JP2000205822A - Image measurement system and its image correcting method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To effectively prevent a measurement error from being caused owing to the coordinate deviation and distortion of an image pickup system by eliminating the need to physically adjust respective image pickup systems. SOLUTION: A conversion parameter calculation part 32 calculates conversion parameters for distortion correction by the image pickup devices from image data for correction obtained by picking up images of an object for image correction. A conversion parameter storage part 33 stores the calculated conversion parameters. A distortion eliminating conversion process part 34 performs a correcting process for distortion elimination as to image data for measurement according to the conversion parameters stored in the conversion parameter storage part 33. A surface shape analyzing process part 35 analyzes the surface shape according to the image data for measurement after the distortion elimination.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image measurement system for analyzing a surface shape of a measurement target from image data obtained by measuring the measurement target using a plurality of image pickup means, and an image calibration method therefor. About.

[0002]

2. Description of the Related Art As an image measurement system using a plurality of image pickup devices, for example, a laser interferometer disclosed in Japanese Patent Application Laid-Open No. 2-287107 is known. In this system, light from a light source is split into two orthogonally polarized light beams and illuminated on a reference surface and a measured surface, and orthogonally polarized light beams respectively reflected from the reference surface and the measured surface are reflected by a wavefront splitting optical system. , The light is divided into three light beams, and three interference fringe images having different phase relationships are respectively imaged by the three imaging devices.

In such a system for performing measurement using a plurality of image pickup apparatuses, the fact that the coordinate systems of the image data obtained by the respective image pickup apparatuses match is a great condition for preventing the occurrence of measurement errors. . However, since a plurality of imaging devices are actually arranged at different spatial positions, a shift of a coordinate system unique to each imaging device occurs. Therefore, it is necessary to adjust the displacement of each imaging device. Conventionally, however, the coordinate system of one of the plurality of imaging devices is used as a temporary reference, and each imaging device is moved in the six-axis direction. Adjustment was made with an adjustment mechanism that could be adjusted to

[0004]

However, in the above-mentioned conventional measuring system, the adjustment mechanism in the six-axis direction is large, and the adjustment mechanism is required by multiplying the number of imaging devices by six. Difficulties in the adjustment method and individual differences in the adjustment caused errors. Furthermore,
When there is a distortion or the like peculiar to the imaging system of each imaging device, it is practically impossible to adjust the distortion.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and eliminates the need for physical adjustment of each imaging system, thereby effectively preventing the occurrence of measurement errors due to coordinate shift and distortion of the imaging system. It is an object of the present invention to provide an image measurement system and an image calibration method thereof.

[0006]

An image measuring system according to the present invention comprises a plurality of image pickup means for picking up an image of an object to be measured;
An image measurement processing unit that performs measurement processing on the object to be measured from the plurality of image data obtained by the plurality of imaging units; Image calibration means for correcting a displacement of the coordinate system of each image data caused by the displacement, wherein the image calibration means captures a calibration imaging target including a plurality of points of interest whose coordinate values are known by each imaging means. Conversion parameter calculation means for calculating a conversion parameter for correcting a deviation of the coordinate system from the coordinate value of the point of interest on the image data obtained by performing the calculation, and the known coordinate value; Conversion parameter storage means for storing the obtained conversion parameters, and image data of the object to be measured obtained by each of the imaging means are stored in the conversion parameter storage means. And characterized in that with a correction means for correcting the transformation parameters.

Further, the image calibration method according to the present invention is characterized in that, when performing measurement processing of the object to be measured using a plurality of image data obtained by imaging the object to be measured by a plurality of imaging means, An image calibration method for correcting a shift in the coordinate system of each image data caused by a spatial shift of the image pickup means, wherein the image pickup means for calibration includes a plurality of points of interest whose coordinate values are known. Calculating a conversion parameter for correcting a deviation of the coordinate system from the coordinate values of the point of interest on the image data obtained thereby and the known coordinate values, Correcting the image data of the object to be measured obtained in the above by the calculated conversion parameter.

According to the present invention, the position of each imaging system of a plurality of imaging units is not adjusted by a mechanical adjustment mechanism, but the deviation of the coordinates of each imaging system from the image data obtained by each imaging unit. And the conversion parameters for correcting the distortion are calculated, and the actual measurement image is corrected using the conversion parameters. Therefore, physical adjustment of each imaging system is not required, and the coordinate shift and the distortion of the imaging system due to the displacement and distortion. The occurrence of a measurement error can also be effectively prevented.

Means for generating optical interference fringes by measuring light from the surface of the object to be measured having different polarization planes and reference light from the reference surface, and converting the phase of the optical interference fringes generated by the means to Means for simultaneously shifting the light beams by a predetermined amount to be incident on each of the image pickup means at the same time. The present invention can be applied to a laser interferometer that performs the above.

[0010]

Preferred embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a diagram showing a configuration of a Fizeau interferometer using a phase shift according to one embodiment of the present invention. The coherent light beam emitted from the light source 1 is converted into parallel light that is expanded by the expansion optical system 2, passes through the non-polarization type half mirror 3 whose reflectance and transmittance do not depend on the polarization direction, and passes through the reference surface 4. Is reflected, and the rest is transmitted and radiated to the surface 6a to be measured 6 via the quarter-wave plate 5. The measurement light reflected from the test surface 6a passes through the quarter-wave plate 5 again, and its polarization plane is rotated by 90 ° with respect to the polarization plane of the reference light. 3, a three-segment optical system 1 including non-polarizing half mirrors 7 and 8 and a total reflection mirror 9
0 divides the optical path into three optical paths 11, 12, and 13.

The reference light and the measurement light traveling on the first optical path 11 pass through the half-wave plate 14 and the polarizing plate 15, and
The reference light and the measurement light traveling on the second optical path 12 are 1
The reference light and the measurement light that pass through the 波長 wavelength plate 16 and the polarizing plate 17 and travel along the third optical path 13 pass through only the polarizing plate 18. Accordingly, in the imaging devices 21, 22, and 23, the phase is π through the optical systems 24, 25, and 26, respectively.
Three interference fringe images shifted by / 2 are simultaneously captured.

FIGS. 2A, 2B and 2C are diagrams showing interference fringe patterns picked up by the image pickup devices 21, 22 and 23, respectively. Each interference fringe pattern has a phase shifted by α. The image data of these patterns is input to the computer 30 and is subjected to processing for surface shape analysis. That is, each of the imaging devices 21, 22,
Assuming that the luminance at the two-dimensional position (x, y) of the image obtained at 23 is I1, I2, and I3, respectively, the phase φ of the wavefront
It is known that (x, y) is obtained as in the following equation (Optical Shop Testing, Edit by Malacara:
"Three-Step Algorithms").

[0013]

(Equation 1)

At this time, if distortions, coordinate shifts, or the like unique to each of the imaging devices 21, 22, and 23 occur, the measurement accuracy decreases. Therefore, in this system, each imaging device 2
It has an image calibration function for correcting software distortions 1, 2, and 23 in software.

FIG. 3 is a functional block diagram showing an image measurement processing system realized by the computer 30 having such an image calibration function. The image measurement processing system includes an image processing unit 31, a conversion parameter calculation unit 32, a conversion parameter storage unit 33, a distortion cancellation conversion processing unit 34, and a surface shape analysis processing unit 35. Image processing unit 31
Performs filter processing, feature extraction processing, center-of-center calculation processing, and the like on image data obtained by each of the imaging devices 21, 22, and 23. The conversion parameter calculation unit 32 includes the
A conversion parameter for distortion correction is calculated from image data for calibration obtained by imaging an imaging target for image calibration for each of 1, 22, and 23. The conversion parameter storage unit 33 stores the calculated conversion parameters. Distortion cancellation conversion processing unit 3
4 performs a correction process for eliminating distortion on the image data for measurement based on the conversion parameters stored in the conversion parameter storage unit 33. The surface shape analysis processing unit 35 performs a surface shape analysis process based on the image data for measurement whose distortion has been eliminated.

Next, the image calibration processing and the surface shape measurement processing of this system will be described in detail. FIG. 4A is a flowchart for explaining the image calibration processing, and FIG. 4B is a flowchart for explaining the surface shape measurement processing. First, in the image calibration process, an image of a calibration target is captured (S1). For example, as shown in FIG. 5, an object to be calibrated includes a plurality of figures (circles in this example) 41 having a predetermined area that are separated from each other. However, the coordinate value of the figure center (center of gravity position) P of each figure 41 is known. This figure 4
The imaging target 42 including 1 is set at the position of the measurement target 6, and image data obtained from each of the imaging devices 21, 22, and 23 is captured (S2). Then, the captured image data is appropriately subjected to a filtering process, a feature extraction process, a center line detection process, and the like, to obtain the coordinates of the center line P, and the coordinate values of the obtained center line P and the original coordinate values are calculated. Conversion parameters for correcting the rotation, scale, translation, and distortion unique to each of the imaging devices 21, 22, and 23 are calculated from the displacement amounts as follows (S3).

[0017] That is, as shown in FIG. 6, the ideal coordinate position of FIG center P of each shape _{41 (x i, y i)} (i = 1~
N), assuming that the actually observed coordinate position is (x _i ′, y _i ′), the transformation formula including rotation, scale, translation, and distortion between them is the affine transformation formula shown in the following Expression 2. Can be used.

[0018]

(Equation 2)

Here, the conversion parameters a ₁ , a ₂ , b ₁ ,
b ₂ , c ₁ , and c ₂ can be calculated by obtaining the minimum value of the error E shown by the following equation (3) (least square method).
The obtained conversion parameters are stored in the conversion parameter storage unit 3.
3 is stored.

[0020]

(Equation 3)

On the other hand, at the time of measurement, as shown in FIG. 4 (b), an image of the object to be measured is taken by the image pickup devices 21, 22, 23 (S11), and three image data are taken in (S12). After appropriately performing a process such as feature extraction, a distortion elimination conversion process is executed (S13). The distortion elimination conversion process is executed for each image data using the conversion parameters stored in the conversion parameter storage unit 33. A surface shape analysis process is performed using the image data from which the distortion has been eliminated (S14).

In the above system, the possible conversion parameters include, for example, those shown in FIGS. 7 (a) and 7 (b).
As shown in FIGS. 4A and 4B, the parallel movement amounts L1 and L2 along the cross section A orthogonal to the optical axis C of the measurement optical path, and as shown in FIGS. Tilt amount θ1, θ
2. A rotational movement amount φ around the optical axis with reference to the cross section A as shown in FIG. 7E, and a focus position correction when the light beam is a non-collimated light beam as shown in FIG. Amount F and the like.

If there is such a so-called tilt or large rotational deviation, the interval between the centers P of the calibration imaging target 42 is determined, for example, in order to determine the up, down, left, and right directions of the imaging devices 21, 22, 23. As shown in FIG. 8, it is desirable to make the value a non-integer multiple. This makes it easier to check the direction of the mark in the coordinate system when the landmark plate or the like is imaged by the imaging devices 21, 22, and 23. However, the pitch of the figure core P does not need to be a non-integer multiple depending on how to give an irregular pitch of the figure core P of the calibration imaging target 42 or how to attach a mark (for example, to provide a direction mark).

As for the conversion parameters, for example, as shown in FIG. 9, the above-mentioned affine transformation or the like may be applied to each triangle formed by three adjacent cores P to obtain the conversion parameters. When the conversion parameters are calculated for each triangle in this way, the conversion processing within each triangle uses a linear conversion processing such as an affine transformation, but since the conversion parameters are different for each triangle, the nonlinear conversion processing is performed for the entire image. There is an advantage that a dynamic effect can be obtained.

The structural advantages of the interferometer according to the embodiment of the present invention described above are as follows. That is,
Japanese Patent Application Laid-Open No. 2-287107 discloses a method of extracting phase information corresponding to a difference in shape of a test surface from a reference surface while maintaining a polarization plane in an orthogonal relationship. For measuring distortion of a transmitted wavefront, a Mach-Zehnder interferometer is used. A Michelson-type interferometer is shown for surface shape measurement using reflected light. However, if a Michelson-type interferometer is used as an interferometer for measuring the surface shape, the light split into the reference surface side and the test surface side after passing through the polarizing semi-transparent mirror passes through independent optical paths, so that the measurement environment Is susceptible to fluctuations in In addition, there is also a disadvantage that it is not suitable for high-accuracy measurement because it is affected by the surface accuracy of the optical components that are interposed independently.

On the other hand, a Fizeau interferometer having a common optical path and a small number of optical parts is generally used as an interferometer for measuring a surface shape with high accuracy. The features of this interferometer are that it is hardly affected by fluctuations due to the common optical path type, and that it does not receive the surface accuracy of optical components other than the reference surface. However, since the Fizeau interferometer is an optical system that generates and interferes with the reference light and the measurement light on the same polarization plane, phase shift interference fringes are simultaneously generated using the polarizing plate as described above. It cannot be applied to systems that cause

On the other hand, in the above-described apparatus, 1 in FIG.
When passing through the 波長 wavelength plate 5, the light is polarized by 45 degrees, and the detection surface 6a
After being reflected by, the light again passes through the 波長 wavelength plate,
The reflected light from the reference surface 4 is orthogonal to the light by 90 degrees, and the light can be extracted independently without interfering with the common optical paths. As a result, compared to the apparatus of Japanese Patent Application Laid-Open No. 2-287107, which is shown as a two-dimensional information acquisition apparatus by a Michelson interferometer, it is less susceptible to disturbances such as air fluctuations, and the generation of phase shift interference fringes can be expected with high accuracy. And a system can be constructed with a small number of optical components.

[0028]

As described above, according to the present invention,
Rather than adjusting the position of each imaging system of a plurality of imaging units by a mechanical adjustment mechanism, a conversion parameter for correcting a coordinate shift or distortion of each imaging system from image data obtained by each imaging unit. Is calculated and the actual measurement image is corrected using this conversion parameter. Therefore, physical adjustment of each imaging system is not required, and the occurrence of measurement errors due to coordinate shift and distortion of the imaging system is effectively prevented. It has the effect that it can be done.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a Fizeau interferometer according to one embodiment of the present invention.

FIG. 2 is a diagram showing three interference fringe images captured by the interferometer.

FIG. 3 is a functional block diagram of an image measurement processing system in the interferometer.

FIG. 4 is a flowchart showing an image calibration process and an image measurement process of the image measurement processing system.

FIG. 5 is a diagram showing a calibration imaging target used in the image calibration process.

FIG. 6 is a diagram illustrating an actual coordinate system and an imaging coordinate system including distortion.

FIG. 7 is a diagram illustrating an example of conversion parameters for correcting a shift of an imaging coordinate system.

FIG. 8 is a diagram showing another example of a calibration imaging target.

FIG. 9 is a diagram for explaining another example of conversion parameter calculation.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Light source, 3, 7, 8 ... Non-polarization type half mirror, 4 ... Reference surface, 5, 16 ... 1/4 wavelength plate, 6 ... Measurement object, 9 ...
Total reflection mirror, 10 ... split optical system, 11, 12, 13 ...
Optical path, 14 1/2 wavelength plate, 15, 17, 18 polarizing plate, 21, 22, 23 imaging device, 30 computer, 31 image processing unit, 32 conversion parameter calculation unit
33: conversion parameter storage unit; 34: distortion elimination conversion processing unit; 35: surface shape analysis processing unit.

Continued on the front page (72) Inventor Kazuhiko Kawasaki 430 Kamiyokoba, Tsukuba-shi, Ibaraki F-term (reference) in Mitutoyo Co., Ltd. EE00 EE08 FF04 FF52 FF61 GG04 HH03 JJ03 JJ05 JJ19 JJ26 LL12 LL33 LL36 LL46 QQ18 QQ23 QQ31

Claims (7)

[Claims] 1. An image measurement system comprising: a plurality of imaging means for imaging a measurement target; and an image measurement processing means for measuring the measurement target from a plurality of image data obtained by the plurality of imaging means. The image measurement processing unit includes an image calibration unit for correcting a deviation of a coordinate system of each image data due to a spatial deviation of the plurality of imaging units, and the image calibration unit has a known coordinate value. The coordinate value of the point of interest on the image data obtained by imaging the calibration imaging target including a plurality of points of interest by each imaging unit, and the known coordinate value, the deviation of the coordinate system, Conversion parameter calculation means for calculating a conversion parameter for correction, conversion parameter storage means for storing the calculated conversion parameter, and the subject obtained by each of the imaging means Image measurement system, characterized in that the image data of the target is obtained and a correction means for correcting by the conversion parameter storing means in the storing transform parameters. 2. A means for generating optical interference fringes from measurement light from the surface of the object to be measured having different polarization planes and reference light from a reference surface; 2. The image measurement system according to claim 1, further comprising: a unit that shifts the phase by a predetermined amount with respect to each other, and simultaneously makes each of the plurality of imaging units incident on each of the imaging units. 3. A method for measuring a measurement target by using a plurality of image data obtained by imaging the measurement target with a plurality of imaging means, wherein a plurality of image data are captured by a plurality of image data. An image calibration method for correcting a displacement of a coordinate system of each image data caused by the calibration method, wherein a calibration imaging target including a plurality of points of interest whose coordinate values are known is imaged by each imaging unit, and obtained by this. From the coordinate values of the point of interest on the image data and the known coordinate values, a conversion parameter for correcting the displacement of the coordinate system is calculated, and the measurement target of the measurement target obtained by each of the imaging units is calculated. An image calibration method for an image measurement system, wherein image data is corrected using the calculated conversion parameter. 4. An object to be calibrated, which includes a plurality of figures each having a predetermined area separated from each other and having a known center coordinate value. 3. An image calibration method for the image measurement system according to 3. 5. The image calibration method according to claim 4, wherein the intervals between the centers are non-integer times. 6. The conversion parameter is a translation amount along a section orthogonal to the optical axis of the measurement optical path, an inclination amount of a normal line based on the section, and a rotational movement about the optical axis based on the section. The image calibration method according to any one of claims 3 to 5, wherein the amount is at least one of a focus position correction amount when the light beam is a non-collimated light beam. 7. The image calibration method according to claim 4, wherein the conversion parameter is calculated for each triangle formed by three adjacent figure cores.