JP2008003000A - Method of calibrating image measuring device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To inexpensively provide a method for calibrating an image measuring device for highly accurately calibrating the image measuring device by obtaining correction data to each point of image data. <P>SOLUTION: The calibration method comprises: imaging (image data acquiring process (ST30)) edges at different positions in an imaging region of a CCD camera while relatively moving a stage and the CCD camera in an x direction by preparing a calibration workpiece having the straight edge and mounting it on the stage so that the edge is parallel to a y axis of a machine coordinate system; detecting (imaging point detecting process) a position for imaging the edge from a drive amount of a drive mechanism; detecting (edge detecting process) the edge in the imaged image data to output the coordinate of the edge so as to obtain two-dimensionally widely distributed data; and calculating (displacement amount calculating process) each imaging point by regarding a displacement amount between an edge coordinate obtained by edge detection on the image data and an edge position obtained from the drive amount of the drive mechanism as an edge position displacement amount. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a calibration method for an image measuring apparatus. For example, the present invention relates to a calibration method for an image measurement apparatus that calibrates an error included in image data in an image measurement apparatus that measures the shape of a workpiece based on image data captured by an imaging unit.

An image measuring machine 10 is known as an apparatus for measuring the shape of an object to be measured.
As shown in FIG. 25, the image measuring machine includes a stage 11 on which the workpiece W is placed,
A light source (not shown) that illuminates light toward the workpiece W, an objective lens that collects reflected light from the workpiece W, a CCD camera 12 that captures an image formed through the objective lens, and an image from the CCD camera 12 An image processing unit 13 that performs image processing such as edge shape detection and edge detection of the workpiece W based on the data (for example, Patent Documents 1 and 2).
The CCD camera 12 can be moved in the X direction and the Y direction by a drive mechanism 14, and the drive mechanism 14 is driven and controlled by a drive control circuit 15. The driving amount of the CCD camera 12 by the driving mechanism 14 is detected by a driving sensor 16.

A method for detecting the coordinates of the measurement points on the workpiece based on the image data obtained by imaging the workpiece W with the image measuring machine 10 will be briefly described.
For example, consider a case where the workpiece W placed on the stage is imaged by the CCD camera 12 and the coordinates of the stain D on the workpiece surface are detected. First, the CCD camera 12 is moved by the drive mechanism 14 to put the spot D in the imaging area. At this time, the position of the CCD camera 12 is detected by the drive sensor 16 in advance. Then, the work W is imaged by the CCD camera 12 and the image data 17 is acquired. Then, for example, the image data 17 shown in FIG. 26 is obtained.

Next, a spot D is detected in the image data. That is, a pixel (pixel) corresponding to the spot D in the image data is detected. When a pixel corresponding to the spot D is detected, the coordinate value of that pixel is output. At this time, the pixel coordinate value of the pixel of the spot D is output with the center of the image data 17 as the origin O. For example, it is assumed that the pixel coordinate value of the pixel corresponding to the spot is (x _p , y _p ). Here, since the position of the CCD camera 12 is detected by the drive sensor 16, the position of the CCD camera 12 on the machine coordinate system is known. That is, the center position O of the image data 17 is already known in the machine coordinate system. On the other hand, only the pixel coordinate value (x _p , y _p ) from the center O of the image data is known for the position of the spot D. Accordingly, when the pixel coordinate value (x _p , y _p ) at the position of the spot D is multiplied by the pixel pitch which is the size (size) of one pixel (one pixel), the center O of the image data 17 to the spot D is obtained. Is calculated. For example, the pixel position x _p in the X direction is 300, the pixel pitch is assumed to be 20 [mu] m, the actual position of the stain as measured from the center O of the image data 17 is calculated by the following equation.

(Actual distance from the center of the image data) = (pixel position x _p ) × (pixel pitch)
= 300 (pixel) x 20 (μm / pixcel)
= 6000μm = 6mm

  Since the center position O of the image data 17 is obtained from the output value of the drive sensor 16, the actual position of the spot D is obtained as follows.

(Spot position) = (position of CCD camera) + (actual distance from the center of the image data)
= (CCD camera position) + (pixel position x _p ) × (pixel pitch)

  The above has been a description of the X direction, but it goes without saying that the same applies to the Y direction.

  By the way, in such an image measuring machine 10, calibration (calibration) is performed in advance for high accuracy, and correction data for correcting the measurement error to a true value is obtained. Then, the correction data is used to correct the measurement data obtained by the measurement to a true value. Thereby, the error which a machine has is processed like software, and very highly accurate measurement data are obtained.

An example of a parameter that affects measurement accuracy is the pixel pitch of image data. Since the pixel pitch also serves as a parameter for converting the pixel coordinate value to the actual distance, it is necessary to calibrate accurately.
Here, conventionally, when calibrating the pixel pitch, a scale having a known length is prepared, and the result of measuring the length of the scale is compared with a true value to obtain a pixel pitch that is the size of one pixel. I was looking for.

JP 2000-205822 A JP 2001-66112 A

However, there is an aberration of the objective lens as a factor causing an error affecting measurement data.
If there is an aberration in the objective lens, the image data 17 captured by the CCD camera 12 includes errors such as distortion. Since the aberration of the objective lens is different for each lens location, the error appearing in the image data 17 is also different for each area. Then, due to the aberration of the objective lens, the pixel pitch differs for each area of the screen data. For example, as shown in FIG. 27, when the tall distortion is strong as shown in (B) with respect to the ideal imaging of (A), a line passing through the central portion (y = yc) and a line passing through the peripheral portion. The pixel pitch is different from (y = ys). Therefore, even if the pixel pitch is obtained only from the result of measuring one reference length as described above, an optimal pixel pitch cannot be obtained for each area of the image data. Therefore, the pixel pitch cannot be calibrated with high accuracy, and there is a problem that measurement errors still remain.

Here, as shown in FIG. 28, a chart in which calibration points with known coordinates are two-dimensionally arranged is prepared, and the correction data at each point is obtained by comparing the result of measuring each calibration point with the true value. In principle, it is also possible.
However, it is very expensive to prepare a chart as shown in FIG. 28 for each magnification of the objective lens and to accurately obtain the coordinate value of each calibration point.
In addition, it is technically difficult to create a chart for a high-magnification objective lens.
Accordingly, there has been a demand for a method that is inexpensive and obtains highly accurate correction data for each area of image data.

  An object of the present invention is to provide a calibration method for an image measuring apparatus that is inexpensive and obtains correction data for each point of the image data to calibrate the image measuring apparatus with high accuracy.

  The calibration method for an image measuring apparatus according to the present invention includes a drive mechanism that moves an imaging unit having an objective lens relative to a stage on which an object to be measured is placed, on the image data captured by the imaging unit. In the calibration method of the image measuring apparatus for detecting the measurement point and calibrating the output value of the image measuring apparatus that outputs the position of the measurement point in the two-dimensional coordinate system composed of the x coordinate and the y coordinate set in the machine, A calibration work placing step for placing the calibration work on the stage in a state in which the direction of the edge of the calibration work having a straight edge is parallel to the y-axis direction of the two-dimensional coordinate system; and the stage by the drive mechanism An image data acquisition step of relatively moving the imaging unit in the x-axis direction and imaging the edge at different positions in an imaging region of the imaging unit; An imaging point detection step for detecting a relative position between the imaging means and the calibration work when the edge is imaged in an image acquisition step based on a driving amount of the drive mechanism, and an image data acquisition step. An edge detection step of performing edge detection based on the acquired image data, an edge position output step of outputting an x coordinate value of each edge at a plurality of preset y coordinates, and an edge position output step A shift amount calculation step of calculating a shift amount when each x coordinate value of the output edge is compared with an actual edge position based on the imaging point as an edge position shift amount, and the edge position at each position in the imaging region A correction data calculation step of calculating correction data in the x direction at each position on the image data based on the magnitude of the shift amount; A correction data storage step of storing in the corresponding positions paired correction data calculated by the data calculating step, further comprising the features.

In such a configuration, first, a calibration work having a straight edge is prepared.
Such a calibration work can be easily prepared because it requires only one straight edge.
Then, the calibration work is placed on the stage in a state where the edge is parallel to the y-axis of the machine coordinate system (calibration work placement step).
Next, the stage and the imaging unit are relatively moved in the x-axis direction by the driving unit.
That is, only the x coordinate changes in the image data while the edge remains parallel to the y axis. Then, edges are imaged at different positions within the imaging area of the imaging means (image data acquisition step). For example, when the driving mechanism is driven at a pitch that divides the imaging region into 10 equal parts in the x direction and 10 edges are imaged at 10 equal positions, 10 pieces of image data are obtained. At this time, the position where the edge is imaged is detected from the drive amount of the drive mechanism (imaging point detection step). Next, edge detection is performed in the captured image data (edge detection step). Then, the coordinates of the edge are output.

Here, since the edge is set in a state parallel to the y direction, the x coordinate is simultaneously detected for a plurality of y coordinates for one edge. Therefore, x coordinate values are output for a plurality of preset y coordinates (edge position output step). For example, y = y ₁ , y ₂ , y ₃ ... Are determined in advance, y = y ₁ and the intersection of the edge, y = y ₂ and the intersection of the edge, y = y ₃ and the edge Output the intersection of ... Since the edge is straight, if the objective lens of the image pickup means has no aberration, the x coordinate values output for each edge should be the same. The x coordinate value will be different.

  In this way, the imaged edge is detected at each imaging point, and x-coordinate values are output for a plurality of set y-coordinates. Then, data that is widely distributed two-dimensionally in the image data is obtained.

Here, since the relative position between the imaging unit and the calibration work is known in advance based on the drive amount of the drive mechanism in the imaging point detection step, the actual imaged edge is determined based on the relative position between the imaging unit and the calibration work. Can be calculated.
If there is no distortion in the image data picked up by the image pickup means, the coordinate value of the edge detected based on the image data should match the edge position specified from the drive amount of the drive mechanism.
However, when there is aberration in the objective lens of the image pickup means, the image data has distortion or the like, so the edge position obtained from the edge coordinates obtained by edge detection on the image data and the drive amount of the drive mechanism Deviation occurs.
Therefore, this deviation amount is calculated for each imaging point as an edge position deviation amount (deviation amount calculation step). Then, the shift amount in the x direction on the image data is obtained for a plurality of imaging points distributed two-dimensionally in the field of view.
As described above, since the shift amount of each point on the image data is obtained, correction data for correcting the detection value obtained from the image data to a true value is calculated based on these data (correction data calculation step). .
The obtained correction data is stored in a predetermined memory in pairs with corresponding points on the image data.

  When actually measuring a workpiece, the workpiece is imaged by an imaging means to obtain image data. Then, each measurement point is detected based on this image data. The correction value is corrected using the correction data corresponding to the detection value, and the corrected result is output. Then, the correct value with the corrected lens aberration is output.

  According to such a configuration, a calibration work having one edge is prepared, and this is moved to image the edges at a plurality of positions in the imaging field of view, so that data distributed two-dimensionally in the imaging region can be obtained. obtain. Since the influence of lens aberration on image data has a two-dimensional characteristic that it varies depending on the location, according to the present invention, image data can be calibrated two-dimensionally using one edge. That is, by capturing an image of one edge at different locations, it is possible to acquire the same image data as when capturing a calibration chart having a plurality of lines. Conventionally, since calibration was performed based on the result of measuring one reference length, it was not possible to calibrate an error distributed two-dimensionally such as lens aberration. Alternatively, when using a chart in which calibration points are obtained in two dimensions in advance, there is a problem that the cost for preparing such a chart is high. For example, it is very difficult for all users to prepare the calibration chart as described above for each magnification of the objective lens.

  In this regard, according to the present invention, it is very simple because it is only necessary to prepare a calibration work having one edge. Of course, it is not necessary to prepare a calibration work for each magnification, and one calibration work can correspond to all magnifications of the objective lens. Then, correction data for two-dimensionally correcting the image data can be obtained by this calibration work, and high-precision calibration can be realized while being inexpensive. In particular, when the user performs the calibration work by himself / herself, a special calibration chart is not required, and the advantage of being easy to perform greatly contributes to the convenience of the user.

Note that the length of the edge of the calibration work is not particularly limited, and may be a certain length in the y direction of the imaging region. For example, it may be about half the y-direction length of the imaging area, or about two-thirds the y-direction length of the imaging area, and of course may be longer than the y-direction length of the imaging area.
Also, the Y direction and the X direction are the coordinate axis directions of the two-dimensional coordinate system. For example, which of the vertical axis and the horizontal axis is the Y axis or the X axis is only a relative one. Of course, the calibration method of the present invention can also be applied to the Y-axis and the X-axis.

  In the present invention, the edge position output step includes an edge coordinate extraction step of extracting the edge position detected in the edge detection step on the pixel coordinates, and a pixel coordinate of the edge extracted in the edge coordinate extraction step in advance. An edge position calculating step of calculating an actual distance from the center of the image data to the edge by multiplying a set pixel pitch, and the correction data calculating step includes a pixel pitch correcting step of correcting the pixel pitch. The pixel pitch correction step includes an edge position shift for plotting the edge position shift amount calculated in the shift amount calculation step on a two-dimensional coordinate with respect to the pixel coordinates of the edge extracted in the edge coordinate extraction step. A regression line calculation step for calculating a regression line for linear regression of the points plotted in the quantity plotting step and the edge position deviation amount plotting step And a pixel pitch correction data calculation step for calculating pixel pitch correction data for correcting a pixel pitch in the x direction at each position on the image data based on the slope of the regression line, and the correction data storage step Preferably includes a pixel pitch correction data storage step for storing the pixel pitch correction data calculated in the pixel pitch correction data calculation step in pairs with each position of the image data.

In such a configuration, when the edge position shift amount is calculated for each point in the shift amount calculation step, the pixel pitch of the image data is calibrated based on the edge position shift amount. In calibrating the pixel pitch, first, the edge position deviation amount obtained for each pixel coordinate is plotted on two-dimensional coordinates (edge position deviation amount plotting step). In this case, position error data at each point (pixel coordinates) of the two-dimensionally distributed in the image data has been demanded. For example, each point on the focusing on the line y = y ₁ y = y ₁ The imaging point and the edge position deviation amount are plotted for. For example, each pixel coordinate value at which an edge is detected is taken on the horizontal axis, and the amount of edge position deviation with respect to each pixel coordinate is taken on the vertical axis. When the points are plotted in this way, the edge shift amount tends to increase in the peripheral region of the image data with respect to the vicinity of the center of the image data. For example, when a pixel pitch larger (or smaller) than the optimum pixel pitch is set, the error from the actual distance increases as the pixel coordinate value becomes farther from the origin. Therefore, in the regression line calculation step, a regression line that linearly regresses the plotted points is calculated. Such a regression line can be obtained by a general least square method. Then, a pixel pitch correction amount is calculated based on the obtained slope of the regression line, and the correction amount is corrected to obtain the pixel pitch in the x direction. That is, a pixel pitch obtained by correcting the inclination of the regression line from the currently set pixel pitch is obtained and used as pixel pitch correction data (pixel pitch correction data calculation step). The pixel pitch correction data thus obtained is stored, and detection based on the corrected pixel pitch is performed in detecting the position on the image data in actual measurement.

According to such a configuration, it is possible to calibrate a different pixel pitch for each location from the central region of the image data to the vicinity of the periphery according to the location.
For example, when there is lens aberration, the pixel pitch, which is the minimum unit of detection accuracy, differs between the central part of the image data and the peripheral part of the image data. Since only one representative value of the pixel pitch was calibrated simply by measuring, the calibration was not accurate.
In this respect, according to the present invention, the pixel pitch can be corrected in detail according to each position, so that highly accurate calibration can be performed.

In addition, after performing an edge position shift amount plotting step for one line (for example, y = y ₁ ) and calculating a regression line (regression line calculating step), the line is changed and y = y ₂ and y = y ₃ Of course, it is preferable to obtain the pixel pitch for each calibration line in the same manner.

  In the present invention, the correction data calculation step includes a position correction parameter calculation step for calculating a position correction parameter for correcting a positional deviation on the image data, and the pixel pitch calculated by the pixel pitch calculation step is used as an edge position. Based on the imaging point, the pixel pitch correction step for calculating the corrected edge position by correcting from the position of each edge output in the output step, and the corrected edge position calculated in the pixel pitch correction step A residual calculation step of calculating a residual of both in comparison with an actual edge position, and the correction data storage step includes the residual calculated in the residual calculation step in each image data And a position correction parameter storing step of storing as a position correction parameter in the x direction at the position.

In such a configuration, after correcting the pixel pitch, a position correction parameter for correcting the positional deviation is calculated.
First, since the pixel pitch after correction is obtained by the pixel pitch calculation step, the edge position is newly calculated using the pixel pitch after correction (pixel pitch correction step). Then, by comparing the edge position calculated after correcting the pixel pitch with the actual edge position based on the imaging point, an error (residual) remaining with the pixel pitch corrected is calculated (residual). Difference calculation step). This error is considered to be caused by distortion or the like generated in the image data due to the aberration of the objective lens, and the amount of error varies depending on each position of the image data. The residual obtained for each point is paired with the corresponding position and stored as position correction data.

  According to such a configuration, it is possible to obtain correction data that corrects a positional shift that differs depending on the location of the image data in accordance with the location. A very high-accuracy image measurement result can be obtained by appropriately correcting an error that varies depending on the location of the image data due to lens aberrations using correction data corresponding to the location. Further, since the position correction parameter is calculated based on the data after correcting the pixel pitch, it is correction data based on a value reflecting only the positional deviation. As described above, the position correction parameter for optimally correcting the measurement error can be calculated by a series of calibrations in which the position correction parameter is calculated following the correction of the pixel pitch.

  In the present invention, it is preferable that the image data acquisition step is repeatedly executed a plurality of times, and the edge position output step outputs an average value of edge positions acquired at the same imaging point.

  According to such a configuration, the correction data is obtained based on the average value of the results obtained by a plurality of times of imaging, so that, for example, more optimal correction data can be obtained without being influenced by a single abnormal detection error. Obtainable.

  In the present invention, in the calibration method of the image measuring apparatus, it is preferable to perform calibration in the y direction by exchanging the x direction and the y direction.

  According to such a configuration, correction data in both the x direction and the y direction is obtained, so that an accurate position on a two-dimensional coordinate in which the detection value is corrected in both the x direction and the y direction can be obtained.

  In the present invention, it is preferable that the calibration method of the image measuring apparatus is executed for each magnification of the objective lens to perform calibration for each magnification of the objective lens.

In such a configuration, if the objective lens is different, the aberration is naturally different. Therefore, calibration is performed for each objective lens to obtain correction data.
Here, if a calibration chart is prepared for each objective lens, it is very expensive. However, in the present invention, it is only necessary to prepare a calibration work having an edge for calibration. Even if calibration is performed for each lens, the cost does not increase.

DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments of the present invention will be illustrated and described with reference to reference numerals attached to respective elements in the drawings.
(First embodiment)
A first embodiment according to a calibration method for an image measuring apparatus of the present invention will be described.
First, the configuration of the image measuring apparatus to be calibrated will be briefly described.
FIG. 1 is a diagram illustrating a configuration of the image measurement apparatus 100.
The image measuring apparatus 100 includes an image measuring machine main body 200, a control unit 300, an input unit 111, and an output unit 112.

  The image measuring machine main body 200 includes a base 211 that serves as a base, a stage 212 that is disposed on the base 211 and on which a work W is placed, a CCD camera 213 that captures an image of the work W, and a CCD camera 213. And a workpiece W are moved three-dimensionally, and a drive sensor 219 (see FIG. 4) for detecting the drive amount of the drive mechanism 214 is provided.

The CCD camera 213 has an objective lens (not shown) inside.
The objective lens is configured by combining a plurality of lenses so as to correct aberrations as much as possible. However, it is impossible to completely correct all aberrations, and residual aberrations remain. Examples of such lens aberration include coma, astigmatism, and curvature of field. For example, when a grating as shown in FIG. 2 is imaged through a lens with aberration, errors and distortions occur as shown in FIG.
Here, the main object of the present invention is to correct the error of the image data due to the aberration of the objective lens at low cost and with high accuracy.

The drive mechanism 214 includes a Y slider 215 that is slidable in the Y-axis direction along the side edge of the stage 212, and an X slider 218 that is supported by the Y slider 215 and is slidable in the X direction. It is equipped with. The Y-slider 215 is a portal frame and has two beam supports that stand on both side ends of the stage 212 and have a height in the Z direction perpendicular to the stage 212 and are slidable in the Y-axis direction. 216 and a beam 217 supported on the upper end of the beam support 216 and having a length in the X direction.
The X slider 218 is slidably provided on the beam 217. The CCD camera 213 is attached to and held by the X slider 218.

  The Y slider 215 and the X slider 218 are driven in the Y axis direction and the X axis direction by a predetermined drive motor, respectively.

The drive sensor 219 includes a Y-axis sensor 220 that detects the amount of movement of the portal frame in the Y-axis direction, and an X-axis sensor 221 that detects the amount of movement of the X slider 218 in the X-axis direction.
Such a drive sensor 219 can be constituted by, for example, a linear encoder.

  The control unit 300 operates a motion controller 400 that controls the operation of the image measuring machine main body 200, operates the image measuring machine main body 200 via the motion controller 400, and processes image data acquired by the image measuring machine main body 200. And a host computer 500 for determining the size and shape of the workpiece W.

  FIG. 4 is a diagram showing an internal configuration of the motion controller 400 and the host computer 500. The motion controller 400 is imaged by a drive control circuit 410 that controls the drive mechanism 214, a drive counter 420 that counts detection signals from the drive sensor 219 and detects the drive amount of the drive mechanism 214, and a CCD camera 213. An image data capturing unit 430 that captures image data.

The host computer 500 includes a calibration operation command unit 510, a data storage unit 520,
The image processing unit 530 includes an arithmetic processing unit 540, a memory 570, a correction data storage unit 580, a correction processing unit 591, and a CPU (central processing unit) 592.

The calibration operation command unit 510 includes a drive command unit 511 and an imaging command unit 512.
The data storage unit 520 includes an image data storage unit 521, an imaging point storage unit 522, and an edge coordinate storage unit 523.
The image processing unit 530 includes an edge detection unit 531 and an edge coordinate extraction unit 532.
The arithmetic processing unit 540 includes an average value calculation unit 541, an edge position calculation unit 542, a deviation amount calculation unit 543, a pixel pitch calculation unit 550, and a position correction parameter calculation unit 560.

The pixel pitch calculation unit 550 includes an edge position deviation amount plot unit 551, a regression line calculation unit 552, and a pixel pitch correction data calculation unit 553.
The position correction parameter calculation unit 560 includes a pixel pitch correction unit 561 and a residual calculation unit 562.

  The correction data storage unit 580 includes a pixel pitch correction data storage unit 581 and a position correction parameter storage unit 582.

  The operation of each functional unit will be described later with reference to a flowchart.

A method for calibrating an image measuring apparatus having such a configuration will be described.
Note that the case of performing calibration in the X direction will be described as an example. However, in the case of performing calibration in the Y direction, it is only necessary to replace X and Y in the following description.

  In the calibration method of the present embodiment, as shown in the flowchart of FIG. 5, the calibration work placing step (ST100) for placing the calibration work Wc having the edge E on the stage 212, and conditions such as the calibration precision level A calibration condition setting step (ST200) for setting the data, a data acquisition step (ST300) for imaging the calibration work Wc and acquiring data as a basis for calibration, calculation of coordinate data by processing the captured image data, etc. An image processing step (ST400) to be performed, a correction data calculation step (ST500) for calculating correction data by comparing an error in the image data with a true value, and a correction data storage step (ST600) for storing the calculated correction data And.

First, in the workpiece placing step (ST100), the calibration workpiece Wc is placed on the stage 212.
FIG. 6 is a diagram illustrating an example of the calibration work Wc.
The calibration work Wc has a straight edge E, and for example, a plate whose half is covered with a chrome mask can be used. When placing the calibration work Wc on the stage 212, for the convenience of calibration in the X direction, the edge E is set in a state parallel to the Y axis as shown in FIG.
In the case of performing calibration in the Y direction, the edge E may be made parallel to the X direction.

After setting the calibration workpiece Wc in the workpiece placement step (ST100), the calibration conditions are set in ST200 (calibration condition setting step).
FIG. 8 is a flowchart showing the procedure of the calibration condition setting step.
As calibration conditions, a calibration line setting step (ST210), an imaging edge interval setting step (ST220), and a repetition count setting step (ST230) are performed.
Each step will be briefly described.

In the calibration line setting step (ST210), as shown in FIG. 9, a plurality of lines Lc parallel to the X axis are set in the imaging region 601. For example, as shown in FIG. 9, five lines Lc may be set at regular intervals so that the Y direction of the imaging region 601 is equally divided into six. In this case, since the calibration line Lc is a line parallel to the X axis, the y coordinate value of the line may be set and inputted. For example, the values y ₁ , y ₂ , y ₃ , y ₄ and y ₅ may be input. The fineness of the pitch of the calibration line Lc determines the density of the calibration points in the y direction in the subsequent calibration procedure.

In the imaging edge interval setting step (ST220), as shown in FIG. 10, a plurality of lines parallel to the y-axis are set in the imaging region 601.
For example, 19 lines may be set at regular intervals so that the X direction of the imaging region 601 is equally divided into 20. In this case, an interval between lines may be input.
The imaging edge interval set here is the interval for imaging the edge E in the subsequent calibration procedure, and determines the density of calibration points in the X direction.

  In the repetition number setting step (ST230), the number of times the edge E is imaged at the same position is set. In calibration based only on a single imaging result, accurate calibration may not be possible due to the influence of abnormal data. Of course, it is preferable to use the average value of data obtained by multiple imaging. is there. Therefore, the number of times the edge E is imaged at the same position in the repetition number setting step (ST230) is set, and the number of data used for calculating the average value is set. The calibration conditions set in this way are stored in the memory 570.

After setting the calibration conditions in this way, a data acquisition step is executed in ST300. The data acquisition process will be described with reference to the flowchart of FIG.
In the data acquisition step (ST300), first, an image data acquisition step (ST310) is performed. This image data acquisition step (ST310) includes a driving step (ST311) and an imaging step (ST312), and is executed by a command from the calibration operation command unit 510. In order to execute the image data acquisition step (ST310), the calibration operation command unit 510 first reads the imaging edge interval among the calibration conditions set in the memory 570. Then, the drive command unit 511 of the calibration operation command unit 510 outputs a command to move the CCD camera 213 in the x direction by the set imaging edge interval to the drive control circuit 410. Then, the drive mechanism 214 is driven by the control pulse from the drive control circuit 410, and the CCD camera 213 is moved in the X direction by the imaging edge interval (see FIG. 12).

For example, initially, as shown in FIG. 13A, it is assumed that the edge E is located on the right end side (+ X direction side) in the imaging region 601.
When the CCD camera 213 is displaced in the + x direction by the imaging edge interval from this state, the edge E is displaced in the −X direction in the imaging region 601 as shown in FIG. Then, when the drive mechanism 214 is driven and the CCD camera 213 is moved by the imaging edge interval, the imaging command unit 512 of the calibration operation command unit 510 commands the image data capturing unit 430 to acquire image data. . Then, imaging is performed by the CCD camera 213 (imaging process ST312), and image data is captured.

  The image data 602 thus captured is stored in the image data storage unit 521 in the image data storage step (ST320).

  Further, simultaneously with the capture of the image data 602 (ST320), the detection of the imaging point is also performed (imaging point detection step ST330). The position of the CCD camera 213 when the edge E is imaged can be detected from the count value of the drive counter 420. The position of the CCD camera 213 corresponds to the center O of the image data 602. At the time of imaging, the drive amount of the CCD camera 213 is detected from the drive counter 420, and the position of the CCD camera 213 (the center position of the image data) is calculated based on the amount of movement. Are stored in the imaging point storage unit 522 as imaging points.

  In this manner, the movement of the CCD camera 213 by the drive mechanism 214 (ST310) and the imaging by the CCD camera 213 (ST312) are performed. Then, after one stroke from the end to the end of the imaging region 601 is completed (ST340: YES), until the repetition of the number of repetitions set in the memory 570 is completed (ST350: YES), the image data An acquisition step (ST310) is executed.

  When the image data 602 is acquired in this way, for example, as shown in FIGS. 13A, 13B, 13C, image data 602 whose edge positions differ by a constant pitch is acquired. It will be.

After the image data 602 is acquired in the data acquisition step (ST300), the image processing step (ST400) by the image processing unit 530 is performed.
The image processing step (ST400) will be described with reference to the flowchart of FIG.
In the image processing step (ST400), an edge detection step (ST410) for detecting an edge in the image data 602, an edge coordinate extraction step (ST420) for extracting the coordinates of the detected edge E, and each point for the extracted edge coordinates. An average value calculating step (ST430) for calculating an average value in FIG. 5, an edge position calculating step (ST440) for calculating an edge position by calculation based on the image data 602, and the edge position calculated in ST440 as an actual edge A deviation amount calculating step (ST450) for calculating the deviation amount when compared with the position as the edge position deviation amount is executed.

In the edge detection in the edge detection step (ST410), the edge detection unit 531 detects the edge E running in the y direction in each image data 602. The image data 602 is captured at different imaging points, and a plurality of image data 602 is also captured at the same imaging point for calculating the average value. Edge detection is performed on all the image data 602. I do. By this edge detection, for example, the edge E in the image data 602 shown in FIGS. 13A, 13B, and 13C is detected.
Then, the pixel corresponding to the edge E is specified.

In the edge coordinate extraction step (ST420), the pixel coordinates of the detected edge E are extracted by the edge coordinate extraction unit 532.
Here, pixel coordinates are extracted in correspondence with the coordinates of y set as the calibration line Lc in the calibration condition setting step (ST200). In the calibration line setting step (ST210), as shown in FIG. 9, when a plurality of calibration lines Lc parallel to the X axis are set, the x coordinate value at the intersection of these calibration lines Lc and the edge E is extracted. (See FIG. 15). Since the edge E is imaged at a plurality of different positions in the imaging region 601 and a plurality of calibration lines Lc are set at different y coordinates, they are acquired as different images for easy understanding. When the images of the edge E are overlapped and expressed as shown in FIG. 16, the coordinate value of the lattice point that is the intersection of the plurality of edges E and the calibration line Lc in the image data 602 is calculated.

Since the calibration line Lc is a set value, the y coordinate value of one calibration line Lc is of course the same value.
On the other hand, the edge E is detected from the image data. Since the edge E of the calibration work Wc is straight, if there is no aberration in the objective lens, the edge E should be detected in the image data 602 while maintaining straightness. That is, when x coordinate values are extracted with a plurality of y coordinates at one edge E, the x coordinate values should all be the same.
However, since there is aberration in the objective lens, the image data 602 has distortion or the like, and even if the straight edge E of the calibration work Wc is detected, the distorted edge E is detected in the image data. It will be. Then, when x coordinates are extracted with a plurality of y coordinate values for one edge E, a value shifted by the amount of lens aberration is obtained.

  Edge E detection (ST410) and edge coordinate extraction (ST420) are performed for each piece of image data, and the obtained data is sent to the arithmetic processing unit 540.

Next, in the average value calculation step (ST430), the average value is calculated.
In the imaging of the edge E, since a plurality of image data is captured at the same imaging point for calculating the average value, the x coordinate value calculated for the same y coordinate in the plurality of image data captured at the same imaging point is used. Calculate the average value. For example, assuming that three pieces of image data are acquired at the same imaging point as shown in FIGS. 17A, _17B , and _17C , average values of the x coordinate values x _a , x _b , and x _c are calculated. The average value calculation step (ST430) is executed by the average value calculation unit 541.

  Next, in the edge position calculation step (ST440), the edge position is calculated by the edge position calculation unit 542. Here, the position of the CCD camera 213 has already been detected as an imaging point, and this corresponds to the center position O of the image data 602. The pixel coordinate value of each edge E has already been extracted by the edge coordinate extraction step (ST420). Therefore, first, when the actual distance from the center O of the image data 602 to the edge E is calculated based on the pixel coordinate value, and the distance is added to (or subtracted from) the center coordinates of the image data, the actual state of the edge E is calculated. Will be calculated.

Here, in calculating the actual distance from the center of the image data 602 to the edge E based on the pixel coordinate value, the pixel coordinate value is multiplied by a pixel pitch which is the size of one pixel (pixel). As the pixel pitch, an initially set value may be used, or a pixel pitch obtained by previous calibration may be used.
In this manner, the edge position based on each pixel coordinate value extracted in the edge coordinate extraction step (ST420) is calculated.

  Of course, since the calibration work Wc is not moved in a state of being placed on the stage 212, if the edge position is accurately obtained, all of the x coordinates thereof should be the same value. It is.

Next, in the shift amount calculation step (ST450), the shift amount calculation unit 543 calculates the edge position shift amount. Here, the edge position based on the image data 602 is calculated by the edge position calculation step (ST440). On the other hand, the actual position of the imaged edge can be detected as follows, for example.
For example, an edge is imaged in accordance with the center O of the image data 602.
Since the imaging point and the edge position coincide with each other at this position, the distance between the center of the image data and the actual edge can be detected by tracking the movement amount of the CCD camera from the count value of the drive counter. Therefore, an actual deviation amount between the edge position detected based on the image data and the actual edge position is calculated. At this time, as shown in FIG. 16, the pixel coordinates of the edge E are calculated for grid points that are two-dimensionally distributed in the image data. Calculate the amount. That is, the deviation amount in the x direction is calculated from the edge positions detected from the image data 602 and the actual edge E for all the calibration points. The calculated edge position deviation amount is stored in pairs with the pixel coordinates of the grid point (calibration point).

  The correction data calculation step (ST500) is executed based on the edge position deviation amount calculated for each calibration point in this way. As shown in FIG. 18, a pixel pitch calculation step (ST510) by the pixel pitch calculation unit 550 and a position correction parameter calculation step (ST520) by the position correction parameter calculation unit 560 are performed.

  The pixel pitch calculation step (ST510) includes an edge position deviation amount plotting step (ST511) by the edge position deviation amount plotting unit 551, a regression line calculation step (ST512) by the regression line calculation unit 552, and a pixel pitch correction data calculation unit 553. And a pixel pitch correction data calculation step (ST513).

In the edge position deviation amount plotting step (ST510), the edge position deviation amount calculated in the deviation amount calculation step (ST440) is plotted on two-dimensional coordinates.
FIG. 19 is an example in which the amount of edge position deviation is plotted on a graph.
In plotting points on the graph, data is plotted for each calibration line Lc. For example, the case where the calibration line Lc where y = y ₁ is calibrated will be described. A calibration point whose edge is detected on the calibration line Lc is read out, and the x-coordinate value of each calibration point is taken on the horizontal axis. FIG. 19 shows an example in which the x direction of the image data 602 is 640 pixels, and edge detection is performed by dividing the image data 602 into 20 equal parts. Then, the deviation amount calculated in the deviation amount calculation step (ST450) for each calibration point is plotted on the vertical axis. In FIG. 19, it can be seen that the amount of deviation increases as the detection of the edge position proceeds from the center of the image data 602 to the periphery. It can be seen that the pixel coordinate value is shifted in the positive direction from the actual value in the positive region and the pixel coordinate value is shifted in the negative direction from the actual value in the negative region. This is because the pixel pitch is set larger than the actual value. Therefore, a correction value for calibrating the pixel pitch to the actual ideal value is calculated next.

In the regression line calculation step (ST512), a regression line l for linear regression of each data point is calculated based on the graph created in the edge position deviation amount plotting step (ST511). Such a regression line l can be easily calculated using a least square method.
FIG. 20 is a diagram in which the calculated regression line l is superimposed on the graph.

  Such plotting of the edge position deviation amount (ST511) and calculation of the regression line l (ST512) are executed for all the calibration lines Lc.

When the regression line l is calculated for each calibration line Lc, the pixel pitch correction data calculation step (ST513) is performed next.
In the pixel pitch correction data calculation step (ST513), the slope of the regression line l is extracted. Then, the corrected pixel pitch is calculated by adding the inclination of the regression line l to the currently set pixel pitch. For example, in the examples of FIGS. 19 and 20, it is assumed that a regression line l represented by the following equation is obtained. The edge position deviation amount is represented by E.

E = 8.01 × 10 ^-6 x _p + 3.20 × 10 ^-4

Since the inclination is 8.01 × 10 ⁻⁶ , the corrected pixel pitch can be obtained by subtracting this inclination from the currently set pixel pitch. Such pixel pitch correction is executed for each calibration line Lc.

  Following the pixel pitch correction (ST510), a position correction parameter calculation step (ST520) is performed, and a pixel pitch correction step (ST521) by the pixel pitch correction unit 561 and a residual calculation step (ST521) by the residual calculation unit 562 ( ST522) is executed.

  In the pixel pitch correction step (ST521), the edge position is calculated using the pixel pitch correction value already calculated. That is, it is equivalent to redoing the above-described edge position calculating step (ST440) using the already calculated pixel pitch. Then, the pixel pitch is corrected. Subsequently, in the residual calculation step (ST522), a deviation amount between the newly calculated edge position and the actual edge position is calculated. That is, it is equivalent to redoing the above-described deviation amount calculation step (ST450) using the newly calculated edge position. FIG. 21 shows the edge position deviation amount corrected for the pixel pitch in this way.

As shown in FIG. 21, even after correction for the pixel pitch, the edge position deviation amount remains as a residual. This is because the image data is distorted due to the aberration of the objective lens in addition to the pixel pitch, and such misregistration has a different value for each point. Therefore, a correction value is set according to each point. is necessary. Therefore, the residual deviation is calculated for each calibration point.
Residuals are calculated for all calibration points. Since the residual calculated in this way becomes a position shift (measurement error) at each calibration point, these values are used as position correction parameters.

  After the correction data calculation step (pixel pitch calculation step ST510, position correction parameter calculation step ST520), the calculated correction parameters are stored in the correction data storage unit 580 in the correction data storage step (ST600). That is, the pixel pitch correction value is stored in the pixel pitch correction data storage unit 581 in the pixel pitch correction data storage step (ST610), and the position correction parameter is stored in the position correction parameter storage unit 582 in the position correction parameter storage step (ST620). Calibration mode ends.

  Since the above description is calibration in the X direction, when the calibration in the Y direction is continued, the direction of the calibration work Wc is changed so that the edge E is parallel to the X axis (see FIG. 22). ). Then, the same method may be executed by replacing X in the above calibration method with Y. Further, when a plurality of objective lenses are prepared for each magnification, calibration in the X direction and Y direction is executed for all the lenses, and the calculated correction parameters (pixel pitch and position correction parameters) are stored as correction data. Stored in the unit 580.

Next, the case where the workpiece W is image-measured using the image measuring apparatus 100 calibrated in this way will be briefly described.
A workpiece is placed on the stage 212 and imaged. The acquired image data has distortion corresponding to the aberration of the objective lens. For example, image data as shown in FIG. 23A is acquired. The pixel coordinates of the measurement point are detected in the captured image data. Subsequently, the pixel coordinate is multiplied by the pixel pitch, and further, the position is corrected by the positional deviation correction parameter. Then, as shown in FIG. 23B, the lens aberration is corrected.
Then, an optimally corrected measurement value is obtained.
Here, with respect to the measurement point P deviated from the calibration point, correction data of the calibration point surrounding the measurement point can be interpolated and used (see FIG. 24).
Such correction processing is executed by the correction processing unit 591 using the correction data in the correction data storage unit 580.

(1) In this embodiment, since it is only necessary to prepare a calibration work Wc having one edge E, it is very simple. Then, by imaging the calibration work at a plurality of positions in the imaging area, data distributed two-dimensionally in the imaging area can be obtained. That is, by capturing an image of one edge E at different locations, it is possible to acquire the same image data as when capturing a calibration chart having a plurality of lines. The influence of lens aberration on image data has a two-dimensional characteristic that it varies depending on the location. According to this embodiment, image data can be calibrated two-dimensionally using a single edge E. Further, it is not necessary to prepare a calibration work for each magnification, and one calibration work Wc can cope with all magnifications of the objective lens.
That is, correction data for two-dimensionally correcting image data can be obtained by the calibration work Wc, and high-precision calibration can be realized while being inexpensive. In particular, when the user performs the calibration work by himself / herself, a special calibration chart is not required, and the advantage of being easy to perform greatly contributes to the convenience of the user.

(2) Different pixel pitches can be calibrated according to the location. When there is lens aberration, the pixel pitch, which is the minimum unit of detection accuracy, will be different between the central part of the image data and the peripheral part of the image data, but the calibration is distributed two-dimensionally in the image data. Since the pixel pitch can be corrected in detail according to each area based on the points, high-precision calibration can be performed.

(3) It is possible to obtain a position correction parameter for correcting a position shift that differs depending on the location of the image data in accordance with the location. Further, since the position correction parameter is calculated based on the data after correcting the pixel pitch, the position correction parameter is correction data based on a value reflecting only the positional deviation. As described above, the position correction parameter for optimally correcting the measurement error can be calculated by a series of calibrations in which the position correction parameter is calculated following the correction of the pixel pitch. A very high-accuracy image measurement result can be obtained by appropriately correcting an error that varies depending on the location of the image data due to lens aberrations using correction data corresponding to the location.

It should be noted that the present invention is not limited to the above-described embodiment, and modifications, improvements, and the like within the scope that can achieve the object of the present invention are included in the present invention.
In the above-described embodiment, the case where the stage is fixed and the CCD camera moves has been described as an example. However, the present invention can be applied even when the CCD camera is fixed and the stage moves.
That is, it is only necessary that the calibration work can be imaged while the CCD camera and the calibration work having an edge are relatively moved.
Further, a predetermined calibration processing program may be installed in the computer, and the computer may be caused to execute each processing step of the image measuring machine calibration method. In this way, if the calibration processing program is installed in a computer and the computer executes each processing step, various parameters can be easily changed.
In order to install the calibration processing program, a memory card, a CD-ROM, or the like may be directly inserted into a computer, or a device that reads these storage media may be connected externally. Further, a calibration processing program may be supplied and installed by communication by connecting a LAN cable, a telephone line or the like to the signal processing device, or may be supplied and installed by wireless.

  The present invention can be used for calibration of an image measuring apparatus.

It is a figure which shows the structure of an image measuring device. The figure which shows an example of the workpiece | work which has a grating | lattice. The figure which shows an example of the image data at the time of imaging a grating | lattice through the lens which has an aberration. It is a figure which shows the internal structure of a motion controller and a host computer. The flowchart which shows the procedure of the calibration method. It is a figure which shows an example of a calibration work. The figure which shows the state which mounted the calibration workpiece | work on the stage. The flowchart which shows the procedure of a calibration condition setting process. The figure which shows the state which set the calibration line Lc parallel to the X-axis in the imaging region. The figure which shows the state which set the some line parallel to a y-axis within an imaging region. The flowchart which shows the procedure of a data acquisition process. The figure which shows a mode that a calibration workpiece | work is imaged, moving a CCD camera. The figure which shows the example of the image data which imaged the edge in a different position. The flowchart which shows the procedure of an image processing process. The figure which shows a mode that the x coordinate value in the intersection of a calibration line and an edge is extracted. The figure which expressed by superimposing the image of the edge E acquired as a different image. The figure which shows a mode that an average value is calculated from three image data imaged at the same imaging point. The flowchart which shows the procedure of a correction data calculation process. The figure which shows an example which plotted the edge position shift amount on the graph. The figure which accumulated the regression line l on the graph. The figure which represented the edge position shift amount by which the part for pixel pitch was corrected on the graph. The figure which changed the direction of the calibration workpiece | work Wc so that the edge E might become parallel to the X-axis. The figure which shows an example (A) of image data before correction | amendment, and an example (B) of image data after correction | amendment. The figure which shows a mode that the correction data of the calibration point surrounding a measurement point are interpolated and used. The figure which shows the structure of the conventional image measuring machine. The figure which shows an example of the image data which imaged the workpiece | work which has a stain. The figure which shows an example of the image data of (B) when tall type distortion appears strongly with respect to the ideal image formation of (A). The figure which shows an example of the chart which arranged the calibration point whose coordinate is known two-dimensionally.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Image measuring machine, 11 ... Stage, 12 ... CCD camera, 13 ... Image processing part, 14 ... Drive mechanism, 15 ... Drive control circuit, 16 ... Drive sensor, 17 ... Image data, 100 ... Image measuring device, 111 ... Input means 112 ... Output means 200 ... Image measuring machine main body 211 ... Stand base 212 ... Stage 213 CCD camera 214 ... Drive mechanism 215 ... Y slider 216 ... Beam support 217 ... Beam 218 ... X slider, 219 ... drive sensor, 220 ... Y axis sensor, 221 ... X axis sensor, 300 ... control section, 400 ... motion controller, 410 ... drive control circuit, 420 ... drive counter, 430 ... image data fetch section, 500: Host computer 510: Calibration operation command unit 511: Drive command unit 512: Imaging command unit 520: Data storage 521 ... Image data storage unit, 522 ... Imaging point storage unit, 523 ... Edge coordinate storage unit, 530 ... Image processing unit, 531 ... Edge detection unit, 532 ... Edge coordinate extraction unit, 540 ... Calculation processing unit, 541 ... Average Value calculation unit, 542 ... edge position calculation unit, 543 ... deviation amount calculation unit, 550 ... pixel pitch calculation unit, 551 ... edge position deviation amount plot unit, 552 ... regression line calculation unit, 553 ... pixel pitch correction data calculation unit, 560: Position correction parameter calculation unit, 561: Pixel pitch correction unit, 562 ... Residual calculation unit, 570 ... Memory, 580 ... Correction data storage unit, 581 ... Pixel pitch correction data storage unit, 582 ... Position correction parameter storage unit , 591... Correction processing unit.

Claims (6)

Provided with a drive mechanism for moving the imaging means having the objective lens relative to the stage on which the object to be measured is placed, and the measurement point is detected on the image data captured by the imaging means and set in the machine In the calibration method of the image measuring machine for calibrating the output value of the image measuring machine that outputs the position of the measurement point in the two-dimensional coordinate system composed of the x coordinate and the y coordinate,
A calibration workpiece placing step of placing the calibration workpiece on the stage in a state in which the direction of the edge of the calibration workpiece having a straight edge is parallel to the y-axis direction of the two-dimensional coordinate system;
An image data acquisition step of relatively moving the stage and the imaging unit in the x-axis direction by the driving mechanism and imaging the edge at different positions in an imaging region of the imaging unit;
An imaging point detection step of detecting a relative position between the imaging unit and the calibration work when the edge is imaged in the image data acquisition step as an imaging point based on a driving amount of the driving mechanism;
An edge detection step for performing edge detection based on the image data acquired in the image data acquisition step;
An edge position output step of outputting an x coordinate value of each edge at a plurality of y coordinates set in advance;
A deviation amount calculating step of calculating, as an edge position deviation amount, a deviation amount when each x coordinate value of the edge output in the edge position output step is compared with an actual edge position based on the imaging point;
A correction data calculation step of calculating correction data in the x direction at each position on the image data based on the magnitude of the edge position deviation amount at each position in the imaging region;
A correction data storage step of storing the correction data calculated in the correction data calculation step in a pair with a corresponding position. In the calibration method of the image measuring device according to claim 1,
The edge position output step includes
An edge coordinate extraction step for extracting the edge position detected in the edge detection step on the pixel coordinates, and an image obtained by multiplying the pixel coordinates of the edge extracted in the edge coordinate extraction step by a preset pixel pitch An edge position calculating step for calculating an actual distance from the center of the data to the edge,
The correction data calculation step includes a pixel pitch correction step of correcting the pixel pitch,
The pixel pitch correction step includes
An edge position deviation amount plotting step for plotting the edge position deviation amount calculated in the deviation amount calculation step on two-dimensional coordinates with respect to the pixel coordinates of the edge extracted in the edge coordinate extraction step;
A regression line calculating step for calculating a regression line for linear regression of the points plotted in the edge position deviation amount plotting step, and
A pixel pitch correction data calculating step for calculating pixel pitch correction data for correcting the pixel pitch in the x direction at each position on the image data based on the slope of the regression line, and
The correction data storage step includes a pixel pitch correction data storage step of storing the pixel pitch correction data calculated in the pixel pitch correction data calculation step in pairs with each position of the image data. Device calibration method. In the calibration method of the image measuring device according to claim 2,
The correction data calculation step includes a position correction parameter calculation step for calculating a position correction parameter for correcting a positional deviation on the image data,
A pixel pitch correction step for correcting the pixel pitch calculated in the pixel pitch calculation step from the position of each edge calculated in the edge position output step to calculate a corrected edge position;
A residual calculation step of calculating a residual between the corrected edge position calculated in the pixel pitch correction step and the actual edge position based on the imaging point,
The correction data storage step includes a position correction parameter storage step of storing the residual calculated in the residual calculation step as a position correction parameter in the x direction at each position on each image data. Calibration method for image measuring apparatus. In the calibration method of the image measuring device according to any one of claims 1 to 3,
The image data acquisition step is repeatedly executed a plurality of times,
The edge position output step outputs an average value of edge positions acquired at the same imaging point. 5. The calibration method for an image measuring apparatus according to claim 1, wherein the calibration is performed in the y direction by exchanging the x direction and the y direction. 6. A calibration method for an image measuring apparatus according to any one of claims 1 to 5, wherein the calibration is performed for each magnification of the objective lens by executing the calibration method for the image measuring machine according to any one of the magnifications of the objective lens.

Images