison and comparison on the display device 11 as described above. become. The result of the comparison processing performed in this way is created as dimensional comparison collation data in the form of a comparison list, and this is output in step S9. Finally, the output result is displayed on the display device 11 (processing S10).

  The present invention makes it possible to more easily remove noise from point cloud data for point cloud data comparison, and to more easily compare point cloud data and design shape data. Data or polygon data can be combined more easily. Therefore, the present invention can be widely used as a useful product in an industrial field where a product manufactured based on the design shape needs to be compared with the design shape.

It is a figure which shows the structure of the point cloud data acquisition apparatus by one Embodiment. It is a figure which shows the structure of the point cloud data comparison apparatus by one Embodiment. It is a figure which shows the flow of a process in a point cloud data comparison apparatus. It is a figure which shows the example of a target object. It is a figure which shows the example of the point cloud data obtained about the target object of FIG. It is a figure which shows the state which looked at the point cloud data of FIG. 5 from the horizontal direction. It is a figure which shows the example of the polygon data obtained by polygonizing the point cloud data of FIG. It is a figure which shows the example of the polygon data after noise removal. It is a figure which shows the example of the point cloud data re-grouped from polygon data. It is a figure which shows the state which aligned point cloud data with design shape data.

Explanation of symbols

1 Point cloud data acquisition device 2 X-axis direction moving mechanism 2e to 6e Encoder (operation amount detection means)
3 Y-axis direction moving mechanism 4 Z-axis direction moving mechanism 5 θ-axis rotating mechanism 6 φ-axis rotating mechanism 8 Sensor 10 Point cloud data comparison device 17 Polygonization processing unit 18 Polygon noise removal processing unit M Object

Claims (6)

In the point cloud data comparison method of acquiring point cloud data for the shape of the object and comparing the point cloud data against the design shape data based on the object,
A point cloud data comparison method comprising a step of creating polygon data from the point cloud data and removing noise from the polygon data. In the point cloud data comparison method of acquiring point cloud data for the shape of the object and comparing the point cloud data against the design shape data based on the object,
When comparing the point cloud data and the design shape data with respect to the dimensions of an arbitrary location, a dimension comparison location can be designated on the design shape data or the point cloud data. Group data comparison method. In a point cloud data comparison device that compares and compares the point cloud data acquired for the shape of the object with the design shape data that is the basis of the object,
A polygonization processing unit for converting the point cloud data into polygon data, and a polygon noise removal processing unit for performing noise removal processing on the polygon data obtained by the processing in the polygonalization processing unit are provided. Point cloud data comparison device. In a point cloud data comparison device that compares and compares the point cloud data acquired for the shape of the object with the design shape data that is the basis of the object,
A comparison processing unit that compares the point cloud data and the design shape data with respect to the dimensions of an arbitrary location is provided, and this comparison processing unit can designate a size comparison location on the design shape data or the point cloud data. A point cloud data comparison apparatus characterized by being configured as described above. Point cloud data is acquired for the shape of the object, and the point cloud data used in the point cloud data comparison for comparing the point cloud data with the design shape data based on the object is acquired. In the point cloud data acquisition device,
A sensor for measuring the surface shape of the object as a collection of many points, and an X-axis direction moving mechanism, a Y-axis direction moving mechanism, a Z-axis direction moving mechanism, a θ-axis rotating mechanism, and a φ-axis rotating mechanism are provided. The sensor can be arbitrarily moved in three directions of X, Y, and Z orthogonal to each other by the X-axis direction moving mechanism, the Y-axis direction moving mechanism, and the Z-axis direction moving mechanism, The θ axis rotation mechanism can arbitrarily rotate the measurement axis in the vertical direction, the object can be arbitrarily rotated in the horizontal direction by the φ axis rotation mechanism, the X axis direction movement mechanism, Each of the Y-axis direction moving mechanism, the Z-axis direction moving mechanism, the θ-axis rotating mechanism, and the φ-axis rotating mechanism is provided with an operation amount detecting means for detecting the operation amount. Point Data acquisition device.   A point cloud data comparison system comprising the point cloud data comparison device according to claim 3 and the point cloud data acquisition device according to claim 5 in combination.

Images