JP2011064550A - Surface profile measuring method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a surface profile measuring method which can measure the surface profile of an object under measurement using a probe easily and with high precision. <P>SOLUTION: The surface profile measuring method for measuring the surface profile of the object under measurement by moving the probe and the object under measurement relatively includes: a first acquisition process for acquiring first surface profile data in a first area of the object under measurement; a second acquisition process for acquiring second surface profile data in a second area of the object under measurement at least part of which overlaps with the first surface profile data; an approximate curve acquisition process S20 for acquiring an approximate curve by performing fitting with an approximate function on the data of the area overlapping with the second surface profile data out of the first surface profile data; a correction amount calculation process S30 for calculating a correction amount for the second surface profile data using the approximate curve; a coordinates conversion process S40 for performing coordinates conversion on the entire second surface profile data on the basis of the calculated correction amount; and a process for combining the first surface profile data and the second surface profile data on which the coordinates conversion has been performed. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a measurement method for measuring the surface shape of a test object, and more particularly to a measurement method for measuring the surface shape of a test probe by bringing the stylus into contact with the surface of the test object.

  Conventionally, as a device for evaluating the surface shape of optical-related elements such as lenses and molds, a contact scanning measurement method has been adopted in which a stylus (probe) follows the surface of an object such as an optical-related element. A surface shape measuring device is used.

  In such a surface shape measuring apparatus, when the test object is scanned by the drive mechanism while the probe is in contact with the test object, the probe follows the shape of the test object, and the probe is The attached stylus shaft is displaced. The shape of the test object can be obtained by acquiring the position coordinates of the stylus shaft to be displaced with a displacement meter and the position coordinates of the scanned test object with a displacement meter in time series.

  However, for example, when measuring the shape of a test object having a hemispherical surface with such a surface shape measuring apparatus, as shown in FIG. 7, the peripheral area A110 of the test object 110 has a strong inclination. As a result of an increase in the displacement amount per unit scanning amount of the probe 111, there is a problem that the measurement accuracy is likely to be lowered. Another problem is that the measurement error increases due to the bending deformation of the probe 111.

In order to solve this problem, Patent Document 1 describes a surface shape measuring apparatus including a probe having a different axial direction with respect to a test object. In the apparatus of Patent Document 1, since the probe axis method is different, the displacement amount per unit scanning amount becomes an appropriate size.
Further, Patent Document 2 discloses a surface shape measuring apparatus that measures the surface shape by scanning the surface shape of the workpiece while controlling the posture of the workpiece, which is a test object, so that the amount of displacement per unit scanning amount is appropriate. Are listed.

JP 2007-121260 A Japanese Patent No. 3923945

  However, in the surface shape measuring apparatus described in Patent Document 1, even if a probe having a different axial direction is provided, it is not necessarily compatible with a workpiece of any shape, and thus cannot be said to have high versatility. Further, when another probe having an axial direction parallel to the test object is provided, there is a problem that the manufacturing cost of the apparatus is increased. Furthermore, it is necessary to calibrate the sensitivity difference between the probes, but there is also a problem that it is difficult to perfectly match the sensitivity.

  Furthermore, when surface shape measurement is performed with the surface shape measuring devices described in Patent Document 1 and Patent Document 2, in order to obtain the surface shape of the entire test object, the probe and the test object are acquired in different positions. A work (fitting) is required to connect the plurality of partial surface shape data. In this case, since it is necessary to consider the tilt amount of each partial surface shape data, there is a problem that it is difficult to improve the accuracy because noise is likely to be introduced by simple fitting between data point sequences.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a surface shape measurement method that can easily and accurately measure the surface shape of a test object using a probe. .

  In the surface shape measurement method of the present invention, a probe is brought into contact with a test object, and the probe and the test object are moved relative to each other to cause the probe to scan the surface of the test object. A surface shape measuring method for measuring the surface shape of the test object, the first obtaining step of obtaining first surface shape data which is the surface shape data of the first region of the test object, and the first surface shape data A second acquisition step of acquiring second surface shape data, which is surface shape data of a second region of the test object at least partially overlapping, and the second surface shape of the first surface shape data An approximation curve acquisition step of obtaining an approximate curve by applying an approximation function to data in a region overlapping with the data, and a correction amount calculation for calculating a correction amount of the second surface shape data using the approximate curve Process and calculation Based on the corrected amount, the coordinate conversion step of converting the entire second surface shape data, the first surface shape data, and the second surface shape data subjected to the coordinate conversion are integrated. And a process.

  According to the surface shape measurement method of the present invention, when obtaining a correction amount necessary for fitting, an adjustable geometric parameter such as shift and tilt can be freely set by fitting an approximate function to point sequence data. Therefore, the versatility for fitting is great.

  In the second acquisition step, the test object and the probe may be changed to a different positional relationship from the first acquisition step. In this case, it is possible to acquire the first surface shape data and the second surface shape data while appropriately adjusting the tilt angle of the test object with respect to the probe and suppressing the occurrence of errors.

  The approximate function may be a NURBS function. In this case, since it is difficult to be affected by dust and foreign matter, and fitting can be performed by interpolating the missing data, an interpolated value or an approximate value corresponding to the actual surface shape of the test object can be calculated. In addition, since the weighting can be changed for each point with respect to a part of the point sequence data of the surface shape data to be fitted, the accuracy of the surface shape data acquired by suitably removing noise is improved. Can be improved.

In the approximate curve acquisition step, when the difference between two adjacent point sequence data among the point sequence data of the first surface shape data to be fitted is a predetermined value or more, the two point sequence data The weight of the NURBS function for one side may be reduced, or the weight of the point sequence data having a period corresponding to a predetermined frequency is changed among the point sequence data of the first surface shape data to be fitted. Also good.
In these cases, noise and the like can be automatically excluded regardless of the user's sequential determination, so that the efficiency of fitting can be further improved.

In the correction amount calculation step, the correction amount may be calculated so that a sum of squares of a difference between the approximate curve and the second surface shape data in a region corresponding to the approximate curve is minimized. Moreover, the surface shape measuring method of the present invention includes a second approximate curve acquisition step of acquiring a second approximate curve by fitting the second surface shape data in a region corresponding to the approximate curve with an approximate function. In addition, the correction amount may be calculated so that a correlation coefficient between the approximate curve and the second approximate curve is maximized in the correction amount calculation step.
In these cases, the process for calculating the correction amount can be simplified.

  According to the method for measuring a surface shape of the present invention, the surface shape of a test object can be easily and accurately measured using a probe.

It is a figure which shows an example of the surface shape measuring apparatus which performs the surface shape measuring method of one Embodiment of this invention. It is a flowchart which shows the flow of the surface shape measuring method using the surface shape measuring apparatus. (A)-(c) is a figure which shows operation | movement of the same shape measuring apparatus in the same surface shape measuring method. It is a figure which shows an example of the area | region setting of the partial surface shape data in a to-be-tested object. (A) is a figure which shows an example of the same partial surface shape data, (b) and (c) are figures which show an example of fitting of the approximate function with respect to the said partial surface shape data, respectively. It is a figure which shows an example of fitting of the approximation function with respect to the partial surface shape data. It is a figure which shows the positional relationship of the probe and test object in the conventional probe type surface shape measuring apparatus.

An embodiment of the present invention will be described with reference to FIGS. Unless otherwise specified, “front” and “back” refer to the negative and positive directions of the Z-axis shown in FIG. 1, respectively.
FIG. 1 shows an example of a surface shape measuring apparatus (hereinafter simply referred to as “shape measuring apparatus”) 1 that performs the surface shape measuring method (hereinafter simply referred to as “shape measuring method”) of the present embodiment. FIG. The shape measuring apparatus 1 includes a stylus part 10 to which a probe 11 is attached, a test object holding part 20 on which a test object 100 is supported, and a Y-axis drive mechanism 30 for moving the stylus part 10. And an X-axis drive mechanism 40 for moving the object holding part 20 and a personal computer 50 for controlling the operation of the entire shape measuring apparatus 1 and processing the acquired surface shape data.

The stylus part 10 includes a probe 11, a hydrostatic bearing 12 through which the probe 11 is inserted, and an inclination angle adjustment part 13 that adjusts the inclination angle of the probe 11.
A probe 11 having a known configuration can be appropriately selected and employed, and is inserted into the hydrostatic bearing 12 so as to be able to slide smoothly within the hydrostatic bearing 12. The hydrostatic bearing 12 is fixed on the base flat plate 14 with the probe 11 facing forward. A displacement meter 15 for detecting the amount of displacement of the probe 11 is attached to the rear end side of the probe 11.

  The inclination angle adjustment unit 13 includes a front fulcrum 13A and an expansion / contraction part 13B that can expand and contract in the Y-axis direction, and changes the inclination angle of the probe 11 to contact the surface of the test object 100. As the expansion / contraction part 13B, a mechanism including a piezoelectric actuator or a screw and an angle adjustment member can be employed.

The specimen holding unit 20 includes a base 21 attached on the base 2 and a swinging part 22 attached to the base 21. The test object 100 is fixed to the swinging portion 22 and is disposed to face the probe 11. The oscillating part 22 is oscillatable with respect to the base 21, and by changing the positional relationship between the oscillating part 22 and the base 21, a test object attached to the oscillating part 22 as will be described later. The angle formed by 100 and the probe 11 can be changed.
In the present embodiment, the surface to be measured is a spherical object, but the shape of the surface to be measured is not particularly limited.

  The Y-axis drive mechanism 30 is provided between the stylus part 10 and the base 2 and moves in the Y-axis direction to maintain the inclination angle of the probe 11 set by the inclination angle adjustment part 13. The stylus part 10 can be moved up and down as it is. A displacement meter 31 for detecting the amount of movement of the Y-axis drive mechanism is provided behind the Y-axis drive mechanism 30.

  The X-axis drive mechanism 40 is provided between the base 21 of the test object holding unit 20 and the base 2 and moves in the X-axis direction so that the entire test object holding unit 20 is moved in the X-axis direction. Can be moved. In front of the X-axis drive mechanism 40, a displacement meter 41 for detecting the displacement of the specimen holding unit 20 in the X-axis direction is provided.

  The personal computer 50 includes an input unit 51 that receives user input, instructions, and the like, and a display unit 52 that displays various types of information. The personal computer 50 is connected to the tilt angle adjusting unit 13, the swinging unit 22, the Y-axis drive mechanism 30, and the X-axis drive mechanism 40, and can control the operations of these mechanisms. The personal computer 50 is also connected to each of the displacement gauges 15, 31, and 41, receives these detection values, reconstructs them into a plurality of partial surface shape data of the test object 100, and will be described later. A process of acquiring the surface shape data of the entire test object 100 by connecting the partial surface shape data is performed.

The operation of the shape measuring apparatus 1 having the above configuration will be described below. FIG. 2 is a flowchart showing a flow of the shape measuring method of the present embodiment using the shape measuring apparatus 1.
This shape measurement method was obtained by obtaining a partial shape data acquisition step S10 for acquiring a plurality of partial surface shape data of the test object 100, and an approximate curve acquisition step S20 for acquiring an approximate curve for a part of the partial surface shape data. A correction amount calculation step S30 for calculating the correction amount of each partial surface shape data based on the approximate curve, a coordinate conversion step S40 for performing coordinate conversion of the partial surface shape data based on the obtained correction amount, and after the coordinate conversion And a partial shape data integration step S50 for obtaining the surface shape data of the entire test object by joining the partial shape data.

  First, in step S <b> 10, surface shape data of the test object 100 is acquired using the probe 11. The details are substantially the same as the data acquisition using a known probe type surface shape measuring apparatus, but are as follows.

  First, the user supports and fixes the test object 100 on the swinging part 22 of the test object holding part 20 so that the surface on which the shape measurement is performed faces the stylus part 10. Next, when the user operates the inclination angle adjusting unit 13 to move the rear portion of the base flat plate 14 upward, the base flat plate 14 is inclined with the fulcrum 13A as a fulcrum. The probe 11 held by the hydrostatic bearing 12 slides smoothly in the hydrostatic bearing 12 by its own gravity. By sliding, the tip of the probe 11 moves forward and comes into contact with the surface of the test object 100 and stops.

  In this state, the personal computer 50 drives the X-axis drive mechanism 40 and the Y-axis drive mechanism 30 to move the test object 100 and the probe 11 relative to each other, thereby causing the probe 11 to scan the surface of the test object 100. . Detection values of the displacement meters 15, 31 and 41 are input to the personal computer 50.

  In the case of a test object having a spherical surface, such as the test object 100, the area near the center of the test object faces the test object 100 and the probe 11 as shown in FIG. In this state, the surface can be scanned by the probe 11 without any problem.

  On the other hand, in the region near the periphery of the test object 100, the surface inclination becomes large. Therefore, when scanning is performed in the positional relationship as shown in FIG. 3A, an error is likely to occur and the error is likely to be large. . Therefore, the user swings the swinging portion 22 via the personal computer 50, changes the posture of the test object 100 as shown in FIG. 3B and FIG. The scanning is performed by adjusting the angle formed with respect to. 3 (a) to 3 (c) are views of the test object 100 and the like as viewed from above. Therefore, in FIGS. 3 (b) and 3 (c), the swinging portion 22 is in the horizontal direction. However, if there is a region with a large tilt angle in the vertical direction on the surface of the test object, the swinging portion 22 is swung in the vertical direction as necessary. After that, scanning with the probe 11 may be performed.

  In this way, after the surface of the test object 100 is scanned by the probe 11, the personal computer 50 detects the detection values received from the displacement meters 15, 31 and 41 for each positional relationship between the probe 11 and the test object 100. A plurality of partial surface shape data are obtained by reconstructing them together.

  FIG. 4 shows an example of partial surface shape data. In FIG. 4, the probe 11 and the test object are operated by operating the region R <b> 1 near the center scanned with the probe 11 and the test object facing the surface of the test object 100 and the rocking unit 22. The partial surface shape data of a total of five regions, ie, the regions R2 to R5 around the region R1 scanned in a state where the positional relationship is changed.

  The range of each partial surface shape data is set so as to share an overlapping region with at least one other partial surface shape data so that fitting can be performed in a later step. Note that the areas R1 to R5 shown in FIG. 4 are examples, and the number and position of the areas can be freely set as long as each partial surface shape data shares an overlapping area with at least one other partial surface shape data. It does not matter.

In the subsequent approximate curve acquisition process of step S20, the user becomes a fitting index from among the partial surface shape data of the arbitrary region acquired in step S10, from the regions overlapping with other partial surface shape data. Extract the index area. Then, the personal computer 50 fits the partial surface shape data of the index region with an approximate curve, and acquires an approximate curve as a fitting reference.
The user can freely select an arbitrary area, but for convenience of explanation, in the following, a partial surface of the area R1 (first area) obtained by facing the probe 11 and the test object 100 in front of each other An example of fitting the surrounding regions R2 to R5 (second region) using the shape data (first surface shape data) as master data will be described.

  As shown in FIG. 4, the region R1 has overlapping regions SR1 to SR4 that overlap with two of the regions R2 to R5. When an index region is extracted from such a region, fitting with two regions can be performed. For example, if the index region in the overlapping region SR1 is used, fitting with the regions R2 and R5 sharing the overlapping region SR1 can be performed. Similarly, if the index region in the overlapping region SR3 is used, the fitting with the regions R3 and R4 can be performed. Therefore, if the index areas are set in the overlapping areas SR1 and SR3 or SR2 and SR4 facing each other, all areas can be fitted.

  As the index region, for example, a region having a fine scratch or the like can be used. There is no particular limitation on the shape etc. that the index area should have, but if the same shape happens to be in the overlap area where extraction is performed, there is a possibility that an error will occur in fitting as a result of erroneous setting of the index area. A region including a shape having asymmetry is preferable.

  After setting the index region, the user takes out the portion of the index region set in the overlapping region SR1 from the partial surface shape data of the region R1, performs fitting using the NURBS function shown in Equation 1, and A curve indicating the approximate curved surface (approximate curve) of the surface shape data is acquired as a function. In addition to the NURBS function, the fitting method includes a least square method, a linear / nonlinear programming method, a search method, and the like, and any method may be used for fitting.

  In the NURBS function, weighting can be performed for each coordinate by changing the value of ω. Therefore, when it can be determined that the value of a certain coordinate is an incorrect value due to a measurement error or dust attached to the surface of the test object 100, the user does not give a weight to the coordinate, or more than other coordinates. By reducing the weighting, an approximate curve that is more faithful to the surface shape of the test object can be obtained.

  For example, in the case of the surface shape data DM of the index area shown in FIG. 5A, when it can be determined that the point sequence data C1 is an incorrect value due to dust or the like, in the fitting using a normal polynomial or the like, FIG. ), The approximate curve A1 including the noise of the point sequence data C1 is acquired. However, when the weight of the point sequence data C1 is reduced using the NURBS function, the point sequence as shown in FIG. The personal computer 50 obtains an approximate curve A2 more faithful to the surface shape of the test object 100, in which the influence of the data C1 is eliminated or corrected.

The determination as to whether or not the value is an illegal value may be made sequentially by the user looking at the surface shape data and may be made via the input unit 51 of the personal computer 50, or the conditions for the determination may be set in a program or the like. Automatic discrimination may be performed. For example, it is determined that an outlier that falls outside a predetermined range is an incorrect value, or when the difference between adjacent point sequence data is greater than or equal to a predetermined value, the point sequence data that is further away from the average value of the point sequence data Conditions such as judging an illegal value. In this way, the work efficiency can be further improved as compared with the case where the user sequentially determines.
In the same procedure, the personal computer 50 acquires an approximate curve of the index region in the overlapping region SR3 of the region R1.

In the subsequent correction amount calculation step of Step S30, the personal computer 50 first extracts the surface shape data of the index region in the overlapping region SR1 from the partial surface shape data of the regions R2 and R5, and the approximate curve acquired in Step S20 and the relevant curve. The correction amounts of the peripheral regions R2 and R5 are calculated by the method of least squares so that the sum of squares of the difference from the surface shape data is minimized. The correction amount is associated with, for example, parallel movement amounts dx, dy, dz along the x-axis, y-axis, and z-axis, and tilt amounts da, db, dc around the respective axes and a shift in the optical axis direction. Seven types of similar magnifications P for enlargement / reduction deformation can be set. Thereby, the correction amounts of the regions R2 and R5 are calculated.
At this time, since the partial surface shape data of the index region of the reference region R1 is functionalized by obtaining an approximate curve, the tilt amount and the like can be easily calculated by performing differentiation and the like easily. In addition, other correction parameters (for example, enlargement / reduction) can be freely introduced.
Further, the personal computer 50 extracts the surface shape data of the index region in the overlapping region SR3 from the partial surface shape data of the regions R3 and R4, and calculates the correction amounts of the regions R3 and R4 in the same procedure. In this way, different correction amounts are calculated for all of the regions R2 to R5 integrated by fitting.

  In the coordinate conversion process of step S40, the personal computer 50 converts all the coordinate values of the partial surface shape data of the regions R2 to R5 based on the correction amount of each region calculated in step S30.

  Finally, in step S50, the personal computer 50 integrates the coordinate-converted partial surface shape data of the regions R2 to R5 and the partial surface shape data of the region R1 so that the respective overlapping regions are overlapped. Perform shape data fitting. Thus, the surface shape data of the entire test object 100 is acquired, and the series of operations is completed.

According to the shape measurement method of the present embodiment, in the approximate curve acquisition step S20, an approximate curve of partial surface shape data in the index region in the region R1 is acquired using the NURBS function, and in the correction amount calculation step S30, the approximation is performed. Based on the curve, the correction amounts of the other regions R2 to R5 are calculated. Then, after the coordinate conversion of the regions R2 to R5 based on the correction amount is performed in the coordinate conversion step S40, fitting of all the partial shape data is performed in step S50 to obtain the surface shape data of the entire test object 100. The
Therefore, impurities such as dust and incorrect values due to measurement errors can be eliminated, data integration based on more accurate approximate curves can be performed, and highly accurate surface shape data can be obtained, and correction parameters can be obtained by functionalization. The degree of freedom increases and fitting with high versatility can be performed.

In addition, although the rocking portion 22 is swung to change the posture of the test object 100 and the partial surface shape is measured, the correction of each region is performed without using the displacement amount information of the rocking portion 22. The amount can be calculated with high accuracy for fitting. Therefore, it is possible to apply the shape measuring method of the present embodiment to a shape measuring device in which an encoder or the like is not provided in the swinging portion 22.
Further, even when a mechanical error is likely to occur in the behavior of the swinging portion 22, such an error is appropriately eliminated, and highly accurate surface shape measurement can be performed.

  When the shape measuring method of the present invention is implemented using a shape measuring apparatus having a configuration capable of acquiring the displacement amount of the oscillating portion 22, the acquired displacement amount is used as an initial value when calculating the correction amount. May be used. In this way, the calculation time for calculating the correction amount can be shortened, and an error such as convergence to an incorrect extreme value can be avoided, thereby improving accuracy.

  In the present embodiment, an example is described in which only the approximate curve of the index region in the reference region R1 is acquired, and the correction amount is calculated using the point sequence data in the index regions of the other regions R2 to R5 and integrated. However, instead of this, an approximate curve is acquired in the same manner for the index regions of other regions (second approximate curve acquisition step), and the correction amount is set so that the correlation coefficient between the two approximate curves is maximized. May be calculated and integrated.

In the present embodiment, the example using the least square method for calculating the correction amount has been described. However, the present invention is not limited to this, and other methods such as a linear / nonlinear programming method and a search method may be used.
Further, the weighting of the coordinates when obtaining the approximate curve may be performed using a so-called frequency filter that changes the weighting of the point sequence data having a period corresponding to a specific frequency. In this case, the noise removal operation can be automated, and the surface shape measurement can be performed more easily.

Furthermore, although the above-described NURBS function does not have the merit, data integration may be performed by obtaining an approximate curve using a general polynomial or the like. Even in this case, it is possible for the user to identify and remove obvious noise and improve fitting accuracy.
However, when the NURBS function is used, the weighting for the coordinates is not an alternative of 1 or zero, and can be set steplessly from zero to one. Therefore, for example, in the case where the point sequence data C1 of the data DM shown in FIG. 5A may be noise, but cannot be determined to be completely noise, the NURBS function is used, as shown in FIG. Thus, it is also possible to acquire the approximate curve A3 obtained by performing intermediate weighting on the point sequence data C1. In this way, finer adjustments can be made to the partial surface shape data to perform more accurate surface shape measurement.

  Although the embodiments of the present invention have been described above, the technical scope of the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the present invention.

  For example, in each of the above-described embodiments, the example in which all the master data of the index area is extracted from the same area R1 has been described, but instead, the index is different in each overlapping area between each area and another adjacent area. A correction amount may be calculated by extracting an area and using an approximate curve for an index area that is different for each area.

  At this time, for example, the partial shape data of the adjacent regions in the clockwise direction or the counterclockwise direction are fitted using the index region existing in the overlapping region of the two, and the final region is integrated to obtain the entire surface shape data. At the time of acquisition, the correction amount may also be calculated for the first reference region, and all regions may be corrected based on the detected correction amount error.

11 Probe 100 Test object R1 region (first region)
R2, R3, R4, R5 region (second region)
S20 Approximate curve acquisition step S30 Correction amount calculation step S40 Coordinate conversion step

Claims (7)

A surface shape for measuring the surface shape of the test object by causing the probe to scan the surface of the test object by bringing the probe into contact with the test object and relatively moving the probe and the test object. A measuring method,
A first acquisition step of acquiring first surface shape data which is surface shape data of a first region of the test object;
A second acquisition step of acquiring second surface shape data that is surface shape data of a second region of the test object at least partially overlapping with the first surface shape data;
Of the first surface shape data, an approximate curve acquisition step of acquiring an approximate curve by applying an approximation function to data of an area overlapping with the second surface shape data;
A correction amount calculating step of calculating a correction amount of the second surface shape data using the approximate curve;
A coordinate conversion step of performing coordinate conversion on the entire second surface shape data based on the calculated correction amount;
Integrating the first surface shape data and the second surface shape data subjected to the coordinate transformation;
A method of measuring a surface shape, comprising:   The surface shape measurement method according to claim 1, wherein in the second acquisition step, the test object and the probe are changed to a different positional relationship from the first acquisition step.   The surface shape measurement method according to claim 1, wherein the approximate function is a NURBS function.   In the approximate curve acquisition step, when the difference between two adjacent point sequence data among the point sequence data of the first surface shape data to be fitted is a predetermined value or more, one of the two point sequence data 4. The surface shape measuring method according to claim 3, wherein the weighting of the NURBS function with respect to is reduced.   The weighting of the point sequence data having a period corresponding to a predetermined frequency among the point sequence data of the first surface shape data to be fitted is changed in the approximate curve acquisition step. The measuring method of the surface shape as described.   In the correction amount calculating step, the correction amount is calculated so that a sum of squares of a difference between the approximate curve and the second surface shape data in a region corresponding to the approximate curve is minimized. The surface shape measuring method according to claim 1. A second approximate curve obtaining step of obtaining a second approximate curve by performing fitting with an approximate function on the second surface shape data in a region corresponding to the approximate curve;
2. The surface shape measurement according to claim 1, wherein in the correction amount calculation step, the correction amount is calculated so that a correlation coefficient between the approximate curve and the second approximate curve is maximized. Method.

Images