JP2000329514A - Method for correcting interference fringe image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an image correcting method for a device for simultaneously measuring phase-shift interference fringes capable of automatically adjusting the contrast and luminance of three simultaneously obtained interference fringe images through the use of electric image pickup data. SOLUTION: In a light-wave interferometer to create interference fringes by sample light from a surface to be detected and reference light from a reference surface, the contrast value and bias value of each section are obtained from interference fringe image data acquired by an image pickup processing means to pick up the image of interference fringes, and the interference fringe image data are corrected and converted so as to cancel the mutual difference between the contrast values and the bias values between the sections to perform image analysis or to display the interference fringe images through the use of the interference fringe image data after the correction and conversion.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a method for correcting an image of an interference fringe acquired by an interferometer, and more particularly to a method for correcting an optical and electrical systematic noise in an error of measuring an interference fringe image.

[0002]

2. Description of the Related Art As is well known, an optical interference fringe method using a light interference phenomenon is used as one of methods for knowing a precise unevenness state of a surface to be measured. A phase shift method using a phase difference between the sample light and the reference light is used. In other words, this phase shift method precisely compares and measures the sample light and the reference light with the phase difference angle.
A plurality of interference fringe screens are obtained by changing the phase of the interference fringes over time by, for example, moving the reference plane in the optical axis direction.

However, in the phase shift method for obtaining different phase states over time, a certain time is required at the stage of capturing interference fringes that are the basis of the analysis. The results are affected by air fluctuations and vibrations during this time, making it difficult to maintain the reliability of the analysis results and moving the surface to be measured relative to the interference optical system. Analysis was difficult in some cases.

In order to solve the above-mentioned problem of the phase shift method, conventionally, a large number of interference fringe images obtained by changing the phases of the wavefronts of the reflected light from the test surface and the reference surface by a fixed amount are simultaneously obtained. An apparatus for simultaneous measurement of phase shift interference fringes as shown in FIG. 10 for imaging is proposed, for example, in Japanese Patent Laid-Open No. 2-287107. That is, in the simultaneous measurement apparatus for phase shift interference fringes in FIG. 10 entitled “Two-dimensional information acquisition apparatus”, the light from the laser light source A is expanded by the lens system B into parallel light, and the polarization semi-transparent mirror C , The light is divided into two orthogonally polarized lights. On these polarization paths, quarter-wave plates D and E are arranged, respectively, and the reference plane F and the test plane G are illuminated by circularly polarized light.
Also, a so-called “Michelson interferometer” is formed in which the reflected light from the surface G to be tested is again converted into linearly polarized light through the quarter-wave plates D and E, and then incident on the polarizing semi-transparent mirror C again.

The two linearly polarized light beams from the "Michelson-type interferometer" are passed through a quarter-wave plate H and turned into clockwise and counterclockwise counterclockwise circularly polarized lights. After these circularly polarized light beams are split into three light beams by the two beam splitters BS1 and BS2 and the total reflection mirror M of the phase shift imaging system indicated by "I", three phase-shifted interference fringe images are simultaneously formed. Is obtained. That is, these split light beams are passed through the polarizing plates J1, J2, J3 and then imaged on three cameras, ie, imaging processing means K1, K2, K3.
That is, in this optical system, the polarization directions of the polarizing plates J2 and J3 are 45 degrees and 90 degrees with respect to the polarizing direction of the polarizing plate J1.
By inclining by three degrees, three interference fringe images (screens) having different phases by 90 degrees can be obtained.

By the way, in the above-mentioned "two-dimensional information acquisition apparatus", three imaging detectors K1, K2, K
Three interference fringe images phase-shifted by 3 are obtained at the same time, and it is premised on image analysis that the contrast value and the bias value of these interference fringe images are equal to each other. Therefore, the “two-dimensional information acquisition device” as described above
In the above, the optical intensity data is optically adjusted by interposing an attenuation plate in the optical path of the imaging processing means K1, K2, K3 and adjusting the density of the attenuation plate located in each optical path.

Further, in the above-described apparatus, various optical elements existing on each optical path and image pickup processing means K1, K2, K3.
As a result, systematic noise is generated, and common noise or separate unique noise is superimposed on each of the three interference fringe images. As a result, the analysis accuracy of the three phase-shifted interference fringe images has been reduced. Noise generated in the interferometer includes noise and interference noise generated due to local nonuniformity (increase or decrease) in reflectance and transmittance due to contamination or defects of the optical element. Interference noise is noise that occurs when a plurality of reflected lights generated at interfaces having different refractive indexes interfere with each other. As a general method to reduce interference noise, additional treatment such as wedge processing and anti-reflection coating is applied to the optical element,
In the final optical adjustment stage of the interferometer, a method has been adopted in which noise is reduced.

[0008]

However, in such an image correction method of the conventional apparatus for simultaneous measurement of phase shift interference fringes, it is not only necessary to individually adjust physical optical components, but also to adjust the physical optical components individually. Due to the deviation of the optical components, the variation of the characteristics of the optical components, and the late contamination, the unevenness of the light intensity when passing through each optical system is superimposed on the imaged data, which poses a problem in accurate image analysis. Furthermore, there is a limit in reducing optically or electrically generated noise by additional measures or adjustments to the optical element. That is, it has been difficult to obtain an image with uniform contrast and bias (light intensity) and no noise within one interference fringe image or between a plurality of interference fringe images.

Particularly, in the simultaneous measurement apparatus for phase shift interference fringes, since three interference fringe images are simultaneously obtained by independent optical systems, the obtained interference fringe images have a uniform contrast and a uniform bias. Noise reduction is a very important issue in image mutual analysis. The purpose of the present invention is
In the interference fringe image measurement as described above,
In view of the image correction method for electrical systematic noise, it is necessary to obtain an image correction method for obtaining an image free of noise with uniform contrast and light intensity within one interference fringe image or among a plurality of interference fringe images. is there.

[0010]

In order to achieve this object, the present invention provides an optical interferometer for generating an interference fringe by using a sample light from a test surface and a reference light from a reference surface to image the interference fringes. The contrast value and the bias value of each part are obtained from the interference fringe image data obtained by the imaging processing means, and the interference fringe image data is corrected so as to eliminate the mutual difference between the contrast value and the bias value between the respective parts. The present invention proposes an image correction method for interference fringes that performs image analysis or interference fringe image display using the converted and corrected interference fringe image data. By correcting these contrast values and bias values, uniform interference fringe image data can be obtained, so that image analysis is facilitated and an easy-to-view screen can be obtained even when displaying interference fringe images.

Further, each of the portions divides the one interference fringe image data into a plurality of portions and assigns the data, so that the contrast value and the bias value of the interference fringe image in one screen can be made uniform. In addition, by assigning each of the portions to each of the plurality of interference fringe image data, for example, in a simultaneous phase shift interference fringe measurement apparatus that simultaneously obtains three interference fringe image data, the three interference fringe image data obtained The contrast value and bias value between them can be made uniform.

Further, according to the present invention, in a light wave interferometer for generating an interference fringe by using a sample light from a test surface and a reference light from a reference surface, an imaging processing means for imaging the interference fringe includes the sample light and the reference light. Instead of light, when a noise measurement image of a pre-planned light intensity distribution of light intensity at which the imaging processing means is not saturated is supplied, the noise measurement image is captured and stored as unique noise distribution data, Using the interference fringe image data obtained by imaging the interference fringes and the specific noise distribution data, a noise distribution included in the interference fringe image data is obtained each time, and these noise distributions are obtained from the interference fringe image data. The present invention proposes an image correction method for interference fringes in which image data is analyzed or interference fringe images are displayed using the corrected interference fringe image data. Since interference fringe image data with a good S / N ratio can be obtained by reducing noise and interference noise by these corrections, image analysis is easy and highly accurate, and it is easy to see even when displaying an interference fringe image. You can get the screen.

Further, the specific noise distribution data is obtained by obtaining a difference between a noise measurement image having a known light intensity distribution and image data obtained when the noise measurement image is taken, thereby obtaining accurate noise distribution data. Can be obtained. Further, the specific noise distribution data is obtained by decomposing the image data obtained when the noise measurement image is captured into spatial frequency components and extracting and extracting the spatial frequency components of the noise. Can be obtained. Alternatively, the noise measurement image may be obtained by shifting one or both of the optical axis of the sample light from the test surface and the optical axis of the reference light from the reference surface to such an extent that interference does not occur. This makes it possible to easily obtain a noise measurement image. The noise measurement image can be obtained more easily by obtaining the noise measurement image from only one of the sample light from the test surface and the reference light from the reference surface.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The embodiments of the present invention will be described below in detail with reference to FIGS. 1 to 6 show a first embodiment of the present invention, and show an example in which correction conversion of a contrast value and a bias value is applied to a simultaneous measurement apparatus for phase shift interference fringes.

FIG. 1 shows an example of an optical system of an apparatus for simultaneous measurement of in-phase interference fringes, in which a sample light and a reference light which are optically free from interference are produced by a "Michelson interferometer". That is, the light from the laser light source 1A is expanded by the lens system 2A to be parallel light, and is split into two orthogonal polarized lights by the polarizing semi-transparent mirror 3A.

On these optical paths, quarter-wave plates 4A and 5A are arranged, respectively, and these circularly polarized lights are radiated on the reference surface 6A and the test surface 7A. The reflected light from the reference surface 6A and the test surface 7A is transmitted through the quarter-wave plates 4A and 5A again, and is again incident on the semi-polarizing mirror 3A as linearly polarized light. The linearly polarized light beam in this state is a non-interfering light beam in which two light beams that have phase information corresponding to the difference in the shape of the test surface 7A with respect to the reference surface 6A and are orthogonally polarized are superimposed.

Thereafter, the light beam obtained by superimposing the two light beams is passed through a quarter-wave plate, and becomes an original light beam that has been turned into counterclockwise and clockwise counterclockwise circularly polarized lights. here,
Laser light source 1A, lens system 2A, polarizing semi-transparent mirror 3A, 1 /
4 wavelength plates 4A, 5A, reference surface 6A, test surface 7A, 1/4
The wave plate 8A constitutes an original light beam generating means. Next, these original light beams are divided into three light beams by a light beam splitting means, that is, a composite prism indicated by a reference numeral "9A". That is, the composite prism 9A, into which the original light beam is incident from the interface 9a, forms a first beam splitter 10A and a direction perpendicular to the first beam splitter 10A at an angle of 45 degrees with respect to the optical axis of the original light beam. It has a built-in second beam splitter 11A.

The first beam splitter 10A divides the transmitted light at a ratio of "2" to the intensity "1" of the reflected light with respect to the intensity "1" of the reflected light. Is arranged in a direction perpendicular to the optical axis in a direction parallel to a plane including the light incident on the beam splitter 10A and the reflected light, and is transmitted through the first polarizing plate 12A. The reflected light causes interference. Then, in the second beam splitter 11A, into which the transmitted light from the first beam splitter 10A is incident, the light beam is split at a ratio of the intensity of the transmitted light to "1" with respect to the intensity of the reflected light "1".

The exit end face 9c from which the reflected light from the second beam splitter 11A is emitted includes the light that enters the beam splitter 11A and the light that is reflected in a plane whose transmission axis is perpendicular to the optical axis. A second polarizing plate 13A set in a direction perpendicular to the plane is provided.
The reflected light having passed through causes interference. The exit end surface 9d from which the transmitted light from the second beam splitter 11A is emitted includes light incident on the beam splitter 11A and light reflected on the exit axis surface in a plane perpendicular to the optical axis in the transmission axis direction. A third polarizing plate 14A set at an angle of 45 degrees with the plane is arranged, and the transmitted light transmitted through the third polarizing plate 14A causes interference.

As described above, the transmission axis direction of the first to third polarizing plates is such that, in the plane perpendicular to the optical axis of the light passing through the polarizing plate, the light is incident on the beam splitter immediately before. Are set to be parallel, perpendicular, and 45 degrees to the plane including the reflected light and the reflected light, so that the phase difference between the sample light and the reference light of the interference fringes caused by the respective interferences is 9 degrees.
It differs by 0 degrees. That is, based on the phase difference between the sample light and the reference light of the interference fringes when the transmission axis direction is parallel, the phase is delayed by 90 degrees from the clockwise circularly polarized light beam when the transmission axis direction is vertical. A linearly polarized light beam can be extracted, and a linearly polarized light beam having a phase advanced by 90 degrees can be extracted from the counterclockwise circularly polarized light beam. Therefore, the phase difference between the sample light and the reference light is 1
80 degrees. When the transmission axis direction is 45 degrees, a right-handed circularly polarized light beam can extract a linearly polarized light beam with a phase delayed by 45 degrees, and a left-handed circularly polarized light beam can extract a linearly polarized light beam with a 45-degree advanced phase. Therefore, the phase difference between the sample light and the reference light is 90 degrees.

Here, the first to third polarizing plates 12A, 12A
A and 14A respectively constitute interference means. These interference fringes having different phases are obtained by imaging processing means 15A, 16A, 17 constituted by a CCD solid-state scanning element or the like.
An image is taken almost simultaneously by A and image processing is performed. Since the interference fringe image captured by the imaging processing unit 17A is inverted left and right with respect to the interference fringe images captured by the imaging processing units 15A and 16A, the position inversion conversion of the imaging processing unit 17A is performed. The left / right reversal process is performed by the unit. The imaging processing means 15A, 16A, 17A
Are arranged such that the distance from the interface 9a is equal to the center of the original light flux.

FIG. 2 shows each of the imaging processing means 15A, 16
A and 17A are interference fringe image data (screens) recorded in this embodiment. In this embodiment, these interference fringe images (screens) are used.
The light intensity of the digital data is corrected by the following method, and the light intensity between the interference fringe images is adjusted. At the time of the correction processing, focusing on the light intensity change of one line as shown in FIG. 3 in the two-dimensional digital data of the interference fringe image (screen), as shown in FIG. Is a substantially periodic change curve β in which the shape of the test surface and the amount of inclination of the test surface are reflected in the light intensity of the interference fringes. That is, in this change curve β, as shown in FIG. 5, the maximum value IMAX and the minimum value IMIN
Will exist. Therefore, IMA obtained here
Using X and IMIN, the contrast value and bias value of the interference fringe image can be calculated by a known method.

Therefore, the contrast value and the bias value of the interference fringe image data of the imaging processing means 15B are matched with reference to certain interference fringe image data (for example, data obtained by the imaging processing means 15A). By repeating the operation of matching the contrast value and the bias value of the interference fringe image data of the means 15C with the contrast value and the bias value of the interference fringe image data of the imaging detector 15A, respectively, the manufacturing of the optical components and the imaging processing means is repeated. It is possible to eliminate a difference for each image caused by the difference.

First, two imaging processing means 15A and 16A
The information included in the interference fringes I1 and I2 obtained by the following equation is: I1 (x, y) = Bias1 (x, y) + Amp1 (x, y) · cos (φ (x, y)) The first equation I2 (x , y) = Bias2 (x, y) + Amp2 (x, y) · cos (φ (x, y + α). The intensity of the interference light is spatially uniform, and Assuming that the contrast value (amplitude) and bias value of the generated interference fringes are spatially uniform, I1 (x, y) = Bias1 + Amp1 · cos (φ (x, y)) The first expression 'I2 ( x, y) = Bias2 + Amp2 · cos (φ (x, y) + α) Equation (2) Assuming that the intensity of the light generating the interference fringes is uniform over the entire screen, the contrast and bias are It can be artificially calculated from the light intensity data of the interference fringes I1 and I2 as follows.

When the contrast is [C], the definition formula of the contrast is as follows.

(Equation 1) Becomes

Here, when the interference fringe image shown in FIG. 2 is obtained, one line of light intensity information which is substantially orthogonal to the interference fringe as shown in FIG. 3 is obtained as shown in FIG. Profile information. In this profile information, as shown in FIG. 5, if the intensity information of the light intensity having the maximum and the minimum is substituted into the definition expression as the maximum value IMAX and the minimum value IMIN, the contrast [C] is obtained, and the bias value is obtained. [B] is obtained in the form of an average value obtained by dividing the integral value of the data of FIG. 3 by the number of data points.

Bias1 = B1, Bias2 = B2, and Amp1 =
By substituting the relation of C1.Bias1 and Amp2 = C1.Bias2 into the first and second equations, I1 (x, y) = B1 + C1.B1.cos (.phi. (X, y)) The third equation I2 (x , y) = B2 + C2.B2.cos (.phi. (x, y) +. alpha.).

With respect to equation (4),

(Equation 2) Multiply by

(Equation 3) Is obtained.

Next, similarly, when the following numerical value 2 is added to both sides of the fifth equation,

(Equation 4) The fifth equation is

(Equation 5) Thus, the bias value and the contrast value of the interference fringe I2 can be made equal to the bias and the contrast of the interference fringe I1.

That is, the correction for matching the interference fringe I2 with the interference fringe I1 may be performed by multiplying the interference fringe I2 by the numerical value 1 and adding the numerical value 2 thereto. In other words, all the biases and contrasts of a plurality of image data can be corrected and converted by using this algorithm.

In the description of the first embodiment of the present invention, three interference fringe image data captured by the simultaneous phase shift interference fringe measuring apparatus shown in FIG. Although the method of correcting and converting contrast and bias has been described, in this method,
It is assumed that the contrast and bias in one image data are uniform. On the other hand, when the contrast and the bias in one image data are not uniform, the image data (screen) can be appropriately divided and the same correction conversion can be performed between the divided partial image data. For example, as shown in FIG.
Even when the image data is divided into four parts left and right, profile information of one line substantially orthogonal to the interference fringes is obtained in each of these partial image data, and the above-described bias and contrast correction conversion is performed between the partial image data. Good. When the bias value and the contrast value of the interference fringe in one interference fringe image are different from each other in the partial portion, the image data (screen) is appropriately divided so that the bias value and the contrast value are the same between the divided images. By performing the correction, it is possible to display a uniform and easily visible interference fringe.

FIGS. 7 to 9 show a second embodiment of the present invention, in which noise distribution is corrected for interference fringe image data.

In the case of the second embodiment of the present invention, a noise distribution corresponding to the light intensity distribution is obtained from the digital data of the interference fringe image (screen) obtained from the imaging processing means 15A, for example, by the following method. Noise of interference fringe images due to interference noise due to unnecessary reflected light and dirt on the optical component surface can be reduced. That is, generally speaking, since the dirt on the optical parts affects the optical transmittance and the reflectance, the magnitude of the noise generated by the dirt is proportional to the intensity of the incident light, and the interference noise is also reduced. Similarly, it depends on the intensity of the reflected light on the surface of the optical component, and in the interference fringe image data obtained by the imaging processing means, the noise distribution is corrected using the fact that noise is proportional to the light intensity. is there.

Although the present invention originally deals with two-dimensional image data, here, as shown in FIG. 7A, one line of light intensity data in a two-dimensional image The description will be made using a one-dimensionally schematic example of the noise generated from the processing means. As shown in FIG. 7B, when imaging the measurement light 1 having a spatially gentle intensity change, when interference noise that varies spatially at a high frequency occurs in an optical component or an imaging processing unit, When light having a large spatial intensity change such as an interference fringe, for example, the measurement light 2 in FIG. 7C is imaged using the same image processing means with the same optical component interposed, the noise intensity also changes to the input intensity distribution. Accordingly, the value greatly changes as shown in the figure.

Focusing on such a fact, in the second embodiment, for example, the interference fringe image data including the intrinsic noise obtained from the imaging processing means 15A and the intrinsic noise are used.
It is proposed that a noise distribution included in image data is obtained each time, and these noise distributions are subtracted from the image data to obtain interference fringe image data corrected for noise distribution. In the following description, noise distribution correction for one interference fringe image will be described. However, for example, when applied to a simultaneous measurement apparatus for phase shift interference fringes, the same applies to other two interference fringe images obtained at the same time. However, the noise distribution correction processing may be performed in the same manner.

FIG. 10 is a noise distribution correction flowchart in the second embodiment. In the noise distribution correction processing, a noise measurement image of a planned light intensity distribution is taken, and
It is assumed that the characteristic noise distribution included in the image data is measured and stored in advance (steps S1 and S2
). Here, the noise measurement image and the specific noise distribution are set to I2 (x, y) and I3 (x, y), respectively, and an interference fringe image used for shape calculation is captured. Let I1 (x, y) (step S3). In this case, the captured interference fringe image data includes the inherent noise generated by the optical components and the image processing unit.

However, the noise distribution included in I1 (x, y) is different from I3 (x, y), and I3 (x, y) is apparently changed according to the light intensity distribution of I1 (x, y). The intensity is modulated. Therefore, the noise distribution included in I1 (x, y) is obtained using the data of I1 (x, y), I2 (x, y), and I3 (x, y). Assuming that the result is I4 (x, y), I4 (x, y) can be obtained by the calculation of the following equation (step S4).

(Equation 6)

When I4 (x, y) data is obtained, I1 (x, y)
By subtracting the intensity of I4 (x, y) from, the following interference fringe image I5 (x, y) with reduced noise can be obtained (step S5).

(Equation 7)

FIG. 9 is a one-dimensional model of an example of data obtained by the numerical processing shown in FIG. This data example is, for a two-dimensional image, a certain one in the light intensity distribution.
This corresponds to the light intensity distribution data of the line. In the one-dimensional model of FIG. 9, the measurement light 1 and the measurement light 2 of FIG. 7 are a noise measurement image and an interference fringe image including a specific noise, respectively. In FIG. 9, imaging information I1 including noise is obtained by using a noise measurement image I2 and a specific noise distribution I3.
(Interference fringe image data), an intermediate noise distribution I4 is obtained, and as a result, good light intensity distribution information I5 with reduced noise (interference fringe image data corrected for noise distribution)
Can be understood.

In the above-described second embodiment of the present invention,
Although the noise distribution included in the imaging data is obtained each time, these noise distributions are subtracted from the imaging information, and the noise distribution corrected interference fringe image data is obtained.
In the present invention, a unique noise distribution can be obtained from a noise measurement image also by the following correction processing. 1) An inherent noise distribution can be obtained by taking a difference between a noise measurement image having a known light intensity distribution and light intensity distribution data obtained by imaging the image. In other words, as a result of using light A (x, y) having a known two-dimensional light intensity distribution and capturing this light with an interferometer, a two-dimensional light intensity distribution of the form B (x, y) is obtained as a noise measurement image. When the data is obtained, the inherent noise distribution data I3 can be obtained by the calculation of I3 = B (x, y) -A (x, y).

2) The intensity distribution data obtained when the noise measurement image is captured may be decomposed into spatial frequency components, and the spatial frequency components of the noise may be extracted to create a unique noise distribution. This is because, when the spatially varying frequency component of the noise contained in the noise measurement image I2 is known, the spatial light intensity change of the two-dimensional light intensity distribution data I2 obtained from the noise measurement image is obtained by Fourier series expansion. The characteristic noise distribution data I3 can be obtained by decomposing into spatial frequency components and filtering only the frequency components corresponding to noise.

The above-described noise measurement image I2 can be easily generated by the following method. 1) By blocking either the sample light from the test surface or the reference light from the reference surface and capturing the other, a noise measurement image including a specific noise distribution can be obtained. 2) If one or both of the optical axis of the sample light from the test surface and the optical axis of the reference light from the reference surface are imaged with eccentricity to such an extent that no interference occurs, the characteristic noise distribution is included. Since light in the interfered state can be obtained, this may be used as a noise measurement image. 3) In addition, a noise measurement image including a specific noise distribution can be obtained by removing a polarizing plate or the like for generating interference fringes and taking an image.

In the first embodiment, the image picked up by the apparatus for simultaneously measuring phase shift interference fringes shown in FIG.
A method for correcting and converting the contrast and bias between each image data for one interference fringe image data is shown. Further, when the contrast and bias in one image data are not uniform, the image data (screen) is appropriately divided. Although the method of performing the same correction conversion between the divided partial image data has been described, the correction conversion may be performed between a plurality of image data after performing the correction conversion within one image data. Alternatively, conversely, after performing the correction conversion between a plurality of image data, the correction conversion may be performed between the respective image data.

[0044]

As is apparent from the above description, according to the interference fringe image data correction method of the present invention, the correction conversion of the contrast and bias of the interference fringe image and the noise correction are performed using the electric imaging data. By doing so, the light intensity can be made uniform and interference fringe image data with a reduced S / N ratio with reduced noise can be obtained, so that image analysis is facilitated and even when an interference fringe image is displayed. An easy-to-view screen can be obtained.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of an optical system of an apparatus for simultaneously measuring phase shift interference fringes according to a first embodiment of the present invention.

FIG. 2 is an interference fringe screen captured by an imaging processing unit.

FIG. 3 is an explanatory diagram of division of an interference fringe screen.

FIG. 4 is a conceptual diagram showing the relationship between the light intensity of interference fringes taken into an image processing unit and a coordinate position.

FIG. 5 is an explanatory diagram of a maximum light intensity and a minimum light intensity position.

FIG. 6 is an explanatory diagram of another example of division of an interference fringe screen.

FIGS. 7A, 7B, and 7C are comparison diagrams illustrating a relationship between optical noise and electric noise of an imaging screen.

FIG. 8 is a flowchart of image pickup signal processing.

FIG. 9 is a schematic diagram of an example of a numerical operation result.

FIG. 10 is an explanatory diagram of an optical system of a conventional two-dimensional information acquisition device.

[Explanation of symbols]

 Reference Signs List 1A Laser light source 3A Polarizing semi-transparent mirror 4A, 5A Quarter wave plate 6A Reference surface 7A Test surface 9A Wavefront splitting optical system 10A First beam splitter 11A Second beam splitter 12A First polarizing plate 13A Second polarizing plate 14A Third Polarizing plate 15A-17A Imaging processing means

 ──────────────────────────────────────────────────の Continuing from the front page (72) Inventor Hiroshi Nobino F-term (reference) 2F064 AA09 CC01 EE01 FF01 GG22 GG23 GG32 GG38 GG53 HH03 HH08 JJ02 1 at 430 Kamiyokoba Tsukuba, Ibaraki

Claims (9)

[Claims] In an optical interferometer for generating an interference fringe by using a sample light from a test surface and a reference light from a reference surface, each part is obtained from interference fringe image data acquired by an image processing means for imaging the interference fringes. Determine the contrast value and the bias value, between the respective parts, to correct the interference fringe image data so as to eliminate the difference between the contrast value and the bias value, using the interference fringe image data after the correction conversion, An image correction method for interference fringes, wherein image analysis or interference fringe image display is performed. 2. The method according to claim 1, wherein each of the parts divides one piece of the interference fringe image data into a plurality of parts and allocates the divided data. 3. The method according to claim 1, wherein each of the portions is assigned to each of the plurality of interference fringe image data. 4. The interference fringe image according to claim 3, wherein said imaging processing means includes a plurality of imaging detectors, and acquires a plurality of said interference fringe image data by said plurality of imaging detectors. Correction method. 5. An optical interferometer for generating an interference fringe by using a sample light from a test surface and a reference light from a reference surface, wherein an imaging processing unit for imaging the interference fringes replaces the sample light and the reference light. When a noise measurement image of a predetermined light intensity distribution of light intensity that does not saturate the imaging processing unit is supplied, the noise measurement image is captured and stored as unique noise distribution data, and the interference fringe is captured. Using the obtained interference fringe image data and the specific noise distribution data, a noise distribution included in the interference fringe image data is obtained each time, and these noise distributions are subtracted from the interference fringe image data to perform correction. And performing image analysis or interference fringe image display using the corrected interference fringe image data. 6. The characteristic noise distribution data is obtained by a difference between a noise measurement image having a known light intensity distribution and image data obtained when the noise measurement image is captured. Image correction method for interference fringes. 7. The characteristic noise distribution data is obtained by decomposing imaging data obtained when the noise measurement image is captured into spatial frequency components and extracting noise spatial frequency components. 5. The method for correcting an interference fringe image according to 5. 8. The noise measurement image according to claim 1, wherein one or both of the optical axis of the sample light from the surface to be inspected and the optical axis of the reference light from the reference surface do not cause interference. 6. The method for correcting interference fringes according to claim 5, wherein the image is obtained by shifting. 9. The interference fringe image according to claim 5, wherein the noise measurement image is obtained from only one of the sample light from the test surface and the reference light from the reference surface. Correction method.