JP5445848B2 - Object shape evaluation device - Google Patents

Description

  The present invention relates to an object shape evaluation apparatus that evaluates the shape of a measurement object by aligning a large number of measurement points corresponding to the shape of the measurement object and a large number of reference points corresponding to the measurement points.

  As a conventional technique for evaluating the measurement result for a measurement object by aligning (matching) the measurement point group and the reference point group, the position of the object to be measured is combined with stereo vision and an ICP (Iterative Closest Point) algorithm. An image measurement method that accurately measures the posture is known (for example, see Patent Document 1). In this stereo image measurement method, a measurement point cloud is obtained by stereo-processing a plurality of camera images of the three-dimensional shape of the measurement target, and is visible from the known shape information of the measurement target and the relative position and orientation of the stereo camera. The model point cloud is determined according to the spatial density of the measurement point group using only the predicted shape information of the visible part, and matching is performed between the measurement point group and the model point group using an ICP algorithm. The position / orientation corresponding to the model group having the smallest evaluation function is adopted as the measurement result.

  However, the ICP algorithm is a technique for positioning or merging position data that does not include an exceptional value in principle (problem for obtaining the most congruent congruent transformation of both data). Therefore, if there is a non-overlapping part or an additional part of data included in the distance data set to be positioned, there arises a problem that these function as exception values and hinder convergence to the correct state.

  In order to solve such problems of the ICP algorithm, an ICP positioning method in which M estimation is introduced has been proposed (see, for example, Non-Patent Document 1). In this M-estimated ICP positioning method, when the congruent transformation is successively converged while updating the correspondence between two point groups (for example, a measurement point group and a reference point group), a residual amount (for example, a measurement point and a reference point) The weight is set according to the distance. By changing the evaluation value in positioning for each corresponding point by this weight, it is possible to suppress the adverse effect due to the exceptional value. However, when measuring the three-dimensional shape of the measurement object, the measurement error increases due to the measurement point being shifted with respect to the inclined surface of the measurement object due to the measurement technical conditions of the measurement apparatus. For penetrating regions such as holes and cutouts, the measurement data itself is missing, resulting in a problem that the measurement error further increases.

Japanese Patent Laying-Open No. 2008-14691 (paragraph number 0006, FIG. 1) Shunichi Kaneko et al. “Robust ICP positioning method with M estimation”, Journal of Precision Engineering V0l.67, No.8, 2001, 1276-1280

  In view of the above situation, the object of the present invention is to provide a measurement point group and a reference point even if a measurement object has a specific measurement region in which the measurement error increases due to the measurement technical conditions of the measurement apparatus. To provide an object shape evaluation apparatus in which joint conversion with a group is performed appropriately and quickly, and as a result, object shape evaluation of a measurement object is performed appropriately.

In order to achieve the above object, the characteristic configuration of the object shape evaluation apparatus according to the present invention includes a measurement point data input unit that inputs measurement point data including position information of a large number of measurement points corresponding to the shape of the measurement object, A reference data storage unit that stores reference point data including position information of a large number of reference points corresponding to the reference shape of the measurement object, and a sequential convergence process for sequentially converging the distance between the corresponding measurement point and the reference point An alignment processing means for aligning the measurement point and the reference point based on the measurement point, and the measurement point data after the alignment processing by the alignment processing means and the reference point data A shape evaluation unit that evaluates a shape, and the alignment processing unit is generated by an edge detection unit that detects an edge point corresponding to an edge of the measurement object from measurement point data. Edge alignment is performed by aligning an edge point group and an edge reference point group that is edge reference point data for the edge of the measurement object stored in the reference data storage unit, and the edge alignment is performed. between the reference point obtained edge congruent transformation parameters the measuring point data overall measurement points using stomach line alignment toward the reference point, adjacent point distance between said measuring points adjacent or neighboring A weight coefficient determining means for determining a distance weight coefficient between adjacent points based on a distance between adjacent points, and using the distance weight coefficient between adjacent points when obtaining a sequential convergence evaluation value in the sequential convergence process. .

According to this feature configuration, the edge measurement point group that is the measurement point corresponding to the edge of the measurement object extracted from the obtained measurement point data by the edge detection process, and the edge reference corresponding to the edge of the measurement object Edge alignment is performed using point clouds. It is important to define the outer shape of the edge measurement point group and the object to be measured, and unlike a hole or notch, it expresses a clear substantive shape and is easy to detect accurately by measurement. Therefore, by this edge alignment, the edge measurement point group accurately overlaps the edge reference point group. Further, alignment is performed so that the measurement point group of the entire measurement point data is aligned with the reference point group of the entire reference point data using the edge joint conversion parameter used for such edge alignment. This makes it possible to superimpose satisfactory measurement point groups and reference point groups even for measurement objects that have a specific measurement area where the measurement error, such as inclined surfaces, holes, and notches, increases. The shape deviation of the measurement object from the reference shape can be detected from the overlay result.
Further, according to this configuration, between any adjacent points that are distances between adjacent points (measurement points or reference points) belonging to the point group in either the measurement point group or the reference point group. A weighting factor is assigned based on the distance. The assigned weighting factor is used to determine the sequential convergence evaluation value in the sequential convergence process that sequentially converges the distance between the corresponding measurement point and the reference point, and the effect of the distance between adjacent points on the sequential convergence evaluation value The degree is adjusted. For example, there is a slight measurement point deviation at a measurement point where the distance between adjacent points is larger than the other points, or at a measurement point corresponding to a reference point where the distance between adjacent points is larger than the others. There may be a large difference in the position coordinate value between the measurement point and the reference point. This is because, when the measurement surface by a laser beam or the like is curved or inclined at an angle close to a right angle with respect to the measurement reference surface, a slight deviation of the measurement scanning pitch leads to a large deviation of the measurement point, This is because the positional correspondence between the reference point and the measurement point is deteriorated. In such a measurement situation, a smaller weighting factor is assigned to a point having a large distance between adjacent points, and the influence of the point on the convergence evaluation value is sequentially suppressed, so that the measurement point group and the reference point group Appropriate congruent conversion between can be realized. As a result, the object shape of the measurement object is properly evaluated.

  All measurement point groups obtained by setting measurement point groups other than the edge measurement point group in the edge measurement point group aligned with the edge reference point group overlap with all reference point groups fairly accurately, but there is no edge measurement. When the ratio of the surface area is large, there are cases where it is not sufficient to match only the edge points. Therefore, in one preferred embodiment of the present invention for accurately matching the measurement point group of the entire measurement point data with the reference point group of the entire reference point data, the measurement using the edge joint conversion parameter is performed. After the alignment of the points toward the reference point, alignment is further performed between the measurement points of the entire measurement point data and the corresponding reference points. As a result, since all the measurement point groups and all the reference point groups are collated, the accuracy is improved, but since the alignment is performed using the edge points in advance, all the measurement point groups and all the reference point groups are Can be performed without waste.

  Further, as a further characteristic configuration in the object shape evaluation apparatus according to the present invention, the weighting factor determination means determines a distance weighting factor between corresponding points based on a distance between corresponding points between the corresponding measurement point and a reference point. It is also preferable to use the distance weight coefficient between corresponding points when the alignment processing means obtains the sequential convergence evaluation value in the sequential convergence processing. This feature is a robust ICP (Iterative Closest Point) positioning method known as a so-called M estimation method. Since it is explained in detail in Non-Patent Document 1, detailed explanation here is omitted, but the distance weight factor between corresponding points determined based on the distance between corresponding points between the corresponding measurement point and the reference point is It is used when obtaining the successive convergence evaluation value in the successive convergence process, and the degree of influence on the successive convergence evaluation value is adjusted by a measurement point having a large distance between corresponding points (so-called outlier). Thereby, in this preferable object shape evaluation apparatus, the distance weight between corresponding points determined based on the distance between corresponding points between the corresponding measurement point and the reference point, and the distance between adjacent points in the same point group. With the distance weight coefficient between adjacent points determined based on the measurement point data estimated to introduce an error, the influence of the measurement point data can be eliminated or reduced. As a result, the object shape evaluation of the measurement object is performed more appropriately.

  In determining the distance weight between adjacent points based on the distance between adjacent points by the weight coefficient determining means, either the distance between adjacent points in the measurement point group or the distance between adjacent points in the reference point group may be adopted. When the distance between adjacent points in the reference point group is adopted, the distance weight coefficient between adjacent points can be obtained in advance before measurement. When there are a plurality of measurement objects of the same shape, it is effective because the distance weight coefficient between adjacent points obtained from common reference point data (reference point group) can be used in each object shape evaluation.

  The object shape evaluation apparatus according to the present invention assigns a smaller weight coefficient to a point having a large distance between adjacent points, thereby suppressing the influence of the point on the successive convergence evaluation value. Therefore, a slight shift in the measurement scanning pitch leads to a large shift in the measurement point, which is particularly suitable for the object shape evaluation of the measurement system in which the position correspondence between the reference point and the measurement point is deteriorated. Therefore, when the measurement point data of the object shape evaluation apparatus according to the present invention is three-dimensional position data extracted from a distance image obtained by processing a captured image of the measurement object irradiated with slit light. Suitable for

It is a schematic diagram which shows the principle of alignment with the measurement point group using the edge point which concerns on this invention, and a reference point group. It is a schematic diagram which shows the process added to the principle of the alignment in FIG. An example of an object shape evaluation apparatus according to the present invention is a schematic diagram schematically showing. It is a functional block diagram of an evaluation unit. It is a functional block diagram of a surface evaluation module. It is a functional block diagram of a defect evaluation module. It is a functional block diagram of a reference data change module. It is a flowchart which shows the flow of the whole control in an object shape evaluation apparatus. It is a flowchart which shows an evaluation calculation process. It is a flowchart which shows an edge extraction routine. It is a flowchart which shows an edge position alignment routine. It is a flowchart which shows an all-points alignment routine. It is a flowchart which shows a defect determination routine. It is a flowchart which shows the shape difference determination routine by mold correction. It is a flowchart which shows the shape difference determination routine by mold correction. It is a flowchart which shows a defect and type | mold correction location determination routine. It is a flowchart which shows a reference data change routine. It is a schematic diagram which shows typically position alignment with a measurement point group and a reference point group. It is the comparison graph which compared the alignment by all point collation, and the alignment which combined edge collation and all point collation. It is a comparison graph which shows the comparison calculation result of the repetition position alignment performed using various weighting factors.

The basic principle of the surface defect evaluation employing the alignment algorithm between the measurement point group using the edge points and the reference point group in the object shape device according to the present invention will be described with reference to FIG.
In this example, in order to obtain the three-dimensional position data of the surface of the measurement object, a light cutting method that performs three-dimensional measurement based on the principle of triangulation is used (# 01). In this light cutting method, a slit object is irradiated with slit light, a band-like light that is curved according to the surface shape of the object to be measured is photographed with a camera, and the point sequence X is obtained from the imaging position in the photographed image. , Y, Z values are calculated. The three-dimensional position data of the measurement points on the entire surface of the measurement object scanned by the slit light is generated as the light cutting pixel position information, and is developed in the work memory as the measurement point data for the evaluation of the surface unevenness described below. (# 02).

  As preparations for the evaluation of surface irregularities, all reference data, which is three-dimensional data serving as a reference for the entire surface of the measurement object, and edge, which is reference data for the edges formed on the surface of the measurement object It is assumed that reference data (which can be created by applying edge detection filter processing to all reference data) is stored.

  First, an edge detection process is performed by applying an edge detection filter such as a Sobel filter to the measurement point group which is the measurement point data developed in the work memory (# 03). The edge measurement point group generated by the edge detection process is associated with the edge reference point group read from the edge reference data stored in advance, and the edge measurement point group is aligned with the edge reference point group. (# 04). For this alignment, a method based on an ICP (iterative closest point) method, which will be described in detail later, is suitable. In this method, E (R, t), R: rotation matrix, t: calculation for minimizing the translation vector is repeated as a joint transformation parameter in the alignment until a predetermined threshold value is reached (# 05). When the alignment of the edge measurement point group and the edge reference point group is completed, and finally the joint conversion parameter that brings the edge measurement point group close to the edge reference point group is obtained, all the measurement points are Alignment is performed so that all measurement point groups from the data approach all reference point groups from all reference data (# 06). When the measurement conditions are good, the measurement point group and the reference point group at the portion finished as intended are almost overlapped. If the measurement point and the reference point are separated by a predetermined distance or more, it is considered that there is a possibility of a defect, and defect information specifying the location is obtained (# 07).

  The above is the basic surface defect evaluation processing procedure employed in the present invention, but using the joint conversion parameters obtained using only the edges, all the measurement point groups from all the measurement point data are set to all the standards. Even if all-point positioning is performed so as to be close to all the reference point groups from the data, there is a case where sufficient overlay is not achieved depending on the shape of the measurement object, measurement conditions, and the like. In such a case, as shown in FIG. 2, it is necessary to perform alignment by further associating all measurement point data with all reference point groups from the state in which all the points are aligned. First, from the state of step # 06 in FIG. 1, all measurement point groups and all reference point groups are associated (# 06a), and the alignment that minimizes the congruent transformation parameter: E (R, t) is predetermined. Repeat until the threshold is reached (# 06b). When the sufficiently converged alignment is obtained, the entire alignment is completed (# 06c). When the measurement point and the reference point are separated by a predetermined distance or more in the finally aligned state, it is considered that there is a possibility of a defect, and defect information specifying the location is generated (# 08).

  An example of an object shape evaluation apparatus that employs an alignment algorithm for a reference point group including an edge reference point and a measurement point group including an edge measurement point group as described above will be described. The object shape evaluation apparatus in this embodiment is set so as to inspect a groove cross section of a measurement object in which a large number of straight deep grooves are aligned on the surface. FIG. 3 is a perspective view schematically showing the configuration of such an object shape evaluation apparatus.

  The object shape evaluation apparatus includes a measurement device unit 1, a measurement object table 2, and a controller 3 that performs measurement operation control on the measurement device unit 1 and the measurement object table 2 and evaluates the measurement result. The measurement apparatus unit 1 includes, as main components of the measurement system, a laser-type slit light source unit 11a that generates slit light and an imaging unit 11b that captures an area of the measurement object irradiated with the slit light. ing. The slit light source unit 11a and the imaging unit 11b are integrally assembled as a measurement head. The measurement object table 2 includes a base 23 and a portal frame 22 erected on the base 23 as main components. As a positioning mechanism for the measurement object, a rotary table 21 for placing the measurement object and a Y-direction scanning mechanism for moving the portal frame 22 in the Y-axis direction are provided on the base 23, and the measurement head is moved to the X-axis. An X-direction scanning mechanism that moves in the direction is provided on the portal frame 22. The X-direction scanning mechanism and the Y-direction scanning mechanism enable XY plane scanning of the surface of the measurement object. Note that an elevating mechanism for elevating and lowering the slit light source unit 11a and the imaging unit 11b is provided in the measurement head.

  The slit light source unit 11 a is a laser slit parallel light projector, and includes a laser slit projector 12 and a cylindrical lens 13. The slit light that spreads out in a fan shape from the laser slit projector 12 is converted into parallel light parallel to the slit optical axis by the cylindrical lens 13 and irradiates the measurement object.

  The imaging unit 11 b includes a telecentric lens group 15, an imaging unit 16 including a large number of light receiving elements (CCD and CMOS) arranged in a plane, and a P polarizing plate 14 arranged on the subject side of the telecentric lens group 15. And. The telecentric lens group 6 is used for imaging a plurality of uneven surfaces on a measurement object without a blind spot at an angle of view of zero. The slit light from the slit light source unit 2 is applied to the surface of the measurement object, and the reflected light reflected there passes through the P polarizing plate 14 and the telecentric lens group 15 along the imaging optical axis of the imaging unit 11b. Thus, it is set to reach the imaging unit 16. In this embodiment, an extremely narrow angle of about 11 degrees is adopted as an intersection angle at which the slit optical axis of the slit light source unit 2 intersects with the imaging optical axis of the imaging unit 11b. Therefore, when the imaging surface of the imaging unit 16 is in a posture perpendicular to the imaging optical axis, the measurement depth range along the slit optical axis direction exceeds the depth of field of the telecentric lens group 15. An out-of-focus area occurs in the depth range. In order to avoid this, the imaging surface of the imaging unit 16 is tilted with respect to the imaging optical axis so as to create a tilt angle, and the depth of field is gained by the principle of tilt shooting. As a result, a photographed image without blur can be acquired even in a measurement range that is greater than the depth of field of the telecentric lens group 15.

  The controller 3 is substantially formed as a computer unit, and includes a light source control unit 31, an image memory 32, an image processing unit 33, a three-dimensional measurement data calculation unit 34, an evaluation unit 4 particularly related to the present invention, and a lifting mechanism. A control unit 35, a rotary table control unit 36, an X direction scanning control unit 37, and a Y direction scanning control unit 38 are provided. The rotary table control unit 36, the X-direction scanning control unit 37, and the Y-direction scanning control unit 38 control the operations of the rotary table 21, the X-direction scanning mechanism, and the Y-direction scanning mechanism, respectively. Set to an appropriate measurement position in the (Y plane). The lifting mechanism control unit 35 adjusts the height of the slit light source unit 11a and the imaging unit 11b with respect to the measurement object.

  The captured image (image data) sent from the imaging unit 11 b to the controller 3 is developed in the image memory 32. Furthermore, if necessary, the image processing unit 33 performs image processing such as coordinate conversion, level correction, and edge detection, and the light cutting line S by the slit light is detected. The three-dimensional measurement data calculation unit 34 is known from the coordinate value of the light cutting line S detected by the image processing unit 33 because the irradiation point and irradiation angle of the slit light, and the angle formed by the slit optical axis and the imaging optical axis are known. By calculating based on the surveying method, it is possible to obtain a large number of measurement point data (distance images) corresponding to the optical cutting line S, that is, the three-dimensional cross-sectional shape of the measurement object forming a plurality of linear deep grooves. . The distance image here is measurement data in which a three-dimensional position coordinate value is assigned to a pixel as a measurement point. Instead of the calculation based on the triangulation method, a method using a table storing the calculation result may be adopted. The measurement point data generated by the three-dimensional measurement data calculation unit 34 is transferred to the evaluation unit 4.

As shown in FIG. 4, the evaluation unit 4 includes a surface evaluation module 5, a defect evaluation module 6, a reference data change module 7, and a data storage unit 8. The surface evaluation module 5 evaluates the surface shape of the measurement object by aligning the measurement point group with the reference point group using an alignment algorithm such as ICP. The defect evaluation module 6 evaluates the surface defect of the measurement object based on the alignment result of the measurement point group output from the surface evaluation module 5 to the reference point group. The reference data change module 7 has a function of changing the reference data through the shape measurement when the shape change within the allowable range of the product occurs due to factors such as partial correction of the mold. Such a change in the reference data is determined in units of a predetermined number of measurement objects. In this embodiment, both the surface evaluation module 5 and the defect evaluation module 6 function as shape evaluation means.
The data storage unit 8 includes a reference data storage unit 81, a defect information storage unit 82, and a mold correction information storage unit 83. The reference data storage unit 81 stores reference point data indicating the surface shape of the measurement object. The reference data is divided into edge reference point data and all reference point data. The total reference point data is data indicating an ideal finished shape set so as to correspond to the measurement points for each of the predetermined blocks divided in advance over the entire measurement target of the measurement target. The edge reference point data is obtained by extracting only the reference points corresponding to the edges in the outer shape defined by the reference point group from the reference point group developed from all the reference point data. The defect information storage unit 82 is defect information that is information on a collection of measurement points that are determined to be defective by satisfying a predetermined condition among measurement points that are more than a predetermined distance from the reference point in a state where the alignment is completed. Is stored. As will be described in detail later, the mold correction information storage unit 83 stores mold correction information, which is information about a portion where the mold may be corrected through the surface defect evaluation process.

  As shown in FIG. 5, the surface evaluation module 5 includes a measurement point data input unit 51, an edge detection unit 52, a point group association unit 53, a weight calculation unit 54, a convergence evaluation unit 55, and a congruent conversion parameter. A generation unit 56 and an alignment execution unit 57 are included. The measurement point data input unit 51 receives measurement point data from the three-dimensional measurement data calculation unit 34. The edge detection unit 52 expands the measurement point group in the working memory based on the measurement point data, and detects an edge measurement point corresponding to the edge on the surface of the measurement object using an edge detection filter such as a Sobel filter. Then, edge measurement point data including an edge measurement point group is generated. The point group associating unit 53 associates the measurement points with the reference points based on the above-described alignment algorithm in units of the predetermined blocks. In order to make the explanation easy to understand, the point group associating unit 53 includes an edge point group associating unit 53a in which the alignment processing target is an edge measurement point group and an edge reference point group, and the alignment processing target is all measurement points. The calculation function is the same, although it is divided into a group and an all-point group associating unit 53a as an all-reference point group. As will be described in detail later, the weight calculation unit 54 obtains a weighting coefficient used for the alignment process. The weight coefficient w used here is a value obtained by multiplying the distance weight coefficient between adjacent points: γi and the distance weight coefficient between corresponding points: ρi. The joint conversion parameter generation unit 56 obtains joint conversion parameters for aligning the measurement point group with the associated reference point group. The convergence evaluation unit 55 calculates a sequential convergence evaluation value when the measurement point group attempts to converge and move to the reference point group using the congruent transformation parameter, and the movement of the measurement point group sequentially converges to the reference point group. Evaluate whether or not The alignment execution unit 57 converts the position coordinates of the measurement point group using the joint conversion parameter that is generated by the joint conversion parameter generation unit 56 and evaluated to be converged by the convergence evaluation unit 55. In this embodiment, the point group associating unit 53, the weight calculating unit 54, the convergence evaluating unit 55, the joint conversion parameter generating unit 56, and the alignment executing unit 57 are between the corresponding measurement point group and the reference point group. An alignment processing means 50 is configured to align the measurement point group and the reference point group based on the sequential convergence process for sequentially converging the distance.

The alignment process in the alignment processing unit 50 will be schematically described.
(1) Input measurement point data from the three-dimensional measurement data calculator 34.
(2) generating an alignment result between the edge measurement point group and the edge reference point group associated by the edge point group association unit 53a as edge alignment data;
(3) Based on the edge alignment data, the alignment result of all measurement point groups and all reference point groups associated by the all point group association unit 53b is generated as all alignment data.
(4) Output all alignment data to the defect evaluation module 6;
It becomes.

  As shown in FIG. 6, the surface defect evaluation module 6 that functions as a surface defect evaluation unit includes an erroneous correspondence measurement point group extraction unit 61, a defect determination unit 62, a mold correction evaluation unit 63, and a mold correction location information registration unit. 64 and a defect feature amount calculation unit 65. The erroneous corresponding measurement point group extraction unit 61 erroneously selects a measurement point where the distance between corresponding points between the corresponding measurement point and the reference point is greater than the first threshold based on the surface defect determination condition described in detail later. A miscorresponding measurement point group (miscorresponding region) is extracted by regarding it as a corresponding measurement point. The defect determination unit 62 determines that the number of neighboring measurement points whose distance from an adjacent measurement point is smaller than the second threshold among the erroneous measurement points included in the extracted erroneous measurement point group is less than the third threshold value. When the area is large, it is determined that there is a possibility of a surface defect in the area including the neighboring measurement point, and defect data describing the defect information is output.

  The defect feature amount calculation unit 65 calculates the defect feature amount such as the defect size and the center of gravity of the defect (surface defect: turning, dent, peeling, etc.) determined by the defect determination unit 62, and the three-dimensional position of the defect. The defect information is stored in the defect information storage unit 82 as defect information. If the object to be measured is a product that is manufactured by a mold, and if a part of the mold is modified during lot production, the shape of the product that has been successfully manufactured does not match the shape of the reference point data. The problem arises that discrepancies are considered defects. In order to solve this problem, the mold correction evaluation unit 63 refers to the defect information to determine whether the defect determined by the defect determination unit 62 is caused by partially correcting the used mold. Has a function to check. If the mold correction evaluation unit 63 evaluates that the defect area corresponding to the received defect data is not due to the mold correction, the defect determination unit 62 is notified that there is no mold correction, so the defect determination unit 62 Is determined to be a defect. When the mold correction evaluation unit 63 evaluates that the defect area corresponding to the received defect data is due to mold correction, the mold correction location information registration unit 64 also outputs an evaluation result indicating that the defect area is due to mold correction. The mold correction location information registration unit 64 registers the type correction location information generated in response to the evaluation result in the type correction location information storage unit 83.

  In this embodiment, the degree of similarity between the reference point group corresponding to the defect determined by the defect determining unit 62 and the measurement point group is further calculated, and this similarity is used as defect information for the defect information storage unit 82. A similarity calculation unit 66 is provided. Therefore, the defect feature amount calculation unit 65 can refer to this similarity when checking whether or not the determined defect is caused by partially correcting the used mold. If the similarity can directly or indirectly represent the similarity between the surface shape defined by the reference point group included in the defect and the surface shape defined by the measurement point group included in the defect Good. As the direct similarity, the equivalent ellipse major axis length corresponding to the defect size (the similarity becomes larger as the length is approximated), the indirect similarity is the reference point group and the measurement point group. Average correspondence distance (the shorter the distance, the higher the degree of similarity can be considered).

  In the case where the object to be measured is a product in which a mold is manufactured, dimensional variation occurs in the mold due to wear or the like when a predetermined number of products are manufactured. Such a dimensional variation causes a deviation between the measurement point data and the reference data even though the product is normally manufactured. In order to avoid judging this deviation as a defect, it is preferable to correct the reference data of the region where such a deviation occurs with the measurement point data of a product manufactured normally. The reference data change module 7 provided for this purpose includes a data input unit 71, a measurement point data shape evaluation unit 72, and a reference data change unit 73 as shown in FIG. The data input unit 71 receives the measurement point data generated by the surface evaluation module 5 and the shape data that is product design data, converts the data into a form that can be internally processed, and transfers the data to the measurement point data shape evaluation unit 72. When the measurement point data shape evaluation unit 72 compares the measurement point data with the product design data and evaluates that the product is within the allowable range of the design dimensions, the reference data change unit 73 accesses the reference data storage unit 81 to Change the reference data with the measurement point data.

  A basic flow of the shape evaluation process of the measurement object using the object shape evaluation apparatus configured as described above will be described below with reference to the flowcharts shown in FIGS. The object to be measured here is a rectangular plate body with a number of straight deep grooves formed on the surface, and the measurement area is about 400 mm × 300 mm. This measurement area is divided into 100 mm × 15 mm measurement blocks. Four measurement blocks are scanned in one X-axis direction scan, and measurement point data representing the three-dimensional cross-sectional shape position of the linear deep groove in the measurement unit defined by the scanning pitch and imaging resolution is acquired and measured. Divide into blocks and store in memory. Each time one X-axis direction scan is completed, the Y-axis direction is moved at a predetermined pitch, and the X-axis direction scan for the next measurement block is performed in the reverse direction. By repeating such scanning in the X-axis direction and movement in the Y-axis direction, the measurement point data of the linear deep groove in the entire measurement region is acquired. Further, taking into account the generation of the measurement blind spot, the measurement in the same measurement region is performed again with the measurement object rotated by 90 degrees. In addition, the evaluation of the measurement result of the measurement object using the acquired measurement point data, that is, the inspection of the measurement object is performed in each block unit, and the inspection result in each block unit is put together to make a final synthesis. A determination is made.

  In order to perform shape evaluation in this apparatus, a measurement object is set on the rotary table 21 (# 1). When a measurement start button (not shown) is operated (# 2 Yes branch), measurement is started. First, the laser slit projector 12 is turned on by the light source controller 31, and the slit light is irradiated (# 3). The X-direction scanning mechanism, the Y-direction scanning mechanism, and the rotary table 21 are operated so that the left edge of the first measurement block that is the measurement start point is irradiated with the slit light (# 04).

  Scanning in the X-axis direction is performed while moving the X-direction scanning mechanism at a constant speed in the positive direction (# 5). At the same time, the image data from the imaging unit 11b is transferred to the image memory 32 (# 6). This scanning in the X-axis direction and acquisition of image data are performed until the slit light reaches the side edge of the measurement object. When the slit light reaches the side edge of the measurement object (# 7 Yes), scanning in the X-axis direction is stopped (# 8). When scanning in the X-axis direction is stopped, measurement point data based on the optical cutting method is calculated, and evaluation calculation processing is performed to determine the presence or absence of surface defects from the measurement point data and reference point data (from # 9 to # 24). ). This evaluation calculation process will be described in detail later. Actually, when the X-axis direction scanning is stopped in step # 08, the Y-direction scanning mechanism is operated while the evaluation calculation process is being executed, and the Y-axis direction shift is performed at a predetermined pitch (# 25). ). When the shift in the Y-axis direction is completed, a check is made as to whether scanning in the X-axis direction still remains (# 26).

  If scanning in the X-axis direction still remains in step # 26 (Yes branch at # 26), the direction of scanning in the X-axis direction is reversed (# 27), and the process returns to step # 5 to perform scanning in the X-axis direction. Do. If scanning in the X-axis direction does not remain in the check of step # 26 (# 14 No branch), the laser slit projector 12 is turned OFF and the slit light irradiation is stopped (# 28). Further, it is checked whether or not it is necessary to rotate the rotary table 21 by 90 degrees in order to complement the measurement point data of the unmeasurable part accompanying the generation of the 90-degree rotational blind spot (# 29). When it is necessary to rotate the rotary table 21 by 90 degrees (# 29 Yes branch), the rotary table 21 is rotated by 90 degrees, and the process returns to step # 3 again to repeat this measurement. If the additional rotation of 90 degrees is not sufficient, another 90 rotations (180 degree position and 270 degree position with respect to the initial posture position) are performed every 90 degrees. If it is not necessary to rotate the turntable 21 by 90 degrees (# 29 No branch), comprehensive determination is performed based on the defect evaluation results in all measurement blocks (# 30). In this comprehensive determination, a defect position display diagram in which the position of the defect is marked on the overall view showing the entire measurement object can be output through a monitor or a print. When the comprehensive determination is completed, the shape evaluation process for the measurement object is completed. If the inspection of a predetermined number of measurement objects is completed at this time, the reference data changing module 7 described above is normally manufactured. A reference data changing routine for correcting the measured point data of the product is performed (# 31).

  This reference data changing routine is shown in FIG. First, in order to calculate the number of measurement objects (the number of measurement workpieces) manufactured with the same mold and inspected by this object shape evaluation apparatus, each time the shape evaluation process for the measurement object is completed ( # 800). If the number of workpieces to be measured is equal to or less than the set number (# 801 No branch), this routine is terminated because it is not necessary to change the reference data. If the number of workpieces is less than the set number (# 801 Yes branch), whether or not the measurement point data of this measurement object is suitable for correcting the reference point data, that is, a defect in the shape evaluation process for this measurement object. Is detected (# 802). If a defect is detected (# 802 No branch), this measurement object is regarded as not suitable for correcting the reference point data, and this routine is terminated. When no defect is detected (# 802 Yes branch), a shape check is further performed based on measurement data (# 803). In the shape check based on the measurement data, the measurement point data is compared with the product design data, and the shape is regarded as normal because the product is within the allowable range of the design dimension. If it is determined by the measurement data shape check that the shape is not normal (# 804 No branch), it is determined that the measurement object is not suitable for correcting the reference point data, and this routine is terminated. When the measurement data shape check determines that the shape is normal (Yes in # 802), the measurement object is regarded as not suitable for correcting the reference point data, and the measurement point data of the measurement object is used as a reference. Change to point data (# 805).

Next, the evaluation calculation process which is the process routine according to the present invention shown in FIG. 9 will be described.
First, the image processing unit 33 of the controller 3 processes the image data transferred from the imaging unit 16 to generate the optical cutting line pixel position information (# 09). From this light section line pixel position information, the three-dimensional measurement data calculation unit 34 uses a table storing the relationship between the pixel position and the three-dimensional position calculated from the pixel position based on the triangulation method. Based on the cutting line pixel position information, the three-dimensional coordinate value of the measurement object is read out, and this value is transferred to the memory address associated with each measurement point as measurement point data (# 10). Of course, instead of using a table, a three-dimensional coordinate value may be obtained and used as measurement point data by performing calculations based on the triangulation method using the light section line pixel position information each time.

  When the measurement point data is obtained, an edge extraction routine for extracting edge measurement points from the measurement point data and generating edge measurement point data is executed (# 11). In this edge extraction routine, as shown in FIG. 10, all measurement point data is read and developed in the working memory (# 100). At this time, a specified value equal to or lower than the measurement lower limit value is assigned as a dummy to the missing data point caused by the inability to measure in all measurement point data (# 101). An edge detection filter (differential filter such as a Sobel filter) is applied to the value in the height direction (Z-axis direction) over the entire area of the measured block to calculate a differential value (# 102). By assigning this differential value using a predetermined threshold value, for example, “0” or “1”, binarized data in which edges and non-edges are separated is generated. The measurement point cloud data is taken out and stored in the memory (# 102), and this routine is terminated.

  Subsequently, alignment between the edge measurement point data and the edge reference point data is performed by the alignment processing means 50. Before that, in the object shape evaluation process performed on the preceding measurement object, The presence / absence of additional reference data set based on the existence estimation of the corrected portion is checked (# 12). If additional reference data exists, the area defined by the additional reference data (additional reference point group) may correspond to the mold correction area of the mold. Here, first, the defect determination when there is no additional reference data will be described.

  If the additional reference data does not exist in the check of step # 12 (# 12 Yes), the edge measurement point group obtained by the edge extraction routine and the edge reference point data read from the reference data storage unit 81 are used. The used edge alignment routine is executed (# 13).

  Prior to describing the alignment routine, an alignment algorithm between the measurement point group and the reference point group employed in the present invention will be schematically described with reference to FIG. In this alignment algorithm, a method based on the ICP algorithm is adopted as a sequential convergence process for aligning the reference point group and the measurement point group. In this ICP algorithm, as schematically shown in FIG. 18, each point of the reference point group (reference point data) is set to the corresponding point of the closest measurement point group (measurement point data), and each corresponding point The congruent transformation parameter that minimizes the sum of squares of the distance is estimated and converged sequentially. In such alignment by the ICP algorithm, in the case of shape measurement in which measurement points are determined at an equal measurement scan pitch as shown in FIG. 18, a slight scan pitch shift may occur depending on the orientation of the measurement surface. This leads to a large shift of measurement points (in FIG. 18, measurement points S4, S5, S6 apply to this). Therefore, it is not preferable to use a measurement point at which an error between an assumed position coordinate and an actual position coordinate is expected to be large and a reference point corresponding thereto as it is for the successive convergence evaluation in the successive convergence process. In order to avoid such problems, the improved ICP algorithm applied in the present invention incorporates weighting factors as described below.

  In FIG. 18, a reference point group (reference point data): a reference point included in M is indicated by m, and a subscript (natural number) is given corresponding to the measurement order. Measurement point group (measurement point data): The measurement point included in S is indicated by s, and a subscript (natural number) is given corresponding to the measurement order in the same manner as the reference point. In the sequential convergence process using the conventional ICP algorithm for positioning the measurement point group over the reference point group, the closest reference point in the reference point group is set as the corresponding point for each measurement point in the measurement point group. A joint transformation parameter (R, t) that minimizes the sum of squares of the distances between corresponding points is obtained. Here, R is a rotation matrix and t is a translation vector. At that time, among the corresponding combinations of reference points and measurement points, if the combination of the above-mentioned measurement points where a large error is expected is handled in the same way as other combinations, there is a possibility that it will be affected by errors that cannot be ignored. is there. Therefore, the distance weight coefficient between adjacent points is assigned based on the distance between adjacent points of the measurement point or the reference point, and the adverse effect is suppressed. Here, the weighting coefficient for the distance between the corresponding points is calculated based on the distance between the adjacent points between the reference points. Of course, the distance between the corresponding points is determined based on the distance between the adjacent points between the measurement points. A weighting factor for the distance may be calculated.

Three-dimensional coordinates _{(x i, y i, z} i) the reference point having: m _i between adjacent point distance: d _i is calculated by the Pythagorean theorem, the square is
d _i ² = x _i ² + y _i ² + z _i ²
The distance weight coefficient between adjacent points: γ _i is the weight function Γ,
It is obtained by γ _i = Γ (d _i ² ).
For example, it is convenient that the weight function: Γ is a threshold function as follows.
When d _i ² is a threshold value: d _th or more, γ _i = 0.01
When d _i ² is less than the threshold: d _th γ _i = 1
The threshold value: d _th is determined appropriately according to the characteristics of the reference point group and the measurement point group, and corresponding to the measurement point where the error between the assumed position coordinate and the actual position coordinate is expected to be large. The adverse effect of the reference point on the successive convergence evaluation can be suppressed.

It should be noted that it is useful to use M estimation as disclosed in the above-mentioned non-patent document in the alignment based on such an ICP algorithm. The ICP algorithm in which M estimation is introduced will be briefly described with reference to FIG. The introduction of M estimation is to use the distance weight factor between corresponding points determined based on the distance between corresponding points between the corresponding measurement point and the reference point for the successive convergence evaluation in the successive convergence process. For example, assuming that the distance between corresponding points is e _i , the distance weight coefficient between corresponding points: ρ _i is
ρ _i = Ρ (e _i ) Again, the weight function: Ρ can be a threshold function such as:
When | e _i | is less than the setting width: Bi,
^{_{ρ i = (B i 2/}} 2) (1- (1- (e i / B i) 2)
When | ei | exceeds the set width Bi
^{_{ρ i = (B i 2/}} 2).

When the distance weight coefficient between adjacent points: γ _i and the distance weight coefficient between corresponding points: ρ _i are also used for successive convergence evaluation, the total weight coefficient: w is the distance weight coefficient between adjacent points: γ _i and the corresponding points. It can be derived from a function having a distance weight coefficient: ρ _i as a parameter. Thus, sequential convergence process sequential convergence evaluation value in: J, if the number of corresponding points is N, between the corresponding point distance: e _i and between the corresponding point distance weighting factor: [rho _i with the neighboring point distance weight factor: gamma An evaluation function with _i as a parameter can be derived using H.
J = (1 / N) ΣH (e _i , ρ _i , γ _i )
In order to simplify the calculation, if the total weight coefficient: wi is multiplied by each weight coefficient,
J = (1 / N) ΣH (e _i , w _i ), w _i = ρ _i × γ _i
It becomes.

Next, a basic method for efficiently determining a specific surface defect such as turning up from a reference point group (reference point data) and a measurement point group (measurement point data) aligned using the algorithm described above. The algorithm will be described with reference to the lower schematic diagram of FIG.
First, between the corresponding point distance: e _i is preset to have the corresponding point distance threshold (first threshold): extracts the continuous measurement point group to be T _A or as erroneous corresponding area. In FIG. 2, measurement points S ₄ , S ₅ , S ₆ are extracted. Further, the distribution density of the measurement points belonging to the erroneous correspondence area is calculated. In order to easily calculate the distribution density, for example, the distance between adjacent points of the measurement points: d _i can be used. In other words, adjacent point distance: d _i between the adjacent points is preset distance threshold (second threshold) regarded as near measurement point the measurement point as a T _L or more. As a result, corresponding points (neighboring measurement points) having a distribution density greater than or equal to a predetermined value are extracted. Further, it is checked whether or not the number of neighboring measurement points is equal to or greater than a preset neighboring point threshold value (third threshold value): T _C. When the number of neighboring measurement points is larger than the neighboring score threshold, a region defined by this neighboring measurement point group, that is, a surface region including this neighboring measurement point group is determined as a surface defect.

In the edge alignment routine (# 13) based on the above-described alignment algorithm, as shown in FIG. 11, first, edge measurement point data is read and the edge measurement point group is developed in the memory ( In step # 200, the edge reference point data is read and the edge reference point group is developed in the memory (# 201). The measurement point and the reference point are associated with each other so as to have the smallest distance between corresponding points (# 202). A weighting factor based on the distance between corresponding points and a weighting factor based on the distance between adjacent points are calculated and multiplied to obtain a total weighting factor (# 203). Further, a convergence evaluation value is calculated sequentially using the total weight coefficient and the distance between corresponding points (# 204). Based on the calculated successive convergence evaluation value, it is checked whether or not the associated corresponding point group is successively converged (# 205). When convergence does not occur (# 205 No branch), the process returns to step # 202 to associate the measurement point with the reference point again. In the case of convergence (# 205 Yes branch), a congruent transformation parameter: E (R, t) is generated (# 206) so that the corresponding point group matches as much as possible. The position coordinate of the edge measurement point group is converted using the generated joint conversion parameter, and the edge measurement point group and all the measurement point groups are moved (# 207, # 208). That is, the entire measurement point group is moved using the joint conversion parameter generated using the edge measurement point group. An average correspondence distance is calculated from the measurement point group and the reference point group after movement (# 209), and the calculated average correspondence distance is compared with a preset threshold value to determine whether the end condition is satisfied. Check (# 210). It is convenient to additionally set an upper limit on the number of repetitions in this check step. When the termination condition is not satisfied (# 210 No branch). Returning to step # 202, the measurement point after movement and the reference point are associated with each other. If the end condition is satisfied (# 210 Yes branch), this edge alignment routine is ended.

  When the edge alignment routine (# 13) ends, the all-point alignment routine is executed (# 14). This all-point alignment routine is shown in FIG. In this all-point alignment routine, after all the measurement point forces are moved using the joint transformation parameters generated in the edge alignment routine, a precise alignment is further performed using all measurement point groups and all reference point groups. It is preferable to execute it to do it. Further, when there is no clear edge in the measurement block or when satisfactory edge measurement data cannot be obtained, the edge alignment routine is omitted and only the all-point alignment routine is performed. Compared with the edge alignment routine (# 13), the all-point alignment routine (# 14) has a relationship between the edge measurement point group and the edge reference point group between the all measurement point group and the all reference point group. However, since the substantial processing procedure is the same, description thereof is omitted here.

  When the alignment routine is completed, the defect evaluation module 6 executes a defect determination routine for detecting a surface defect area using the surface defect determination algorithm described above (# 15). This defect determination routine is shown in FIG. First, using a threshold value between corresponding points, a process of extracting a collection of measurement points (specific measurement point group) whose distance between corresponding points is equal to or greater than the threshold value as a miscorresponding region (# 400). . If no erroneous correspondence area is extracted (# 401 No branch), this defect determination routine is terminated. When an erroneous correspondence region is extracted (# 401 Yes branch), a set of points whose distance between adjacent points between measurement points included in the erroneous correspondence region is equal to or smaller than a preset threshold value between adjacent points is set as one set. Label as a body (# 402). Further, a feature amount (defect feature amount) is calculated for each aggregate to which a label is assigned (# 403). Examples of the feature amount include the center of gravity of the aggregate as a defect possibility, the equivalent ellipse major axis length, the equivalent ellipse major axis angle, and the fillet diameter. When the feature amount is larger than the set value (# 404 Yes branch), it is regarded as a defect, and the feature amount is recorded in the defect information storage unit 82 as a defect together with the three-dimensional position and the like. If the feature amount is smaller than the set value (# 404 No branch), it is not regarded as a defect, the feature amount is not recorded, and this routine ends.

Next, a shape difference determination routine based on mold correction is performed to determine whether or not the area where the assembly determined to be a defect in the defect determination routine is an area that is regarded as a mold correction location corrected by the mold. The shape difference determination routine may specifically take various forms, but here, two forms will be described.
First, the shape difference determination routine (# 16A) of the first embodiment is shown in FIG. The target shape of the measurement object manufactured by the mold before the mold correction can be set based on the reference point group. Further, the deviation between the measurement point group and the reference point group at the mold correction position in the measurement object manufactured by the mold after the mold correction occurs at the same position. Therefore, among the defects determined by the defect determination routine, the defects that are continuously generated at the same location are the deviations from the target shape due to the mold correction, and are not regarded as defects, and can be distinguished from each other. This is the gist of the shape difference determination routine (# 16A). Therefore, first, the feature amount of the defect determined by the previous measurement object is read (# 500), and the defect feature amount determined by the current measurement object is read (# 501). Within a predetermined range of feature values selected from the defect feature amounts of defects at substantially the same position as the previous time and this time, for example, the center of gravity of the defect, the equivalent ellipse major axis length, the equivalent ellipse major axis angle, the fillet diameter, etc. It is checked whether they match (# 502). If they match (# 502 Yes branch), it is determined that the defective part is a mold correction part (# 503). If they do not match (# 502 No branch), it is determined that the defect part is not a mold correction part. (# 504), this routine is terminated.

  The shape difference determination routine (# 16B) of the second embodiment is shown in FIG. This shape difference determination routine is manufactured with the shape of the mold correction location (that is, the measurement shape defined in the measurement point group) in the measurement object manufactured with the mold after the mold correction, and the mold before the mold correction. The fact that the shape itself is similar to the target shape of the measurement object (that is, the measurement shape defined in the measurement point group) is used. Therefore, first, reference point data including edge reference point data and all reference point data is read (# 600), and defect information including the feature amount of the defect determined by the current measurement object is read (# 601). . Based on the feature amount of the defect and the reference point data corresponding to the defect, the defect is generated on a vertical side surface (generally parallel to the measurement light) or the bottom or top surface (generally to the measurement light) It is determined whether it has occurred (# 602). In the case of a vertical side surface (# 602 Yes branch), an edge reference point group corresponding to an area equivalent to the defect portion obtained from the defect information is extracted from the reference point data (# 603), and edge alignment limited to the defect area is performed. A routine is executed (# 604). If it is not a vertical side surface (# 602 No branch), all reference point groups corresponding to the same area as the defect portion obtained from the defect information are extracted from all reference point data (# 605), and all positions limited to the defect area A matching routine is performed (# 606). That is, here, the alignment is limited to a narrow area called a defect area, but the edge alignment routine is applied to the defect area on the vertical side surface where the measurement point data is likely to be unstable.

  When the alignment by any of the alignment routines is completed, the similarity between the reference point plane defined by the reference point group and the measurement point plane defined by the measurement point group is calculated (# 607). This similarity is calculated using, for example, the equivalent ellipse major axis length of the defect region, the average correspondence distance, and the like as parameters. It has been experimentally confirmed that it is preferable that the similarity function is set so that the similarity becomes higher as the equivalent ellipse major axis length becomes larger and the similarity becomes lower as the average corresponding distance becomes larger. . The calculated similarity is compared with a predetermined threshold (# 608). If the similarity is high (# 608 Yes branch), it is determined that the defective part is a mold correction part (# 609), and if the similarity is low (# 609 No branch), it is determined that the defective part is not a mold correction part. (# 610), and this routine is finished.

  Returning to the flowchart of the evaluation calculation process in FIG. 9, when the shape difference determination routine (# 16A or # 16B) by the above-described mold correction ends, is there a place that is regarded as a mold correction place in the shape difference determination routine by the mold correction? A check is made (# 17). When there is a mold correction location (# 17 Yes branch), the type correction location information is recorded in the type correction location information storage unit 83, and the area of the type correction location or the entire measurement block including the type correction location is recorded. Are newly registered in the reference data storage unit 81 as new reference point data (# 18), and this routine is terminated. Note that the mold correction location information storage unit 83 may also be used as the reference data storage unit 81.

  As described above, when the mold correction location information or the additional reference data is recorded in Step # 18, the No branch is made in the check of Step # 12 in the subsequent evaluation calculation processing for the measurement object. When No branch is made here, the alignment using the edge measurement point data and the edge reference point data before the mold correction using the above-described alignment algorithm, the edge measurement point data and the edge reference point data after the mold correction are used. Alignment is executed (# 19). Further, alignment using all measurement point data and all reference point data before mold correction, and alignment using all measurement point data and all reference point data after mold correction are executed (# 20).

When the alignment is completed, a defect / mold correction location determination routine is executed to determine whether the product is a defect or a mold correction location even if products before and after mold correction are mixed (# 21).
This defect / die correction point determination routine (# 21) is shown in FIG. First, from the reference point data before mold correction, a set of measurement point groups (specific measurement point group) where the distance between corresponding points is equal to or greater than the threshold using the distance threshold between corresponding points is set as a miscorresponding region. Extract (# 700). The extracted miscorresponding area is subjected to defect detection by a process similar to the defect determination process shown in FIG. 13, and when a defect is found, it is recorded in the defect information storage unit 82 as defect information: E1 (# 705). Since steps # 701 to # 704 here correspond to steps # 401 to # 404 of the defect determination process, description thereof is omitted here.

  Furthermore, from the reference point data after mold correction, a set of measurement point groups (specific measurement point group) where the distance between corresponding points is equal to or greater than the threshold value is extracted as a miscorresponding region using the distance threshold between corresponding points (# 706). Processing similar to steps # 701 to # 704 (steps # 707 to # 710) is performed on the extracted erroneous correspondence area, and when a defect is found, defect information: E2 is stored in the defect information storage unit 82. It is recorded (# 711).

If the defect found in the process so far is not a shape difference due to mold correction but a real defect, it can be found as a defect from the defect determination process using both reference point data before and after mold correction. . Therefore, the defect determined that the defect information: E1 and the defect information: E2 belong to the common area within the predetermined range (# 712 Yes branch) is finally determined as a defect (# 714). In addition, if the defect found in the process so far is found as a defect only from the defect determination process using the reference point data before the mold correction, it is considered that the shape is not a real defect but a shape difference due to the mold correction. . Thus, defect information: E1 only determined to belong to the region (# 713Yes branches) have been defective portion is determined by the type of correcting portions, measured product measurement work number is the number is counted up (# 715) .

  When the defect / mold correction location determination routine is completed and the flow returns to the flowchart of the evaluation calculation process in FIG. 9, whether the number of products after mold correction is equal to or less than the set value, that is, products before and after mold correction are mixed. Is determined (# 22). If it is equal to or smaller than the set value (# 22 Yes branch), it is determined that the product before and after the mold correction is mixed, and this routine is terminated. If it exceeds the set value (# 22 No branch), it is considered that there is no longer any product before the mold correction, and the reference data before the mold correction is deleted (# 23). Further, the number of measured workpieces, which is the number of measured products, is reset (# 24), and this evaluation calculation processing routine is terminated.

By performing object shape evaluation processing as described above, especially edge alignment, the measurement technology for measuring objects with insufficient measurement point data from the side or bottom, etc., where sufficient reliability could not be obtained with the conventional technology. However, the shape can be evaluated with high reliability. For example, the result of an experiment performed on an artificial measurement object having no side surface is shown in the graph of FIG. In the alignment using only the entire alignment, even if the alignment is repeated 10 times or more, the corresponding point average distance of 30 microns or more remains. On the other hand, in the alignment that combines the edge alignment and the total alignment, an effective alignment is realized such that the corresponding point average distance is substantially equal to zero from the initial alignment.
Further, in the case of a mold product such as an aluminum die cast, even if the shape changes due to wear of the mold, the measurement point data of a normal product can be sequentially registered and changed as reference point data. With this, even if the shape of the mold changes due to mold correction, etc., even if the product before mold correction and the product after mold correction are mixed, a highly reliable object is introduced by introducing a process to distinguish the mold correction location from the defect. Shape evaluation is obtained.

Hereinafter, another embodiment is illustrated.
(1) The edge reference point data may not be generated in advance and stored in the reference data storage unit, but may be generated from all reference point data by edge detection filter processing or the like when used.
(2) In the above-described embodiment, the ICP algorithm with weight calculation is employed as the alignment algorithm. However, a general ICP algorithm may be employed, or other alignment algorithms may be employed.
(3) In the above embodiment, the distance weighting factor between adjacent points is calculated in the alignment routine. However, when this distance weighting factor between adjacent points is obtained based on the distance between adjacent points of the reference point group, it is calculated in advance. The calculation speed is increased by making a table in advance.
(4) In the above embodiment, many threshold values set in advance as the determination conditions are used. However, the threshold values are not set in advance, and the surface shape of the measurement object or the measurement result A configuration may be adopted in which the threshold value is calculated from the statistical characteristics.
(5) In the above embodiment, the weight function: Γ _i for obtaining the distance weight coefficient between adjacent points: γ _i is a binary function determined according to the distance between adjacent points: d _i , but the following (a), Other weight functions as listed in (b) and (c): Γ can also be used.
Here, the distance between adjacent points normalized by the measurement resolution is defined as follows,
Di ² = (x _i / Δx) ² + (y _i / Δy) ² + (z _i / Δz) ²
Here, Δx: measurement resolution in the x direction Δy: measurement resolution in the y direction Δz: measurement resolution in the z direction.
(a) γ _i = Γ (p) = 1 / p ² , where p = d _i ² or p = D _i ²
(b) γ _i = Γ (p) = 1 / p, where p = d _i ² or p = D _i ²
(c) γ _i = Γ (p) = 1 / exp (p), where p = d _i ² or p = D _i ²
As described above, various distance weight coefficients between adjacent points: γ _i can be used.
FIG. 20 shows a normal M-estimated ICP (black diagram plot line) that does not use a weight coefficient, a weight coefficient (white square plot line) in the binarization function used in the above-described embodiment, and the above (a ), (B), and (c) show the results of the comparison operation of repeated alignment performed. The weighting factor in (a) is the reciprocal of the square of the distance between adjacent points, and is indicated by a white triangular plot line, and the weighting factor in (b) is the reciprocal of the distance between adjacent points, indicated by a peke plot line, The weighting factor in (c) is indicated by the exp function and indicated by a black circle plot line. In FIG. 20, the horizontal axis indicates the number of repetitions of alignment, and the vertical axis indicates a value corresponding to the corresponding point average distance as an evaluation value for comparative evaluation. From this comparison calculation result, the distance weighting coefficient between adjacent points in (a) is the best evaluation value, and the matching error in alignment is small. At that time, p = a p = D _i ² instead of d _i ^2, i.e. using a normalized between adjacent point distance further verification error without being affected by the resolution can be expected be less in measurement resolution.

  The present invention can be used for object shape evaluation based on measurement point data using various three-dimensional measurement apparatuses.

4: Evaluation unit 5: Surface evaluation module (shape evaluation means)
6: Defect evaluation module (shape evaluation means)
7: Reference data change module 8: Data storage unit 81: Reference data storage unit 82: Defect information storage unit 83: Mold correction information storage unit 50: Registration processing means 52: Edge detection unit

Claims (5)

A measurement data input unit for inputting measurement point data including position information of a large number of measurement points corresponding to the shape of the measurement object, and a reference point including position information of a large number of reference points corresponding to the reference shape of the measurement object A reference data storage unit for storing data, and an alignment processing means for aligning the measurement point and the reference point based on a sequential convergence process for sequentially converging the distance between the corresponding measurement point and the reference point; Shape evaluation means for evaluating the shape of the measurement object based on the measurement point data and the reference point data after the alignment processing by the alignment processing means,
The alignment processing means includes an edge point group generated by an edge detection unit that detects an edge point corresponding to an edge of the measurement object from measurement point data, and the measurement object stored in the reference data storage unit Edge position alignment is performed for aligning an edge reference point group that is edge reference point data for an object edge, and measurement points of the entire measurement point data are obtained using edge congruence conversion parameters obtained by the edge alignment. There have rows aligned toward the reference point,
A weight coefficient determining means for determining a distance weight coefficient between adjacent points based on a distance between adjacent points between adjacent measurement points or a distance between adjacent points between adjacent reference points; An object shape evaluation apparatus used when a weighting factor is used to obtain a sequential convergence evaluation value in the sequential convergence process .   2. The object according to claim 1, wherein after the alignment of the measurement point toward the reference point using the edge joint conversion parameter, the alignment of the measurement point of the entire measurement point data and the corresponding reference point is further performed. Shape evaluation device. The weighting coefficient determining means determines a distance weighting coefficient between corresponding points based on a distance between corresponding points between the corresponding measurement point and the reference point, and the distance weighting coefficient between corresponding points is the alignment processing means. There object shape evaluation apparatus according to claim 1 or 2 used to obtain a sequential convergence evaluation value in the sequential convergence process. The object shape evaluation apparatus according to claim 3 , wherein the distance weight coefficient between adjacent points is determined based on a distance between adjacent points of the reference point. The measurement point data, in any one of claims 1 is a three-dimensional position data retrieved from the acquired distance image by processing the captured image of the measurement object illuminated by the slit light 4 The object shape evaluation apparatus described.

Images