JP2008032449A - Method and device for measuring three dimensional shape and calibration matter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for measuring three dimensional shape capable of measuring the three dimensional shape with a good precision even though the deviation from the parallel between the coordinate axes of the camera coordinate system and the rotation axis of the object to be measured is remaining a little. <P>SOLUTION: Before measuring the shape of the object to be measured, a device for measuring the three dimensional shape calibrates the deviation between the camera coordinate system and the reference coordinate system using the reference work. Regarding the remaining deviation after calibration, the affection of the coordinate conversion is eliminated by the angle correction matrix Mθ. After the coordinate conversion by the Mθ, the coordinate conversion is performed by the axis matching matrix MA for making the rotation axis of the measurement object included in a coordinate plane of the coordinate system. The converted coordinate values for every prescribed rotation angle of the measurement object OB are composited, and the whole external shapes of the measurement object are obtained. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a method and apparatus for irradiating a measurement object with light and measuring the three-dimensional shape of the measurement object based on information obtained from the reflected light, and measuring the three-dimensional shape of the measurement object. The present invention relates to a calibration object used for the above.

  When measuring the three-dimensional shape of the measurement object by irradiating the measurement object with light and receiving the reflected light, the three-dimensional shape of the entire measurement object is obtained by measuring while rotating the measurement object. Can be measured. Such a measuring method is in principle the same as a method of measuring the three-dimensional shape of the entire measuring object by fixing the measuring object and rotating the shape measuring device around the measuring object.

  For example, in Patent Document 1, a three-dimensional shape is measured from various angles by irradiating a measurement target with line light from a predetermined angle and rotating the measurement target, and from this three-dimensional shape data and rotation angle data, Alignment of 3D shape data for each rotation angle is performed. After the alignment, the position of the rotation axis in the coordinate system of the shape measuring device is changed to correct the positional deviation of the three-dimensional shape data for each rotation angle, and the measurement object The three-dimensional data showing the three-dimensional shape of is obtained.

In the shape measurement method described in Patent Document 1, since the rotation axis of the measurement object is orthogonal to the light irradiation plane, it is necessary to move the line light in the height direction of the measurement object. On the other hand, in the shape measuring apparatus described in Patent Document 2, since the rotation axis of the measurement object is substantially parallel to the irradiation plane of the line light, the irradiation position of the line light is fixed and the measurement object is rotated. It is possible to measure the three-dimensional shape of the entire measurement object simply by doing. For this reason, a means for moving the line light is not required, and the manufacturing cost of the shape measuring apparatus can be reduced.
JP-A-10-074272 JP 05-272927 A

  In the shape measuring apparatus described in Patent Document 2, accurate shape measurement cannot be performed when the coordinate axis of the camera coordinate system and the rotation axis of the measurement object are deviated from parallel. In this case, even if the operator adjusts the posture of the shape measuring instrument so that there is no parallel shift between the coordinate axis of the camera coordinate system and the rotation axis of the object to be measured, it should be completely free of deviation. It is difficult. Therefore, there is a problem that the accuracy of the three-dimensional shape measurement is deteriorated due to a slight parallel shift.

  The present invention has been made to address the above-described problems, and its purpose is to achieve good accuracy even when some deviation from the parallel between the coordinate axis of the camera coordinate system and the rotation axis of the measurement object remains. 3D shape measurement method and 3D shape measurement device capable of measuring 3D shape with the above, and a reference coordinate system in which the rotation axis of the measurement object and one coordinate axis are parallel when performing the above method Another object of the present invention is to provide a calibration object used for detecting the amount of deviation between the reference coordinate system and the camera coordinate system.

  In order to achieve the above object, the present invention is characterized in that the measurement object is rotated and the line light is substantially parallel to the rotation axis of the measurement object from the shape measuring device capable of emitting the line light toward the measurement object. Is a three-dimensional shape measurement method for measuring the three-dimensional shape of an object to be measured by receiving reflected light of the irradiated light, which is a coordinate system defined by a shape measuring instrument and whose coordinate plane is line light. A displacement amount detecting step for detecting a displacement amount between a camera coordinate system parallel to the irradiation plane of the light source and a reference coordinate system in which one coordinate axis is parallel to the rotation axis of the measurement object; and An irradiation step of irradiating the line light from the shape measuring instrument toward the measurement object while rotating around, a light receiving step of receiving reflected light of the line light irradiated on the measurement object, and a predetermined measurement object For each rotation angle Based on the light reception information obtained in the light receiving step, coordinate values indicating the position on the surface of the measurement object irradiated with the line light are set for each predetermined rotation angle as a group of coordinate values in the camera coordinate system. A coordinate value acquisition step to be acquired, and a group of correction coordinate values obtained by performing coordinate conversion on the basis of the deviation amount of the group of coordinate values acquired for each of the predetermined rotation angles in the coordinate value acquisition step. A coordinate conversion step acquired for each predetermined rotation angle, and a group of the correction coordinate values for each predetermined rotation angle obtained in the coordinate conversion step are synthesized based on the predetermined rotation angle, and a measurement object A three-dimensional shape measuring method including a synthesizing step of acquiring a group of synthesized coordinate values indicating the three-dimensional shape.

  Another feature of the present invention is that a rotation driving means for rotating the measurement object around the rotation axis, a rotation angle detection means for detecting a rotation angle of the measurement object rotated by the rotation driving means, A shape measuring device that emits line light substantially parallel to the rotation axis of the measurement object, receives reflected light of the line light irradiated on the surface of the measurement object, and outputs light reception information, and the rotation angle detection A three-dimensional shape measuring apparatus including a rotation angle detected by the means and a data processing apparatus to which the light reception information output from the shape measuring instrument is input, wherein the data processing apparatus is input from the rotation angle detecting means The coordinate value indicating the position on the surface of the measurement object irradiated with the line light based on the received light information input from the shape measuring device at every predetermined rotation angle of the rotation angle to be Measurement A coordinate value acquisition means for acquiring at each predetermined rotation angle as a group of coordinate values in a camera coordinate system which is a coordinate system defined by a vessel and whose irradiation plane of line light is parallel to one coordinate plane; and the coordinate values The coordinate value group acquired for each predetermined rotation angle by the acquisition means is coordinate-converted based on a deviation amount between a reference coordinate system defined by a calibration object stored in advance and the camera coordinate system, A coordinate conversion unit that obtains a group of corrected coordinate values in which the shift amount is corrected for each predetermined rotation angle, and a group of the corrected coordinate values for each predetermined rotation angle obtained by the coordinate conversion unit. And synthesizing means for synthesizing based on the predetermined rotation angle and acquiring a group of synthesized coordinate values indicating the three-dimensional shape of the measurement object.

  According to the above invention, before actually measuring the three-dimensional shape of the measurement object by the shape measuring instrument, the camera coordinate system defined by the shape measuring instrument and the reference coordinate system defined by the calibration object Obtain the amount of deviation. Thereafter, the shape measuring instrument irradiates the measurement target with line light, and acquires a group of coordinate values (point group data) indicating the surface shape of the measurement target in the camera coordinate system. Further, a group of corrected coordinate values obtained by performing coordinate conversion on the acquired point group data based on the deviation amount is acquired for each rotation angle of the measurement object. Then, a group of correction coordinate values acquired for each rotation angle of the measurement object is synthesized based on the rotation angle at that time, and the three-dimensional shape of the entire measurement object is measured. In this way, the amount of deviation between the camera coordinate system and the reference coordinate system is acquired before measuring the shape, and after measuring the shape of the measurement object, the measurement is performed by correcting based on the amount of deviation. An accurate three-dimensional shape of the object can be measured.

  In the above-mentioned invention, the line light is light in which a light passing region emitted from the shape measuring instrument is a flat surface (this plane is also referred to as an irradiation plane in this specification), and linear light is parallel. Or can be formed by swinging around the light source, or linear light can be spread in a planar shape by a cylindrical lens. The camera coordinate system is a coordinate system defined by the shape measuring instrument, and the irradiation plane of the line light is a coordinate plane (for example, a plane including two coordinate axes (for example, the x axis and the z axis) of the camera coordinate system (for example, parallel to the xz plane). Preferably, the irradiation plane of the line light is one coordinate plane (for example, xz plane) of this camera coordinate system. The reference coordinate system is a coordinate system defined by a calibration object. In the present invention, one coordinate axis (for example, the x axis) of the reference coordinate system may be parallel to the rotation axis of the measurement object. Further, the shift amount includes an angle shift between the camera coordinate system and the reference coordinate system, and a position shift. The angle shift refers to a shift angle between the direction of each coordinate axis in the reference coordinate system and the direction of each coordinate axis in the camera coordinate system. Examples of the angular deviation include an angular deviation θx around the x axis, an angular deviation θy around the y axis, and an angular deviation θz around the z axis. The positional deviation refers to a deviation from a case where the coordinate axis origin of the reference coordinate system and the coordinate axis origin of the camera coordinate system are in a relative positional relationship as a reference. Examples of the positional deviation include a positional deviation Δx in the x-axis direction, a positional deviation Δy in the y-axis direction, and a positional deviation Δz in the z-axis direction.

  The deviation amount detecting step includes a pre-irradiation step of irradiating line light from the shape measuring instrument toward a calibration object having a shape that can define a reference coordinate system, and the calibration step in the pre-irradiation step. A pre-light receiving step for receiving reflected light of the line light irradiated on the object, and a calculating step for calculating a deviation between the camera coordinate system and the reference coordinate system based on the light receiving information obtained in the pre-light receiving step; It is good to include. The calibration object that is irradiated with the line light prior to measuring the shape of the measurement object can define a reference coordinate system. Therefore, the line light is irradiated onto the calibration object, and the reflected light is received. Is calculated and the three-dimensional shape of the calibration object in the irradiation plane is calculated, the deviation between the camera coordinate system and the reference coordinate system can be calculated based on the three-dimensional shape.

  In the present invention, it is only necessary to determine the position of the rotation axis of the measurement object. However, when the position of the rotation axis is not determined, the coordinate value group cannot be coordinate-converted by the coordinate conversion step. In this case, the three-dimensional shape measurement method of the present invention further includes a rotation axis calculation step, and the rotation axis position can be calculated by this rotation axis calculation step. This rotation axis calculation step irradiates line light from a shape measuring instrument toward a fixed point definition object capable of defining a predetermined fixed point from the outer shape, and coordinates of the fixed point from reflected light of the line light irradiated to the fixed point definition object Calculating the value is performed at a plurality of rotation angles while rotating the fixed point defining object by a predetermined angle around the rotation axis of the measurement object, and the measurement object is based on the calculated coordinate values of the plurality of fixed points. The rotation axis is calculated. The calculation method is a method of detecting a fixed point in a fixed point definition object from various angles rotated about the rotation axis center, and calculating the rotation axis center from the locus of the fixed point at that time. By this rotation axis calculation step, the position of the rotation axis of the measurement object is calculated. Since the rotation axis of the measurement object is parallel to one coordinate axis of the reference coordinate system, the position of the rotation axis calculated as described above indirectly represents the deviation between the reference coordinate system and the camera coordinate system. But there is. Therefore, the coordinate value group can be coordinate-converted based on the amount of deviation between the camera coordinate system and the reference coordinate system and the position of the rotation axis of the measurement object.

  In the coordinate conversion step, one coordinate axis (for example, the x axis) of the coordinate system of the camera coordinate system and the measurement object are determined based on an angle shift between the camera coordinate system and the reference coordinate system among the shift amounts. A first coordinate conversion step for converting the group of coordinate values to obtain an angle correction coordinate value so that the rotation axes of the two are parallel, and the relative amount between the camera coordinate system and the reference coordinate system among the deviation amounts Measurement on one coordinate plane (for example, xz plane) of the camera coordinate system based on the deviation of the target position or the amount of deviation and the position of the rotation axis of the measurement object calculated in the rotation axis calculation step A second coordinate conversion step of converting the angle correction coordinate value group so as to include the rotation axis of the object and acquiring the correction coordinate value. According to this, the coordinate value by the camera coordinate system is first converted into the angle correction coordinate value when the direction of the coordinate axis of the camera coordinate system coincides with the direction of the coordinate axis of the reference coordinate system by the first coordinate conversion step. At this stage, the rotation axis of the measurement object is parallel to one coordinate plane (for example, xz plane) of the camera coordinate system, but may not be included in one coordinate plane of the camera coordinate system. If the rotation axis of the measurement object is not included in one coordinate plane of the camera coordinate system, when the measurement object is rotated around the rotation axis and a group of coordinate values is synthesized based on the rotation angle, Since the shift between the rotation axis and one coordinate plane of the camera coordinate system is synthesized without being corrected, the accuracy of the obtained three-dimensional shape is deteriorated. Therefore, in the second coordinate conversion step, the relative position shift between the two coordinate systems, or the shift amount and the rotation axis calculation step, is calculated so that the rotation axis of the measurement object is included in one coordinate plane of the camera coordinate system. The angle correction coordinate value is coordinate-transformed based on the position of the rotation axis of the measurement object. Thereby, when a measurement object is rotated around a rotation axis and a group of coordinate values is synthesized based on the rotation angle, a highly accurate three-dimensional shape can be obtained.

  In this case, the second coordinate conversion step includes a relative position shift between the camera coordinate system and the reference coordinate system, or a position of the rotation axis of the measurement object calculated in the shift amount and the rotation axis calculation step. And rotating the camera coordinate system to coordinate-convert the group of angle correction coordinate values so that the rotation axis of the measurement object is included in one coordinate plane of the camera coordinate system. A value may be acquired. Alternatively, the second coordinate conversion step includes a relative position shift between the camera coordinate system and the reference coordinate system, or the amount of shift and the position of the rotation axis of the measurement object calculated in the rotation axis calculation step. By moving the camera coordinate system based on the above, the group of angle correction coordinate values is transformed so that the rotation axis of the measurement object is included in one coordinate plane of the camera coordinate system, and the correction coordinate value May be obtained. In order to include the rotation axis of the measurement object in one coordinate plane of the camera coordinate system, the coordinate axis of the measurement object is included in the coordinate plane by rotating the coordinate axis of the camera coordinate system and the coordinate axis of the camera coordinate system. Two methods are conceivable, in which the rotation axis of the object to be measured is included in the coordinate plane by translating in the direction of the coordinate axis. With either method, the rotation axis of the measurement object can be included in one coordinate plane of the camera coordinate system.

  The rotation axis calculation step calculates an equation of a circle that passes through the calculated coordinate values of the plurality of fixed points, and measures a straight line that goes straight to the plane including the calculated circle and passes through the center of the calculated circle It can be calculated as the rotation axis of the object. Further, if the fixed point defining object is a sphere and the fixed point is the center of the sphere, the fixed point can be easily defined and the rotation axis can be calculated.

  In addition, as the calibration object used in the deviation amount acquisition step, a first plane portion that defines a z-axis direction whose normal direction is one coordinate axis of the reference coordinate system, and the first plane portion are formed. A second plane part that defines the direction of the x-axis, which is a coordinate axis that is orthogonal to the z-axis, and a plane that is orthogonal to the z-axis and on which the first plane part is formed It is preferable to have a third plane portion and a fourth plane portion that form intersecting lines on different planes, and that the x-axis is parallel to the rotation axis of the measurement object. In the shift amount detection step, the first plane portion, the second plane portion, the third plane portion, and the fourth plane portion are line light based on the light reception information obtained in the preliminary light reception step. The equation of the straight line including the cutting line when irradiating is selectively calculated according to the amount of deviation to be obtained. Since each plane portion can define any coordinate axis of the reference coordinate system, how the line light is irradiated onto such a plane portion can be obtained from the above linear equation. Therefore, the shift amount can be detected based on the linear equation. In this case, the normal direction of the first plane part and the second plane part may be perpendicular to the rotation axis of the measurement object. The intersection line between the third plane part and the fourth plane part may be parallel to the y-axis in the reference coordinate system.

  Further, the calibration object is parallel to the first plane part and is formed at a position spaced apart from the first plane part by a predetermined distance, and is parallel to the second plane part. It is more preferable to have an additional second plane part formed at a position spaced apart from the second plane part by the predetermined distance. If the calibration object is used, in the deviation amount detection step, based on the light reception information obtained in the preliminary light reception step, the first plane portion, the second plane portion, the additional first plane portion, An equation of a straight line including a cutting line when the line light is irradiated to the additional second plane part is obtained on the camera coordinate system, and then a straight line including the cutting line of the line light irradiated to the first plane part and the above The distance p1 from the straight line including the cutting line of the line light irradiated to the second plane part and the straight line including the cutting line of the line light irradiated to the additional first plane part and the additional second plane part are irradiated. The distance p2 from the straight line including the cut line of the line light is obtained, and the shift amount can be uniquely determined based on the ratio of the distance p1 and the distance p2.

  Hereinafter, embodiments of the present invention will be described. The three-dimensional shape measurement method in the present embodiment is divided into a calibration step for calibrating the displacement of the shape measuring device in advance and a shape measurement step for actually measuring the shape of the measurement object. A calibration process is performed, followed by a shape measurement process. First, the calibration process will be described. FIG. 1 shows an overall schematic diagram of a shape measurement system when a calibration process is performed. As shown in FIG. 1, the shape measuring system of the present embodiment includes a shape measuring device 10 and an attitude control device 20. The shape measuring instrument 10 measures the three-dimensional surface shape of an object located on the front side of the object and outputs information representing the measured three-dimensional surface shape. For example, according to the triangulation method using laser light. It measures the three-dimensional surface shape of an object.

  In the shape measuring instrument 10, irradiation means for irradiating a measurement target with laser light is housed in a housing 11. The irradiating means scans the laser light emitted from the light source based on the emission command from the controller 12 by rotating a galvano mirror, a polygon mirror or the like, or creates line light by passing it through a cylindrical lens or the like. Irradiate the line light toward the measurement object. The light receiving means for receiving the reflected light of the laser light irradiated on the measurement object is also housed in the housing 11. As this light receiving means, a light receiving sensor in which a CCD (charge coupled device) or the like is arranged is used.

  The posture control device 20 includes a support unit 21, a position adjuster 22 attached to the upper end of the support unit 21, and an angle adjuster 23 disposed on the upper surface of the position adjuster 22. The column unit 21 includes an outer cylinder 211 and an inner cylinder 212 inserted into the outer cylinder 211 so as to be movable in the axial direction. The inner cylinder 212 is rotatable in the direction around the axis within the outer cylinder 211. A screw hole 211a is formed in the side periphery of the outer cylinder 211, and a bolt 211b is attached to the screw hole 211a. Therefore, the axial movement and rotation of the inner cylinder 212 in the outer cylinder 211 can be restricted by rotating the bolt 211 b and bringing the tip of the bolt 211 b into contact with the outer periphery of the inner cylinder 212.

  As shown in the figure, the position adjuster 22 is configured in a rectangular parallelepiped shape, and a first slit 221 and a second slit 222 are formed on two side surfaces of the four side surfaces. In addition, a first lever 223 and a second lever 224 are attached to the position adjuster 22, and the first lever 223 is connected to a link mechanism (not shown) incorporated in the position adjuster 22 through the first slit 221. The second lever 224 is connected to the link mechanism through the second slit 222. The link mechanism is connected to the lower surface of the angle adjuster 23 so that the angle adjuster 23 can be moved in the front-rear and left-right directions. Therefore, by moving the first lever 223 in the horizontal direction along the first slit 221, the angle adjuster 23 connected to the link mechanism is against the measurement object facing the front of the shape measuring instrument 10. Move left and right. Further, by moving the second lever 224 in the horizontal direction along the second slit 222, the angle adjuster 23 moves in the front-rear direction with respect to the measurement object.

  In the present embodiment, the angle adjuster 23 is formed by overlapping two disk-shaped members (a lower disk member 231 and an upper disk member 232) at a predetermined interval. Both the lower disk member 231 and the upper disk member 232 have the same disk shape. The bottom disc member 231 has three bottomed holes 231a formed at equal intervals in the circumferential direction. The bottomed hole 231a opens at the upper surface of the lower disk member 231 and is closed at the lower surface. Three screw holes 232a are formed through the upper disk member 232 at equal intervals in the circumferential direction. The bottomed holes 231a and the screw holes 232a are aligned in the vertical direction between the bottomed holes 231a and the screw holes 232a when the lower disk member 231 and the upper disk member 232 are vertically stacked. The lower disk member 231 and the upper disk member 232 are respectively formed so as to be able to take a state to be performed. And both the disk members 231 and 232 are overlapped so as to be in the above state. Further, a bolt 233 is attached to the bottomed hole 231a and the screw hole 232a whose axial centers coincide with each other in the vertical direction.

  As the bolt 233 rotates, the screwed state of the bolt 233 and the screw hole 232a changes, and the relative height position of the upper disk member 232 with respect to the lower disk member 231 is changed. Therefore, the inclination of the upper disk member 232 is adjusted by independently rotating the bolts 233 attached to the bottomed holes 231a and the screw holes 232a. As shown in the figure, since the shape measuring instrument 10 is placed on the upper surface of the upper disk member 232, the inclination of the shape measuring instrument 10 with respect to the measurement object can be adjusted by rotating each bolt 233. .

  The measurement object is a reference work (calibration object) as shown in the figure in the calibration process. The reference workpiece 40 includes a shaft part 41, a support part 42, and a measurement part 43. The shaft portion 41 has a long rod shape, and support portions 42 are attached to both ends thereof. The support part 42 is substantially T-shaped as shown in the figure. Further, the reference workpiece 40 is supported by two long plate members 31, 31 arranged facing each other. The long plate-like members 31, 31 are formed in a rectangular parallelepiped shape so as to stand in the vertical direction, and a cutout portion 31a having a semicircular cross section is formed on the upper surface thereof. The support portion 42 of the reference workpiece 40 has a T-shaped vertical portion arranged in the cutout portion 31 a of the long plate-shaped member 31, and a lower surface of the T-shaped horizontal portion of the long plate-shaped member 31. It is placed on the top surface. Accordingly, the reference workpiece 40 is supported by the long plate-like member 31 so as not to rotate, and the orientation of the reference workpiece 40 is determined.

  As shown in FIG. 1, the measuring unit 43 is formed in a flat plate shape, and a surface other than the surface facing the shape measuring instrument 10 is formed flat, but on the surface facing the shape measuring instrument 10. Has six plane portions. These surfaces are a first plane part 43a, a second plane part 43b, a third plane part 43c, a fourth plane part 43d, a fifth plane part 43e and a sixth plane part 43f from the left in the figure, and a shape measuring instrument These surfaces are arranged in a row in the horizontal direction when viewed from the 10 side. 2A is a front view of the measurement unit 43, FIG. 2B is a plan view, and FIG. 2C is a left side view.

  In FIG. 2A, the first flat portion 43a formed at the left end in the drawing and the sixth flat portion 43f formed at the right end in the drawing are formed on the same plane. Further, in the first plane portion 43a and the sixth plane portion 43f, the direction of the normal vector n1 is directed to the shape measuring instrument 10 in a state where the reference workpiece 40 is supported by the long plate member 31 as shown in FIG. They are arranged vertically so as to be in the horizontal direction. Here, the direction of the normal vector n1 defines the direction of the z axis in the reference coordinate system. The sixth plane portion 43f is provided for accurately calculating the first straight line L1 described later, and is omitted if the first straight line L1 can be calculated with high accuracy only by the first plane portion 43a. Also good.

  In FIG. 2A, the second plane part 43b formed on the right side of the first plane part 43a and the fifth plane part 43e formed on the left side of the sixth plane part 43f are formed on the same plane. ing. Further, as shown in FIG. 1, the second plane part 43 b and the fifth plane part 43 e are inclined in the z-axis direction of the reference coordinate system defined by the reference workpiece 40 from the upper end to the lower end in the figure. In the present embodiment, as shown in FIG. 1, the lower end sides of the second plane portion 43 b and the fifth plane portion 43 e are inclined so as to protrude from the upper end side with respect to the shape measuring instrument 10, but this relationship is reversed. But it ’s okay.

  Further, when the plane on which the first plane portion 43a and the sixth plane portion 43f are formed is the virtual plane A, and the plane on which the second plane portion 43b and the fifth plane portion 43e are formed is the virtual plane B, the virtual plane A The plane A and the virtual plane B are different planes, and the direction vector n2 of the line of intersection of the planes A and B is in a relationship orthogonal to the normal vector n1 of the first plane portion 43a and the sixth plane portion 43f. That is, the second plane part 43b and the fifth plane part 43e rotate the first plane part 43a and the sixth plane part 43f by a predetermined angle φ (see FIG. 2C) around an axis orthogonal to the z-axis. There is a relationship. For this reason, the direction of the direction vector n2 defines the direction of the x axis orthogonal to the z axis. The direction of the x axis coincides with the axial direction of the shaft portion 41 of the reference workpiece 40. Since the directions of the x axis and the z axis are defined in this way, the direction of the y axis orthogonal to these axes is also defined. Note that the fifth plane portion 43e is provided for accurately calculating the second straight line L2 described later, and is omitted if only the second plane portion 43b can calculate the second straight line L2 with high accuracy. May be.

  The 3rd plane part 43c is a surface formed right next to the 2nd plane part 43b in Fig.2 (a), and as shown to FIG.1 and FIG.2 (b), it is z-axis direction from the illustration left end to a right end It is inclined to. Further, the fourth plane portion 43d is a surface formed on the right side of the third plane portion 43c in FIG. 2A, and, like the third plane portion 43c, from the left end to the right end in FIG. 2B. Inclined in the z-axis direction. In the present embodiment, as shown in FIG. 1, the right end of the third plane portion 43c is inclined so as to protrude toward the shape measuring instrument 10 with respect to the left end, and the left end of the fourth plane portion 43d is relative to the right end. Are inclined so as to protrude toward the shape measuring instrument 10. Therefore, an intersection line 43g between the third plane part 43c and the fourth plane part 43d is formed so as to protrude in the z-axis direction with respect to the shape measuring instrument 10. The intersecting line 43g may be a plane orthogonal to the z axis and formed in a plane different from the virtual plane A.

  The normal vector n1 and the direction vector n2 define the directions of the x-axis, y-axis, and z-axis in the reference coordinate system, but the positions of the x-axis, y-axis, and z-axis in the reference coordinate system are the origins of the coordinate axes in the reference coordinate system. It is defined by defining as follows. The coordinate axis origin in the reference coordinate system is parallel to the normal vector n1 and is a straight line (z-axis) parallel to the normal vector n1 passing through the intersection of the plane including the central axis of the shaft portion 41 of the reference workpiece 40 and the intersection line 43g. The origin is the point where the direction center line intersects the plane on which the first plane portion 43a and the sixth plane portion 43f are formed. This defines the reference coordinate system.

  In the present embodiment, the third plane portion 43c is a plane rotated by a predetermined angle ψ (see FIG. 2B) around the y axis with respect to the first plane portion 43a and the sixth plane portion 43f. The fourth plane portion 43d is a plane rotated by a predetermined angle ψ around the y axis with respect to the first plane portion 43a and the sixth plane portion 43f and on the opposite side to the third plane portion 43c. Therefore, the intersection line 43g between the third plane part 43c and the fourth plane part 43d is a straight line parallel to the y-axis. However, the intersecting line 43g is not necessarily parallel to the y axis.

  As shown in FIG. 1, a controller 12 and a data processing device 13 are electrically connected to the shape measuring instrument 10. The controller 12 and the data processing device 13 have a microcomputer such as a CPU, ROM, RAM, and hard disk as main components. Based on the input instruction from the operator input from the input device 14, the controller 12 emits a laser beam from the shape measuring instrument 10 and operates the shape measuring instrument 10 so that the reflected light of the laser beam can be received. Control. The data processing device 13 receives light reception information from the shape measuring instrument 10 and control information of the shape measuring instrument 10 from the controller 12, and based on these information, a reference coordinate system and a coordinate system defined by the shape measuring instrument 10 and And the calculated deviation amount (angle, length) is output to the display device 15. The display device 15 includes a display or a printer, and displays or prints the shift amount input from the data processing device 13 on the display. The coordinate system defined by the shape measuring instrument 10 is referred to as a camera coordinate system in this specification.

  Next, a description will be given of the control when the deviation amount between the camera coordinate system and the reference coordinate system is acquired using the reference workpiece 40 configured as described above. First, when an operator inputs an instruction to execute acquisition of the deviation amount of the shape measuring instrument 10 to the controller 12 via the input device 14, the controller 12 instructs the shape measuring instrument 10 to emit line light. Then, the line light is irradiated from the shape measuring instrument 10 toward the reference workpiece 40 (pre-irradiation step). Here, the irradiation plane of the line light is substantially horizontal to the direction along the shaft portion 41 of the reference workpiece 40, that is, the x-axis of the reference coordinate system. However, it does not have to be strictly horizontal. The line light irradiated on the reference workpiece 40 is reflected at the irradiation point of the reference workpiece 40. The reflected light is received by the light receiving device in the shape measuring instrument 10 (pre-light receiving step). Information on the reflected light received in this way is input to the data processing device 13.

The data processing device 13 calculates the distance ln from the laser light source to the irradiation point of the reference workpiece 40 for each of the points at predetermined intervals of the line light based on the input light reception information by the principle of triangulation. In addition, since the direction from the laser light source toward the irradiation point is also known from the received light information, the data processing device 13 calculates the coordinate value of the irradiation point based on the distance ln and the direction. The coordinate system used at this time is a camera coordinate system. In the camera coordinate system, the origin position when measuring the distance ln to the irradiation point of the reference workpiece 40 is the origin of the coordinate axis, and the laser light irradiation direction when the laser light irradiation angle ξ is 0 ° is the z-axis direction. The direction in which the laser beam irradiation direction changes perpendicular to the z-axis is defined as the x-axis direction, and the z-axis is defined by a coordinate axis having the direction perpendicular to the x-axis direction as the y-axis direction. Therefore, the irradiation plane of the line light is an xz coordinate plane of the camera coordinate system, and the y-direction component is 0 in the component of the coordinate value of the irradiation point. Therefore, the data processing device 13 calculates a plurality of coordinate points N (x _n , 0, z _n ) in the horizontal direction of the reference workpiece 40 by the above calculation. The coordinate value of the coordinate point N (x _n , 0, z _n ) can be calculated based on the distance ln and the irradiation angle ξ. For example, the coordinate point of the irradiation point with the distance ln and the irradiation angle ξ can be calculated as (ln · cos ξ, 0, ln · sin ξ).

Next, the data processing device 13 determines the four straight lines (first straight line L1, second straight line L2, third straight line) from the coordinate value N (x _n , 0, z _n ) of the irradiation point of the line light on the surface of the reference workpiece 40. The equation of L3, the fourth straight line L4) is calculated. Here, the coordinate value N (x _n , 0, z _n ) of the irradiation point is the first, second, third and fourth of the reference workpiece 40 as shown in FIG. 3 on the xz coordinate plane of the camera coordinate system. The fifth and sixth plane portions 43a, 43b, 43c, 43d, 43e, and 43f are acquired as point cloud data obtained by scanning the respective cross-sectional outlines. In this case, when the irradiation point is switched from the cross-sectional outer shape of the first flat surface portion 43a to the cross-sectional external shape of the second flat surface portion 43b, or from the cross-sectional outer diameter of the second flat surface portion 43b to the cross-sectional outer diameter of the third flat surface portion 43c. When the irradiation point is switched, a large shift occurs in the coordinate point at the time of switching. Therefore, when this deviation (distance between adjacent coordinate points) is less than a predetermined value, it is recognized as point group data indicating the cross-sectional outline within the same plane part, and when the deviation is greater than a predetermined value, it is within the same plane part. It is recognized that it is not point cloud data indicating the cross-sectional outer shape of. By performing such recognition for each adjacent coordinate point, the point group data G1 belonging to the cross-sectional outline of the first plane portion 43a, the point group data G2 belonging to the cross-sectional outline of the second plane portion 43b, and the third plane portion 43c. Point group data G3 belonging to the cross-sectional outline of the fourth plane portion 43d, point group data G4 belonging to the cross-sectional outline of the fourth plane portion 43d, point group data G5 belonging to the cross-sectional outline of the fifth plane portion 43e, and the cross-sectional outline of the sixth plane portion 43f Point cloud data G6 can be acquired.

After the point cloud data is acquired by dividing the cross sectional outline of each plane portion as described above, the data processing device 13 acquires the cross section of the point cloud data G1 and the sixth plane portion 43f belonging to the cross sectional outline of the first plane portion 43a. The equation of the first straight line L1 is calculated by the least square method using the point group data G6 belonging to the outer shape. Further, the equation of the second straight line L2 is calculated by the least square method using the point group data G2 belonging to the cross-sectional outline of the second plane portion 43b and the point group data G5 belonging to the cross-sectional outline of the fifth plane portion 43e. Further, the equation of the third straight line L3 is calculated by the least square method using the point cloud data G3 belonging to the cross-sectional outline of the third plane portion 43c. Further, the equation of the fourth straight line L4 is calculated by the least square method using the point group data G4 belonging to the cross-sectional outline of the fourth plane portion 43d. The equations of the first, second, third and fourth straight lines L1, L2, L3 and L4 are expressed as the following formulas 1, 2, 3, and 4, respectively. It should be noted that the first to fourth straight lines may be obtained by selectively calculating only those necessary for each deviation amount described later, that is, depending on the deviation amount to be obtained.

  Based on the calculation result of the straight line equation, the data processing device 13 calculates the deviation angle θx. As shown in FIG. 4, the shift angle θx is an angle formed by a line segment LS1 indicating the incident direction of the laser beam from the shape measuring instrument 10 to the reference workpiece 40 in the yz plane of the reference coordinate system and the z axis. The deviation angle θx indicates an angle at which the shape measuring instrument 10 is inclined around the x axis of the reference coordinate system. The calculation of θx is performed as follows.

First, the data processing device 13 calculates the coordinate point C (x _c , z _c ) of the intersection of the third straight line L3 and the fourth straight line L4. The coordinate point C can be solved by combining the equation of the third straight line L3 (Equation 3) and the equation of the fourth straight line L4 (Equation 4), and the solution is as shown in Equation 5 below.

Next, the data processing device 13 calculates the distance S between the coordinate point C (x _c , z _c ) and the first straight line L1. The distance S is obtained from the formula of the distance between the point and the straight line, and is expressed by the following formula 6.

Subsequently, the data processing device 13 calculates the shift angle θx using the distance S. In this case, the shift angle θx is an angle formed by the line segment LS1 indicating the incident direction of the laser beam on the reference workpiece 40 in FIG. 4 and the z axis defined by the reference workpiece 40 as described above. The coordinate point C of the intersection point is a coordinate point of the intersection point of the line segment LS1 and the intersection line 43g in FIG. 4, and the distance S is the first plane portion 43a and the sixth point from the coordinate point C in the line segment LS1. The distance to the plane on which the plane portion 43f is formed is shown. Since the shortest distance from the intersection line 43g to the plane on which the first flat surface portion 43a and the sixth flat surface portion 43f are formed can be measured from the reference workpiece 40, if this distance (shortest distance) is m, the deviation angle The sine of θx can be expressed by the ratio of the distance S and the distance m, and using this relationship, the deviation angle θx can be calculated as shown in Equation 7 below.

  The deviation angle θx obtained by the calculation is displayed on the display device 15. The operator adjusts the inclination of the shape measuring instrument 10 by rotating the bolt 233 attached to the angle adjuster 23 so that the displayed deviation angle θx becomes zero. As a result, the shift angle θx is adjusted. Since the above equation 7 can be directly calculated from the equations of m and the first, third and fourth straight lines L1, L3 and L4, these coefficients are used after the equations of each straight line are determined. If the deviation angle θx is directly calculated by Equation 7, the above Equations 5 and 6 can be omitted.

  Since the distance S and the distance m have only positive values in the above formula 7, θx is always a positive value, and the incident direction of the laser light on the reference workpiece 40 is θx above the z axis defined by the reference workpiece 40. It cannot be determined from the calculated shift angle θx whether it is shifted or shifted downward by θx. Therefore, the operator needs to calculate θx while repeatedly adjusting the deviation angle θx, and to grasp the direction in which the deviation angle θx is displaced.

  Next, the data processing device 13 calculates the deviation angle θz. As shown in FIG. 6, the deviation angle θz is an angle formed by a line segment LS2 indicating a cutting line by the laser beam from the shape measuring instrument 10 to the reference workpiece 40 in the xy plane of the reference coordinate system and the x axis. The deviation angle θz indicates an angle at which the shape measuring instrument 10 is inclined around the z-axis of the reference coordinate system. The calculation of the shift angle θz is performed as follows.

ここで、θ_L1の正接tanθ_L1は第一直線Ｌ１の傾き（−ａ₁／ｂ₁）に等しく、θ_L2の正接tanθ_L2は第二直線Ｌ２の傾き（−ａ₂／ｂ₂）に等しいので、上記式８は第一直線Ｌ１および第二直線Ｌ２の方程式から計算することができる。なお、θzが０でない場合は、図６に示すようにレーザー光が基準ワーク４０に対して斜めに照射していることになるので、第二平面部４３ｂおよび第五平面部４３ｅにおける照射スポットのｚ軸方向位置がｘ軸方向に沿って変化し、第一および第六平面部４３ａおよび４３ｆとの相対距離が変化する。このため第二直線Ｌ２が第一直線Ｌ１に交差する。 First, the shape measuring instrument 10 irradiates the reference workpiece 40 with laser light again. By this laser light irradiation, as shown in FIG. 5, the first, second, third, fourth, fifth and sixth plane portions 43a, 43b, 43c, 43d in the xz coordinate plane of the camera coordinate system are used. , 43e, and 43f, point cloud data G1 to G6 obtained by scanning the respective cross-sectional outlines are obtained. Based on these point cloud data, the data processor 13 calculates an angle θ _L1-L2 formed by the first straight line L1 and the second straight line L2. The angle θ _L1-L2 is obtained when the angle formed between the first straight line L1 and the x axis is θ _L1 and the angle formed between the second straight line L2 and the x axis is θ _L2 on the xz coordinate plane of the camera coordinate system. , equal to the angle theta _L1 - [theta] _L2 of the difference between the angle theta _L1 and the angle theta _L2. Therefore, θ _L1−L2 is calculated by the following equation 8 by the addition theorem of trigonometric functions.

Here, the tangent tan θ _L1 of θ _L1 is equal to the slope (−a ₁ / b ₁ ) of the _first straight line L1, and the tangent tan θ _L2 of θ _L2 is equal to the slope (−a ₂ / b ₂ ) of the second straight line L2. The above equation 8 can be calculated from the equations of the first straight line L1 and the second straight line L2. If θz is not 0, the laser beam is applied obliquely to the reference workpiece 40 as shown in FIG. 6, so that the irradiation spot of the second plane part 43b and the fifth plane part 43e is irradiated. The z-axis direction position changes along the x-axis direction, and the relative distance between the first and sixth plane portions 43a and 43f changes. For this reason, the second straight line L2 intersects the first straight line L1.

ここで、ｎは、基準ワーク４０のｙ軸方向長さ、ｐは、第二平面部４３ｂおよび第五平面部のｚ軸方向への変化量である（図６参照）。ｎおよびｐは、基準ワーク４０から測定することができる。上記式９は、以下のように導出される。 Subsequently, the data processing device 13 calculates θz by executing the calculation of Equation 9 below.

Here, n is the length of the reference workpiece 40 in the y-axis direction, and p is the amount of change in the z-axis direction of the second plane portion 43b and the fifth plane portion (see FIG. 6). n and p can be measured from the reference workpiece 40. Equation 9 is derived as follows.

First, when θz is not 0 and the laser beam is irradiated obliquely with respect to the reference workpiece 40 as shown in FIG. 6, the y-axis of the irradiation point H at the left end of the second plane portion 43b in the drawing is shown. the height T _H along is different from the height T _J along the y-axis of the irradiation point J shown right end of the fifth flat portion 43e. The difference between both heights is ΔT (= T _H −T _J ). In addition, since the second plane portion 43b and the fifth plane portion 43e are formed to be inclined in the z-axis direction, the depth length (the first length along the z-axis is different if the height position along the y-axis is different. The distance between the plane portion 43a and the plane on which the sixth plane portion 43f is formed is also different. Therefore, as shown in FIG. 7 which is an enlarged view of the reference workpiece 40 viewed from the yz coordinate plane, the depth length at the irradiation point H (the irradiation point H and the first plane portion 43a and the sixth plane portion 43f are formed). is the distance between the plane are) depth length in D _H and the irradiation point J (the distance between the plane irradiation point J and the first flat portion 43a and the sixth flat portion 43f is formed) D _J are different. Let ΔD be the difference between the two depth lengths.

ここで、図６に示すように、第二平面部４３ｂから第五平面部４３ｅまでのｘ軸方向長さをｑとすると、ｑ’はｑ／cosθzで表されるが、近似的にcosθz＝１とすると、上記式１０は式１１のようになる。

また、相似の関係から、ΔＤとΔＴについて、式１２の関係が成立する。

また、θzの正接tanθzは、図６から明らかなように下記式１３のように表すことができる。

したがって、式１１を式１２に代入してΔＤを消去し、さらにその式を式１３に代入してΔＴを消去することにより、上記式９が導かれる。 Here, when the distance between the irradiation points H-J on the irradiation plane is q ′, the depth length changes by ΔD while the laser light travels on the reference workpiece 40 from the irradiation spot H to the irradiation spot J. become. A straight line passing through the irradiation points H and J can be represented by a straight line L2 in FIG. 5, the depth length D _H is the distance between the point H and the straight line L1 in FIG. 5, and the depth length D _J is FIG. Is equal to the distance between the point J and the straight line L1. In FIG. 5, the change rate (ΔD / q ′) of the depth length ΔD with respect to q ′ is the change in the z-axis direction (the direction perpendicular to the straight line L1) in the reference coordinate system of the straight line L2 with respect to the straight line L1. This is the same as the value divided by the amount of change in the direction of the straight line L1. Therefore, this change rate is represented by the difference in slope between the first straight line L1 and the second straight line L2, that is, the tangent of θ _L1 -L2. Therefore, the following formula 10 is established.

Here, as shown in FIG. 6, when the length in the x-axis direction from the second plane portion 43b to the fifth plane portion 43e is q, q ′ is expressed by q / cos θz, but approximately cos θz = Assuming 1, Equation 10 above becomes Equation 11.

Further, from the similar relationship, the relationship of Expression 12 is established for ΔD and ΔT.

Further, the tangent tan θz of θz can be expressed as the following Expression 13 as is apparent from FIG.

Therefore, by substituting Equation 11 into Equation 12 to eliminate ΔD, and further substituting the equation into Equation 13 to eliminate ΔT, Equation 9 is derived.

The deviation angle θz is calculated as described above. The calculated deviation angle θz is displayed on the display device 15. The operator rotates each bolt 233 of the angle adjuster 23 so that the displayed deviation angle θz becomes 0, and adjusts the inclination of the shape measuring instrument 10 with respect to the reference workpiece 40. As a result, the shift angle θz is adjusted. In addition, tanθ _L1-L2 in the above equation 9 can also be obtained directly from the equations of the first straight line L1 and the second straight line L2. In this case, the calculation of Equation 8 can be omitted.

  In addition, the data processing device 13 calculates the shift position Δy. As shown in FIG. 9, the shift position Δy includes a center line lx along the x-axis direction of the reference workpiece 40 in the xy plane of the reference coordinate system, and a cutting line of the laser light irradiated to the reference workpiece 40. This displacement position Δy is an amount of displacement of the shape measuring instrument 10 in the y-axis direction with respect to the reference workpiece 40.

In calculating the shift position Δy, first, the reference workpiece 40 is irradiated with laser light. By this irradiation, as shown in FIG. 8 in the xz coordinate plane of the camera coordinate system, the first, second, third, fourth, fifth and sixth plane portions 43a, 43b, 43c, Point group data G1 to G6 obtained by scanning the respective cross-sectional outlines 43d, 43e, and 43f are obtained. Then, the distance t between the first straight line L1 and the second straight line L2 is calculated. Here, it is assumed that the shift angle θz is corrected, and the first straight line L1 and the second straight line L2 are in a substantially parallel relationship. The distance t can be obtained by executing the calculation of the following equation (14).

上記式１５は、以下のように導出される。まず、図９に示すように、距離ｔは、基準座標系のｙ−ｚ平面における、第二平面部４３ｂおよび第五平面部４３ｅ上へのレーザー光の照射点Ｋから、第一平面部４３ａおよび第六平面部４３ｆが形成される平面までの距離である。一方、Δyは、基準座標系のｙ−ｚ平面における基準ワーク４０の中心線lxが第二平面部４３ｂおよび第五平面部４３ｅを通過する点Ｌと、上記点Ｋとの間の距離のｙ軸方向成分である。したがって、相似の関係から、ｎ：ｐ＝（ｎ／２−Δy）：ｔとなる。この関係を変形することにより、上記式１５が導かれる。 Next, the data processing device 13 calculates the shift position Δy by executing the calculation of Equation 15 below.

The above equation 15 is derived as follows. First, as shown in FIG. 9, the distance t is determined from the irradiation point K of the laser beam on the second plane part 43b and the fifth plane part 43e in the yz plane of the reference coordinate system, to the first plane part 43a. And the distance to the plane on which the sixth plane portion 43f is formed. On the other hand, Δy is a distance y between the point L where the center line lx of the reference workpiece 40 in the yz plane of the reference coordinate system passes through the second plane part 43b and the fifth plane part 43e, and the point K. It is an axial component. Therefore, n: p = (n / 2−Δy): t from the similar relationship. The above formula 15 is derived by modifying this relationship.

  In this way, the shift position Δy can be calculated. The calculated displacement position Δy is displayed on the display device 15, and the operator moves the inner cylinder 212 of the column unit 21 in the axial direction so that the displayed displacement position Δy becomes 0, and the height of the column unit 21 is increased. Adjust the height. As a result, the shift position Δy is adjusted. Note that the shift position Δy can also be directly calculated from the equations of the first straight line L1 and the second straight line L2, and in this case, the calculation of the above equation 14 can be omitted.

  Further, the data processing device 13 calculates the deviation angle θy. As shown in FIG. 11, the deviation angle θy indicates the incident direction of the laser light when the laser light is incident on the intersecting line 43g of the reference workpiece 40 from the shape measuring instrument 10 in the xz coordinate plane of the reference coordinate system. The angle formed by the line segment LS3 and the z-axis, and the deviation angle θy indicates the angle at which the shape measuring instrument 10 is inclined about the y-axis of the reference coordinate system.

In order to calculate the deviation angle θy, first, the reference work 40 is irradiated with laser light from the shape measuring instrument 10. By this irradiation, the first, second, third, fourth, fifth, and sixth plane portions 43a, 43b, 43c, and 43d of the reference workpiece 40 are shown on the xz coordinate plane of the camera coordinate system as shown in FIG. , 43e, and 43f, point cloud data G1 to G6 obtained by scanning the respective cross-sectional outlines are obtained. The data processing device 13 calculates the third straight line L3 and the fourth straight line L4 from the obtained point group data, and further calculates the intersection C (x _c , z _c ) between the third straight line L3 and the fourth straight line L4. . The intersection point C can be expressed as Equation 5 above. Next, a fifth straight line L5 that passes through the intersection C and is perpendicular to the first straight line L1 is calculated. The fifth straight line L5 is orthogonal to the first straight line L1 and can pass through the intersection point C and can be expressed as the following Expression 16.

上記式１７は、上記式１６によって表される第五直線Ｌ５および上記式１によって表される第一直線Ｌ１が共に点Q(x_q,y_q)を通ることに基づいて導かれる。 Next, the data processing device 13 calculates the coordinate point Q (x _q , y _q ) of the intersection point of the fifth straight line L5 and the first straight line L1 by executing the calculation of Expression 17 below.

The above equation 17 is derived based on the fact that the fifth straight line L5 represented by the above equation 16 and the first straight line L1 represented by the above equation 1 both pass through the point Q (x _q , y _q ).

上記式１８は、以下のようにして導かれる。第一直線Ｌ１，第三直線Ｌ３，第四直線Ｌ４および第五直線Ｌ５がカメラ座標系のｘ−ｚ座標平面において図１０に示す状態のように計算された場合、第三直線Ｌ３と第四直線Ｌ４との交点Cと、第五直線Ｌ５と第一直線Ｌ１との交点Qとを結ぶ線分CQと、カメラｘ−ｚ座標系におけるｚ軸とのなす角がずれ角度θyである。この場合、ずれ角度θyの正弦は、線分ＣＱの長さと、座標点Ｃと座標点Ｑとのｘ軸方向成分の差x_c-x_qとの比によって表される。ここで、線分ＣＱの長さは、基準ワーク４０の交線４３ｇから第一平面部４３ａおよび第六平面部４３ｆが形成される平面までの距離ｍであり（図１１参照）、基準ワーク４０から測定することができる。したがって、ずれ角度θyの正弦sinθyは、（x_c−x_q）／ｍと表すことができる。式１８は、この関係に基づき導かれる。 After obtaining the point Q (x _q , y _q ), the data processing device 13 calculates θy by executing the following equation 18.

The above equation 18 is derived as follows. When the first straight line L1, the third straight line L3, the fourth straight line L4 and the fifth straight line L5 are calculated as shown in FIG. 10 on the xz coordinate plane of the camera coordinate system, the third straight line L3 and the fourth straight line The angle between the line segment CQ connecting the intersection C with L4 and the intersection Q between the fifth straight line L5 and the first straight line L1 and the z axis in the camera xz coordinate system is the shift angle θy. In this case, the sine of the deviation angle θy is represented by the ratio between the length of the line segment CQ and the difference x _c -x _q of the x-axis direction component between the coordinate point C and the coordinate point Q. Here, the length of the line segment CQ is the distance m from the intersection line 43g of the reference workpiece 40 to the plane on which the first plane portion 43a and the sixth plane portion 43f are formed (see FIG. 11). Can be measured from Therefore, the sine sin θy of the deviation angle θy can be expressed as (x _c −x _q ) / m. Equation 18 is derived based on this relationship.

The deviation angle θy calculated as described above is displayed on the display device 15, and the operator rotates the inner cylinder 212 of the column unit 21 so that the displayed deviation angle θy becomes zero. As a result, the shift angle θy is adjusted. Note that x _c -x _q in the above equation 18 can also be directly obtained from the equations of the first straight line L1, the third straight line L3, and the fourth straight line L4. Therefore, x _c -x _q is directly calculated from the equations of the first straight line L1, the third straight line L3, and the fourth straight line L4, and θy is calculated without obtaining the fifth straight line L5 and the intersection point Q based on the equation. be able to. In this case, calculation of the above formula 16 and the above formula 17 can be omitted.

Further, the data processing device 13 calculates the shift position Δx. As shown in FIG. 13, the shift position Δx is a shift amount in the x-axis direction of the shape measuring instrument 10 from the center line lz along the z-axis direction of the reference workpiece 40 in the xz plane of the reference coordinate system. . In calculating the displacement position Δx, first, the reference workpiece 40 is irradiated with laser light. By this irradiation, as shown in FIG. 12 in the xz coordinate plane of the camera coordinate system, the first, second, third, fourth, fifth and sixth plane portions 43a, 43b, 43c, Point cloud data obtained by scanning the respective cross-sectional outlines 43d, 43e, and 43f is obtained. The data processing device 13 obtains the third straight line L3 and the fourth straight line L4 based on these point group data, and further determines the coordinate point of the intersection C (x _c , z _c ) between the third straight line L3 and the fourth straight line L4. calculate. The coordinate point of the intersection point C can be calculated as shown in Equation 5 above.

Next, the data processing device 13 calculates the shift position Δx. In this case, if the time x-axis component at the point of intersection C is zero is a reference position of the x-axis direction of the shape measuring device 10, the x-axis direction component x _c is the x-axis direction as is the deviation of the intersection C Indicates the amount. Therefore, x _c acquired on the xz coordinate plane of the camera coordinate system as shown in FIG. 12 is set as the shift position Δx. The displacement position Δx calculated in this way is displayed on the display device 15, and the operator adjusts the first lever 223 of the position adjuster 22 so that the displayed displacement position Δx becomes 0 or a predetermined value. . As a result, the shift position Δx is adjusted.

Further, the data processing device 13 calculates the shift position Δz. As shown in FIG. 13, the displacement position Δz is from the shape measuring instrument 10 to the plane on which the first plane portion 43a and the sixth plane portion 43f of the reference workpiece 40 are formed in the xz plane of the reference coordinate system. and the z-axis direction distance z _1, which is a difference between a predetermined reference distance z _0. In calculating the displacement position Δz, first, the reference workpiece 40 is irradiated with laser light. By this irradiation, as shown in FIG. 12 in the xz coordinate plane of the camera coordinate system, the first, second, third, fourth, fifth and sixth plane portions 43a, 43b, 43c, Point cloud data obtained by scanning the respective cross-sectional outlines 43d, 43e, and 43f is obtained. The data processing device 13 obtains the third straight line L3 and the fourth straight line L4 based on these point group data, and further determines the coordinate point of the intersection C (x _c , z _c ) between the third straight line L3 and the fourth straight line L4. calculate. The coordinate point of the intersection point C can be calculated as shown in Equation 5 above.

Next, the data processing device 13 calculates the displacement position Δz. In this case, as shown in FIG. 12, the distance z ₁ is the distance in the z-axis direction from the shape measuring instrument 10 to the intersection C and the first plane portion 43 a and the sixth plane portion from the intersection line 43 g in the reference workpiece 40. 43f is the sum of the distances to the plane on which 43f is formed. The z-axis direction distance from the shape measuring instrument 10 to the intersection point C is the z-axis direction component z _c of the intersection point C, and the distance m from the intersection line 43g to the plane on which the first plane part 43a and the sixth plane part 43f are formed. Can be understood by measuring the reference workpiece 40. Therefore, z ₁ = z _c + m. Therefore, Δz can be obtained by executing the calculation of z ₁ + m−z ₀ . The displacement position Δz calculated as described above is displayed on the display device 15, and the operator adjusts the second lever 224 of the position adjuster 22 so that the displayed displacement position Δz becomes zero. As a result, the shift position Δz is adjusted.

  As described above, in the calibration process, based on the light reception information obtained in the prior light reception step, the shift amount of the camera coordinate system with respect to the reference coordinate system (shift position Δx, Δy, Δz, shift angle θx, θy, θz) Is calculated (calculation step). The operator adjusts the posture of the shape measuring instrument 10 while confirming the numerical values displayed on the display device 15 so that these deviation amounts become substantially zero. At this time, the amount of deviation other than θx may be roughly adjusted (it is not necessary to adjust until the amount of deviation is exactly zero). However, since the shift angle θx is expressed by a positive value even if it is shifted in the negative direction, it is not known which is shifted. Therefore, the adjustment is repeated until only θx becomes 0 within the range of measurement accuracy. After such an operation is finished, the minute shift amounts (shift angles θy, θz, shift positions Δx, Δy, Δz) that remain finally are detected (shift amount detection step). Remember. Then, the calibration process ends.

  The calibration process corresponds to the shift amount detection step of the present invention. In this case, in the above example, after calculating the deviation amounts in the calculation step, the operator adjusts the posture of the shape measuring instrument 10 so that the calculated deviation amount is adjusted to 0 or minute. However, if all the deviation amounts calculated in the calculation step are originally minute, no adjustment by the operator is required. After the calibration process is completed, the reference workpiece 40 is removed from the long plate-like member 31, and the actual measurement object is attached to the long plate-like member 31, and the shape measurement step is performed.

  FIG. 14 shows an overall view of the shape measuring system in the shape measuring step. As shown in the figure, the measuring object OB is arranged facing the front surface of the shape measuring instrument 10. In the present embodiment, the measurement object OB is a rotating object having an axis, such as a crankshaft used in a vehicle, but in the present embodiment, it is a rod-like member for the sake of simplicity. A first fixing jig 51 is attached to one end of the measurement object OB, and a second fixing jig 52 is attached to the other end. As shown in the drawing, both fixing jigs 51 and 52 are formed in a columnar shape and are arranged to face each other, a hole is formed in the central portion on the facing side, and both ends of the measuring object OB are located at the opposite ends. It is inserted into this hole. The hole is devised in shape and the like so that the measurement object OB does not rotate. When the measurement object OB is attached in this way, the central axis (rotation axis) of the measurement object OB is the same axis as the shaft portion 41 of the reference workpiece 40 in the calibration process.

  Further, as shown in the figure, a motor 53 is attached to the notch 31a of one long plate member 31, and a support member 54 is attached to the notch 31a of the other long plate member 31. . The motor 53 and the support member 54 are attached so as not to rotate within the notch 31a. The output shaft 53 a of the motor 53 is connected to the surface of the first fixing jig 51 that is not facing the second fixing jig 52. Further, a rod-like protrusion 52 a is formed from the surface of the second fixing jig 52 that does not face the first fixing jig 51, and this protrusion 52 a is formed on the support member 54. It is inserted into a bearing (not shown) such as a bearing. For this reason, the second fixing jig 52 is rotatably supported by the support member 54.

  The shape measuring instrument 10 is electrically connected to the controller 12 and the data processing device 13. The controller 12 and the data processing device 13 have a microcomputer such as a CPU, a ROM, and a RAM as main components. The controller 12 receives an input command from the input device 14 and information from the data processing device 13 and outputs a control command signal to the shape measuring instrument 10 and the motor 53 based on these. The data processing device 13 receives control information from the controller 12, light reception information from the shape measuring instrument 10, rotation information from the motor 53, etc., executes data processing calculation based on these information, and outputs the calculation result. Output to the display device 15. The controller 12, the data processing device 13, the input device 14, and the display device 15 may be shared with those used during the calibration process. Further, the posture control device 20 shown in the figure is the same as the posture control device 20 of FIG. 1 described in the calibration process.

In the above configuration, when an operator inputs a signal for measuring the shape of the measurement object OB to the controller 12 through the input device 14, the controller 12 outputs a drive command signal to the shape measuring instrument 10 and the motor 53. As a result, the shape measuring instrument 10 irradiates the line light toward the measurement object OB, and the motor 53 is driven to rotate the measurement object OB (irradiation step). The reflected light of the line light irradiated to the measuring object OB is received by the light receiving means in the shape measuring instrument 10 (light receiving step), and the received light information is input to the data processing device 13. Based on the received light reception information, the data processing device 13 calculates the distance from the shape measuring instrument 10 to the irradiation point and the direction of the irradiation point for each point at a predetermined interval of the line light. And the coordinate value in the camera coordinate system of the irradiation point of the measuring object OB is calculated from the calculated distance and direction. As a result, point cloud data (hereinafter referred to as cross-sectional shape data) of coordinate values Xa (x _a , 0, z _a ) representing the shape of the cross-sectional outer shape obtained by cutting the measurement object OB at the portion irradiated with the line light is obtained. Can do.

  An encoder is housed in the motor 53, and a pulse signal is generated as the motor 53 is driven. This pulse signal is input to the data processor 13. The data processing device 13 calculates the rotation angle of the measurement object OB based on this pulse signal. The cross-sectional shape data is calculated for each predetermined rotation angle (for example, every 5 °) of the measurement object OB.

The data processing device 13 calculates cross-sectional shape data for each predetermined rotation angle until the measurement object OB makes one rotation, that is, until the rotation angle calculated based on the pulse signal from the motor 53 reaches 360 °. Then, the coordinate value Xa (x _a , 0, z _a ) of the cross-sectional shape data obtained for each predetermined rotation angle is subjected to coordinate conversion by the coordinate conversion matrix M, and the corrected coordinate value Xb (x _b , y after coordinate conversion) is converted. _b , z _b ) is obtained (coordinate conversion step). Since the coordinate transformation matrix M can be expressed as a product of the angle correction matrix Mθ and the axis alignment matrix MA, the following Expression 19 is established.

上記式２０において、θy、θzは、キャリブレーション工程にてずれ角度θyおよびθzを校正した後になお残る微小なずれ角度である。キャリブレーション工程において、ずれ角度θy、θzが表示装置１５に表示されると、作業者が目視によりずれ角度を０にするように形状測定器１０の傾きを調整し、ずれ角度θyやθzを校正するが、一つのずれ角度を調整すると他方のずれ角度が増えたりして、なかなか全てのずれ角度を０に調整するのは難しい。従って、上記キャリブレーション工程では、ずれ角度θyおよびずれ角度θzを厳密に０にすることまで要求せず、粗調整として、ずれ角度θyおよびθzが所定の角度範囲内に収まればよいものとしている。なお、基準ワークを形状測定器１０で測定することにより得られたずれ角度θyとθzは、他方のずれ角度が小さい場合は、本来のずれ角度からのずれは無視できるほど小さいが、他方のずれ角度が大きい場合は本来のずれ角度からのずれが大きくなる。このため、Ｍθによって座標値を座標変換する場合には、キャリブレーション工程によりずれ角度θyとθzをできるだけ小さな値としておくのがよい。本実施形態においては、ずれ角度θyとθzの絶対値を１°以下にすれば、精度のよい計算を行うことができる。このようにして校正してなお残るずれ角度θyおよびθｚが上述したようにデータ処理装置１３内に記憶されており、上記式２０ではこの記憶されたずれ角度θyおよびθｚが使用される。なお、角度補正行列Ｍθに使用される角度θyとθzは、ずれ角度θyとθzを補正して０°にするための角度であるので、基準ワークを形状測定器１０で測定することにより得られたずれ角度θyとθzとは符号が逆になっている。 Further, the angle correction matrix Mθ is expressed as the following Expression 20.

In the above equation 20, θy and θz are minute deviation angles still remaining after the deviation angles θy and θz are calibrated in the calibration process. In the calibration process, when the deviation angles θy and θz are displayed on the display device 15, the operator adjusts the inclination of the shape measuring instrument 10 so that the deviation angle becomes zero by visual observation, and calibrates the deviation angles θy and θz. However, when one shift angle is adjusted, the other shift angle increases, and it is difficult to adjust all the shift angles to zero. Therefore, in the calibration step, it is not required to set the deviation angle θy and the deviation angle θz to exactly 0, and as a rough adjustment, the deviation angles θy and θz only need to be within a predetermined angle range. The deviation angles θy and θz obtained by measuring the reference workpiece with the shape measuring instrument 10 are so small that the deviation from the original deviation angle is negligible when the other deviation angle is small. When the angle is large, the deviation from the original deviation angle becomes large. For this reason, when the coordinate value is converted by Mθ, it is preferable to set the deviation angles θy and θz as small as possible by the calibration process. In the present embodiment, accurate calculation can be performed if the absolute values of the shift angles θy and θz are 1 ° or less. The deviation angles θy and θz that remain after calibration in this way are stored in the data processor 13 as described above, and the stored deviation angles θy and θz are used in the above equation 20. Note that the angles θy and θz used in the angle correction matrix Mθ are angles for correcting the shift angles θy and θz to 0 °, and are obtained by measuring the reference workpiece with the shape measuring instrument 10. The deviation angles θy and θz are opposite in sign.

  The determinant on the right side of the right side of Equation 20 is a rotation determinant that rotates the coordinate axis about the z axis by a shift angle θz, and the left determinant is a rotation determinant that rotates the coordinate axis about the y axis by a shift angle θy. is there. Therefore, by converting the coordinate value of the cross-sectional shape data using the determinant of Equation 20, even if the camera coordinate system is inclined about the y axis and the z axis with respect to the reference coordinate system, this inclination is corrected. The coordinate value of the shape data in the coordinate system thus obtained can be calculated. Further, the deviation angle θx is calibrated so as to become 0 in the calibration process described above. Therefore, the coordinate transformation by the determinant of Equation 1 corrects all the deviation angles with accuracy, and the coordinate value of the cross-sectional shape data in the coordinate system having no angular deviation with respect to the reference coordinate system can be calculated.

Here, the y axis in the reference coordinate system is the y _base axis, the z axis is the z _base axis, the y axis in the coordinate system (angle correction conversion coordinate system) after being corrected by the angle correction matrix Mθ is the y _Mθ axis, and the z axis is _{Assuming the} z _Mθ axis, the relationship between both coordinate systems can be expressed as shown in FIG. In FIG. 15, the y _base axis is included in a plane including the first plane part 43 a and the sixth plane part 43 f of the reference workpiece 40. Therefore, the rotation center of the measurement object OB is not the origin of the reference coordinate system, and the coordinate axis origin of the reference coordinate system is shifted from the rotation center of the measurement object OB in the z _base axis direction by the radius r. Also, the distance obtained by subtracting the predetermined distance z ₀ from the z-axis direction distance between the coordinate axis origin of the reference coordinate system and the coordinate axis origin of the angle correction conversion coordinate system is a shift position Δz, the y-axis direction distance between the z _base axis and the z _Mθ axis. Shift position Δy.

As shown in the figure, when the z _Mθ axis of the angle correction conversion coordinate system is shifted by Δy in the y _Mθ axis direction from the rotation center of the measurement object OB (the shift position Δy is also adjusted by the operator in the calibration process). However, it cannot be completely adjusted to 0 and a minute shift position Δy is generated), and the xz coordinate plane of the angle correction conversion coordinate system does not include the rotation center of the measurement object OB. When a measurement object OB, which will be described later, is rotated as it is and the coordinate value is coordinate-converted based on the rotation angle, the amount of deviation from the xz coordinate plane of the angle correction conversion coordinate system of the rotation axis of the measurement object OB. Will be converted into uncorrected coordinate values, and the accuracy of the obtained three-dimensional shape will deteriorate. For this reason, the coordinate axis of the angle correction conversion coordinate system is coordinate-converted by the axis alignment matrix MA so that the xz coordinate plane of the angle correction conversion coordinate system includes the rotation center of the measurement object OB. In this case, the coordinate conversion by the axis alignment matrix MA rotates the coordinate axis of the angle correction conversion coordinate system around the x axis so that the xz coordinate plane includes the rotation center of the measurement object OB. Rotation angle at this time is the angle between the z _base axis of the reference coordinate system (z _MA axis in FIG. 15) the z-axis of the coordinate system the angle is corrected by the rotation, the sine of斯Karunasu angle , Expressed as a ratio (Δy / d) of the shift position Δy to the distance d between the coordinate axis origin of the angle correction conversion coordinate system and the rotation center O of the measurement object OB. Further, the cosine of the angle formed is expressed by the ratio of the distance in the z-axis direction from the coordinate axis origin of the angle correction conversion coordinate system to the rotation center O of the measurement object OB with respect to the distance d. Since the z-axis direction distance from the coordinate axis origin of the angle correction conversion coordinate system to the rotation center O of the measurement object OB is represented by z ₀ + Δz + r from FIG. 15, the cosine is (z ₀ + Δz + r) / d. expressed. Further, the distance d is expressed by the following formula 21.

Therefore, by rotating the angle correction conversion coordinate system, the axis alignment matrix MA for allowing the laser light irradiation plane to pass through the rotation center of the measurement object can be expressed as the following Expression 22.

In FIG. 15, if the y coordinate axis in the coordinate system after conversion by the axis alignment matrix MA (hereinafter referred to as the axis alignment conversion coordinate system) is y _MA and the z coordinate axis is z _MA , the axis alignment conversion coordinate system is shown in FIG. Thus, it can be seen that the z _MA axis passes through the center of rotation of the measurement object OB. That is, since the x-axis of the axis alignment conversion coordinate system is parallel to the rotation axis of the measurement object OB as before conversion, the rotation axis of the measurement object OB is included in the xz coordinate plane of the axis alignment conversion coordinate system.

  The coordinate transformations using the two determinants Mθ and MA may be performed separately, but may be performed at once in order to simplify the calculation process. In this way, by converting the coordinate value obtained by the coordinate transformation matrix M (= MA · Mθ), the x axis is parallel to the rotation axis of the measurement object OB, and the z axis is the rotation of the measurement object OB. The cross-sectional shape data of the measurement object OB in the coordinate system passing through the axis can be acquired.

In addition, the axis alignment matrix MA does not rotate the coordinate system, but translates the coordinate system in the y _Mθ axis direction and the z _Mθ axis direction as shown in FIG. Can pass through the center of rotation of the measurement object OB. In this case, the equation 19 and the alignment matrix MA are expressed by the following equations 23-1 and 23-2.

  In the above equation 23-2, the z component may be any value, but when using equation 25 described later, the z component of the above equation 23-2 is used. By performing coordinate conversion based on the above equation 23-2, the z axis of the axis alignment conversion coordinate system passes through the rotation center axis of the measurement object OB. That is, the rotation center axis of the measurement object OB is included in the xz coordinate plane of the axis alignment conversion coordinate system.

In the present embodiment, as described above, the group of coordinate values Xa (x _a , 0, z _a ) of the measured cross-sectional shape data is converted into corrected coordinate values Xb (x _b , y _b , z _b ) using the coordinate transformation matrix M. ) Is converted at every predetermined rotation angle of the measurement object OB. For example, if the predetermined rotation angle of the measurement object OB is 5 °, the coordinate value Xa (x _a , 0, of the shape data acquired when the rotation angle of the measurement object becomes 5 ° due to the rotation of the motor 53. z _a) correction coordinate value Xb (x _b to the coordinate transformation matrix M group, y _b, converted into a group of z _b), the correction coordinate value Xb (x _b, y _b, a group of z _b) Xb (5 °) is stored in the data processing device 13. Next, the group of coordinate values Xa (x _a , 0, z _a ) of the cross-sectional shape data acquired when the rotation angle of the measurement object OB becomes 10 ° is used as a corrected coordinate value Xb (x _b , y _b , z _b ), and this corrected coordinate value Xb (x _b , y _b , z _b ) is stored in the data processor 13 as Xb (10 °). In this way, a group of corrected coordinate values Xb obtained by sequentially converting the coordinates at every rotation angle of 5 ° is stored.

A group of correction coordinate values Xb (x _b , y _b , z _b ) stored for each rotation angle of the measurement object OB is corrected based on a deviation amount between the camera coordinate system and the reference coordinate system. However, since the rotation angle of the measurement object OB is not taken into consideration, a group of coordinate values indicating, on the same coordinate system, cross-sectional shape data for each predetermined rotation angle of the portion where the laser beam cuts the measurement object OB. Only. Therefore, by combining the group of coordinate values stored for each rotation angle of the measurement object OB in consideration of the rotation angle of the measurement object OB when it is measured, the entire shape of the measurement object is first obtained. Can know.

Here, as in this embodiment, fixing the shape measuring instrument 10 and measuring the shape of the measurement object OB while rotating the measurement object OB means that if the viewpoint is placed on the measurement object OB side, This is the same as measuring the shape of the measurement object OB while fixing the measurement object OB and rotating the shape measuring instrument 10 around the rotation axis of the measurement object OB. Therefore, as shown in FIG. 17, it may be considered that the coordinate axis rotates by the same angle in the opposite direction to the direction in which the measurement object OB rotates around the rotation axis. Accordingly, the correction coordinate value Xb (x _b , y _b , z _b ) stored for each predetermined rotation of the measurement object OB is coordinate-transformed by the composite transformation matrices Sa and Sb as shown in the following equation 24, thereby measuring A group of synthesized coordinate values Xc (x _c , y _c , z _c ) indicating the three-dimensional shape of the measurement object OB in the coordinate system fixed with respect to the object OB can be acquired.

上記式において、ζは、モータ５３内のエンコーダによって検出される測定対象物ＯＢの回転角度である。なお、測定対象物ＯＢの回転角度ζはモータ５３側から見て右回りを正とする。すなわち測定対象物ＯＢを基準にした場合、モータ５３側から見て軸合わせ変換座標系の回転角度ζは左回りを正とする。 Here, the combined transformation matrices Sa and Sb are expressed by the following formulas 25-1 and 25-2.

In the above equation, ζ is the rotation angle of the measurement object OB detected by the encoder in the motor 53. Note that the rotation angle ζ of the measurement object OB is positive when viewed clockwise from the motor 53 side. That is, when the measurement object OB is used as a reference, the rotation angle ζ of the axis alignment conversion coordinate system as viewed from the motor 53 side is positive in the counterclockwise direction.

The data processing device 13 converts each group of corrected coordinate values Xb (x _b , y _b , z _b ) obtained for each predetermined rotation angle of the measurement object OB using the above-described composite conversion matrices Sa and Sb. A group of composite coordinate values Xc (x _c , y _c , z _c ) is acquired. The entire shape of the measurement object OB is displayed by displaying all the data of the composite coordinate values Xc (x _c , y _c , z _c ) obtained for each predetermined angle on the display device 15. The entire shape of the displayed measurement object OB is based on a coordinate value group obtained by performing coordinate conversion based on a minute shift amount that cannot be adjusted in the calibration process. For this reason, this embodiment can perform very accurate three-dimensional measurement.

As described above, the three-dimensional shape measurement method of the present embodiment includes a shift amount detection step for detecting a shift amount between the camera coordinate system and the reference coordinate system, while rotating the measurement object OB around the rotation axis. An irradiation step of irradiating the line light from the shape measuring instrument 10 toward the measurement object OB, a light receiving step of receiving reflected light of the line light irradiated on the measurement object OB, and a predetermined rotation angle of the measurement object OB A coordinate value indicating a position on the surface of the measurement object OB irradiated with the line light is represented by a coordinate value Xa (x _a , 0, z _a in the camera coordinate system based on the light reception information obtained in the light reception step every time. ) As a group of coordinate values acquisition step acquired at each predetermined rotation angle, and a group of coordinate values Xa (x _a , 0, z _a ) acquired at each predetermined rotation angle at the coordinate value acquisition step. Correction seat obtained by coordinate conversion based on Value _{Xb (x b, y b,} z b) and the coordinate transformation step of obtaining the group for every predetermined rotation angle of the correction coordinate value for each predetermined rotation angle obtained by the coordinate conversion step Xb (x _b, A synthesis step of synthesizing a group of y _b , z _b ) based on a predetermined rotation angle and obtaining a group of synthesized coordinate values Xc (x _c , y _c , z _c ) indicating the three-dimensional shape of the measurement object OB It is included. As can be seen from the above steps, the deviation amount between the camera coordinate system and the reference coordinate system is acquired before the actual shape measurement, the shape measurement of the measurement object OB is performed, and the coordinate value is obtained. By converting the coordinate value based on the above, the accurate three-dimensional shape of the measurement object OB can be measured.

  Further, in the deviation amount detection step, a pre-irradiation step of irradiating the line light from the shape measuring instrument 10 toward the reference work 40 as a calibration object having a shape that can define a reference coordinate system, and a pre-irradiation step A pre-light receiving step for receiving the reflected light of the line light irradiated on the reference workpiece 40, and a calculating step for calculating a deviation between the camera coordinate system and the reference coordinate system based on the light reception information obtained in the pre-light receiving step. It is supposed to be. For this reason, it is possible to accurately calculate the deviation between the camera coordinate system and the reference coordinate system by analyzing the coordinate value of the portion irradiated on the reference workpiece 40.

  Further, in the coordinate conversion step, first, in the first coordinate conversion step, one coordinate axis of the camera coordinate system and the rotation axis of the measurement object are made parallel based on the angle shift between the camera coordinate system and the reference coordinate system. The coordinate value group is transformed to obtain an angle correction coordinate value, and then, in the second coordinate transformation step, one of the camera coordinate systems is determined based on the relative position shift between the camera coordinate system and the reference coordinate system. A group of angle correction coordinate values is subjected to coordinate conversion so that the coordinate plane includes the rotation axis of the measurement object, and the correction coordinate values are acquired. In this way, the second coordinate conversion step converts the angle correction coordinate value based on the shift of the relative position of both coordinate systems so that the rotation axis of the measurement object is included in one coordinate plane of the camera coordinate system. Therefore, it is possible to accurately measure the three-dimensional shape of the measurement object.

(Second embodiment)
FIG. 18 is an overall configuration diagram of a shape measuring apparatus according to the second embodiment of the present invention. As shown in the figure, the shape measuring instrument 10 in the present embodiment emits line light vertically. The measurement object OB irradiated with the line light is formed in a cylindrical shape standing in the vertical direction, and is placed on the upper surface of the measurement table 32. The measurement table 32 has a support 33 connected to the lower surface thereof, and the support 33 is rotated by a driving means 34 such as a motor connected to the lower part thereof. Therefore, the measurement object OB is rotated around the vertical axis on the measurement table 32 by the rotation of the support 33.

That is, in this embodiment, the line light irradiation plane and the rotation axis of the measurement object OB are formed along the vertical direction, and the shape is measured in this state. As described above, the present embodiment only changes the direction of the shape measurement in the first embodiment. Therefore, in the calibration process, the measurement unit 43 of the reference workpiece 40 is replaced with the first plane unit as shown in FIG. 43a to the sixth flat surface portion 43f are installed on the measurement table 32 so as to be arranged vertically (vertical direction), and laser light is irradiated from the shape measuring instrument 10 to these flat surface portions. In the present embodiment, the side surface where the measurement unit 43 of the reference workpiece 40 is combined with the measurement table 32 is perpendicular to the first plane unit 43a, the second plane unit 43b, the fifth plane unit 43e, and the sixth plane unit 43f. Therefore, the x coordinate axis of the reference coordinate system defined by the reference workpiece 40 is parallel to the rotation axis of the measurement table 32. Therefore, based on the information obtained from the reflected light in each plane part, each shift amount is calculated similarly to the first embodiment, and the angle and position of the shape measuring instrument 10 are adjusted. In this case, since the orientation of the reference workpiece 40 relative to the shape measuring instrument 10 is not fixed as in the first embodiment, the adjustment of θx is performed by rotating the measuring table 32.
Then, after adjusting θx to 0 ° and adjusting other shift amounts to be within the allowable range, a matrix for coordinate conversion is calculated based on the shift amounts as in the first embodiment. In the embodiment, the alignment matrix MA cannot be calculated only by the amount of deviation obtained from the shape measurement of the reference workpiece 40. In other words, in the first embodiment, the positional relationship between the coordinate axis of the reference coordinate system defined by the reference workpiece 40 and the rotation axis of the measurement object is determined. Specifically, the reference coordinate is determined from the coordinate axis origin of the reference coordinate system. Since the point moved by the radius r of the shaft portion 41 in the direction opposite to the z-axis direction of the system is the rotation center, the xz coordinate plane of the angle correction conversion coordinate system includes the rotation axis of the measurement object OB. The axis alignment matrix MA used for coordinate transformation can be calculated as Equation 22 or Equation 23-2.

  On the other hand, in the present embodiment, since the reference workpiece 40 is not positioned with respect to the measurement table 32 in the calibration process, the coordinate axis of the reference coordinate system defined by the rotation center of the measurement table 32 and the reference workpiece 40 is used. The positional relationship with cannot be determined. For this reason, among the deviation amounts obtained from the shape measurement of the reference workpiece 40, the angle correction matrix Mθ can be calculated from the deviation in angle between the camera coordinate system and the reference coordinate system, but the axis alignment matrix MA is calculated. I can't. Therefore, after performing a predetermined calibration by measuring the reference workpiece 40 in order to obtain the rotation center, the sphere 35 is placed on the measurement table 32 via the support portion 36 as shown in FIG. The shape of the sphere 35 is measured.

The sphere 35 has a known radius, and the shape measuring instrument 10 irradiates the sphere 35 with line light so as to pass through the approximate center of the sphere 35. Then, by rotating the support 33, the sphere 35 is rotated at 120 ° intervals, and the shape measurement is performed three times. At this time, when the line light is deviated from the center of the sphere 35, the plane including the center of the sphere 35 and the irradiation plane of the line light are shifted as shown in FIG. FIG. 22 shows a relationship between the plane including the center of the sphere 35 and the irradiation plane of the line light on the yz coordinate plane. In FIG. 22, the equation of a circle including the irradiation point (laser light irradiation cutting line in FIG. 21) is derived from the coordinate value of the irradiation point on the irradiation plane of the line light. Therefore, the coordinate value of the center of the _circle and the radius r _circle can be calculated from the derived circle equation. Further, the radius r _sphere of the sphere 35 is known. Further, as can be seen from FIG. 22, the coordinate value of the center of the actual sphere 35 is separated from the coordinate value of the center of the circle by a distance s (= (r _sphere ² −r _circle ² ) ^1/2 ) in the y-axis direction. Located at the point. In this case, since it is not known whether the distance s is separated in the positive direction of the y-axis or in the negative direction, as shown in FIG. It can be considered as a coordinate value indicating the center coordinate of. The coordinate values of these two points are obtained by calculation.

Further, since the sphere 35 is rotated at intervals of 120 ° and the line light is irradiated at each angle, the line light is irradiated to the sphere 35 three times during one rotation. When the coordinate values of two points considered to be the rotation center are calculated for each irradiation, a total of six coordinate values are calculated. Figure 23 is centered on the six-point diagrams radius showed six circles is r _sphere at y-z coordinate plane. In FIG. 23, a circle centered on a point represented by a square is indicated by a dotted line, and a circle centered on a point represented by a diamond is indicated by a one-dot chain line. Next, three coordinate values are selected by selecting one of the two coordinate values calculated on each irradiation plane, and an equation of a circle passing through these three points is calculated. Since there are two kinds of selection of coordinate points for each irradiation plane, there are eight ways for convenience. The equation of the circle is calculated for all of these eight ways, and the combination of the coordinate points that are closest to 120 ° or 120 ° between the three coordinate points passing through the circle is determined as the center of the sphere 35 in each irradiation plane. To do. In FIG. 23, a circle passing through the three points determined in this way is displayed by a solid line.

  When three points passing through the center of the sphere 35 in each irradiation plane are determined, the coordinate value of the center of the circle passing through the determined three points is obtained. This coordinate value is the coordinate value of the rotation center in the xz coordinate plane of the camera coordinate system of the measurement table 32. A straight line that passes through this coordinate point and is parallel to the x-axis direction of the reference coordinate system becomes the rotation axis of the measurement object OB. In this case, the obtained coordinate value of the rotation center is a coordinate value by the camera coordinate system. On the other hand, in order to use the coordinate value of the rotation center for the calculation of the matrix component of the coordinate transformation matrix of Equation 26 or Equation 27 described later, the coordinates by the angle correction matrix Mθ of Equation 20 are used for the obtained coordinate value of the rotation center. Conversion is performed to obtain coordinate values in the angle correction conversion coordinate system. That is, if the y-coordinate value of the rotation axis of the measurement object obtained as described above is y0 and the z-coordinate value is z0, the coordinate value is converted by the angle correction matrix Mθ, and the angle correction conversion coordinate system is used. A y-coordinate value y1 and a z-coordinate value z1 of the rotation axis of the measurement object are obtained.

As described above, after calculating the displacement amount and the coordinate value of the rotation center of the measurement object OB, the measurement object OB whose shape is actually measured is placed on the measurement table 32 as shown in FIG. The shape measuring step is performed. Then, the coordinate value Xa (x _a , 0, z _a ) calculated by the shape measuring instrument 10 is coordinate-transformed by the coordinate transformation matrix M, and the corrected coordinate value Xb (x _b , y _b , z _b) after the coordinate transformation is obtained. ). In this case, the coordinate transformation matrix M includes an angle correction matrix Mθ and an axis alignment matrix MA. The angle correction matrix Mθ has the same components as those described in the first embodiment. Further, in the case of conversion by the axis alignment matrix MA, the xz coordinate plane is rotated by rotating the x coordinate axis of the angle correction conversion coordinate system for the coordinate value group coordinate-converted by the angle correction matrix Mθ. The axis alignment matrix MA for performing coordinate conversion into a coordinate value group based on a coordinate system including the center is represented by y1 as the y-coordinate component of the coordinates of the rotation center of the measurement object OB in the angle correction conversion coordinate system obtained as described above. If the z-coordinate component is z1, the following equation 26 is obtained.

上記式２６および式２７において、ｄは、角度補正変換座標系の座標軸原点と角度補正変換座標系のｘ−ｚ平面における測定対象物ＯＢの回転中心との間の距離であり、下記式２８で表される。

In addition, the xz coordinate plane is set to the center of rotation of the measurement object by translating the coordinate value group coordinate-transformed by the angle correction matrix Mθ in the coordinate axis direction of the y-coordinate axis and the z-coordinate axis of the angle correction transformation coordinate system. An axis alignment matrix MA for performing coordinate conversion into a coordinate value group based on the coordinate system to be included is expressed by Equation 27 below.

In the above formulas 26 and 27, d is the distance between the coordinate axis origin of the angle correction conversion coordinate system and the rotation center of the measurement object OB in the xz plane of the angle correction conversion coordinate system. expressed.

  As described above, the present embodiment includes the rotation axis calculation step for calculating the coordinate value of the rotation center of the measurement object OB. For this reason, even if the position of the rotation axis of the measurement object OB is unknown, it can be detected. Therefore, the coordinate value group coordinate-transformed by Mθ is used for this step in the step of converting the coordinate value group to the coordinate value group by the coordinate system whose xz coordinate plane includes the rotation center of the measurement object (second coordinate conversion step). Based on the rotation axis position (y1, z1) of the measurement object in the angle correction conversion coordinate system, the axis alignment matrix MA is a deviation amount between the reference coordinate system and the camera coordinate system (the shift used in the angle correction matrix Mθ). Angle) and the rotation axis position (y0, z0) calculated in the rotation axis position calculation step can be determined, and a highly accurate three-dimensional shape can be measured. In this case, the rotation axis calculation step irradiates the fixed point definition object with line light from the shape measuring instrument toward the fixed point definition object (sphere 35) capable of defining a predetermined fixed point (center) from the outer shape. The fixed point coordinate value is calculated from the reflected light of the light at a plurality of points while rotating the fixed point definition object by a predetermined angle around the rotation axis of the measurement object, and based on the calculated coordinate value of the fixed point. The position of the rotation axis of the measurement object is calculated. Since the rotation axis of the measurement object is calculated by such a rotation axis calculation step, the position of the rotation axis can be grasped.

(Third embodiment)
Next, a third embodiment of the present invention will be described. In the first embodiment and the second embodiment, when calculating the deviation amount from the shape measurement data of the reference workpiece 40, the sign of the deviation angle θx is unknown, and which is relative to the direction of the z coordinate axis of the reference coordinate system. In this embodiment, the deviation angle θx is adjusted by devising the shape of the reference workpiece, so that the deviation angle θx is adjusted until the deviation angle θx becomes zero. It is intended to know the positive and negative.

  FIG. 24 is an overall schematic diagram in the calibration process of the shape measuring system employing the reference workpiece 40 in the present embodiment. FIG. 25 is a diagram showing the reference workpiece 40 used in the present embodiment, where (a) is a front view, (b) is a plan view, and (c) is a left side view. As can be seen from the figure, the reference workpiece 40 in the present embodiment is additionally provided with an additional first plane part 43a ′ and an additional second plane part 43b ′ in addition to the first plane part 43a to the sixth plane part 43f. Yes.

  As can be seen from FIG. 25, the additional first plane portion 43a 'and the additional second plane portion 43b' are formed between the second plane portion 43b and the third plane portion 43c. The additional first plane portion 43a 'is parallel to the first plane portion 43a and the sixth plane portion 43f, but is formed on a different parallel plane from the first plane portion 43a and the sixth plane portion 43f. The additional second plane part 43b 'is parallel to the second plane part 43b and the fifth plane part 43e, but is formed on a different parallel plane from the second plane part 43b and the fifth plane part 43e. Moreover, the maximum height (z-axis direction length) of the additional second plane part 43b ′ with respect to the additional first plane part 43a ′ is the second plane part 43b (with respect to the first plane part 43a (or the sixth plane part 43f)). Alternatively, it is the same as the maximum height (z-axis direction length) with respect to the fifth plane portion 43e). That is, as can be seen from FIG. 25 (c), the angle φ1 formed by the second plane portion with respect to the first plane portion 43a and the angle φ2 formed by the additional second plane portion 43b ′ with respect to the additional first plane portion 43a ′ are the same. is there. However, these angles are not necessarily the same.

When the reference workpiece 40 having such a shape is placed on the long plate-shaped member 31 as shown in FIG. 24 and irradiated with line light in the horizontal direction and the shape is measured, the xz coordinate plane of the camera coordinate system is measured. The point cloud data as shown in FIG. 26 is obtained above. Among these point group data, Gn (n = 1, 2, 3, 4, 5, 6, 1 ′, 2 ′) is point group data (n = 1 ′, 2 ′) belonging to the nth plane portion, respectively. In the case of, point group data belonging to the additional nth plane portion). In this point group data, first, from the point group data G1 ′ indicating the irradiation point irradiated to the additional first plane portion 43a ′, the additional first straight line L1 indicating the cutting line of the light irradiated to the additional first plane portion 43a ′. And the additional second straight line L2 ′ indicating the cutting line of the light irradiated on the additional second plane part 43b ′ from the point equation data G2 ′ indicating the irradiation point irradiated on the additional second plane part 43b ′. Calculate the equation. The equations of the additional first straight line L1 ′ and the additional second straight line L2 ′ can be expressed as the following Expression 29 and Expression 30.

  Also, equations for the first straight line L1 and the second straight line L2 are obtained. The equations of the first straight line and the second straight line can be expressed as in the above formulas 1 and 2. Next, a distance p1 between the first straight line L1 and the second straight line L2 and a distance p2 between the additional first straight line L1 'and the additional second straight line L2' are obtained. In this case, assuming that the first straight line L1, the second straight line L2, the additional first straight line L1 ′, and the additional second straight line L2 ′ are all parallel, one straight line that is perpendicular to the first straight line L1 and passes through a predetermined coordinate value. (Hereinafter, this straight line is referred to as a vertical straight line), a point A1 where the vertical straight line and the first straight line L1 intersect, a point A2 where the vertical straight line and the second straight line L2 intersect, and a point where the vertical straight line and the additional first straight line L1 ′ intersect A1 ′, a point A2 ′ where the vertical straight line intersects with the additional second straight line L2 ′ are obtained. The distance between the points A1 and A2 is p1, and the distance between the points A1 'and A2' is p2.

  FIG. 27 is a diagram illustrating the relationship between the line segment LS1 indicating the incident direction of the laser beam from the shape measuring instrument 10 to the reference workpiece 40 and the reference workpiece 40 in the yz coordinate plane of the reference coordinate system. In the figure, the distance between the intersection T between the line segment LS1 and the first plane part 43a and the intersection U between the line segment LS1 and the second plane part 43b is the distance p1, and the line segment LS1 and the additional first plane part 43a. The distance between the intersection point V with 'and the intersection point W between the line segment LS1 and the additional second plane portion 43b' is the distance p2. As shown in the figure, it can be seen that p1 is larger than p2 when the deviation angle θx is an angle that is shifted to the upper side in the drawing with respect to the horizontal line (z axis). On the other hand, it is understood that p1 is smaller than p2 when the deviation angle θx is an angle deviating downward in the figure with respect to the horizontal line (z axis). When the deviation angle θx is 0, p1 and p2 are equal. In this manner, it is possible to determine whether the shift angle θx is shifted up or down with respect to the horizontal line (z axis) based on the magnitude relationship between p1 and p2. If the shift angle θx is shifted upward from the horizontal line and is negative, the shift angle θx can be uniquely determined in the present embodiment.

When the deviation angle θx can be uniquely determined, it is not necessary to adjust the deviation angle θx in the calibration process until the deviation angle θx becomes zero as in the first embodiment. In this case, the adjustment of the shift angle θx in the calibration process can be finished when θx is within a predetermined minute angle (for example, ± 1 °). In the shape measurement step, the shift angle θx may be corrected. In this case, the angle correction matrix Mθ for correcting the shift angle is expressed as the following Expression 31. Note that, similarly to the angles θy and θz, θx used in the angle correction matrix is an angle for correcting the deviation angle θx to 0 °, and thus was obtained by measuring the reference workpiece with the shape measuring instrument 10. The sign is opposite to that of the shift angle θx.

この式においてθxの符号は角度補正行列に使用される角度θxと同じである。また、記号ｐ、ｎおよびｔは式１５のｐ、ｎおよびｔと同じであり、ｒは式２１および式２２と同様、軸部４１の半径である。このようにして、測定対象物の三次元形状を求めることができる。 The determinant of the axis alignment matrix MA for including the center of rotation of the measurement object in the laser light irradiation plane is the same as that shown in the first embodiment, but Δy in the components in the determinant Is not completely adjusted so that the deviation angle θx becomes 0, and correction is necessary. The shift position Δy in the present embodiment is expressed as the following Expression 32.

In this equation, the sign of θx is the same as the angle θx used for the angle correction matrix. Symbols p, n, and t are the same as p, n, and t in Equation 15, and r is the radius of the shaft portion 41 as in Equations 21 and 22. In this way, the three-dimensional shape of the measurement object can be obtained.

  As described above, the reference workpiece 40 (calibration object) used in the first to third embodiments in the calibration process has a z-axis whose normal direction (direction of the normal vector n1) is one coordinate axis. A second plane that defines the direction of the x axis, which is a coordinate axis in which a direction vector n2 parallel to the intersection line between the first plane portion 43a that defines the direction and the plane on which the first plane portion 43a is formed, is orthogonal to the z axis. A portion 43b, and a third plane portion 43c and a fourth plane portion 43d that form an intersection line 43g on a plane that is orthogonal to the z-axis and that is different from the plane on which the first plane portion 43a is formed. The direction of the direction vector n2 is parallel to the rotational axis of the measurement object OB. For this reason, the reference workpiece is irradiated with laser light in the calibration process, a linear equation is obtained from the point cloud data obtained from the reflected light at each plane, and the reference coordinates defined by the reference workpiece from the obtained equation. The amount of deviation between the system and the camera coordinate system defined by the shape measuring instrument 10 can be detected.

  Further, the reference workpiece 40 used in the third embodiment is formed at a position parallel to the first plane portion 43a and spaced from the first plane portion 43a by a predetermined distance in the z-axis direction of the reference coordinate system. One plane portion 43a ′ and an additional second plane portion 43b ′ formed at a position parallel to the second plane portion 43b and spaced from the second plane portion 43b by a predetermined distance in the z-axis direction of the reference coordinate system. . In the shift amount detection step, the first plane portion 43a, the second plane portion 43b, the additional first plane portion 43a ′, and the additional second plane portion 43b ′ are lined based on the light reception information obtained in the prior light reception step. Equations L1, L2, L1 ′, L2 ′ of straight lines including cutting lines when light is irradiated are obtained, and then a distance p1 between the straight lines L1 and L2 and a distance p2 between the straight lines L1 ′ and L2 ′ are obtained. Ask. The ratio between the distance p1 and the distance p2 is that the incident direction of the line light is inclined with respect to the z-axis (upward or downward in FIG. 27), that is, the deviation angle θx is positive or negative. It depends on whether there is. Therefore, the shift angle θx can be uniquely determined based on the ratio between the distance p1 and the distance p2.

In 1st embodiment of this invention, it is the whole schematic diagram of the shape measurement system when performing a calibration process. It is a figure which shows the reference | standard workpiece | work which concerns on 1st embodiment of this invention, (a) is a front view, (b) is a top view, (c) is a side view. It is a figure which shows visually the three-dimensional data of the standard work in a camera coordinate system measured by a shape measuring device when obtaining deviation angle thetax in a calibration process. It is a figure which shows the geometric relationship between the incident direction of a laser beam and the reference | standard workpiece | work for calculating | requiring deviation | shift angle (theta) x. It is a figure which shows visually the three-dimensional data of the standard work in a camera coordinate system measured by a shape measuring device when obtaining deviation angle thetaz at a calibration process. It is a figure which shows the geometrical relationship of the irradiation direction of a laser beam, and a reference | standard workpiece | work for calculating | requiring deviation | shift angle | corner (theta) z. FIG. 6 is an enlarged view showing a geometric relationship between a laser beam irradiation point and a reference workpiece for obtaining a deviation angle θz. It is a figure which shows visually the three-dimensional data of the standard work in the camera coordinate system measured by the shape measuring instrument when obtaining deviation position Δy in a calibration process. It is a figure which shows the geometrical relationship of the irradiation point of a laser beam, and a reference | standard workpiece | work for calculating | requiring deviation | shift position (DELTA) y. It is a figure which shows visually the three-dimensional data of the standard work in the camera coordinate system measured by the shape measuring instrument when obtaining deviation angle thetay in a calibration process. It is a figure which shows the geometric relationship between the incident direction of a laser beam and the reference | standard workpiece | work for calculating | requiring deviation | shift angle | corner (theta) y. It is a figure which shows visually the three-dimensional data of the standard work in the camera coordinate system measured by the shape measuring device when obtaining displacement positions Δx and Δz in the calibration process. It is a figure which shows the geometrical relationship of the incident direction of a laser beam and the reference | standard workpiece | work for calculating | requiring deviation | shift position (DELTA) x and deviation | shift position (DELTA) z. In 1st embodiment of this invention, it is the whole schematic diagram of the shape measurement system when performing a shape measurement process. It is a figure which shows the relationship between a reference | standard coordinate system, an angle correction conversion coordinate system, an axis alignment conversion coordinate system, and the rotation center of a measurement object. It is a figure which shows the relationship between a reference | standard coordinate system, an angle correction conversion coordinate system, an axis alignment conversion coordinate system, and the rotation center of a measurement object. It is the figure which showed the relationship of each coordinate system at the time of synthesize | combining the three-dimensional data by the coordinate system for every predetermined rotation angle. In 2nd embodiment of this invention, it is the whole schematic diagram of the shape measurement system when performing a shape measurement process. In 2nd embodiment of this invention, it is the whole schematic diagram of the shape measurement system when performing a calibration process. In 2nd embodiment of this invention, it is a figure which shows a state when calculating | requiring the rotation center of a measuring object. When calculating | requiring the rotation center of a measuring object, it is a schematic diagram which shows the shift | offset | difference with the plane containing the center of a spherical body, and the irradiation plane of a laser beam. It is a schematic diagram which shows the geometric relationship between the said 2 sphere and the irradiation plane when calculating | requiring the center of two spheres from the irradiation plane of a laser beam. When determining the rotation center of the measurement object, the sphere is irradiated three times per rotation, and six circles centered on the six points obtained from the respective irradiation surfaces, and the measurement object obtained from these circles. It is a schematic diagram showing the rotation center. In 3rd embodiment of this invention, it is the whole schematic diagram of the shape measurement system when performing a calibration process. It is a figure which shows the reference | standard workpiece | work which concerns on 3rd embodiment of this invention, (a) is a front view, (b) is a top view, (c) is a side view. In 3rd embodiment, it is a figure which shows visually the three-dimensional data of the reference | standard workpiece | work in the camera coordinate system measured by the shape measuring device when calculating | requiring deviation | shift angle (theta) x in a calibration process. In 3rd embodiment, it is a figure which shows the geometric relationship between the incident direction of a laser beam and the reference | standard workpiece | work for calculating | requiring deviation | shift angle (theta) x.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Shape measuring device, 12 ... Controller, 13 ... Data processing device, 14 ... Input device, 15 ... Display device, 35 ... Sphere, 40 ... Reference | standard workpiece, 43 ... Measuring part, 43a ... First plane part, 43a '... Additional first plane part, 43b ... second plane part, 43b '... additional second plane part, 43c ... third plane part, 43d ... fourth plane part, 43e ... fifth plane part, 43f ... sixth plane part, 43g ... intersecting line, 51 ... first fixing jig, 52 ... second fixing jig, M ... coordinate transformation matrix, MA ... axis alignment matrix, Mθ ... angle correction matrix, OB ... measurement object, S ... composite transformation matrix

Claims (15)

While rotating the measurement object, irradiate the line light from the shape measuring instrument capable of emitting line light toward the measurement object substantially parallel to the rotation axis of the measurement object, and receive the reflected light of the irradiated light A three-dimensional shape measuring method for measuring a three-dimensional shape of a measurement object,
Deviation between the coordinate system defined by the shape measuring instrument and whose coordinate plane is parallel to the line light irradiation plane, and the reference coordinate system in which one coordinate axis is parallel to the rotation axis of the object to be measured A deviation amount detecting step for detecting
An irradiation step of irradiating line light from the shape measuring instrument toward the measurement object while rotating the measurement object around the rotation axis;
A light receiving step for receiving reflected light of the line light irradiated on the measurement object;
Based on the light receiving information obtained in the light receiving step for each predetermined rotation angle of the measuring object, a coordinate value indicating a position on the surface of the measuring object irradiated with the line light is represented by a coordinate value in the camera coordinate system. A coordinate value obtaining step for obtaining the predetermined rotation angle as a group of:
A group of correction coordinate values obtained by performing coordinate transformation on the group of coordinate values acquired at each predetermined rotation angle in the coordinate value acquisition step based on the deviation amount is acquired at each predetermined rotation angle. A coordinate transformation step;
The corrected coordinate value group for each predetermined rotation angle obtained in the coordinate conversion step is combined based on the predetermined rotation angle to obtain a combined coordinate value group indicating the three-dimensional shape of the measurement object. A synthesis step to
A three-dimensional shape measuring method, comprising: The three-dimensional shape measuring method according to claim 1,
The deviation amount detecting step includes:
A pre-irradiation step of irradiating line light from the shape measuring instrument toward a calibration object having a shape capable of defining a reference coordinate system;
A pre-light-receiving step of receiving reflected light of the line light irradiated on the calibration object in the pre-irradiation step;
A three-dimensional shape measurement method comprising: a calculation step of calculating a deviation between the camera coordinate system and the reference coordinate system based on light reception information obtained in the prior light reception step. In the three-dimensional shape measuring method according to claim 1 or 2,
Irradiating line light from a shape measuring instrument toward a fixed point definition object capable of defining a predetermined fixed point from the outer shape, and calculating a coordinate value of the fixed point from reflected light of the line light irradiated to the fixed point definition object, The fixed-point defining object is rotated at a plurality of rotation angles while rotating by a predetermined angle around the rotation axis of the measurement object, and the position of the rotation axis of the measurement object is calculated based on the calculated coordinate values of the plurality of fixed points. A rotation axis calculating step for
In the coordinate conversion step, coordinate conversion is performed on the basis of the shift amount and the position of the rotation axis calculated in the rotation axis calculation step. In the three-dimensional shape measuring method according to any one of claims 1 to 3,
The coordinate transformation step includes
Based on the angle deviation between the camera coordinate system and the reference coordinate system among the deviation amounts, the coordinate value group is set so that one coordinate axis of the camera coordinate system and the rotation axis of the measurement object are parallel to each other. A first coordinate conversion step of obtaining an angle correction coordinate value by performing coordinate conversion;
Of the deviation amount, the relative position deviation between the camera coordinate system and the reference coordinate system, or based on the deviation amount and the position of the rotation axis of the measurement object calculated in the rotation axis calculation step, A second coordinate conversion step of obtaining the correction coordinate value by performing coordinate conversion of the group of angle correction coordinate values so that the rotation axis of the measurement object is included in one coordinate plane of the camera coordinate system;
A three-dimensional shape measuring method, comprising: In the three-dimensional shape measuring method according to claim 4,
The second coordinate conversion step is based on a relative position shift between the camera coordinate system and the reference coordinate system, or the shift amount and the position of the rotation axis of the measurement object calculated in the rotation axis calculation step. Then, by rotating the camera coordinate system, the group of angle correction coordinate values is transformed so that the rotation axis of the measurement object is included in one coordinate plane of the camera coordinate system, and the correction coordinate values are A three-dimensional shape measuring method, characterized by comprising: In the three-dimensional shape measuring method according to claim 4,
In the second coordinate conversion step, the relative position shift between the camera coordinate system and the reference coordinate system, or the shift amount and the position of the rotation axis of the measurement object calculated in the rotation axis calculation step, By moving the camera coordinate system based on the coordinate conversion of the group of angle correction coordinate values so that the rotation axis of the measurement object is included in one coordinate plane of the camera coordinate system. A three-dimensional shape measuring method, characterized by comprising: In the three-dimensional shape measuring method according to claim 3,
The rotation axis calculation step calculates an equation of a circle that passes through the calculated coordinate values of the plurality of fixed points, and measures a straight line that goes straight to the plane including the calculated circle and passes through the center of the calculated circle A method for measuring a three-dimensional shape, characterized by using a rotation axis of an object. In the three-dimensional shape measuring method according to claim 3 or 7,
The three-dimensional shape measuring method, wherein the fixed point defining object is a sphere, and the fixed point is a center of the sphere. The three-dimensional shape measuring method according to any one of claims 2 to 8,
The calibration object has a coordinate axis in which an intersection line between a first plane portion defining a z-axis direction whose normal direction is one coordinate axis and a plane on which the first plane portion is formed is orthogonal to the z-axis. A second plane part that defines the x-axis direction and a third plane part that forms a line of intersection on a plane that is orthogonal to the z-axis and that is different from the plane on which the first plane part is formed And the fourth plane portion, and the x-axis is parallel to the rotation axis of the measurement object,
In the shift amount detection step, line light irradiates the first flat surface portion, the second flat surface portion, the third flat surface portion, and the fourth flat surface portion based on the light reception information obtained in the preliminary light receiving step. A method for measuring a three-dimensional shape, wherein a straight line equation including a cutting line is selectively calculated according to a shift amount to be obtained, and the shift amount is detected from the calculated straight line equation. The three-dimensional shape measuring method according to claim 9,
The calibration object includes an additional first plane part formed at a position parallel to the first plane part and spaced from the first plane part by a predetermined distance, and parallel to the second plane part and the first plane part. An additional second plane portion formed at a position spaced apart from the two plane portions by the predetermined distance,
In the shift amount detection step, line light is applied to the first plane portion, the second plane portion, the additional first plane portion, and the additional second plane portion based on the light reception information obtained in the preliminary light reception step. The equation of the straight line including the cutting line when irradiating is obtained on the camera coordinate system, and then the straight line including the cutting line of the line light irradiated to the first plane part and the second plane part are irradiated. A distance p1 from a straight line including a cutting line of line light, a straight line including a cutting line of line light irradiated to the additional first plane part, and a cutting line of line light irradiated to the additional second plane part are included. A method for measuring a three-dimensional shape, wherein a distance p2 to a straight line is obtained, and the amount of deviation is uniquely determined based on a ratio between the distance p1 and the distance p2. Rotation drive means for rotating the measurement object around the rotation axis;
Rotation angle detection means for detecting the rotation angle of the measurement object rotated by the rotation drive means;
A shape measuring instrument that emits line light substantially parallel to the rotation axis of the measurement object, receives reflected light of the line light irradiated on the surface of the measurement object, and outputs light reception information;
A rotation angle detected by the rotation angle detection means, and a data processing device to which the light reception information output from the shape measuring instrument is input,
The data processing device includes:
The position on the surface of the measurement object irradiated with the line light is determined on the basis of the received light information input from the shape measuring device for each predetermined rotation angle among the rotation angles input from the rotation angle detection means. Coordinates to be acquired at each predetermined rotation angle as a group of coordinate values in a coordinate system defined by the shape measuring instrument and having a line light irradiation plane parallel to one coordinate plane. A value acquisition means;
The coordinate value group acquired for each predetermined rotation angle by the coordinate value acquisition means is coordinated based on a deviation amount between a reference coordinate system defined by a calibration object stored in advance and the camera coordinate system. A coordinate conversion means for converting and acquiring a group of corrected coordinate values in which the deviation amount is corrected for each predetermined rotation angle;
A group of the corrected coordinate values for each of the predetermined rotation angles obtained by the coordinate conversion means is synthesized based on the predetermined rotation angle, and a group of synthesized coordinate values indicating the three-dimensional shape of the measurement object is obtained. Combining means to obtain;
A three-dimensional shape measuring apparatus comprising: While rotating the measurement object, irradiate the line light from the shape measuring instrument capable of emitting line light toward the measurement object substantially parallel to the rotation axis of the measurement object, and receive the reflected light of the irradiated light In the three-dimensional shape measurement method for measuring the three-dimensional shape of the measurement object, a reference coordinate system in which one coordinate axis is parallel to the rotation axis of the measurement object is defined, and the reference coordinate system and the shape measuring instrument are defined. A calibration object used to detect the amount of deviation from the camera coordinate system in which the coordinate plane is parallel to the irradiation plane of the line light,
A normal line is a coordinate axis in which an intersecting line between a first plane part defining a z-axis direction which is one coordinate axis of the reference coordinate system and a plane on which the first plane part is formed is orthogonal to the z-axis. A second plane part that defines a direction of a certain x axis, a third plane part that forms a line of intersection on a plane that is orthogonal to the z axis and that is different from the plane on which the first plane part is formed, and A calibration object comprising a fourth plane part. The calibration object according to claim 12,
The calibration object according to claim 1, wherein the first plane part and the second plane part have normal directions perpendicular to the rotation axis of the measurement object. The calibration object according to claim 12 or 13,
The calibration object, wherein an intersection line between the third plane part and the fourth plane part is parallel to the y-axis. The calibration object according to any one of claims 12 to 14,
An additional first plane portion formed at a position parallel to the first plane portion and spaced from the first plane portion by a predetermined distance; and parallel to the second plane portion and from the second plane portion to the predetermined plane A calibration object, comprising: an additional second flat surface portion formed at a position separated by a distance.

Images