JPS5940105A - Pattern measuring device - Google Patents

Abstract

PURPOSE:To determine the size of an object to be measured at a high speed, by calculating beforehand the correlation between the shadow of a known reference pattern and the objects to be measured of various sizes and comparing the numerical values and the results of measurement. CONSTITUTION:The pattern of a black stripe 32 has the pattern of the apertures of a shadow mask 15 as a reference pattern. When R, i.e., the ratio between the max. value and min value of the quantity of the incident light to a detector 18 is determined, the width B of the black stripe 32 is determined by using a precalculated R-B curve. A signal processing unit 21 consists basically of a CPU and an ROM. The CPU has the functions of changing the angle of a galvanomirror 13, assigning the address of the ROM by using the ratio R between the max. value and min. value in the section where the angle changes, after the change in the angle attains 1/3 of the aperture pitch (p) of the mask 15, fetching the data stored thereon and transmitting the same to a display device 22 and the function of controlling a moving device 17. The R-B curve is stored in the ROM. When the angle of the mirror 13 is known, the incident angle of the parallel luminous fluxes is assigned and to which stripe 32 the fluxes correspond is known.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field J to which the invention pertains] This invention relates to a pattern measuring device for measuring the dimensions of patterns that are regularly and repeatedly arranged.

[Technical background of the invention and its problems]

``Conventionally, as a device for measuring the dimensions of a pattern, there was a device that used the pattern to be measured as a lens, focused the image on an image pickup device such as a CCD, photoelectrically converted it, and measured the electrical signal. Measurement cannot be performed unless the measurement pattern y image is accurately focused on the image sensor, and an automatic focusing mechanism is required.Automatic focusing mechanisms are generally complex, require sophisticated equipment, and are expensive. Furthermore, when measuring patterns on three-dimensional curved surfaces, it was difficult to increase the measurement speed because it took time to focus.Also, in such measurements, measurement was performed point by point. When the edge of a pattern is irregularly disturbed, it is necessary to measure the slope at multiple points and find the average value. There was a problem.

Also, there is a pattern in which a pattern to be measured is irradiated with rape light, the transmitted light or reflected light is subjected to 7-lier transform using a lens, and the pattern width is measured according to the intensity of a specific frequency component of the Fourier transform pattern. Measuring device (% Gansho 55-27
321). This device had many advantages, such as no need for focusing on the pattern to be measured, and the average value of all patterns included within the beam diameter of the irradiated laser light could be immediately determined. However, when measuring optically inhomogeneous diffusers such as ground glass or patterns printed on opaque materials such as paper, the S/N ratio is low when determining the Fourier transform pattern, making measurement impossible. Ta.

Furthermore, if the object to be measured is limited to the measurement of a black stripe (registered trademark of Tokyo Shibaura Electric Co., Ltd.) pattern on a color Braum tube face plate, a narrowly focused laser beam is scanned over the black stripe and the transmitted light intensity is measured. There is a device that detects the signal using a detector and measures the time width of the signal. (Nonaka et al., Journal of the Society of Television Engineers 34 A2 pp.
, 141-146 (1980)) This device had the disadvantage that it was difficult to focus the laser beam sufficiently narrowly relative to the width of the black stripe, and errors were likely to occur when detecting the width of the electric signal and the shiver turn. Another drawback is that in order to focus the laser beam uniformly on the pattern to be measured, it must be set accurately at the center of curvature of the curved surface of the laser plate. Furthermore, since this device also measures point by point, it is necessary to measure multiple points when measuring irregular patterns that occur near the edges of straight stripes, which are the basic pattern of the black stripe pattern. It was necessary to perform averaging after that, and it also took a lot of time to measure the points. As described above, this device also had many problems.

[Purpose of the invention]

This invention was made in order to eliminate the above-mentioned drawbacks, and it does not require focusing and can directly measure the average pattern size of the measurement area at high speed. It is an object of the present invention to provide a pattern measuring device that can measure even patterns printed on non-uniform media or opaque objects, and can perform high-speed and highly accurate measurements.

[Summary of the invention]

In this invention, a light beam is irradiated onto 11 reference patterns. Furthermore, the pattern to be measured is irradiated with light that passes through this reference pattern and carries the reference pattern information. Next, a photoelectric conversion element placed behind the pattern to be measured converts the amount of light that has passed through the reference pattern and the pattern to be measured into an electrical signal. At this time, only the angle of the light incident on the reference pattern is changed using a galvanometer mirror, a lens, or the like. Then, the shadow of the reference pattern moves on the pattern to be measured. Therefore, an electric signal obtained by photoelectrically converting the amount of light that has passed through the reference pattern and the pattern to be measured becomes a signal representing a correlation function between the shadow of the reference pattern and the pattern to be measured.

Therefore, if the dimensions of this reference pattern are known and the shape of the pattern to be measured is unknown, it is possible to determine the dimensions of the pattern to be measured from the maximum and minimum values of this electrical signal. In other words, by calculating in advance the correlation function between the shadow of this known reference pattern and the measured pattern of each dimension, and comparing this value with the measurement results, it is possible to determine the size of the measured object. becomes.

〔Effect of the invention〕

In the present invention, the size is determined by the correlation between the shadow of the reference pattern and the pattern to be measured. Furthermore, in a system where the distance between the reference pattern and the pattern to be measured is kept substantially constant, this correlation is uniquely determined. Therefore, in a system where the distance between the reference pattern and the measured pattern is kept constant and all or part of the transmitted light is photoelectrically converted by the detector, the position of the illumination system and the position of the detector have little effect. Measurements can be made without being affected. That is, measurement without the need for focusing becomes possible. For example, even if the pattern is on a three-dimensional curved surface, such as the black stripe turned on the inner surface of the face plate of a color cathode ray tube, the reference number (turn) is kept at a constant distance from this curved surface. If you choose an object that is dripping, you can measure it without any focusing.

The correlation signal then calculates the average value of all (turns) within the illuminated area, so even if irregular disturbances occur at the edges of (for example) The average dimensions of the exact pattern can be measured.

Even if the pattern to be measured is patterned on an opaque object such as a diffuser plate or paper, and the transmitted light is diffused, the present invention can determine the maximum and minimum amount of transmitted light between the reference pattern and the pattern to be measured. Since it is determined only by the ratio of values, it is possible to measure without being influenced in any way. In this way, the present invention makes it possible to construct an extremely practical measuring device with many advantages.

[Embodiments of the invention]

Next, an embodiment of the present invention will be described with reference to the drawings. In this embodiment, a pattern of bracket stripes formed on the inner surface of a face plate of a color cathode ray tube is measured.

First, the black stripe will be briefly explained in order to better understand this embodiment. In a normal color television, the basic structure is a mask with a large number of pores (called apertures), that is, a mask with 3 holes formed corresponding to each aperture.
It has a colored luminous surface, and colors are selected by the angle of incidence of each of the three electron beams incident on the aperture of the shadow mask. In this embodiment, a slot mask having vertical rectangular holes is used as the shadow mask.

Such a shadow mask is attached to the face plate of the color cathode ray tube. The inner surface of the faceplate is provided with phosphor strips with a black light absorber or black stripe between each phosphor. At this time, if the widths of the phosphor strips are not equal, the color balance will be disrupted.

---1lii To provide black stripes and phosphor strips on the inner surface of this faceplate, proceed as follows.
Put on the shadow mask.

Next, an ultra-high pressure mercury lamp or the like is provided at one of the three points where the three electron beams are placed, and exposure is performed. This exposure is performed at three points, and black stripes are formed by development. As is clear from the manufacturing method, this black stripe has a period that is one third of the opening pitch of the apertures of the shadow mask.

In other words, the black stripe pattern is.

The aperture pattern of the shadow mask is used as a reference pattern.

After the black stripes are formed in this way, 3
Colored phosphor strips are also provided by photographic printing techniques. That is, after leaving the shadow mask again, a slurry containing one color of phosphor and a photosensitive binder is formed.
Spray it on the inner surface of the face plate, and after drying, apply it as a shadow mask. Then, exposure is performed using an ultra-high pressure mercury lamp or the like, and a phosphor pattern of the first color is formed at a predetermined position by development. Repeat this for the three colors R, G, and H.

Here, the coating is applied in the order of the black stripe and the phosphor strips from the inner surface of the faceplate, so by accurately providing the black stripes, the white balance can be maintained regardless of the coating accuracy of the phosphor strips. , it will be good. This is considered to be one of the advantages of black stripes.

As shown in FIG. 1, such a face plate pattern measuring device uses a laser beam (I) that emits a coherent laser beam, and a laser light source (I) that emits a coherent laser beam.
Collimating lenses (12a) and (12b) that expand the beam diameter of the laser light from ill and converge it again, and this collimating lens (12a).

A galvano mirror (+3) that changes the traveling direction of the laser beam from (12b), a lens (14) whose front focal point is the position where the D laser beam from this galvano mirror 03 converges, and a A shadow mask (IQ) as a reference pattern to which parallel ray bundles are irradiated, a face plate ae of a color cathode ray tube to which this shadow mask (19 is integrally attached, and this face plate) (a moving device surface for moving ill9, A detector that receives the transmitted light from the face plate 0e and converts it into an electrical signal, and an A-D converter that converts the analog signal from this detector (Ie or +11) into a digital signal.
a), a signal processing device ta that performs the processing described below on the digital signal from this A-Di converter (), a display device a3 that visualizes the results of this signal processing device Cυ, and a signal processing device &I) It is controlled by a galvanometer mirror (a drive circuit 1231 that drives a mist and changes the reflection angle of the laser beam).

In this device, the inner surface of the face plate 1161 is irradiated with a parallel beam only through the aperture of the 7-Yado mask O0. That is, face plate 01
A silhouette of a shadow mask θ9 is projected onto the inner surface of the mask 31.

This situation will be explained together with the principle of this invention.

The laser light emitted from the laser light source (1υ) passes through the collimator lenses (12a), (12b), the galvano mirror (1 row), and the galvano mirror (1 row) on the shadow mask 09 which is the reference pattern. The mirror surface of the moat almost forms an image.However, it is not necessary that there is a strict image formation relationship.

The parallel beam from the lens 0 passes through the shadow mask 119 and enters the face plate (151) on which the black stripe is formed.At this time, the light passing through the face plate 151 is converted into an electrical signal. This electrical signal is a correlation signal between the shadow pattern of the shadow mask (19) and the black stripe.

In this state, when the galvanometer mirror (131) is vibrated,
The incident angle of the parallel beam flux to the shadow mask (151) changes.Then, the correlation signal also changes, and from this change, the dimension of the black stripe, which is the pattern to be measured, is measured.

Further, a detailed explanation will be given using FIG. 2. A shadow mask a5 in which the parallel line flux Gl) having a diameter D = 2 to 3φ has an aperture pitch, that is, a slot center spacing p = 750 μm.
incident on . The center spacing of the phosphor strips L=250 μm. For convenience of explanation, the parallel line flux Gυ is the shadow mask t+
51 and face grating ae perpendicularly.

At this time, of the parallel beam 6υ, only the beam 01 that has passed through the slot of the shadow mask 09 is the face plate) (+
It is incident on the inner surface of 61. Of the incident light, the light irradiated on the black stripe 02 does not pass through the face plate (IG. Only the light irradiated on the phosphor strips provided between the black stripes 04 passes through the face plate) (I[9
and reaches the detector 111. Further, a rough diffusion surface (16a) is provided on the inner surface of the face plate ae. This diffusion surface (16a) is a 4-face spray)
(It is provided so that the inside cannot be seen from the surface of IG+. It is preferable that the inside cannot be seen by the viewer, and furthermore, the coloring state is favorable due to the diffusion of light. Also, this Utilizing the rough surface of the diffusion surface (16a),
It also makes it difficult for slurry etc. to peel off. Because of the presence of such a diffusing surface (16a), it is not possible to perform measurements using coherent light and Fourier diffraction. As can be seen from the overall description, this invention is characterized in that it does not use Fourier diffraction, but rather uses a correlation function between two patterns.

In addition, as will be described later, in this example, the ratio of the maximum value and the minimum value of the amount of transmitted light is calculated, so the diffusing surface (16a)
can be measured without being affected by the existence of

Now, to qualitatively explain the change in the amount of transmitted light, the luminous flux (to)
When it covers the black stripe G2, it becomes the minimum.

Conversely, the luminous flux is at its maximum when it is incident on the phosphor strips.

Furthermore, it should be noted that, as shown in FIG. In other words, the average value is measured for a plurality of black stripes (17J) rather than for individual black stripes (boils).

Keeping the above points in mind, we will conduct a detailed analytical discussion. now,
Let f(x) be the shadow mask (151 shadow, that is, the Fresnel diffraction pattern of shadow mask a9), and let the black stripe pattern be g(x) (indicating the distribution of light-transmitting areas). , face plate) passes through the IJIlG and reaches the detector (country).
T) becomes . However, a=D/2 (D is the diameter of the parallel line bundle 0D incident on the shadow mask (one) as described above.

), τ is the shadow mask lIS as shown in FIG.
Parallel ray bundle C incident from the perpendicular direction to the face plate (projected onto the inner surface for 1 second by ) (where τ is the distance traveled from the turn to the shadow mask
tanθ ・・・・・・・・・(2). For the following discussion, we will use the shadow mask (the amount of light when a parallel ray bundle Gυ is incident on one from its normal direction,
■ Normalize ■(τ) by (0) and define this function as φ
(τ), then . Furthermore, since the beam diameter of the parallel ray bundle is sufficiently larger than the pitch p of the shadow mask (15 slots), φ(T) = ...
...(4). φ(τ) may be a ratio of light amounts or a ratio of output signals from the detector (1ε). Also, this φ(τ) does not depend on the beam size (diameter) of the illumination light.

Next, as described above, g(2) representing the pattern of 0 black stripes has a cycle that is one third of the aperture pitch p of the shadow mask α9, and is a binary function of 0 and 1. On the other hand, s f(x) is a Fresnel diffraction pattern, as described above. Taking the above into consideration, when formula (4) is actually calculated, the result is shown in FIG. 3.

It is clear that qualitatively the output from detector ll81 depends on the width of the black stripe trowel. For example, as shown in FIG.
6) The inner black stripe (the edge where the threat is located, the output from the detector (1 second) is minimum.

Conversely, as shown in FIG.
When the center of the parallel beam bundle (1) is located in the area where the sides are coated and the phosphor is coated, the output from the detector (vessel) is maximum. It can be seen from a simple qualitative discussion that the maximum and minimum values are influenced by the center position of the black stripe and the spacing of the black stripe countries.

We will treat this quantitatively. The shadow mask (151) Fresnel diffraction pattern on the inner surface of the face plate (te is
Although it is analytically very complex, it was found that it can be approximated very well using trigonometric functions. For example, as shown in FIG. 6, when the value of q mentioned above is set appropriately, a waveform shown by a solid line U is obtained. When this waveform is approximated by the dotted line - shown in FIG. 6, there is little error in this embodiment.

Therefore, the Fresnel diffraction pattern f (x) is approximated by a trigonometric function, for example, l+amaX. Furthermore, when formula (4) is calculated when the positional relationship is as described above, and the ratio between the maximum value and the minimum value is actually calculated, it becomes as follows. However, a is the aperture size of the shadow mask with zero average, and α is a coefficient when the Fresnel diffraction pattern is approximated by a trigonometric function. When q is around l Q l1m, α=
At 0.768 degrees, there was the best agreement with the calculation using the Fresnel diffraction pattern. This trigonometric function can also be considered as one component of the Fresnel diffraction pattern.

Normally, before measuring the black stripe (2) pattern, the inspection of the shadow mask (15) has been completed and the Sekiguchi size a of the shadow mask u9 is known. Therefore, in equation (5), α + a + P
+ is known.

Therefore, when B is taken as an independent variable and R is illustrated, it becomes as shown in FIG. 7. This diagram contains more useful information. That is, when R, which is the ratio of the maximum value and minimum value of φ(τ), is determined, the width B of the black stripe 0 can be determined using FIG.
will be found.

The above is the basic aspect of this embodiment. Furthermore, the galvanic chisel 9-! shown in FIG. +3) is slightly rotated, the shadow mask f1 of the parallel line bundle is created as shown in FIG.
51 changes, and the parallel ray bundle G)3 moves in a direction inclined by θ from the normal direction of the shadow mask u9. Accordingly, the amount of light reaching the shadow mask also changes, and as shown in FIG. 3, the output waveform of the detector ag also changes along the increasing direction of τ.

The maximum and minimum values of the first three sets then give a width of 2 and 6 blacksts containing red, green, and blue phosphor strips. For example, from the ratio of the value 11° at point ff1) and the value ■2 at point @, the width of the black stripe G2 corresponding to red is the difference between the value 3 at point ff1 and the value ■4 at point (74). From the ratio, the width of 0 black stripes corresponding to green is
The width of the black stripe C33 corresponding to blue can be determined from the ratio of the value 5 at point (79) and the value 6 at point (761). , Face Play) (On the +1 inner surface, phosphor stripes and black stripes C3 are visible when forming.

Furthermore, Shadow Mask a! Diffraction of 19) (The movement amount τ of the turn is expressed by the angle of incidence θ of the parallel ray bundle on the shadow mask (151) according to equation (2). Therefore, φ(τ) shown in FIG. 3 is expressed as θ. expressed as φψ).

Since it is expressed as φ (θ) in this way, as will be described later, the signal processing device C29 and drive circuit Q3 determine which black stripe (34 You can see if it is compatible.

The measurement principle in this embodiment has been explained above, and next, the configuration for realizing this will be explained in detail.

As shown in FIG. 1, the light from the face plate 6Q is input as a digital signal to the signal processing device Qυ via a detector (111, an amplifier tll, and an A-D converter).

The signal processing device υ is basically a CPU (Central
processing unit) and ROM (Re
ad only memory). The functions of the CPU are to detect the maximum value and minimum value each time a signal is input from the A-D converter i'2cJ, and to detect the maximum value and minimum value each time the detection of the maximum value and minimum value is completed. ), and the function of changing the angle of the galvano mirror (131) via the drive circuit 14 every time the detection of the maximum value and the maximum /J% value is completed. After this angle change reaches one-third of the aperture pitch p of the shadow mask (Is), a function that calculates the ratio of the maximum value and minimum value in this section, and a function that uses this ratio to calculate the head address. A function for specifying, a function for extracting data stored at the specified address in this head and sending it to the display device (2'IJ), and a function for moving the data (17).
).

Here, the R-B curve shown in FIG. 7 is stored in the ROMIC. However, as mentioned above, the address specification is
This is done by the ratio R between the maximum value and the minimum value.

Furthermore, once the angle of the galvano mirror (19) is known, the angle of incidence of the parallel beam flux to the shadow mask holder 9 is designated.Then, as described above, it is known which black stripe (34) corresponds to.

The moving device αn moves the face plate (149) to which the shadow mask α9 is attached to change the measurement position.Normally, the measurement does not need to be performed on the entire inner surface of the face plate aQ, but at 10 points. It is sufficient that this moving device ([n) also has the ability to perform position settings to that extent.

The CPU controls the moving device (17) and the drive circuit (substrate), so that the display device (function) can display not only information on the blank drive (not only the width of the blanks but also the measurement position and information). Black stripes can be given different types of characters, and these can be displayed.

Father, this example uses laser light, so
Since the amount of light incident on the detector 1181 is large, it is resistant to external light. In order to operate more stably against external light, @1
In the figure, a dye filter or an interference filter that allows only the laser beam to pass may be used in front of the detector fl&. In FIG. 1, measurements can be made in the same way by using an incoherent light source such as a white ball instead of the laser light #, Gl).

Furthermore, the light beams incident on the shadow mask (15) do not have to be completely parallel.

[Other embodiments of the invention]

FIG. 8 shows another embodiment of the present invention. The embodiment is basically the same as the embodiment shown in FIG. This made measurement possible. That is, the deflection galvanometer mirror (13a) is placed near the center of curvature of the curved surface of the cathode ray tube face plate tL1. Furthermore, this galvano mirror (13a) is rotated by a motor (8υ),
This enables dimensional scanning. (23a) is a drive circuit that controls the galvanometer mirror (13a) and the motor [F]υ with signals from the signal processing device 01). Next, the light is detected by a detector (
[8] Collect light. At this time, since there is a possibility that external light is also collected, only the laser light is received using an interference filter (c). By scanning the laser beam with the galvanometer mirror (13a) in this manner, it is possible to measure the entire surface of the object while keeping it fixed. The other operations are similar to the embodiment shown in FIG.

[Other embodiments of the invention]

FIG. 9 shows another embodiment of the present invention. This example is also the first
This embodiment is essentially the same as the illustrated embodiment, except that multiple light sources are used as a means of changing the direction of light incident on the shadow mask. In other words, as a light source, for example, LE
As shown in FIG. 9, light sources 19υ, +92. (By arranging 931 and emitting light using a switching drive circuit (94) that switches and drives each of them, the direction of the light incident on the shadow mask is changed.By doing this, it is possible to perform measurements at extremely high speed. The rest is exactly the same as the embodiment shown in Fig. 1.If the light emitting elements are arranged not one-dimensionally but as shown in Fig. 10, a black matrix will be formed instead of a black stripe. It is also possible to measure the aperture diameter of the shadow mask.In addition, using a large number of light sources instead of three makes it possible to perform stable measurements against relative angular deviations between the light sources and the shadow mask.Also, as shown in FIG. Furthermore, instead of moving the light source, an optical fiber may be used, and measurements can be made in the same way by using an optical fiber and moving the optical fiber.

In the above several embodiments, the width measurement of a black stripe patterned on the inner surface of a color cathode ray tube has been described, but the present invention is not limited to this. For example, if a reference pattern is prepared in the measurement of a pattern printed on a ceramic substrate, it becomes possible to perform the measurement in the same manner as in this embodiment. It is also possible to measure the pitch of black stripes.

Further, since the present invention requires a superimposed signal of the reference pattern and the pattern to be measured, it is natural that not only the light passing through the pattern to be measured but also the reflected light may be used. As described above, it is natural that any modifications are included in the present invention as long as they do not depart from the spirit of the invention.

[Brief explanation of the drawing]

FIGS. 1 to 7 show one embodiment, with FIG. 1 showing the overall configuration, and FIGS. 2 to M7. aυ... Laser light source (13... Galvano mirror 11... Lens (15... Shadow mask + t5... Face plate 118)... Detector ell)... Signal processing device agent Patent attorney Noriyuki Chika (TJ or 1 person) Figure 1 Figure 2 Figure 3 Figure 7 Figure 4 Figure 5 Figure 6 Figure 8-l Go □ ■ 31- Figure 9 if Figure 10 2 Mu

Claims (1)

[Claims] (17) a light source, a reference pattern on which the light beam from the light source is irradiated, an optical system for changing the direction of incidence of the light beam on the reference pattern, and a reference pattern provided at a predetermined distance from the reference pattern; A pattern to be measured that has a significant correlation with the pattern, a detector that converts the amount of superimposed light between the pattern to be measured and the reference pattern into an electrical signal, and the maximum and minimum values of the electrical signal from this detector are used. A pattern measuring device comprising: a signal processing device that calculates dimensions of a pattern to be measured.