JP2011123018A - Fringe scanning interference fringe measuring method, and interferometer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To facilitate quick and highly-accurate shape measurement by using a simple constitution, concerning a fringe scanning interference fringe measuring method and an interferometer. <P>SOLUTION: The fringe scanning interference fringe measuring method acquires n interference fringe images by shifting one phase increment Δϕ at a time and calculates the shape of a surface to be measured. The method includes: a candidate image acquisition step (step S1) of acquiring a candidate image by being shifted by a phase amount for subdividing the phase increment Δϕ; a brightness change acquisition step (step S2) of acquiring a brightness change on the same position of each candidate image; a phase change calculation step (step S3) of calculating a phase change by using the brightness change; a selection process of an image for interpolation operation (step S4) of selecting as the image for interpolation operation, a plurality of peripheral candidate images having respective phase amounts shown as Δϕ(k-1) (k=1, ..., n); and an interference fringe image calculation step (step S5) of calculating n interference fringe images, by performing interpolation operation by using the image for interpolation operation. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a fringe scan interference fringe measuring method and an interferometer.

Conventionally, in order to measure the shape of a lens surface, a reflective surface, a transmission surface, etc. of an optical element, a light beam is irradiated on a measured surface and a reference surface, and an interference fringe image by light reflected or transmitted is obtained. Various interferometers that measure the surface shape of a surface to be measured by analyzing an interference fringe image are known.
As an analysis method of interference fringe images, a plurality of interference fringe images are acquired by shifting the phase of the optical path difference by relatively moving the measured surface and the reference surface, and from the luminance change at the same position of each interference fringe image A fringe scan method is known in which the phase change is obtained and the shape of the surface to be measured is calculated.
For example, Patent Document 1 discloses a fringe scan method (five bucket method) using five images in which the phase shift amount of each interference fringe image is 0, π / 2, π, 3π / 2, 2π, A fringe scan method (7 bucket method) using seven images with phase shift amounts of 0, π / 2, π, 3π / 2, 2π, 5π / 2, 3π, and a phase shift amount of 0, An interferometer for performing a fringe scanning method (13 bucket method) using 13 images of π / 4, 2π / 4, 3π / 4,..., 12π / 4 and its analytical expression are described. Further, Patent Document 2 describes analytical expressions used for the 7-bucket method and the 9-bucket method, which are fringe scanning methods using seven or nine interference fringe images.

US Pat. No. 5,473,434 JP-A-11-173808

However, the conventional fringe scan interference fringe measuring method and interferometer as described above have the following problems.
In the fringe scanning method, a measurement error occurs when the phase shift amount of each interference fringe image deviates from a predetermined value. For example, in the technique described in Patent Document 1, since a lens having a reference surface is moved with respect to a surface to be measured by a piezo element, it is necessary to perform high-precision movement of λ / 8 by the piezo element.
However, the relationship between the applied voltage and the expansion / contraction amount of the piezo element has, for example, individual nonlinear characteristics as shown by a curve 200 in FIG. 10 and also has a hysteresis characteristic. Therefore, the applied voltage is changed linearly. Driving, there is a problem that a movement error occurs and affects the measurement accuracy.
Therefore, in general, it is necessary to always calibrate the relationship between the applied voltage and the amount of expansion / contraction, so maintenance costs such as calibration costs are incurred, and measurement is not possible during the calibration period. There is.
In the technique described in Patent Document 2, the number of interference fringe images is increased from that in Patent Document 1, so that the measurement accuracy can be improved to some extent, but the movement accuracy of the piezo element needs to be kept high. This is the same as Patent Document 1.
In order to detect the relative movement amount of the reference surface, it may be possible to feedback control the expansion / contraction amount of the piezo element by providing a length measuring means such as a capacitance sensor in these interferometers. In this case, there is a problem that the device configuration and the control method become complicated, and the device becomes large and expensive.

  The present invention has been made in view of the above problems, and provides a fringe scan interference fringe measurement method and an interferometer that can easily and accurately perform shape measurement using a simple configuration. The purpose is to do.

In order to solve the above-described problems, in the invention described in claim 1, the same light beam is measured light reflected or transmitted by the surface to be measured and reference light having a wavefront corresponding to the shape of the reference surface. An interference fringe is formed by dividing the measurement light and the reference light to obtain an interference fringe image, and from this state, the phase difference between the measurement light and the reference light is set to a substantially constant phase. A fringe scan interference fringe measuring method that acquires a total of n (n is an integer of 3 or more) interference fringe images shifted by increments Δφ and calculates the shape of the surface to be measured based on these n interference fringe images. Because
In order to obtain the n interference fringe images, the phase difference between the measurement light and the reference light is shifted from 0 to Δφ · (n−1) or more in a size that subdivides the phase increment Δφ, and n A candidate image acquisition step of acquiring more interference fringe images and storing them as candidate images of the n interference fringe images, and the luminance at the same position of each candidate image acquired in the candidate image acquisition step, Based on the luminance change data when arranged in order of the magnitude of the phase shift amount of each candidate image, based on the phase change calculation step for calculating the phase change of each candidate image, and the phase change calculated in the phase change calculation step Then, among the candidate images, an interpolation calculation image selection that uses a plurality of candidate images in the vicinity of a phase amount of Δφ · (k−1) (where k = 1,..., N) as interpolation calculation images. Process and image selection process for interpolation calculation By performing an interpolation operation was the interpolation operation for image using the interference fringe image calculation step of obtaining the n interference fringe image, and a method comprising.

According to a second aspect of the present invention, in the fringe scan interference fringe measuring method according to the first aspect, the interference fringe image calculating step is configured such that the phase amount is Δφ · (k−1) or less from the candidate image. Thus, a candidate image _FIk that is a candidate image closest to Δφ · (k−1) and a candidate that is a candidate image whose phase amount is larger than Δφ · (k−1) and closest to Δφ · (k−1). When the image F _{Ik + 1} is used and the phase amounts of the candidate images F _{Ik and} F _{Ik + 1} are Δφ · (k−1) −A _k and Δφ · (k−1) + B _k , respectively, by linear interpolation based on), the method of calculating the n-number of the interference fringe image as an interference fringe images G _{k.
G k = (F Ik × B} k + F Ik + 1 × A k) / (A k + B k) ··· (1)
Here, G _k , F _Ik , and F _{Ik + 1} represent two-dimensional image data of each interferometer image.

  According to a third aspect of the present invention, in the fringe scan interference fringe measuring method according to the first or second aspect, the candidate image acquisition step continuously changes an optical path difference of the reference light with respect to the measurement light in a fixed direction. In this case, the candidate image is acquired at a sampling period corresponding to a phase amount of a size that subdivides the phase increment Δφ.

  According to a fourth aspect of the present invention, in the fringe scan interference fringe measuring method according to any one of the first to third aspects, the phase increment Δφ is π / 2.

  The invention according to claim 5 is the fringe scan interference fringe measuring method according to any one of claims 1 to 3, wherein the phase increment Δφ is π / 4.

  In the invention according to claim 6, the same light beam is divided into measurement light reflected or transmitted by the surface to be measured and reference light having a wavefront corresponding to the shape of the reference surface, and the measurement light and the reference Included in the interference measurement optical system for changing an optical path difference of the reference light with respect to the measurement light, an interference measurement optical system that forms an interference fringe with light, an imaging unit that captures an image of the interference fringe An optical element moving mechanism that moves the optical element, and the optical element moving mechanism moves the optical element, and the phase of the optical path difference changes to Δφ · (n−1) (n is an integer of 3 or more) or more. In the meantime, an image acquisition unit that acquires from the imaging unit a plurality of interference fringe images more than n shifted in increments such that the phase of the optical path difference subdivides Δφ, and the image acquisition unit acquires The luminance at the same position of the plurality of interference fringe images The phase change calculation unit that calculates the phase change of the plurality of interference fringe images from the luminance change data when arranged in order of the magnitude of the phase shift amount of the plurality of interference fringe images, and calculated by the phase change calculation unit From the plurality of interference fringe images acquired by the image acquisition unit based on the phase change, a plurality of candidates whose phase amount is near Δφ · (k−1) (where k = 1,..., N) An n-interference fringe image is obtained by performing an interpolation calculation using an interpolation calculation image selection unit that selects an image as an interpolation calculation image and the interpolation calculation image selected by the interpolation calculation image selection unit. An interference fringe image calculation unit that acquires the shape of the surface to be measured from the n interference fringe images calculated by the interference fringe image calculation unit.

  According to the fringe scan interference fringe measuring method and the interferometer of the present invention, an interference fringe image whose phase has been sequentially changed is acquired as a candidate image, and an interpolation calculation is performed on the candidate image, so that the phase shift amount is approximately Δφ from 0. Since the increased n interference fringe images can be calculated, easy and highly accurate shape measurement can be quickly performed using a simple configuration without using a means for controlling the phase shift amount with high accuracy. There is an effect that can be.

It is a schematic block diagram which shows schematic structure of the interferometer which concerns on embodiment of this invention. It is a functional block diagram which shows the function structure of the control means of the interferometer which concerns on embodiment of this invention. It is a flowchart which shows the measurement flow of the fringe scan interference fringe measuring method which concerns on embodiment of this invention. It is a flowchart explaining operation | movement of the candidate image acquisition process of the fringe scan interference fringe measuring method which concerns on embodiment of this invention. It is a schematic diagram of the candidate image acquired at a candidate image acquisition process. It is a typical graph which shows an example of the approximated curve in the phase change calculation process of the fringe scan interference fringe measuring method which concerns on embodiment of this invention. It is the typical graph for demonstrating the image selection process for interpolation calculations of the fringe scan interference fringe measuring method which concerns on embodiment of this invention, and the enlarged view of the A section. Interference fringe image of the test surface in measurement example 1, image display of the wavefront calculation result of measurement example 1 calculated based on the fringe scan interference fringe measurement method of the embodiment of the present invention, and fringe scan interference fringe measurement of comparative example 1 It is the image display of the wave front calculated based on the method. Interference fringe image of the test surface in Measurement Example 2, image display of the wavefront calculation result of Measurement Example 2 calculated based on the fringe scan interference fringe measurement method of the embodiment of the present invention, and fringe scan interference fringe measurement of Comparative Example 2 It is the image display of the wave front calculated based on the method. It is a typical graph which shows an example of the relationship between the applied voltage of a piezoelectric element, and the expansion-contraction amount.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In all the drawings, even if the embodiments are different, the same or corresponding members are denoted by the same reference numerals, and common description is omitted.

An interferometer according to an embodiment of the present invention will be described.
FIG. 1 is a schematic configuration diagram showing a schematic configuration of an interferometer according to an embodiment of the present invention. FIG. 2 is a functional block diagram showing a functional configuration of the control unit of the interferometer according to the embodiment of the present invention.

The interferometer 50 of this embodiment can implement the fringe scan interference fringe measuring method according to the embodiment of the present invention. The interference measurement optical system constituting the interferometer will be described as an example in the case where a Fizeau optical system is used in this embodiment. In addition, although an appropriate shape can be measured as the measured surface 5, a description will be given below using a configuration example in the case of measuring a concave shape.
As shown in FIG. 1, the schematic configuration of the interferometer 50 includes a laser light source 1, a collimating lens 2, a beam splitter 3, a Fizeau lens 4 (an optical element included in the interference measuring optical system), a condensing lens 6, and a CCD 7 (imaging). Part), a piezo element 8 (optical element moving mechanism), a piezo element controller 9, a measurement control part 10, and a display part 11.

The laser light source 1 is a light source that generates coherent light for forming interference fringes. In this embodiment, a light source that generates laser light having a wavelength λ as divergent light is employed. The divergent light generated by the laser light source 1 is converted into parallel light 30 a by the collimator lens 2 and is incident on the beam splitter 3.
The beam splitter 3 reflects and guides the parallel light 30a onto the optical axis of the Fizeau lens 4, and transmits a measurement surface reflected light 30c and a reference surface reflected light 30d, which will be described later, incident from the Fizeau lens 4 side. It is.

The Fizeau lens 4 reflects a part of the parallel light 30a incident on the optical axis by the Fizeau surface 4a to form reference surface reflected light 30d (reference light), and the parallel light 30a incident on the optical axis. It is a lens that transmits the other part as transmitted light 30b and condenses the transmitted light 30b.
The shape of the Fizeau surface 4a is accurately finished in accordance with the ideal shape of the surface 5 to be measured, and constitutes a reference surface for measuring interference fringes.

When measuring the surface shape of at least the surface to be measured 5, the surface to be measured 5 has the optical axis of the surface to be measured 5 coincident with the optical axis of the Fizeau lens 4, and the center of curvature of the surface to be measured 5 is It arrange | positions so that it may correspond with the condensing position of the transmitted light 30b by the Fizeau lens 4. FIG.
In order to realize such an arrangement, the relative positional relationship between the Fizeau lens 4 and the measured surface 5 can be adjusted by a moving mechanism (not shown).
The transmitted light 30b incident on the measured surface 5 is reflected as the measured surface reflected light 30c. At this time, since the optical axes of the surface to be measured 5 and the Fizeau lens 4 coincide with each other, and the center of curvature of the surface to be measured 5 coincides with the condensing position of the Fizeau lens 4, the light beam of the transmitted light 30b is reflected on the surface to be measured. 5, the measured surface reflected light 30c travels backward along the same optical path as the transmitted light 30b, reenters the Fizeau lens 4, and is transmitted toward the beam splitter 3.

  Therefore, the measured surface reflected light 30c and the reference surface reflected light 30d are both light beams formed by dividing the parallel light 30a, which is the same light beam, by the Fizeau surface 4a. Then, the measured surface reflected light 30 c is reflected by the measured surface 5 to be measured light whose wavefront has changed according to the shape of the measured surface 5. On the other hand, the reference surface reflected light 30d is reflected by the Fizeau surface 4a, thereby becoming reference light having a wavefront corresponding to the shape of the Fizeau surface 4a. Further, the measured surface reflected light 30c is reflected by the measured surface 5 and travels backward in the same optical path, so that the optical path length between the Fizeau surface 4a and the measured surface 5 with respect to the reference surface reflected light 30d. The optical path difference is twice that of. Therefore, the measured surface reflected light 30c and the reference surface reflected light 30d form interference fringes due to the optical path difference corresponding to the shape error between the measured surface 5 and the Fizeau surface 4a on the Fizeau surface 4a.

  As described above, the collimating lens 2, the beam splitter 3, and the Fizeau lens 4 constitute a Fizeau interferometer, and the same light flux is reflected or transmitted by the surface to be measured and the shape of the reference surface. This is an example of an interference measurement optical system that is divided into reference light having a corresponding wavefront and that forms interference fringes by the measurement light and the reference light.

The condensing lens 6 forms an image of the interference fringes caused by the measured surface reflected light 30c and the reference surface reflected light 30d on the imaging surface 7a of the CCD 7.
The CCD 7 is an image sensor that photoelectrically converts an interference fringe image formed on the imaging surface 7a at a predetermined video rate. As the video rate, an appropriate value can be adopted as necessary. For example, a video rate of 30 fps can be preferably adopted.
The CCD 7 is electrically connected to the measurement control unit 10, the imaging operation is controlled by the measurement control unit 10, and the image signal captured by the CCD 7 is sent to the measurement control unit 10.

The piezo element 8 is an optical element that moves the Fizeau lens 4 along the optical axis in order to change the optical path difference between the measured surface reflected light 30c and the reference surface reflected light 30d on the Fizeau surface 4a by a minute amount. It is a moving mechanism.
The amount of movement of the piezo element 8 is controlled by changing the applied voltage by a piezo element controller 9 electrically connected to the piezo element 8.

The piezo element controller 9 controls the voltage applied to the piezo element 8 based on a control signal from the measurement control unit 10 and controls the expansion / contraction amount of the piezo element 8.
As shown in FIG. 2, the schematic configuration of the piezo element controller 9 includes a calculation unit 25, a timer 26, a D / A conversion unit 27, and an amplifier unit 28.
The calculation unit 25 generates applied voltage data that linearly increases the applied voltage from 0 V to the command voltage within the rise time in accordance with the rise time and command voltage control signal sent from the measurement control unit 10, and performs measurement. The applied voltage data is sequentially supplied to the D / A converter 27 in accordance with a movement start signal from the controller 10.
The timer 26 supplies a reference clock for sending applied voltage data from the arithmetic unit 25 at appropriate intervals.
The D / A converter 27 converts the applied voltage data supplied by the calculator 25 into a voltage signal. The voltage signal is amplified by the amplifier unit 28 and output to the piezo element 8 as an applied voltage for driving the piezo element 8.
Thus, according to the piezo element controller 9 of the present embodiment, an applied voltage for continuously driving the piezo element 8 in a certain direction can be supplied in accordance with the movement start signal from the measurement control unit 10. It has become.

As shown in FIG. 2, the functional block configuration of the measurement control unit 10 includes an image acquisition unit 20, an arithmetic processing unit 21, a movement control unit 22, and an image storage unit 23.
The image acquisition unit 20 takes in the image signal sent from the CCD 7 for each image frame at the timing instructed by the calculation processing unit 21, converts it into luminance data, and sends it to the calculation processing unit 21 as two-dimensional image data. To do.
The arithmetic processing unit 21 captures and acquires the image data sent from the image acquisition unit 20, stores it in the image storage unit 23 together with information on the acquisition order, and performs arithmetic processing on the image data stored in the image storage unit 23. Then, the interference fringes are analyzed, and the shape of the measured surface 5 is calculated. The calculated shape of the measured surface 5 is displayed on the display unit 11 together with a graph, an intermediate calculation result, and the like accompanying this analysis as necessary.
Further, the arithmetic processing unit 21 sends the rise time and command voltage information to the piezo element controller 9 via the movement control unit 22, sets conditions for the expansion / contraction amount of the piezo element 8, and sends a movement start signal. .
Further, the movement control unit 22 detects that the movement by the piezo element controller 9 has been completed, and notifies the arithmetic processing unit 21 of it.
In the present embodiment, such a measurement control unit 10 is configured by a computer including a CPU, a memory, an input / output interface, an external storage device, and the like, and these functions are realized by executing appropriate control programs by the computer. is doing.

Next, the operation of the interferometer 50 will be described focusing on the fringe scan interference fringe measurement method according to the embodiment of the present invention.
FIG. 3 is a flowchart showing a measurement flow of the fringe scan interference fringe measurement method according to the embodiment of the present invention. FIG. 4 is a flowchart for explaining the operation of the candidate image acquisition step of the fringe scan interference fringe measuring method according to the embodiment of the present invention. FIG. 5 is a schematic diagram of candidate images acquired in the candidate image acquisition step. FIG. 6 is a schematic graph showing an example of an approximate curve in the phase change calculation step of the fringe scan interference fringe measurement method according to the embodiment of the present invention. Here, the horizontal axis represents time, and the vertical axis represents luminance. FIG. 7A is a schematic graph for explaining the image selection process for interpolation calculation of the fringe scan interference fringe measuring method according to the embodiment of the present invention. Here, the horizontal axis represents time, and the vertical axis represents the phase change amount. FIG.7 (b) is an enlarged view of the A section of Fig.7 (a).

In the Fizeau type optical system used in the interferometer 50, when the laser light source 1 is turned on, divergent light of wavelength λ is generated, collimated lens 2 forms parallel light 30a, is reflected by beam splitter 3, and is reflected by Fizeau lens 4. Incident on the axis.
The parallel light 30a is divided by the Fizeau surface 4a, and a part thereof is reflected by the Fizeau surface 4a toward the beam splitter 3 and proceeds as reference surface reflected light 30d. Other light is transmitted as transmitted light 30 b, condensed by the lens action of the Fizeau lens 4, condensed at one point, then guided to the measurement surface 5 and incident in the normal direction of the measurement surface 5. By doing so, it is reflected as the measured surface reflected light 30c. The measured surface reflected light 30c travels backward along the same optical path as the transmitted light 30b, passes through the Fizeau lens 4, and is emitted to the beam splitter 3 side. At that time, an interference fringe image corresponding to the optical path difference between the measured surface reflected light 30c and the reference surface reflected light 30d is formed on the Fizeau surface 4a.
The interference fringe image is formed on the imaging surface 7 a by the condenser lens 6. The interference fringe image is photoelectrically converted by the CCD 7 and sent to the measurement control unit 10 as an image signal and displayed on the display unit 11. The measurer adjusts the relative position between the Fizeau lens 4 and the measured surface 5 to a position where the best interference fringes can be obtained for measurement.
Then, the measurer instructs the start of fringe scan interference fringe measurement (hereinafter abbreviated as interference fringe measurement) through the measurement control unit 10.

In the known fringe scanning method, the phase of the optical path difference between the reference surface and the surface to be measured is changed by a substantially constant phase increment Δφ from the interference fringe image for shape measurement, and a total of n (where n is (An integer of 3 or more) interference fringe images are acquired. These n interference fringe images become interference fringe images in which the phase shift amount is increased by Δφ between 0 to Δφ · (n−1). Therefore, these phase shift amounts are common to the interference fringe images. By applying the relationship with the n luminance values at the position to be applied to the sine wave, the phase (initial phase) of the luminance value of the interference fringe image (phase shift amount is 0) for shape measurement is calculated, Based on the interference fringe image for shape measurement, the phase at each position is calculated. Since this phase corresponds to the amount of deviation from the shape of the reference surface, the surface shape of the surface to be measured 5 can be calculated.
The present interference fringe measurement is characterized by a process of acquiring these n interference fringe images. About another point, since the well-known process which was disclosed by patent document 1, 2, etc. can be employ | adopted, detailed description is abbreviate | omitted, for example.

In this interference fringe measurement, steps S1 to S6 are sequentially executed as shown in FIG. The phase increment Δφ may be a phase increment used for well-known fringe scan interference fringe measurement. When Δφ = π / 2, π / 4, etc., it is known that highly accurate measurement results can be obtained even if the number (number) of interference fringe images is small. Hereinafter, as an example, a case where Δφ = π / 2 will be described.
Step S1 is realized by the flow shown in FIG. 4, and the Fizeau lens 4 is continuously moved in a fixed direction by a distance L during a rising time T set in advance by the piezo element 8 and set in advance. M (where m> n) interference fringe images are sequentially acquired for each sampling period ΔT, and the time series images F ₁ , F ₂ ,... it is F m step of storing as _(see FIG. 5) (candidate image acquisition step).
For this reason, the time-series image F _j (j is an integer satisfying 1 ≦ j ≦ m, the same applies hereinafter) depends on the expansion / contraction characteristics of the piezo element 8, and the measured surface reflected light 30c and the reference surface reflected light 30d The optical path difference is shifted by approximately (λ / 4) · (n−1) / (m−1), that is, the phase shift amount is from 0 to (π / 2) · (n−1) or more. In the state increased by approximately (π / 2) · (n−1) / (m−1). These time-series images F ₁ , F ₂ ,..., F _m are n interference fringe image candidate images used for the calculation of the fringe scan method.
Here, in order to shorten the measurement time, the sampling period ΔT is preferably as short as possible. In the present embodiment, the sampling period ΔT matches the video rate. For example, when the video rate is 30 fps, ΔT = 1/30 (s).
Further, the movement distance L of the Fizeau lens 4 requires a distance corresponding to an optical path difference of (λ / 4) · (n−1). In the Fizeau interferometer of this embodiment, {(λ / 4) · (N-1)} / 2, but considering the variation in the amount of piezo expansion / contraction, it is preferable to make it slightly larger (for example, three times the variation in the amount of piezo expansion / contraction).
In addition, the number m of candidate images is preferably set to a size that can subdivide the phase increment Δφ to such an extent that there is no problem even if a linear change in the phase change between candidate images in the acquisition order is adjacent. In this embodiment, n = 5 and m = 25 are adopted as an example.

First, in step S11, information on the rise time T and the command voltage is sent from the movement control unit 22 to the calculation unit 25 of the piezo element controller 9, and then the movement start signal 100 is sent. Then, on the measurement control unit 10 side, the process proceeds to step S12.
On the piezo element controller 9 side, when information on the rise time T and the command voltage is received from the movement control unit 22, step S20 is executed.
In the present embodiment, the rise time T is set to T = (m−1) · ΔT. For example, in the above numerical example, T = (25−1) /30=0.8 (s).
In step S20, the calculation unit 25 performs setting so that applied voltage data proportional to time is output from 0 V to the command voltage within the rising time T. When the movement start signal is received, the timer 26 is initialized and step S21 is executed.
In step S 21, the applied voltage data is converted into a voltage signal by the D / A converter 27 according to the value of the timer 26, amplified appropriately by the amplifier 28, and applied to the piezoelectric element 8.
In step S22, it is determined whether or not the value of the timer 26 has exceeded the rising time T. If the rising time T has not been exceeded, the process proceeds to step S21.
When the rise time T is exceeded, the movement control unit 22 is notified of the movement end by sending the movement end signal 101, the voltage output is stopped, and the position of the piezo element 8 is returned to the initial state.

In step S12, an interference fringe image is acquired from the CCD 7 in synchronization with the sampling period ΔT, and stored in the image storage unit 23 together with information on the image acquisition order.
In step S13, it is determined whether or not there is a movement end signal 101 from the piezo element controller 9. If the movement end signal 101 is not received, the process returns to step S12.
If the movement end signal 101 has been received, the operation in step S1 is terminated, and the process proceeds to step S2 in FIG.
In this manner, by repeating step S12, m time-series images F ₁ , F ₂ ,..., F _m are stored in the image storage unit 23 while the Fizeau lens 4 is moved by the piezo element 8. It is memorized sequentially.

In step S2, as shown in FIG. 5, each time series image _Fj is called from the image storage unit 23 by the arithmetic processing unit 21, and each luminance at the luminance calculation position 32 which is a common position on each interference fringe image unit 31 is obtained. Get P _j . That is, step S2 constitutes a luminance change acquisition step of acquiring a luminance change at the same position of each candidate image acquired in the candidate image acquisition step.
An example of the relationship between time and luminance P _j is plotted in FIG. The time-series image F _j is acquired at a constant sampling period, and a voltage proportional to time is applied to the piezo element 8. However, since the expansion and contraction characteristics of the piezo element 8 are nonlinear with respect to the applied voltage, the Fizeau lens. The moving pitch of 4 is not equal. Therefore, the graph of the brightness P _j deviates from the sine wave.

Next, in step S3, the operation processing unit 21 calculates an approximate curve 33 (see FIG. 6) of the graph of the luminance P _j. However, in the present embodiment, if the movement amount of the piezo element 8 is linear with respect to time, the approximate curve 33 is expressed by the following equation (2) using the fact that the luminance graph is a sine wave with respect to time. It will be represented by the group (3).
f (i) = D ₁ · i + D ₂ · i ² + D ₃ · i ³ (2)
g (i) = A · sin {f (i)} + B · cos {f (i)} + C (3)
Here, g (i) represents luminance, and f (i) represents the nonlinearity of the piezo element 8 by a cubic equation. D ₁ , D ₂ , D ₃ , A, B, and C are coefficients of the approximate expression. The variable i uses the value of the subscript j corresponding to time as it is. f (i) is the time dimensionless rise time, when representing the phase variation amount with respect to image sequences F _1.
The coefficients D ₁ , D ₂ , D ₃ , A, B, and C can be obtained by performing known arithmetic processing such as Newton's method so as to minimize S in the following equation (4).

As described above, step S3 is a phase change calculation step of calculating the phase change from the luminance change at the same position of each candidate image acquired in the candidate image acquisition step. Further, the arithmetic processing unit 21 constitutes a phase change calculation unit.
In addition, since a graph as shown in FIG. 6 can be plotted if Expressions (2) and (3) are used, in the present embodiment, an actual measurement value of the luminance P _j so that the approximation accuracy can be visually confirmed. At the same time, it is displayed on the display unit 11.

Next, in step S4, based on the phase change calculated in the phase change calculation step, the phase amount is in the vicinity of Δφ · (k−1) (where k = 1,..., N) from the candidate images. An interpolation calculation image selection step of selecting a plurality of candidate images as an interpolation calculation image is performed.
In the present embodiment, the value of f (i) is equal to or less than Δφ · (k−1) and is calculated using Δφ · (k−1) by using the expression (2) obtained in step S3 by the arithmetic processing unit 21. I that is closest to) is calculated as I _k . At this time, I _k +1 is i which is larger than Δφ · (k−1) and closest to Δφ · (k−1). Therefore, f (I ₁ ) ≦ 0 <f (I ₁ +1),..., F (I ₅ ) ≦ 2π <f (I ₅ +1).
For example, when Formula (2) is represented by the curve 41 shown in FIG. 7A, I ₁ = 1, I ₂ = 10, I ₃ = 15, I ₄ = 19, and I ₅ = 23, respectively. Desired. However, in this embodiment, f (1) = 0.
The arithmetic processing unit 21 obtains five interference fringe images to be used for the fringe scan method from these values of I _k and I _k +1, and therefore, for the five sets of interpolation calculations from the time series image F _j . image _{(F 1,} F _2), to select the _{(F 10, F 11),} (F 15, F 16), (F 19, F 20), (F 23, F 24). Hereinafter, it is expressed as an interpolation calculation image (F _Ik , F _{Ik + 1} ).
For this reason, the calculation processing unit 21 constitutes an interpolation calculation image selection unit.
In this step, a graph as shown in FIG. 7 is displayed on the display unit 11, or the brightness value data of the time-series image selected on the graph as shown in FIG. 6 displayed in step S3 is highlighted. May be.

Next, in step S5, the arithmetic processing unit 21 uses the five sets of interpolation calculation images (F _Ik , F _{Ik + 1} ) selected in step S4, and performs an interpolation calculation to linearly interpolate these images to obtain a phase amount Δφ · calculating a (k-1) become the interference fringe image _{G k.}
For example, as shown in FIG. 7B, the phase amount at I ₂ = 10 is f (I ₂ ) = π / 2−A ₂ , and the phase amount at I ₂ + 1 = 11 is f (I ₂ +1). = Π / 2 + B ₂ , and the phase amount approximately linearly changes between them, and therefore the interference fringe image G ₂ is approximated by proportionally distributing the luminances of F ₁₀ and F _11. be able to.
Therefore, the arithmetic processing unit 21 calculates the interference fringe image G _k by the following equation (1), and obtains n interference fringe images.
For this reason, step S5 constitutes an interference fringe image calculation step of acquiring n interference fringe images by performing an interpolation calculation using the interpolation calculation image selected in the interpolation calculation image selection step. . In addition, the arithmetic processing unit 21 constitutes an interference fringe image calculation unit.

A _k = Δφ · (k−1) −f (I _k ) (5)
B _k = f (I _k +1) −Δφ · (k−1) (6)
_{G k = (F Ik × B} k + F Ik + 1 × A k) / (A k + B k) ··· (1)

Here, G _k , F _Ik , and F _{Ik + 1} represent two-dimensional image data of each interferometer image. For example, if the long side direction and the short side direction of the CCD 7 are the x direction and the y direction, respectively, the above calculation needs to be performed for each pixel, so that G _k (x, y), F _Ik (x, y ), F _{Ik + 1} (x, y), but (x, y) is omitted for the sake of simplicity (and so on).

Next, in step S6, using the interference fringe image time series images G ₁ , G ₂ , G ₃ , G ₄ , and G ₅ calculated in step S5, for example, the arithmetic processing disclosed in Patent Document 1 is calculated. This is performed by the processing unit 21 to calculate a wavefront corresponding to the surface shape. Then, the result is displayed as a graphic image on the display unit 11 (see FIGS. 8B and 9B).
The interference fringe measurement is thus completed.

  As described above, in the interference fringe measurement by the interferometer 50, an interference fringe image that is phase-shifted by the phase increment Δφ necessary for the calculation such as the 5-bucket method is acquired due to the movement accuracy or nonlinearity of the piezo element 8. Even if it is not possible, an interference fringe image that is estimated to have a phase amount of Δφ · (k−1) can be calculated by performing an interpolation operation on the obtained candidate images. For this reason, the error of the phase shift amount of n interference fringe images used in the fringe scanning method can be significantly reduced.

Here, the effect | action of this embodiment is demonstrated based on the measurement examples 1 and 2 and the comparative examples 1 and 2. FIG.
FIG. 8A is an interference fringe image of the test surface in Measurement Example 1. FIG. FIG. 8B is an image display of the wavefront calculation result of Measurement Example 1 calculated based on the fringe scan interference fringe measurement method of the embodiment of the present invention. FIG. 8C is an image display of a wavefront calculated based on the fringe scan interference fringe measuring method of Comparative Example 1. FIG. 9A is an interference fringe image of the test surface in Measurement Example 2. FIG. FIG. 9B is an image display of the wavefront calculated based on the fringe scan interference fringe measuring method of the embodiment of the present invention. FIG. 9C is an image display of a wavefront calculated based on the fringe scan interference fringe measuring method of Comparative Example 2.

In measurement example 1, the candidate image acquisition process, the phase change calculation process, the interpolation calculation image selection process, and the interference fringe image calculation process of the present embodiment are sequentially performed, and then five interference fringe images G ₁ , G ₂ , Based on the image data of G ₃ , G ₄ , and G ₅ , the wavefront was calculated by the 5-bucket method, and the result as shown in the graphic image in FIG. 8B was obtained.
The surface accuracy of the test surface 5 in this measurement example is a lens surface having a general accuracy of PV = 0.12λ, and a wavefront that changes smoothly is calculated.
In the measurement example 1, as shown in FIG. 8A, the alignment of the test surface 5 and the interferometer 50 and the optical axis are sufficiently aligned to minimize the number of interference fringes (null state). ) To measure.
In this measurement example, calculations (tilt and power correction) are performed to correct the wavefront tilt and power generated by the lens installation state. (Hereinafter, Comparative Example 1, Measurement Example 2, and Comparative Example 2 are the same).

On the other hand, in Comparative Example 1, after performing the candidate image acquisition step and the phase change calculation step of Measurement Example 1, the five candidates whose phase amounts are closest to 0, π / 2, π, 3π / 2, and 2π, respectively. An image is selected, and the wavefront is calculated by the 5-bucket method based on the image data of these candidate images. For example, in the example of FIG. 8A, if A ₁ <B ₁ , A ₂ > B ₂ , A ₃ > B ₃ , A ₄ <B ₄ , A ₅ <B ₅ , F ₁ , F ₁₁ , F ₁₆ , F ₁₉ , and F ₂₃ were selected, and the wavefront was calculated from these five candidate images, and the results shown in FIG. 8C were obtained.

According to FIGS. 8B and 8C, the measurement example 1 and the comparative example 1 have a result that is not so different.
This is because the candidate image acquisition process of the present embodiment is set to a size that can sufficiently subdivide the phase increment Δφ, so that the result is not much different from the case of performing the interpolation calculation without performing the interpolation calculation. It is done.

Measurement example 2 employs the same test surface 5 as measurement example 1, and as shown in FIG. 9 (a), the same measurement is performed without adjusting the null state as sufficiently as measurement example 1. It is a thing. In this case, as can be seen from the fact that the interference fringes are vertical stripes, the test surface is slightly inclined in the horizontal direction in the figure.
As shown in FIG. 9B, the wavefront calculated in the measurement example 2 can obtain the same result as in FIG. 8B.

On the other hand, in Comparative Example 2, after performing the candidate image acquisition process and the phase change calculation process of Measurement Example 2, the five candidates whose phase amounts are closest to 0, π / 2, π, 3π / 2, and 2π, respectively. An image was selected, and the wavefront was calculated by the 5-bucket method based on the image data of these candidate images. The result shown in FIG. 9C was obtained.
In FIG. 9C, the surface shape of the test surface 5 appears as a whole as in FIG. 9B, but unlike FIG. 9B, the calculated wavefront includes Corresponding to the fringe positions of the interference fringes in (a), a shape that periodically changes in the horizontal direction in the figure appears. This is a pseudo shape that does not originally exist on the test surface 5.

In the 5-bucket method, when there is an error in the phase increment Δφ, if the initial phase is θ, an analysis error ε (θ) that can be approximately expressed by the following equation (16) occurs.
ε (θ) = Asin (2θ) + B · cos (2θ) + C (16)
In the sufficient null state as in Measurement Example 1 and Comparative Example 1, the initial phase θ is substantially constant over the entire lens surface, so that ε (θ) is approximately constant over the entire lens surface and there is an error in Δφ. However, the influence on the calculation result of the wave front is difficult to appear.
Strictly speaking, since the test surface 5 used is PV = 0.12λ, the initial phase θ varies in the range of 0.48π depending on the location of the lens even though it is substantially constant. The degree of the swell component for the period is added to the calculation result of the wavefront.
On the other hand, in the insufficient null state as in measurement example 2 and comparative example 2, since the initial phase θ varies depending on the lens position, ε (θ) also varies depending on the lens position, and when there is an error in Δφ, the wavefront This will affect the calculation results.
As shown in FIG. 9A, since there are three interference fringes in the horizontal direction, there is a phase difference of about 6π at the left end and the right end. It can be read also from FIG.

  Note that Comparative Examples 1 and 2 are methods for selecting an interference fringe image with as little phase amount error as possible, but the conventional method obtains only five interference fringe images by moving the phase increment Δφ. In the case of the technique, the driving error of the piezo element 8 becomes the phase amount error as it is, and this error cannot be corrected. Therefore, it is obvious that the accuracy is inferior to those of the first and second comparative examples.

As described above, according to the fringe scan interference fringe measurement method of the present embodiment, an interference fringe image having an accurate phase increment Δφ necessary for wavefront calculation cannot be obtained due to the movement accuracy of the piezo element 8 or the like. Even in such a case, an interference fringe image having the phase increment Δφ necessary for the wavefront calculation can be calculated by performing an interpolation operation on the candidate image obtained by shifting the phase so as to subdivide the phase increment Δφ.
Therefore, the piezo element 8 is easily calibrated with a simple configuration without calibrating the piezo element 8 or measuring the amount of movement of the piezo element 8 using a length measuring means such as a capacitance sensor and performing feedback control. Accurate measurement can be performed.
In addition, since calibration of the piezo element 8 is not required, there is an advantage that calibration costs and a period during which measurement cannot be performed due to calibration do not occur.
Further, since candidate images are acquired in synchronization with the video rate of the CCD 7, even if the number of candidate images is increased, measurement can be performed in a short time.
In addition, since the piezoelectric element 8 can be employed even if it has a low degree of nonlinearity or has a large variation in nonlinearity, an inexpensive element can be employed.
Further, in this embodiment, since the piezo element 8 is continuously driven in a certain direction, there is no variation in the movement position due to hysteresis characteristics, no position vibration at the time of stopping such as intermittent driving, etc. The corresponding movement position increases monotonically along the time series. Therefore, the measurement resolution can be reliably improved by increasing the number m of candidate images.

  In addition, according to the fringe scan interference fringe measurement method of the present embodiment, even if the phase amount of the correction calculation image deviates from the phase amount necessary for wavefront calculation, a good interference fringe image is estimated by interpolation calculation. Therefore, for example, as in Comparative Examples 1 and 2, the number of acquired candidate images can be reduced compared to the case where candidate images with close phase amounts are selected without performing an interpolation calculation. For this reason, the measurement time can be further reduced and quick measurement can be performed. Further, since the storage space for candidate images can be reduced, an inexpensive apparatus configuration can be achieved.

[Modification]
Next, a modification of this embodiment will be described.
The present modification is different from the above embodiment in that cubic interpolation is adopted as the interpolation calculation in the image selection process for interpolation calculation in the above embodiment. Below, it demonstrates centering on a different point from the said embodiment.

In the image selection process for interpolation calculation of the present modification, cubic interpolation is performed, and each Δφ · (k−1) is obtained using the equation (1) obtained in step S3 of the above embodiment as in the above embodiment. I _k is calculated for the phase amount of. Then, F _Ik−1 , F _Ik , F _{Ik + 1} , and F _{Ik + 2} are selected from the candidate images acquired in step S _{<b> 2} of the above-described embodiment, and the following equations (7), (8), Based on (9), the interference fringe image _Gk is calculated.
In addition, the function f of Formula (9) is the said Formula (2), (alpha) _k of Formula (7) is calculated | required by the appropriate numerical solution method from Formula (9).
For example, when k = 2, as shown in FIG. 7B, F ₉ , F ₁₀ , F ₁₁ , and F ₁₂ are selected as correction calculation images and α ₂ is used.

In this modification, since I _k−1 , I _k , I _{k + 1} , I _{k + 2} are not in the range of j = 1,..., M, the following equations (9), (10 A dummy candidate image is prepared as shown in FIG.

Thus, in this modification, four candidate images F _Ik−1 , F _Ik , F _{Ik + 1} including a candidate image (F _Ik or F _{Ik + 1} ) having a phase amount closest to Δφ · (k−1), based on the F _{Ik + 2,} to calculate the interference fringe image _{G k} by performing the cubic interpolation operation.
Since cubic interpolation is continuous interpolation up to the first-order differentiation, it is possible to calculate an interference fringe image closer in phase amount to Δφ · (k−1) compared to linear interpolation. Fringe scan interference fringe measurement can be performed.
This modification is an example when an interpolation calculation different from linear interpolation is used in the image selection process for interpolation calculation.

Here, the derivation of the equations (7) and (8) will be briefly described.
In the cubic interpolation, as shown in the equation (12), when a control point Q _j (j = 1,..., M) is given, the product sum of the control point Q _j and the cubic polynomial C (j−t) is used. An interpolation value c (t) (where t = 2 to m−1) is obtained. The interval of t <2 and t> m−1 cannot be correctly interpolated by cubic interpolation, but it is sufficient to increase the control points in advance by preparing dummy control points.

Here, the condition that the polynomial C (j−t) satisfies is that the function c (t) passes through the control points Q ₂ ,..., Q _m−1 , and the continuous differentiation (the right derivative and the left derivative of each control point are the same). Is to satisfy the following condition. Therefore, as a cubic polynomial, coefficients a ₀ ,..., A ₃ , b ₀ ,..., B ₃ are determined under the condition of the following expression (14) as in the following expression (13).

Since there are seven conditions in equation (14) and one condition is insufficient for eight unknowns, other unknowns are solved by setting b ₃ = K. Thereby, the following equation (15) is obtained.

The value of K can be appropriately set, but is preferably in the range of -0.5 to -2. When K = −0.5, interpolation is performed in a low-pass manner, and when K = −2, high frequency tends to be emphasized, and usually −1 is often used.
So far, the polynomial C (X) in the equation (12) has been described as taking the product sum with all control points. However, when | X | ≧ 2, C (X) = 0. Focusing on J to J + 1, the only control points that need to be summed with C (X) are Q _j−1 ,..., Q _{j + 2} , the control point Q _j is F _{j, and} C (X) is Cubic (X). Then, Expression (7) is derived.
Cubic (X) in Expression (8) of this modification corresponds to the case where K = −1 in C (X) in Expression (15).

  In the above description, an example in which linear interpolation calculation and cubic interpolation calculation are used as the interpolation calculation has been described. However, the interpolation calculation is not limited to these, and any known interpolation calculation may be used. Good.

Further, in the above description, the phase of the candidate image F ₁ as described in example where the 0, including a null state, in advance to obtain the candidate image in the range of more than 2 [pi, also phase amount 0 including Thus, Δφ · (k−1) (k = 1,..., N) may all be calculated by interpolation.

  In the above description, an example using a Fizeau interferometer as an interferometer has been described. However, the type of interferometer is not limited to the Fizeau type as long as a phase-shifted interference fringe image can be acquired. For example, a Twiman Green interferometer, a Mach-Zehnder interferometer, or the like can be preferably used.

  In the above description, as an example, the interferometer has been described with the configuration of the Fizeau interferometer when the concave surface is used as the measurement surface, but the arrangement of the interferometer when the convex surface or plane is the measurement surface, The configuration is well known, and the fringe scan interference fringe measuring method of the present invention can be used in any case.

  In the above description, the applied voltage is applied to the piezoelectric element in proportion to time. However, the applied voltage may be simply changed monotonously with respect to time.

  In the above description, an example in which candidate images are acquired at regular sampling times ΔT has been described. However, the sampling timing sets the change in the phase amount of the acquired candidate images to an appropriate size. Therefore, it may be set at unequal intervals. For example, the sampling timing may be set so that the phase change changes finely only near the phase shift amount of n interference fringe images.

  In the above description, since the optical element moving mechanism is continuously driven in a certain direction, the candidate images are sequentially acquired in order of the magnitude of the phase shift amount. Since the magnitude relationship may be rearranged in the phase change calculation step, if the candidate image acquisition step can acquire information related to the order of magnitude of the phase shift amount, the acquisition order of candidate images is not necessarily matched to the magnitude of the phase shift amount. It does not have to be.

  In the above description, the examples in the case of using the expressions (1) and (2) as the approximate expressions for calculating the phase change calculated in the phase change calculating step have been described. An approximate expression having an appropriate function form can be adopted according to the movement characteristics of the mechanism. For example, the formula (1) may be a fourth-order or higher order approximation formula for i. Further, the approximate curve 33 in FIG. 6 may be represented by one approximate expression.

  In the above description, an example in which a piezoelectric element is used as the optical element moving mechanism has been described. However, if the optical element can be moved by a minute amount within a range of, for example, about 1 to 2 wavelengths of use, the piezoelectric element can be moved. Is not limited.

DESCRIPTION OF SYMBOLS 1 Laser light source 2 Collimating lens 3 Beam splitter 4 Fizeau lens 4a Fizeau surface 5 Surface to be measured 6 Condensing lens 7 CCD (imaging part)
7a Imaging surface 8 Piezo element (optical element moving mechanism)
9 Piezo element controller 10 Measurement control unit 20 Image acquisition unit 21 Calculation processing unit (phase change calculation unit, interpolation calculation image selection unit, interference fringe image calculation unit)
22 Movement control unit 23 Image storage unit 30a Parallel light (same luminous flux)
30b Transmitted light 30c Measurement surface reflected light (measurement light)
30d Reference surface reflected light (reference light)
50 Interferometer F _j Time-series image (candidate image)
G _k interference fringe image _{F Ik, F Ik + 1,} F Ik-1, F Ik + 2 interpolation operation for image

Claims (6)

The same light beam is divided into measurement light reflected or transmitted by the surface to be measured and reference light having a wavefront corresponding to the shape of the reference surface, and interference fringes are formed by the measurement light and the reference light. Then, an interference fringe image is obtained, and further, from this state, the phase difference between the measurement light and the reference light is shifted by substantially constant phase increment Δφ in total (n is an integer of 3 or more) A fringe scan interference fringe measuring method for obtaining the interference fringe image and calculating the shape of the surface to be measured based on these n interference fringe images,
In order to obtain the n interference fringe images, the phase difference between the measurement light and the reference light is shifted from 0 to Δφ · (n−1) or more in a size that subdivides the phase increment Δφ, and n A candidate image acquisition step of acquiring more interference fringe images and storing them as candidate images of the n interference fringe images;
The phase change of each candidate image is calculated from the luminance change data when the luminance at the same position of each candidate image acquired in the candidate image acquisition step is arranged in the order of the magnitude of the phase shift amount of each candidate image. A phase change calculating step,
Based on the phase change calculated in the phase change calculation step, a plurality of candidates whose phase amount is in the vicinity of Δφ · (k−1) (where k = 1,..., N) are selected from the candidate images. An interpolation calculation image selection step of selecting an image as an interpolation calculation image;
An interference fringe image calculation step of obtaining the n interference fringe images by performing an interpolation calculation using the interpolation calculation image selected in the interpolation calculation image selection step;
A fringe scan interference fringe measuring method comprising: The interference fringe image calculation step includes:
From the candidate image, the phase amount is equal to or less than Δφ · (k−1), and the candidate image _FIk is the candidate image closest to Δφ · (k−1), and the phase amount is Δφ · (k−). 1) Using a candidate image F _{Ik + 1} that is a candidate image that is larger and closest to Δφ · (k−1), the phase amounts of the candidate images F _{Ik and} F _{Ik + 1} are respectively Δφ · (k−1) −A _{2. The k} interference fringe images are calculated as interference fringe images G _k by linear interpolation based on the following equation (1) when _k is Δφ · (k−1) + B _k. The fringe scan interference fringe measuring method described in 1.
_{G k = (F Ik × B} k + F Ik + 1 × A k) / (A k + B k) ··· (1)
Here, G _k , F _Ik , and F _{Ik + 1} represent two-dimensional image data of each interferometer image. The candidate image acquisition step includes
The candidate image is acquired at a sampling period corresponding to a phase amount having a magnitude to subdivide the phase increment Δφ while continuously changing an optical path difference of the reference light with respect to the measurement light in a certain direction. The fringe scan interference fringe measuring method according to claim 1 or 2.   The fringe scan interference fringe measuring method according to claim 1, wherein the phase increment Δφ is π / 2.   The fringe scan interference fringe measuring method according to claim 1, wherein the phase increment Δφ is π / 4. The same light beam is divided into measurement light reflected or transmitted by the surface to be measured and reference light having a wavefront corresponding to the shape of the reference surface, and interference that forms interference fringes by the measurement light and the reference light Measuring optics,
An imaging unit that captures an image of the interference fringes;
An optical element moving mechanism for moving an optical element included in the interference measurement optical system in order to change an optical path difference of the reference light with respect to the measurement light;
While the optical element moving mechanism moves the optical element, the phase of the optical path difference changes by Δφ while the phase of the optical path difference changes to Δφ · (n−1) (n is an integer of 3 or more). An image acquisition unit for acquiring a plurality of interference fringe images more than n shifted in increments so as to be subdivided from the imaging unit;
From the luminance change data when the luminance at the same position of the plurality of interference fringe images acquired by the image acquisition unit is arranged in the order of the magnitude of the phase shift amount of the plurality of interference fringe images, the plurality of interference fringe images A phase change calculator for calculating the phase change of
From the plurality of interference fringe images acquired by the image acquisition unit based on the phase change calculated by the phase change calculation unit, the phase amount is Δφ · (k−1) (where k = 1,... n) an image selection unit for interpolation calculation that selects a plurality of candidate images near the image for interpolation calculation;
An interference fringe image calculation unit that obtains the n interference fringe images by performing an interpolation calculation using the interpolation calculation image selected by the interpolation calculation image selection unit;
With
An interferometer that calculates a shape of a surface to be measured from the n interference fringe images calculated by the interference fringe image calculation unit.