JPS6221046A - Defect inspection for shadow mask - Google Patents

Abstract

PURPOSE:To achieve a higher inspection accuracy, reliability and efficiency, by arranging the image pickup operation in an image pickup system to be performed several times on varied conditions to identify a defect on the basis of the results of detection at each of the operations. CONSTITUTION:A shadow mask 6 is irradiated by a transmission light section composed of an incandenscent lamp 8 to be lighted by a DC power source 9 and a diffusion plate 7 and the area to be inspected is taken with a TV 1. Then, the output signal of a camera 1 is converted from analog into digital with an image processor 2 to make a digital image data, and various image processings thereof are performed at a high speed with a frame memory and an arithmetic unit including addition and subtraction of a screen. A control section3 controls a shadow mask moving mechanism made up of the unit 2, an X-Y state 10 and a drive section 5. When the image data thus processed is observed with a TV monitor 4, the brightness changes locally at a defect alone, thereby facilitating the recognition of the defect and also permitting the identification of the type thereof (by white and black points) depending on the sequence of the inversion between the brightness and the darkness toward the perimeter at the defect.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a defect inspection method for shadow masks used in color cathode ray tubes and the like, and in particular to an optical inspection method that can automatically inspect shadow masks. related to

[Background of the invention]

Traditionally, shadow mask defect inspections have been carried out visually using a ridge eye or a microscope, but this requires a large amount of manpower to inspect a large number of products, and sensory testing is also difficult. Therefore, there was a problem of poor inspection accuracy and reliability.

In order to solve these problems, for industrial products with periodic patterns arranged at equal pitches, such as shadow masks, images can be obtained using microscopic imaging methods that can sufficiently resolve the arrangement units and defect shapes. Methods of detecting and inspecting defects have been proposed, such as by examining the video signal of a defect-free pattern and performing pattern recognition, or by comparing the signal with a signal obtained by similarly photographing a pattern without defects, and some methods have not yet been implemented. However, such methods require mechanical precision depending on the size of the defect to be detected.B.
In order to detect minute defects, high-precision equipment is required, so the equipment is expensive, and since the inspection requires microscopic imaging, the screen size that can be processed at one time is small. There are problems in that it takes a lot of time to inspect the entire pattern.

Furthermore, for products with periodic apertures such as shadow masks, an optical Fourier transform spatial filtering method has been proposed that takes advantage of the light diffraction phenomenon caused by periodic patterns when irradiated with coherent light. B. Although it has excellent inspection speed and detection sensitivity, it requires creating a spatial filter for each pattern to be inspected and requires a precise optical system, making the device expensive.
Furthermore, although defects can be detected, there are many problems such as the inability to determine the size relationship of the defect aperture with respect to a reference value, and a new inspection method has been desired.

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to provide an inspection method capable of solving the above-mentioned problems, inspecting a shadow mask efficiently and with high precision, and discriminating the type of defect.

[Summary of the invention]

In order to achieve this objective, the present invention provides a video signal that contains defect information even if the unit pattern of the shadow mask is not sufficiently resolved or is not resolved at all in a large-area photographic field of view. This is due to the effect of random noise reduction on the video signal by screen addition processing, which is an image processing method, and when the shadow mask to be inspected is moved in a predetermined manner, the video signal of the defect-free portion is almost the same before and after the shadow mask is moved. By subtracting the image data before and after the movement without any change, changes in the video signal due to the unit pattern in the defect-free area and changes in the video signal due to shading in the imaging system and dust in the optical system are erased, and the image data in the defect-free area is removed. It was found that the image data of B was approximately O and the value changed locally as the defect area entered, and the defect was detected based on this, and at the same time, the defect detection requirements specific to shadow masks were satisfied. What are the imaging conditions? The feature is that the final judgment is made after obtaining the defect detection results multiple times.

[Embodiments of the invention] Hereinafter, the present invention will be described in detail based on Examples.

FIG. 1 shows an embodiment of the shadow mask inspection method according to the present invention, in which periodic openings 11? : This shows a method when applying an abnormality in the aperture area of a shadow mask held as a unit pattern to an example of inspection-f.
A transmission illumination section consisting of an incandescent lamp 8 and a diffuser plate 7, which are turned on by a current power source 9, illuminates the shadow mask 6 for the test amount, and the left camera of the TV photographs the inspection area. The image processing device 2 converts the output signal of the TV camera (A/1) into digital image data, and performs various image processing including addition and subtraction on the screen at high speed using a frame memory and an arithmetic unit. 3 is a control unit that includes an image processing device 2 and an XY stage 1.
11f controls a shadow mask moving mechanism composed of a drive unit 5 and a drive unit 5.

Note that in FIG. 2, 12 and 13 represent apertures with defects.

Next, the operation of detecting defects in the shadow mask 6 according to this embodiment will be explained. In order to simplify the explanation, here, we will use shooting conditions in which changes due to the unit aperture of the video signal by the TV left camera can be ignored, for example, shooting conditions in which about 10 unit apertures 11 are included in the shadow mask area corresponding to 1 pixel f. As a condition, move the shadow mask,
A case will be described BAf in which the direction of displacement is the scanning line direction of the TV left camera and the displacement distance of the shadow mask is an integral multiple of the pixel pitch.

First, FIG. 3(a) shows the light transmission weight distribution on a straight line passing through the defective part of the shadow mask, and 14 shows the aperture area larger than that of the normal aperture 11, as shown by the aperture 13 in FIG. 15 shows a change in light transmittance due to a large defect (hereinafter referred to as a white dot), and 15 shows a change in light transmittance due to a defect (hereinafter referred to as a black dot) whose opening area is smaller than the normal aperture 11, as shown in the aperture 12 in FIG. shows.

Next, FIG. 3 fbl is a diagram showing a video signal scanned on the same line as FIG.
Gradual signal changes (shading) due to sensitivity unevenness on the imaging surface, random noise generated in the video signal processing circuit, and local signal changes 16 due to dust attached to the optical system are not shown.

In addition, Fig. 3 tel is a diagram showing the result of screen addition processing by the image processing device 2, and the ratio of the random noise component in Fig. 3 tb> is reduced to l/FW when the number of additions is 7N. Indicates that

Furthermore, Figure 3 + d) shows the result of displacing the shadow mask and not performing screen addition processing, and the signal due to the defect on the shadow mask also moves as the shadow mask moves, but the shading of the imaging system and the optical system This shows that the position of the signal 16 has not changed due to dust or the like.

And, Figure 3 [e) is from Figure 3 [c] to Figure 3 +
dl 17) This shows the result of image data subtraction.The shadow mask is created by eliminating components that do not change due to the shading included in the data in Figure 3 (fcl, fd) and the movement of the shadow mask as shown in 16. Only the signal due to the change in light transmittance and the reduced random noise component remain, and as a result, the signal due to the defect is the difference in value from the average value in the vicinity at a position separated by the number of pixels according to the amount of movement of the shadow mask. It can be seen that the signs are almost the same, the signs are reversed, and the order of reversal is reversed depending on the type of defect (white dot, black dot).

The above describes an example of subtracting between two screen image data that have been subjected to screen addition processing before and after shadow mask displacement, but the screen data of the amount of movement is the same as addition from the addition data before movement. Even if the number of frames is subtracted, the result is exactly the same, and processing can be performed using only one frame memory.

Therefore, if you observe the image data that has been processed above on the TV monitor 4, you can easily recognize the defect because the brightness of the defective area changes locally. The type of defect (white dot, black dot) can be identified by the order of inversion.

In addition, from this image data, subtraction of neighborhood average values or differential processing, especially when the pattern movement is parallel movement, image processing that calculates the difference between corresponding pixels before and after the movement of one point on the pattern is performed. When this is done, as shown in FIG. 3(f), gradual changing components of the image data are removed, and defects can be automatically detected by comparison with a predetermined darkness value.

By the way, 1 and above explain how to erase signals other than those caused by defects by adding, pattern movement, and subtraction of screen data, but depending on the shooting conditions, changes in the video signal due to unit patterns may not be negligible. As explained above, if the pixel pitch is moved by an integral multiple, the unit pattern signal remains in the image data after addition and subtraction, as shown in FIG. 4(b), and minute defects cannot be detected. Note that FIG. 4 (ta) is the data before subtraction.

However, in such a case, for example, for a shadow mask in which the apertures are arranged at equal pitches, set the direction of kt If so, the image data from each defect-free aperture does not change before and after the movement, and therefore, as shown in FIGS. 5(a) and (b), the image data after addition and subtraction is as shown in FIG. 3(e). Results similar to those shown are obtained, making it possible to detect minute defects.

Next, the moving distance of the shadow mask will be explained. 6th
The figure is a diagram explaining the relationship between the optical image of the defect opening 17 and 18, the pixel P, and the relationship between the image data D after addition and subtraction.
In a), the amount of movement of the shadow mask is an integral multiple of the pixel pitch,
Figure 6 (shows an example when bl is not an integer multiple. And,
In FIG. 6(a), the image data D of the defective part is vertically symmetrical with respect to the neighborhood average value, but if it is not an integral multiple, then as shown in FIG.
As shown in the example b), since the rate at which the change in image data D due to the defective image is distributed to adjacent pixels differs before and after the movement, the vertical symmetry with respect to the neighborhood average value is lost, and automatic detection processing is performed. This will cause an error in the above.

Therefore, when moving the shadow mask, it is preferable that the positional relationship between the defect image and the pixel receiving it is the same before and after the movement.For example, when moving in the pixel arrangement direction, an integral multiple of 5 pixel pitch. You will get better results if you set it to . Therefore, if the change in the video signal due to a unit aperture cannot be ignored, if the moving distance is a common multiple of the aperture array pitch and the pixel pitch, image data can be erased by the defect-free apertures mentioned above, and the moving distance can be set to an integer of the pixel pitch. Both conditions of doubling can be satisfied.

By the way, the movement of the shadow mask required in one or more of the inspection methods of the present invention requires erasing information about defect-free openings from the image data after addition and subtraction, and under this condition, if possible, defects can be removed. This means that better results should be obtained if the positional relationship between the image and the pixel that receives it matches before and after the movement. Therefore, as long as these conditions are satisfied, the direction of movement does not necessarily have to coincide with the pixel arrangement direction.

Next, as an embodiment of the present invention, a method for reducing fluctuations in defect signal level caused by the positional relationship between a defect image and a pixel will be described.

FIG. 7(a) shows a state in which the image of the defective opening 17 is located at the center of one pixel P, and FIG. 7(b) shows a state in which the image of the defective opening 17 is located on the contact points of four pixels P. In this way, even if the defect is the same, there are cases where most of the signal change due to the defect is concentrated in one pixel as in Figure 7 (al), and cases where the signal change due to the defect is distributed over four pixels (Figure 7 Fb). In this case, a difference of approximately four times occurs in the defect signal level, resulting in a problem of low reproducibility of defect detection. However, most of the dispersion of such defect signals to surrounding pixels is within a 3 x 3 pixel area centered on the defect image, and the influence on the outside can be ignored, so image processing Using spatial filter processing, which is often used in It is possible to reduce changes in

Next, as yet another embodiment of the present invention, a method for automatically detecting defects and determining the type ρ will be described.

In FIG. 8(a), the direction in which the shadow mask is displaced is the same as the pixel arrangement direction, the moving pressure-distance is set to twice the pixel pitch, and after performing screen addition processing, the shadow mask is moved to the right in the figure.
This is an example of image data resulting from subtraction using the same number of frames as addition. 20 indicates a white spot defect, 21 indicates a local change in image data due to a black spot defect, and the other parts are unit patterns. It represents a change in image data (low frequency change) in response to gradual changes in array pitch, size, etc.

Here, let's take a closer look at the image data of the defective part.

Pixel A (22, 22'), which corresponds to the defect during screen addition, and pixel B (23, 23'), which corresponds to the screen during subtraction, are pixels that have defect information, and the displacement distance of the shadow mask, that is, the pixel corresponding to 2 pixel pitches. Appears several (2 pixels) apart,
If IC is the neighborhood average value of pixel C (24, 24') corresponding to the midpoint of each pixel data IA and IB, then IA
-I. and I, -I.

is a value with opposite sign and almost the same absolute value as B, and the order of this sign corresponds to the type of defect (white dot, black dot), IA-I,
It can be seen that the absolute value of is proportional to the size of the defect. Also,
It can also be seen that the difference between the average value of IA and IB and IC is a sufficiently small ratio with respect to the value of I, -I. Therefore, image data processing such as IA-I, for example, for FIG.
When I(t+t) is calculated, B and ■(j-1)-I(j+1) become IA and I, respectively, as shown in Figure 8+b).
The data indicating the type and size of the defect for the pixel (25, 25') B is shown as B. Also, since low frequency changes are also reduced, the defect can be detected by comparing it with a certain threshold. It becomes possible to determine the type.

However, at this time, as shown in FIG. 8(b), the signs on both sides of the pixel to be detected as a defect are opposite, and the value is IA.
-Since half the data (26, 26') of IB is generated, false defects t will also be detected for defects with a signal level more than twice the threshold value. This causes inconvenience in the judgment of

Therefore, in the above IA, I, IC, ("A+"B)/
2- "c" should not be less than a certain ratio with respect to IA-1B, for example, "ft" with respect to Fig. 8 [a]
+=(” (J −s )+I(J+□))/2−bird)
(However, ri′j) is the neighborhood average value of I(1))
ff is calculated to obtain FIG. 8(c), and compared with a predetermined threshold value, among the defect detection pixels including false defects, the above (step □+I,)/
2-I.

By selecting pixels for which the value is less than a predetermined value, the false defects can be excluded, and stable defect detection and type determination can be performed.

By the way, the above invention is the same as the invention already filed as Japanese Patent Application No. 1983-96593, and therefore, defects in the shadow mask due to irregularities in the shape of the opening in the horizontal projection plane are sufficiently covered. Good detection results can be obtained.

However, in the shadow mask targeted by the present invention, not only the shape of the opening on the horizontal projection plane;
Abnormalities in the cross-sectional shape of the opening and scratches on the surface other than the opening are detected as defect detection target B, and foreign matter such as dust attached to the surface of the opening is detected as another defect. It is required to distinguish between melt and detect.
Therefore, with the above conditions, it is still not possible to obtain sufficient detection results.

Therefore, in the present invention, as described below, defect detection is performed multiple times under different imaging conditions, and the defect is finally determined based on the results.

That is, the present invention is a method for detecting a local change in the brightness of a shadow mask as seen from the lens of an imaging device as a defect when imaging a shadow mask. If the shooting conditions are selected appropriately, it should be possible to detect various defects.
This is what I used.

First, regarding the illumination method, if transmitted light illumination (transmitted bright field illumination) as shown in the first figure is used, a defective aperture with a large opening area such as 13 in the second figure is treated as a white spot and the second
Defect openings with small opening areas such as No. 12 shown in the figure and No. 9
A foreign object located in an obstructing position such as 32 in the figure can be detected as a black spot, and if reflected illumination light (reflected dark-field illumination) as shown in FIG.
It is possible to detect pits, scratches, and foreign objects on the surface of the shadow mask as shown in Figure 33, and even detect foreign objects that block the aperture as shown in Figure 9 (using transmitted dark-field illumination light as shown in Figure 12). 32 can be detected as a white point.

Then, defects were detected using different illumination methods for each of these, and nine types of detected defect positions (white spots, white spots, etc.) were detected.

By examining data such as black dots) and signal levels, it becomes possible to identify the type of defect in more detail.

For example, among the defects detected by transmitted bright-field illumination as described above, defects detected as spots (black spots) darker than the surroundings include defects with a small opening area like 12 in FIG.
There are defects caused by foreign matter as shown in 32 in the figure, but the first
As shown in Figure 2, when defects are detected using transmitted dark field illumination,
For example, a defect with a small opening area like 12 in Figure 2 has almost no sensitivity, whereas a foreign object like 32 in Figure 9 has the property of scattering illumination light. If a white spot is detected with transmitted dark-field illumination at the same position as the defect detected as a black spot with transmitted bright-field illumination, it can be determined that this defect is a foreign object. It is possible to distinguish between defects with small opening areas and foreign objects.

Next, regarding the photographing angle and viewing angle, as mentioned above, the method of the present invention is a method of detecting defects as local changes in brightness as seen from the photographing lens under set photographing conditions,
Therefore, in the case of transmitted light illumination, only a defect in the aperture shape of the shadow mask as seen from the photographing lens is detected as a defect.

Therefore, if the photographing method is set to tilt photographing (a type of oblique photographing) as shown in Fig. 13, the cross-sectional shape of the aperture that cannot be detected when photographing from the vertical direction as shown at 34 in Fig. 11 can be avoided. Since it becomes possible to detect defects, defect detection is performed multiple times from at least two or more directions using the center line N of this tilted photograph in Fig. 13 as the rotation axis, and in any of these cases. If no defect is detected in the shadow mask, it can be determined that the defect 34 shown in FIG. 11 is not present in the shadow mask. Note that in this tilted photography, as is clear from FIG. However, general oblique photography may be used in which the optical axis of the lens 31 and the center of the imaging surface of the TV camera 1 are kept aligned.

By the way, in the shadow mask, as shown in FIG. 14, the shadow mask 6 is also formed into a curved surface such as a part of a cylindrical surface or a part of a spherical surface according to the shape of the phosphor screen of the cathode ray tube 30, or is supported by a support member. In most cases, it is retained.

Therefore, as shown in FIGS. 14 and 15, when setting the viewing angle of the TV camera 1, the incident angle θ of the electron beam with respect to the shadow mask in the cathode ray tube and the angle θ′ of the illumination light contributing to photography are determined. If the angle of view is set so that they match or are close in amount, it will be possible to prevent defects in the shape of the aperture at each point on the shadow mask as seen from the direction in which the electron beam passes, which is the cause of defects in images displayed on color TV cathode ray tubes. Therefore, defective openings in the shadow mask can be detected most appropriately.

Next, when showing the flJ in which the defect inspection of the shadow mask was performed according to the above example, the diameter of the unit opening is 100 μm.
A shadow mask having an arrangement pitch of 300 μm as shown in FIG. 2 is illuminated with transmitted light as shown in FIG. The 'rV camera 1 used was 300X220
When photographing the area (2), the total time of photographing and image processing is only about 5 seconds, and the diameter of the unit aperture is 10 μm.
It was possible to automatically detect defects that differ by m, and even determine the size of the opening relative to its surroundings.

〔Effect of the invention〕

As explained above, according to the present invention, minute defects in a shadow mask are automatically detected by photographing in a wide field of view in which the unit aperture cannot be resolved, and processing the image data. (white dots, black dots).

Furthermore, according to the present invention, since the detection sensitivity is less dependent on the resolution of the imaging system, shadow masks formed or held in three-dimensional shapes such as spherical or cylindrical shapes can be easily inspected. It can be easily automated even for a wide range of shadow mask defects, improving inspection accuracy and
Effects such as improved reliability and efficiency can be obtained.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of the periodic pattern inspection method according to the present invention; FIG. 2 is an explanatory diagram showing an example of the periodic pattern; FIGS. Figure (a),
[b), Figure 5 ia), (b), Figure 6 1aL +
b+, Fig. 7 1a), lb), and Fig. 8(a)~
ic) is an explanatory diagram showing the operation of the present invention, Fig. 9 is an explanatory diagram of foreign matter attached to a shadow mask, Fig. 10 is an explanatory diagram of reflective dark field illumination, and Fig. 11 is an explanatory diagram of various defects appearing on a shadow mask. An explanatory diagram, Fig. 12 is an explanatory diagram of transmitted dark field illumination, Fig. 13 is an explanatory diagram of tilted photography, Fig. 14 is an explanatory diagram showing the state of the shadow mask in the cathode ray tube,
FIG. 15 is an explanatory diagram of the angle of view during photographing. . 1...TV camera, 2...Image processing device, 3...Control unit, 4...TV monitor, 5...Drive mechanism , 6... Shadow mask, 7... Diffusion plate, 8... Lamp,
9...DC power supply, 10... Stage,
11...Unit aperture, 12.13...Defect, 14.15...Transmittance change due to defect, 1
6...Signal change due to dust etc. in optical system, 17.
...Defect image, 27...Threshold value, 28...
... Neighborhood average value, 30 ... Braun tube, 31
... Photographing lens, 32 ... Foreign object, 33
······pit. /N=nー;' Agent Patent Attorney Takeshi Junji Department (and 1 other person, 'Figure 1 Figure 2 ○ ○ OO Figure 3 Figure 4 Figure 5 l8rjA I

Claims (14)

[Claims] (1) Move the shadow mask to be inspected by a predetermined distance in a predetermined direction within the field of view of the video imaging system, and use one frame of first image data from the video signal obtained by capturing the image before the movement. In a shadow mask defect inspection method in which a defect detection process is performed based on third image data obtained by subtracting one frame of second image data from a video signal obtained by later imaging, the video imaging system A defect inspection method for a shadow mask, characterized in that the imaging operation is performed as a plurality of imaging operations under different imaging conditions, and the defect is determined based on the detection results of each of the plurality of imaging operations. (2) Claim 1 is characterized in that both the first image data and the second image data are image data obtained by adding each pixel data of a video signal obtained by capturing images a plurality of times. A method for inspecting defects in shadow masks. (3) In claim 1 or 2, the plurality of imaging operations include, as part of the imaging conditions, each unit aperture of the shadow mask that appears in data for each pixel of the imaged image data. conditions such that the influence of changes in is sufficiently small and negligible, the predetermined direction coincides with the direction in which the pixels are arranged, and the predetermined distance is an integral multiple of the pixel pitch. A method for inspecting defects in a shadow mask, characterized in that the method is configured to commonly include the following. (4) In claim 1 or 2, the predetermined direction coincides with the arrangement direction of the unit apertures of the shadow mask, and the predetermined distance is an integral multiple of the arrangement pitch of the unit apertures. A method for inspecting defects in a shadow mask, comprising: (5) In claim 1 or 2, the predetermined direction coincides with the arrangement direction of the unit apertures of the shadow mask, and the predetermined distance is equal to the arrangement pitch of the unit apertures and the pixel of the image data. A defect inspection method for a shadow mask, characterized in that the shadow mask is configured to be a common multiple of pitches. (6) The shadow mask according to claim 1 or 2, characterized in that the defect detection processing includes addition processing of neighboring pixel data of 3×3 pixels at most. Inspection method. (7) In claim 1 or 2, the defect detection process is configured such that, with respect to the third image data, an image in which one point on the shadow mask corresponds to the first image data is , when the pixel corresponding to the second image data is B, and the pixel corresponding to the midpoint of pixels A and B is C, the neighborhood average value of this pixel C and the average of the data of pixels A and B are A point where the difference in value is sufficiently small compared to the difference between the data of pixels A and B, and the difference between the data of pixels A and B is at least a predetermined level is determined to be a defect, and the difference between the data of pixels A and B is determined as a defect. A defect inspection method for a shadow mask, characterized in that the method is configured to recognize the type and size of a defect based on the sign and absolute value of the difference. (8) In claim 1 or 2, the shadow mask is characterized in that the unit openings are arranged with at least one of the shape, size, and arrangement pitch changing at a predetermined rate. A method for inspecting defects in shadow masks. (9) A defect inspection method for a shadow mask according to claim 1 or 2, wherein the shadow mask has a curved shape. (10) In claim 1 or 2, the shadow mask includes striped openings, the predetermined direction is the direction of the striped openings, and the predetermined distance is between pixels of the image data. A defect inspection method for a shadow mask, characterized in that the shadow mask is configured to be an integral multiple of the pitch. (11) In claim 1 or 2, the plurality of imaging operations sequentially capture the shadow mask from different directions with a normal line passing approximately at the center of the imaging surface of the shadow mask as a rotation axis. A method for inspecting defects in a shadow mask, characterized by including an oblique imaging operation. (12) In claim 1 or 2, the plurality of imaging operations include transmission bright field illumination photography, transmission dark field illumination photography, reflective bright field illumination photography, and reflective dark field illumination photography. A method for inspecting defects in a shadow mask, the method comprising at least two or more imaging operations. (13) In claim 1 or 2, the plurality of imaging operations are performed such that the imaging angle for each point on the shadow mask is approximately equal to the incident angle of the electron beam for each point on the shadow mask. A method for inspecting defects in a shadow mask, the method including the method under imaging conditions. (14) In claim 1 or 2, the plurality of imaging operations include at least transmitted bright-field illumination photography and transmitted dark-field illumination photography, and the defect determination is performed using transmitted bright-field illumination. A defect in a shadow mask characterized in that a part detected as a white spot defect during transmission dark-field illumination imaging among the parts detected as a defect with a small aperture area during imaging is not detected as a defect. Inspection method.