JP3302863B2 - Inspection method and inspection device for perforated plate - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a perforated plate in which a large number of perforated holes are arranged at substantially regular intervals, and which is used for inspecting non-uniformity in light transmittance caused by a dimensional abnormality of the perforated holes. In particular, the present invention relates to an inspection method and an inspection apparatus for inspecting a perforated plate such as a shadow mask for a color cathode ray tube and a color filter for a liquid crystal display panel.

[0002]

2. Description of the Related Art In general, a shadow mask is manufactured by arranging a large number of through-holes in a thin metal plate in a substantially periodic manner using a photo-etching method. This method of manufacturing a shadow mask corresponds to a coating process in which a resist having etching resistance is applied to the main surface of a thin metal plate to form a resist film on the main surface of the thin metal plate, and corresponds to an electron beam passage hole and the like of the shadow mask. A baking step of irradiating the resist film with light through a pattern plate having a predetermined pattern to bake a predetermined pattern on the resist film, and supplying a developing solution to the resist film having the predetermined pattern printed thereon, A developing step of dissolving and removing a predetermined portion of the film to form a resist film in a predetermined pattern and exposing a main surface of the metal sheet to be etched; and an etching solution for the metal sheet having the resist film formed in the predetermined pattern. Is supplied, and the exposed main surface of the metal sheet not covered with the resist film is etched to form a large number of electron beam passage holes. Consisting of an etching step for forming a through hole.

In the above-mentioned coating process, when the resist is not uniformly applied to the main surface of the metal sheet and the thickness of the resist film formed on the main surface is not uniform, or when the resist film of the metal sheet is etched in the etching step. If the liquid is not supplied uniformly, local dimensional abnormalities occur in many through holes formed in the metal sheet.

[0004] As a method of inspecting the local dimensional abnormality of the through hole, a method of inspecting the unevenness of the light transmittance due to the dimensional abnormality such as the shape and the hole diameter of the through hole has been adopted. More specifically, as shown in FIG. 11, the inspector first holds the shadow mask SM by hand, and
To a light table 202 having a built-in light source (not shown).
In front of the light-transmitting inclined table surface 204. Then, the light of the light source is applied to the shadow mask SM via the inclined table surface 204, and the shadow mask SM is visually observed while swinging the shadow mask SM as indicated by an arrow in the drawing. At this time, if a dimensional abnormality such as the shape or diameter of the through-hole is locally generated in the shadow mask SM, the inspector recognizes it as unevenness of the light transmittance passing through the through-hole, and this unevenness is within an allowable range. Is determined based on past experience, and non-defective shadow masks and defective shadow masks are selected. As the unevenness of the light transmittance due to the abnormal size of the through hole, the unevenness of the light density (the unevenness of the light transmittance) is scattered over the entire surface of the shadow mask SM, and the unevenness of the light and shade is partially scattered. And uneven stripes in which shading unevenly occurs in the vertical or horizontal direction.

[0005]

However, the above-described inspection has the following problems due to the visual discrimination of the inspector.

[0006] In addition to the above-mentioned unevenness in light transmittance, the inspector
You will also see the amount of light that passes through the through hole. Further, in the inspection, the unevenness occurring in the non-defective shadow mask SM as the limit sample is not always compared with the unevenness occurring in the shadow mask SM to be inspected. Is compared with the shadow mask SM to be inspected.
This was performed while visually observing only the unevenness. Therefore, a high degree of skill and considerable experience is required to determine the quality of the sample based on the comparison between the limit sample and the unevenness of the shadow mask SM to be inspected. In addition, even if the degree of skill and experience are almost the same, the quality may vary depending on the individual differences of the inspectors and the health condition.

Further, when the light amount distribution is substantially uniform over substantially the entire through-hole region where the through-hole is formed in the shadow mask SM, there is the following problem in determining the quality of the unevenness. It is indispensable to prepare a limit sample that is the basis of the judgment in determining the contrast with the limit sample,
There are limits to preparing limit samples. This is because the manner of occurrence of unevenness is not uniform because it is affected by the quality in each of the above-described steps and the composition of the metal sheet as a raw material. Therefore, when there is a slight shading in the shadow mask SM having a substantially uniform light amount distribution and good unevenness as a whole due to the limitation of the limit sample, the entire unevenness and the like described above are different from those of the non-defective shadow mask SM. In the simple comparison discrimination, the shadow mask SM may be regarded as a non-defective product. However, such a shadow mask SM may be regarded as a defective product in a high-quality cathode ray tube, and there is a case where it is not sufficient to judge the shadow mask SM with a non-defective product by visual inspection.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and it is an object of the present invention to improve the reliability of determining the quality of light transmittance unevenness caused by an abnormal dimensional dimension of a hole in a hole plate such as a shadow mask. With the goal.

[0009]

Means for Solving the Problems and Their Functions and Effects In order to solve such problems, the inspection method for a perforated plate according to the first invention is directed to a perforated plate in which a large number of through holes are arranged substantially periodically. In a method for inspecting a perforated plate for inspecting unevenness in light transmittance caused by a dimensional abnormality of a perforated hole, the perforated plate is irradiated with light from one main surface side so that the other side of the perforated plate is exposed to light. An imaging step of imaging from the main surface side of the camera and obtaining gradation data of the captured image; and removing low-frequency components appearing in distributions of individual data belonging to the gradation data, and standardizing data from the gradation data. A low-frequency removal step of generating the data, a variation calculation step of scanning the normalized data for each predetermined area group, and calculating a degree of variation of individual data belonging to the area group for each of the area groups; Maximum of the degree of the variation obtained for each group And it calculates an average value, having a presentation step of presenting the difference between the maximum value and the average value the calculation.

In order to solve the above-mentioned problems, the inspection apparatus for a perforated plate according to the second aspect of the present invention relates to a perforated plate in which a large number of through holes are arranged in a substantially periodic manner. In an inspection apparatus for a perforated plate for inspecting unevenness in light transmittance caused by the irradiation, an irradiating means for supporting the perforated plate and irradiating the supported perforated plate with light from one main surface side thereof, Imaging means for imaging the light-irradiated perforated plate from the other main surface side and obtaining gradation data of the picked-up image; and a low-frequency component appearing in a distribution of individual data belonging to the gradation data. Low-frequency removing means for generating normalized data from the gradation data, and scanning the normalized data for each predetermined region group to determine the degree of variation of individual data belonging to the region group. A variation calculating means for calculating for each group; It calculates the maximum degree of serial variation and the mean value, having a presenting means for presenting the difference between the maximum value and the average value the calculation.

In the method for inspecting a perforated plate according to the first invention or the apparatus for inspecting a perforated plate according to the second invention having the above configuration, light is irradiated from one main surface of the perforated plate to transmit the light. Let through. Thus, light is transmitted from the through-holes of the perforated plate, and a light-dark distribution is obtained on the other main surface side of the perforated plate according to the density of the arrangement of the through-holes. This light / dark distribution is captured as gradation data, which is the density value of a pixel, through imaging of a perforated plate.

The tone data is subjected to removal of low-frequency components, and the low-frequency components appearing in the distribution of individual data belonging to the tone data are removed, thereby obtaining normalized data. In this normalized data, by removing the above low frequency components,
The data distribution as a low frequency component, that is, a gradual distribution of lightness and darkness according to the density of the arrangement of the perforations in the perforated plate is removed, and unevenness in light transmittance caused by abnormal dimensions of the perforations appears. .
The unevenness of the light transmittance shown in the standardized data is captured for each area group by scanning the standardized data for each predetermined area group, and is reflected in the degree of variation of the data calculated for each area group. That is, the unevenness of the light transmittance shown on the perforated plate is grasped as unevenness for each predetermined region group, and is regarded as the degree of variation of data for each region group. Then, even if the unevenness of the light transmittance is slightly different in shading, the data area is narrowed in the area group, so that the degree of data variation is emphasized. On the other hand, the average value of the degree of variation obtained for each region group is an index of unevenness of the perforated plate scanned in this region group. Therefore, by presenting the difference between the maximum value and the average value of the degree of variation obtained for each region group, the degree of data variation can be emphasized and recognized. It is possible to make a good determination.

In the configuration of the first aspect of the present invention, the determining step of determining whether or not the unevenness of the perforated plate to be inspected is good or bad based on a difference between the calculated maximum value and the average value, Instead of the presenting step or together with the presenting step.

Further, in the configuration of the second invention,
A determination unit for determining the quality of the unevenness of the perforated plate to be inspected based on a difference between the calculated maximum value and the average value is provided instead of the presentation unit or together with the presentation unit.

In the inspection method or inspection apparatus for a perforated plate having such a configuration, the quality of the perforated plate is determined based on the difference between the maximum value and the average value of the degree of variation obtained for each region group. This determination can be made in a state where the degree of data variation is emphasized and with a difference from the index of unevenness. Therefore, according to the inspection method or the inspection apparatus for the perforated plate having these configurations, it is possible to accurately detect the unevenness of the light transmittance caused by the dimensional abnormality of the through-hole regardless of the density of the unevenness. Thus, the reliability of the determination of the quality of the unevenness can be improved.

In order to solve the above-mentioned problems, a third aspect of the present invention relates to a method for inspecting a perforated plate. In the inspection method of the perforated plate for inspecting the unevenness of the light transmittance that occurs, the perforated plate is irradiated with light from one main surface side to image the perforated plate from the other main surface side, An imaging step of obtaining gradation data of the captured image; and a low-frequency removal step of removing low-frequency components appearing in a distribution of individual data belonging to the gradation data and generating normalized data from the gradation data. Scanning the standardized data for each predetermined region group, and calculating a degree of variation of individual data belonging to the region group for each of the region groups; and a step of calculating the variation obtained for each of the region groups. The maximum value of the degree and the frequency are calculated, and the calculated It has a presentation step of presenting the difference in the degree of variation in the degree and the maximum frequency variation of Daine, the.

According to a fourth aspect of the present invention, there is provided an inspection apparatus for a perforated plate in which a large number of perforated plates are arranged substantially periodically. In an inspection apparatus for a perforated plate for inspecting unevenness in light transmittance caused by the irradiation, an irradiating means for supporting the perforated plate and irradiating the supported perforated plate with light from one main surface side thereof, Imaging means for imaging the light-irradiated perforated plate from the other main surface side and obtaining gradation data of the picked-up image; and a low-frequency component appearing in a distribution of individual data belonging to the gradation data. Low-frequency removing means for generating normalized data from the gradation data, and scanning the normalized data for each predetermined region group to determine the degree of variation of individual data belonging to the region group. A variation calculating means for calculating for each group; Serial calculates the maximum value and frequency of the degree of variation, having a presenting means for presenting the difference in the degree of variation in the degree and the maximum frequency variation of the maximum value the calculation.

In the inspection method for a perforated plate according to the third invention or the inspection apparatus for a perforated plate according to the fourth invention having the above-described structure, light is irradiated from one main surface of the perforated plate to transmit the light. Let through. Thus, light is transmitted from the through-holes of the perforated plate, and a light-dark distribution is obtained on the other main surface side of the perforated plate according to the density of the arrangement of the through-holes. This light / dark distribution is captured as gradation data, which is the density value of a pixel, through imaging of a perforated plate.

The tone data is subjected to removal of low-frequency components, and the low-frequency components appearing in the distribution of individual data belonging to the gradation data are removed, thereby obtaining normalized data. In this normalized data, by removing the above low frequency components,
The data distribution as a low frequency component, that is, a gradual distribution of lightness and darkness according to the density of the arrangement of the perforations in the perforated plate is removed, and unevenness in light transmittance caused by abnormal dimensions of the perforations appears. .
The unevenness of the light transmittance shown in the standardized data is captured for each area group by scanning the standardized data for each predetermined area group, and is reflected in the degree of variation of the data calculated for each area group. That is, the unevenness of the light transmittance shown on the perforated plate is grasped as unevenness for each predetermined region group, and is regarded as the degree of variation of data for each region group. Then, even if the unevenness of the light transmittance is slightly different in shading, the data area is narrowed in the area group, so that the degree of data variation is emphasized. On the other hand, since the degree of variation at the maximum frequency of the degree of variation obtained for each area group is obtained for a number of area groups, it is an index of unevenness for the perforated plate scanned in this area group. Become. Therefore, by presenting a difference between the degree of variation of the maximum value of the degree of variation and the degree of variation of the maximum frequency obtained for each region group, the degree of the degree of variation of data can be emphasized and recognized. It is possible to accurately determine the degree of transmittance unevenness.

In the configuration of the third aspect of the present invention, whether or not the unevenness of the perforated plate to be inspected is good or bad is determined based on a difference between the calculated degree of variation of the maximum value and the degree of variation of the maximum frequency. Is provided instead of or together with the presenting step.

Further, in the configuration of the fourth aspect,
A determination unit that determines the quality of the unevenness of the perforated plate that is the inspection target based on the difference between the degree of variation of the calculated maximum value and the degree of variation of the maximum frequency,
Instead of the presenting means or together with the presenting means.

In the inspection method or inspection apparatus for a perforated plate having these configurations, the perforated plate is inspected based on the difference between the degree of variation of the maximum value and the degree of variation of the maximum frequency obtained for each region group. Since the quality of the unevenness is determined, the determination can be made in a state where the degree of the variation of the data is emphasized and with the difference from the index of the unevenness. Therefore, according to the inspection method or the inspection apparatus for the perforated plate having these configurations, it is possible to accurately detect the unevenness of the light transmittance caused by the dimensional abnormality of the through-hole regardless of the density of the unevenness. Thus, the reliability of the determination of the quality of the unevenness can be improved.

[0023]

Other Embodiments of the Invention The above-described invention can adopt other embodiments as described below. According to a first aspect, in the above invention, low-frequency components appearing in a distribution of individual data belonging to the grayscale data are removed, and generation of normalized data from the grayscale data is performed. A median filter having a filter window of m rows and n columns (m and n are natural numbers) is smoothed to obtain smoothed data,
The normalization data is generated by dividing the gradation data by the smoothed data.

In the first aspect, the low-frequency components appearing in the distribution of the individual data belonging to the gradation data are effectively removed by smoothing using a median filter, and the smoothing from which the low-frequency components have been removed is performed. Data. Then, through the division of the gradation data by the smoothed data, the data distribution as a low-frequency component and the gradual distribution of lightness and darkness in accordance with the density of the arrangement of the perforations of the perforated plate are removed, and the transparency of the normalized data is removed. It shows the unevenness of light transmittance caused by the abnormal size of the hole. Therefore, according to the first aspect, it is also possible to accurately detect the unevenness of the light transmittance caused by the dimensional abnormality of the through hole regardless of the density of the unevenness, and the reliability of the determination of the quality of the unevenness is improved. Can be increased.

According to a second aspect, in the configuration of the second invention and the fourth invention, the calculating means scans the normalized data for each of the area groups in a partial range. Means are provided for overlapping the groups.

According to the inspection apparatus of the second aspect, when scanning of the normalized data is performed so that the area groups do not overlap, there is a case where light transmittance unevenness exists over the preceding and following area groups. However, the unevenness can be detected with high accuracy, and the reliability of the quality determination of the unevenness can be further improved.

According to a third aspect, in the configuration of the second invention and the fourth invention, the calculating means determines the size of the area group when scanning the normalized data. There is an area group setting means for setting according to the type.

According to the inspection apparatus of the third aspect, the irregularity is included in the region group so that the degree of the variation of the data is effectively emphasized regardless of the type of the irregularity of the light transmittance. Can be. For example, when the unevenness of the light transmittance is the entire unevenness scattered almost over the entire surface of the perforated plate or the partially unevenly scattered light, the square size (a row × a Column; a is a natural number) or a rectangular group (b rows × c columns; b and c are natural numbers). Further, in the case of vertical unevenness in which unevenness of light transmittance is generated in a streak shape in the vertical direction, a rectangular size (d rows × e columns; d, e: Natural number d <e)
In the region group, the vertical unevenness can be effectively included at the position where the vertical unevenness occurs. Further, in the case of lateral unevenness in which light transmittance unevenness occurs in a streak shape in the horizontal direction, a rectangular size (f rows × g columns; f, g
Is a natural group of f> g), and the horizontal unevenness can be effectively included at the position where the horizontal unevenness occurs. Therefore, according to the inspection apparatus for a perforated plate having these configurations, it is possible to accurately detect the unevenness regardless of the type of the unevenness, and it is possible to further enhance the reliability of the determination of the quality of the unevenness.

[0029]

Next, an embodiment of the present invention will be described based on an embodiment of a shadow mask inspection apparatus. First, the external configuration of the shadow mask inspection apparatus 30 of this embodiment will be described. As shown in the front view of FIG. 1, the shadow mask inspection device 30 includes an optical measurement device 40 for imaging the shadow mask SM and obtaining image data, and a data processing device 50 for performing various data processing based on the image data. It is composed of

The optical measuring device 40 has a surface plate 41, and an opening for passing illumination light is formed substantially at the center on the upper surface of the surface plate 41. A diffusion plate 43 (for example, a glass plate, a plastic plate, etc.) that diffuses and transmits light is placed on the upper surface of the surface plate 41.
A light source 44 is arranged below the light source 3. This light source 4
For example, a high frequency lighting type fluorescent lamp is used as 4. Note that a mask plate 45 is placed on the upper surface of the diffusion plate 43, and a shadow mask SM is adhered and fixed to the mask plate 45 with an adhesive tape or the like.

A stand support arm 46 extending upward from the surface plate 41 is provided.
6 is provided with a so-called cantilever camera holding beam 47. The camera holding beam 47 is attached to the stand support arm 46 so as to be adjustable by an adjusting mechanism (not shown).
The camera 49 and the camera 49 can be integrally moved in the vertical direction (Z direction in the figure). The CCD camera 49 is attached to the camera holding beam 47 so as to be adjustable, and can move the CCD camera 49 in the left-right direction (X direction in the figure). For this reason, the CCD camera 49 can be moved up and down, left and right so that the regions to be imaged by the shadow masks of various sizes coincide with the imaging regions of the CCD camera 49.

The CCD camera 49 is a CCD camera in which CCD elements are two-dimensionally arranged, and is capable of digitally outputting a 10-bit digital image in a grayscale range. Note that CC
From the D camera 49, an image of 1,534 pixels × 1,024 pixels can be obtained.

The data processing device 50 includes a display 52 for displaying an image described later, an image processing device 54 for performing various image processing, and a camera control device 62 for controlling the gain and shutter speed of the CCD camera 49. I have.

Next, the electrical configuration of the shadow mask inspection apparatus 30 will be described with reference to the block diagram of FIG. The light emitted from the light source 44 sequentially passes through the diffusion plate 43 and the through holes of the shadow mask SM and enters the CCD camera 49. At this time, the shadow mask SM is closely attached and fixed to the mask plate 45, and the through-hole region SMe, which is a region where the through-hole of the shadow mask SM is formed, is formed by the opening 45 of the mask plate 45.
The peripheral region of the shadow mask SM that is exposed from a and has no through hole is masked. Therefore, the CCD camera 49 captures an image of the shadow mask SM with respect to the through-hole area SMe, and two-dimensionally arranged gray-scale images (multi-valued images)
As image data Di. This image data Di is
After being captured by the image processing device 54 via the camera control device 62 and subjected to image processing described later, various images are displayed on the display 52.

The image processing device 54 is a general-purpose high-speed image processing device that performs various processes according to programs registered in advance. The image processing device 54 includes a CPU 66 that performs other predetermined processes in addition to various image processes and the like. Main storage device 68 such as RAM for temporarily storing data such as data Di
And a keyboard 70 for inputting data, an auxiliary storage device 72 for storing data such as image data Di, for example, a flexible disk device, etc., and a printer 74 for outputting inspection results and the like. They are connected to each other via a bus line 64. Further, a camera control device 62 is connected to a bus line 64 of the image processing device 54.
And the display 52 are connected.

Next, the unevenness inspection processing performed by the shadow mask inspection apparatus 30 of this embodiment will be described with reference to the flowcharts of FIG.

FIG. 3 is a flowchart showing the outline of the inspection processing of the unevenness of the shadow mask SM performed in the present embodiment. When the processing is started, initialization necessary for performing the subsequent processing is performed. (Step S100). For example, display of an inspection start initial screen on the display 52, clearing of a memory area used in processing described later, and the like are performed.

When this initialization is performed, the display 52
Displays an initial setting screen, and the input portions of various items (initial setting items) required in the course of the subsequent inspection processing remain without data input on the initial setting screen.

Subsequent to step S100, initial settings are made for various items required in the course of inspection (step S110). In this initial setting, data is sequentially input to the initial setting items on the initial setting screen displayed without inputting data in the initialization processing. That is, as shown in the flowchart of FIG. 4 showing the detailed processing of the initial setting processing, the output file name of the inspection result (step S11
1) The mask type of the mask plate 45 used to mask the shadow mask SM to be inspected (step S11)
2) The name of the inspector (step S113) is sequentially input.
These are displayed in the inspection result displayed or printed out after the inspection is completed. The output file is a file for storing inspection results for each shadow mask SM to be inspected, and has a data format of a sample ID (for example, a manufacturing serial number) of the shadow mask SM to be inspected, an inspector name , Inspection date, etc.

When these inputs are completed, the process stands by until the inspection start switch is pressed (step S114). When the inspection is started, the output file is opened (step S115) to prepare for the output of the inspection result. It should be noted that, in this output file, the results of discrimination regarding non-uniformity described later are output, and the data is stored. After that, the input initial setting data such as the output file name is output (step S116), the initial setting data is displayed on the initial setting screen of the display 52, and the result display screen is set (step S117). Subsequently, a result display screen is displayed on the display 52 for displaying the result of the quality determination of the unevenness of the shadow mask SM to be inspected (step S118). The result of the quality determination of the unevenness is automatically input according to the determination result in the unevenness determination process described later, and the items include the type of the unevenness (entire unevenness, partial unevenness, vertical unevenness, and horizontal unevenness) and its unevenness. This is the result of pass / fail determination for each unevenness.

After the initial setting, as shown in FIG. 3, the operator waits for the inspector to complete the setting of the shadow mask SM on the diffusion plate 43 on which the mask plate 45 is placed. It waits for an input start instruction issued by the user (step S120). If there is an input start instruction, a transmission image of the shadow mask SM to be inspected is fetched by a series of processes from the subsequent step S122, and various processes for unevenness inspection are performed.
First, following step S120, the light source 44 is controlled to be turned on, and the CCD camera 49 is set to match the imaging conditions when the CCD camera 49 captures a transmitted image of the shadow mask SM (step S12).
2). More specifically, the shutter speed setting and gain setting of the CCD camera 49 are performed based on predetermined imaging conditions such as shutter speed, gain, focus, and the like, and the through-hole area SMe of the through-hole of the shadow mask SM to be inspected. , Focus adjustment, etc., and Z of the CCD camera 49
Shaft position adjustment is performed.

The Z-axis position adjustment is performed by manually operating an adjustment mechanism (not shown) to vertically move the camera holding beam 47 and the CCD camera 49 integrally with respect to the stand support arm 46, and to adjust the shutter speed. Settings, gain adjustment, and the like are performed by the camera control device 62. Also,
The focus adjustment is performed by manually adjusting the lens unit 77 of the CCD camera 49. In this case, the focus of the CCD camera 49 is adjusted to be slightly blurred in order to remove moire that occurs between the arrangement of pixels of the CCD camera 49 and the periodic arrangement of the through holes of the shadow mask SM.

When the imaging conditions of the CCD camera 49 are set in this way, the CCD camera 49 starts imaging the shadow mask SM on the diffusion plate 43, and the light emitted from the light source 44 and transmitted through the through holes of the shadow mask SM. A transmission image (hereinafter, referred to as a raw image) is captured by the CCD camera 49 (step S124). The CCD camera 49 outputs the raw image of the shadow mask SM to the main storage device 68 via the camera control device 62 and the bus line 64 as grayscale data as digital data of 1,534 pixels × 1,024 pixels and a gradation range of 10 bits.

In general, the distance between a large number of through-holes formed in the shadow mask SM is narrow at the center of the shadow mask and widens toward the periphery. This is because the arrangement of the through holes of the shadow mask SM is designed in consideration of forming the flat shadow mask SM into a dome shape. From this, the shadow mask S
The above-described gradation data, which is the data of the transmission image of M, is a light-dark pattern (hereinafter, referred to as “bright” at the center of the shadow mask SM and becomes darker toward the peripheral direction, according to the density of the arrangement of the through-holes of the shadow mask SM. This pattern is referred to as grade) is data represented as a shade range.
In addition, the gradation data reflects the shading based on the unevenness of the shadow mask SM.

Then, the gradation data of the raw image is sent to the image processing device 54 via the camera control device 62. In addition to the data output to the image processing device 54, the camera control device 62 displays the raw image on the display 52. Thereby, the imaging area of the CCD camera 49 can be confirmed.

The output signal (gradation data) of the raw image obtained by the CCD camera 49 is superimposed with a signal resulting from uneven light emission distribution caused by the light source 44 and the diffusion plate 43 itself and uneven sensitivity of the image pickup system. . Therefore, the above step S
Subsequent to capturing the raw image at 124, shading correction is performed by dividing the image data (gradation data) by the reference data (gradation data) having the same number of data (step S126). The reference data is gradation data of an image captured in advance with only the mask plate 45 mounted on the diffusion plate 43 without mounting the shadow mask SM, and is stored in the main storage device 68 in advance. Therefore, this reference data is read at the time of shading correction.

Subsequently, a processing target area SMe0 in which stripe unevenness is determined in the unevenness inspection processing is set (step S).
128). That is, as shown in FIG. 5, the through-hole region SMe of the through-hole of the shadow mask SM is a region whose periphery is curved. Therefore, in the region of this curved periphery, matching of the number of data when performing a deviation calculation described later is performed. Crumble. Therefore, in advance in step S128, a rectangular processing target area SM
e0 is defined as a plane area defined by the X axis and the Y axis. Next, the set processing target area SMe0
A median filter (hereinafter abbreviated as M / F) having a filter window of 31 pixels × 31 pixels (hereinafter simply referred to as 31 × 31; the same applies hereinafter) is obtained by performing the shading correction on the gradation data of The image data is corrected using the image data (step S130), and standardized data is obtained by removing low frequency components from the grayscale data by M / F.

Here, the contents of the image correction processing in step S130 will be described. In this image correction, the following smoothing processing and normalization processing are performed. First, in the smoothing process, data (gradation data) that fills each data window of the 31 × 31 filter window is arranged in the order of its size to create a data sequence, and data at the center position of the data sequence is output. The processing to be a value is defined as one processing unit. Then, for the shading-corrected gradation data for the processing target area SMe0, for example, the filter window is moved one pixel at a time in a predetermined direction while changing the M / F filtering area, and the processing in one processing unit is performed. Data area of the above gradation data (processing target area SMe0)
Repeat over.

By this smoothing processing by the median filter, noise having a high spatial frequency is effectively reduced from the gradation data after shading correction. Further, smoothed data of an image in which small fluctuations are smoothed is obtained. In the smoothed data, although the grade and large fluctuation of the shadow mask SM, which is a low-frequency component, are reflected, the unevenness of the shadow mask SM is reduced or eliminated as noise or small fluctuation.

In the subsequent normalization processing, the gradation data after the shading correction is divided by the smoothed data after the smoothing processing by the M / F to obtain the normalized data for the raw image. In this division operation, if the value of the data of the smoothed data is zero, the division by the corresponding data is not performed, and the corresponding data of the normalized data is set to a value of zero. Since the standardized data thus obtained has undergone the division of the gradation data by the smoothed data, the grade and large fluctuation (low-frequency component) of the shadow mask SM reflected as a shade range on the smoothed data are removed. , Data representing the unevenness of the light transmittance caused by the abnormal size of the through hole.

Since the standardized data indicates the unevenness of the light transmittance as described above, it is possible to display an uneven image on the display 52 based on the standardized data. In this case, the following processing is performed when displaying the uneven image. First, since the normalized data is real number data and is difficult to handle in image processing, the data is converted to integer data by multiplying it by a predetermined integer, for example, 4000, to obtain normalized integer data. Next, in order to display an image based on the normalized integer data, that is, the unevenness of light transmittance caused by the dimensional abnormality of the through-hole as an image of density range data,
A visualized image conversion is performed on the standardized integer data, and an uneven image is displayed on the display 52.

Here, the contents of the processing until the standardized data is obtained from the gradation data will be schematically described together with the contents of the processing for displaying the uneven image. As shown in FIG.
The raw image G picked up by the CCD camera 49 is
After the processing of 124 to 126, the gradation data GD is subjected to shading correction. The graph in the figure shows the gradation data GD in the image along the horizontal axis of the raw image. On the other hand, the smoothed data M / FD obtained by smoothing the grayscale data GD by M / F by a part of the process (smoothing process) in step S130, as shown in FIG. The data is reduced or removed from the gradation data GD as a small fluctuation, and becomes data reflecting only the grade of the shadow mask SM.

Next, after the remaining processing (normalization processing) of step S130, normalized data KD obtained by dividing the gradation data GD by the smoothed data M / FD is obtained. Then, the uneven image is displayed through integer data conversion of the standardized data KD and visualization image conversion. As shown in the figure, this unevenness image is an image of only the unevenness of the shadow mask SM to be inspected, and is an image in which the unevenness is enhanced because the grade is removed.

Subsequent to the above-described image correction, a process of determining unevenness in light transmittance is performed (step S140). In the unevenness determination process, as shown in the flowcharts of FIG. 7 and subsequent figures, a determination process is performed for each type of unevenness (whole unevenness, partial unevenness, vertical unevenness, and horizontal unevenness). That is, as shown in the flowchart of FIG. 7, first, a determination process for the entire unevenness is performed (step S150), and thereafter, a determination process for the partial unevenness (step S170), a determination process for the vertical unevenness (step S190), The determination process (step S210) for the lateral unevenness is sequentially performed. The details of the unevenness determination processing are the same, although the detailed processing contents used in the processing vary depending on the unevenness to be determined, which will be described later. Therefore, the details of the determination process (step S150) for the entire unevenness will be described below as an example.

As shown in the flowchart of FIG. 8, when the determination process for the entire unevenness is started (step S150), first, the size of a processing window used for data collection in the determination process is selected (step S152).
That is, even if the unevenness is the whole, the unevenness of the light transmittance is not uniform in the state of the spots in the through-hole area SMe of the shadow mask SM, and the size of the unevenness in the spots is not uniform. Therefore, the size of the processing window is selected to cope with various overall unevenness. Specifically, a processing window having a small size is used for an image having small individual unevenness, and a processing window having a large size is used for an image having large individual unevenness. In this embodiment, 20 × 20, 30 × 30, 40 × 45,
Processing windows having window sizes of 60 × 60, 80 × 90, and 240 × 180 (all of which are pixels) were selected. This is because if such a square or rectangular processing window is selected, individual unevenness in the overall unevenness can be included in the processing window regardless of its size.

After setting the window size in this way, the window scanning and data calculation / storage are performed for one of the processing window SWs (step S).
154). That is, as shown in FIG.
W to be processed while shifting W by a half pitch thereof
While scanning vertically and horizontally over the area of Me0, each time,
Using the individual data of the gradation data belonging to the processing window SW, the standard deviation σ of these data is obtained from the following equation and stored.

[0057]

(Equation 1)

In the above formula, f (x, y) is the density value (gradation data value) at the coordinates (x, y) in the processing window SW, and av is the entire processing target area SMe0 of the shadow mask SM. It is an average density value (gradation data value), and N is the total number of data in the processing window SW (the total number of pixels, 400 if the processing window SW is a 20 × 20 processing window). Note that the value of the average density av is calculated after the setting of the processing target area SMe0 in step S128 and the image correction in step S130 are performed on the gradation data that has been captured in step S124.
In the present embodiment, the smoothing data M
Since the gradation data GD is divided by / FD,
The value of the average density av is 1.

As is clear from this equation, the standard deviation σ
Represents the degree of variation of the data belonging to the processing window SW. If the value of the standard deviation σ is large, the degree of unevenness is large at the scanning position in the processing window SW. The standard deviation σ is calculated and stored every time the processing window SW is scanned until the processing window SW is scanned over the entire processing target area SMe0 as described above.

Thereafter, all the stored standard deviations σ
, The average value σav, the maximum value σmax, and the most frequent value σpeak are obtained (step S156). Average value σ
av is obtained by averaging each time the standard deviation σ is calculated, and the maximum value σmax is a magnitude comparison each time the individual standard deviation σ is calculated or a magnitude comparison after all the standard deviations σ are calculated. Is required. Also, for the most frequent value σpeak, a histogram shown in FIG. 10 is created by plotting the value of the standard deviation σ and its frequency each time the standard deviation σ is calculated. The value of the deviation σ is obtained as the most frequent value σpeak.

Next, the absolute value | σmax-σav | of the difference between the maximum value σmax and the average value σav and the maximum value σmax
absolute value of the difference between x and the most frequent value σpeak | σmax−σ
peak | is calculated and stored (step S1).
58). In this case, the calculated absolute value | σma of these differences
x-σav | and | σmax-σpeak | are stored in association with the size of the processing window SW and the type of unevenness (whole unevenness), and are also displayed on the display 52. Therefore, the inspector determines the maximum value σ for the standard deviation σ
absolute value of the difference between max and average value σav | σmax−σa
v | and the absolute value | σmax−σpeak | of the difference between the maximum value σmax and the most frequent value σpeak |

Since the average value σav of the standard deviation σ for each processing window SW is an average value, the average value σav serves as an index of unevenness in the entire processing target area SMe0 scanned by the processing window SW. Also, the most frequent value σpeak obtained from the standard deviation σ of each processing window SW is the value of the standard deviation σ obtained most in various places in the processing target area SMe0. The index serves as an index of unevenness in the entire scanning target area SMe0. On the other hand, the processing window S with a narrow data area
The maximum value σmax of the standard deviation σ obtained with respect to W emphasizes this unevenness to some extent because the data area becomes narrow in the processing window SW even if the overall unevenness is only slightly different in shading. Will be represented. Therefore, the absolute value of the difference | σmax−σav |
The inspector receiving the presentation of | σmax−σpeak | can recognize the degree of the overall unevenness as the magnitude of the absolute value of the difference, and therefore, the unevenness of the light transmittance regardless of the density of the unevenness or the skill of the inspector. The degree can be accurately determined.

Subsequent to the calculation and storage of the absolute values in step S158, these absolute values are stored in the respective discrimination maps (whether the overall unevenness is good or bad) stored in the main storage device 68 in advance.
σmax-σav | and a map associating | σmax-σpeak |) with the quality of overall unevenness, and determines the quality of overall unevenness, and stores the result in the main storage device 68. (Step S160).
In this case, the result of the pass / fail determination of the overall unevenness is determined by the absolute values | σmax−σav | and | σmax−σpeak |
Is stored in association with the size of the processing window SW and the type of unevenness (whole unevenness).

Thereafter, it is determined whether or not the determination of the overall unevenness has been completed for all the processing windows SW selected in step S152 (step S162). If a negative determination is made, the processing window SW is changed (step S).
164), the processing from step S154 is performed in the processing window SW after this change. On the other hand, step S1
If an affirmative determination is made in step 62, the process proceeds to the next process, in this case, the process for determining partial unevenness in FIG. 7 (step S170).

In the processing for determining partial unevenness in step S170, after selecting processing window sizes corresponding to various partial unevennesses, each selected processing window SW performs the same processing as the above-described overall unevenness determination. (Steps S154 to S164). The processing window SW selected in the partial unevenness determination processing is 10 × 10, 20 × 2
This is a processing window having window sizes of 0, 30 × 30, 40 × 40, and 50 × 50. This is because if such a square-sized processing window is selected, individual unevenness in partial unevenness can be included in the processing window regardless of its size.

Step S190 following step S170
In the vertical unevenness determination process described above, after selecting processing window sizes corresponding to various vertical unevennesses, the same processing as the above-described overall unevenness determination is performed in each of the selected processing windows SW (steps S154 to S154). 164). The processing window SW selected in this vertical unevenness determination processing has a vertically long rectangular size (10
× 60, 5 × 60, 5 × 90, 10 × 90 and 15 ×
90) is a processing window of the window size. This is because if such a rectangular processing window is selected, individual unevenness in vertical unevenness can be included in the processing window regardless of the width and length of the unevenness.

Step S190 following step S190
In the horizontal unevenness determination processing in 210, after selecting processing window sizes corresponding to various horizontal unevennesses, the selected processing window SW performs the same processing as the above-described overall unevenness determination (step S154). 164).
The processing window SW selected in the horizontal unevenness determination processing is as follows.
This is a processing window having a window size of a horizontally long rectangular size (60 × 10, 60 × 5, 80 × 10, 80 × 5, and 120 × 10) along the horizontal direction which is the direction in which unevenness occurs. This is because if such a rectangular processing window is selected, individual unevenness in the horizontal unevenness can be included in the processing window regardless of the width and length of the unevenness.

After the pass / fail judgment of each unevenness is performed in the processing up to step S210, the process returns to the processing of FIG. 3, and the pass / fail judgment results in steps S150 to S210 are displayed on the display 52 (step S22).
0). In this case, the type of the processing window SW and the type of the unevenness (whether the whole unevenness, the partial unevenness, the vertical unevenness, and the horizontal unevenness) that caused the pass / fail result and the absolute value | σmax−σ
av |, | σmax−σpeak | are also displayed on the display 52. After that, another shadow mask S
It is determined whether or not it is necessary to continuously perform the above-described inspection on M based on the pressing state of a predetermined switch (step S230). If the inspection is to be continued, the processing from step S120 described above is repeated. Note that these items in the above determination result are stored in the auxiliary storage device 72 at the time of the corresponding process together with the sample ID of the corresponding shadow mask SM to be inspected, and the other shadow masks SM are determined in step S230. Is performed, the display image on the display 52 is erased and the main storage device 68 of various data temporarily stored in the process of the determination process is prepared in preparation for the inspection of the other shadow mask SM. From the memory.

As described above, in the shadow mask inspection apparatus 30 according to the present embodiment, the entire unevenness, the partial unevenness, the vertical unevenness, and the horizontal unevenness occurring in the processing target area SMe0 are processed by the corresponding processing window SW. Each time scanning is performed, a standard deviation σ is obtained from density value data (gradation data) belonging to the processing window SW. Then, the maximum value σmax, the average value σav, and the most frequent value σpeak are obtained from the standard deviation σ obtained for each scan of the processing window SW, and the absolute value of the difference | σmax−σav |, | σm
ax-σpeak | is used to determine the quality of the above-described various non-uniformities. The maximum value σmax of the standard deviation σ used in the pass / fail judgment is expressed by emphasizing the degree of unevenness through narrowing of the area to which the data belongs, and the average value σmax of the standard deviation σ
av and the most frequent value .sigma.peak are indexes of unevenness in the entire processing target area SMe0 through averaging or selection of the most frequent value. Therefore, according to the shadow mask inspection apparatus 30 of the present embodiment, the various unevennesses described above can be accurately detected even if the shading of the unevenness is slight in the shadow mask SM to be inspected. It is possible to improve the reliability of determining whether or not various irregularities (irregularities in light transmittance) occur due to abnormal dimensions of the through holes in the shadow mask SM.

Further, in the shadow mask inspection apparatus 30 of the present embodiment, a processing window SW of such a size that the corresponding unevenness is effectively included for each of the entire unevenness, the partial unevenness, the vertical unevenness and the horizontal unevenness. Set. Therefore, regardless of the type of the unevenness, each unevenness can be detected with high accuracy, and the reliability of the quality determination of the unevenness can be further improved.

Further, in the shadow mask inspection apparatus 30 of this embodiment, when the processing window SW is scanned over the processing target area SMe0, the processing window SW is scanned while being shifted by a half pitch thereof. The standard deviation σ was calculated. Therefore, every time the processing window SW is scanned, the first
The / 2 window area is overlapped (see FIG. 9). For this reason, it is possible to obtain the standard deviation σ of the unevenness generated on the periphery of the processing window SW by including the unevenness in the processing window SW at the time of scanning the next processing window SW. Therefore, according to the shadow mask inspection apparatus 30 of the present embodiment, in addition to the density of the unevenness, the unevenness can be detected with higher accuracy regardless of the location where the unevenness occurs. Can be further improved.

Further, since the result of the quality judgment for each unevenness becomes accurate, the following effects can be obtained. That is, the shadow mask S to be inspected
If there is a defect in M, the quality control of each step (coating step, baking step, developing step, etching step, etc.) in the photo-etching method of the shadow mask SM together with the type of the irregularity and the discrimination result of the irregularity. And the quality can be improved through this.

The shadow mask S to be inspected is
In addition to the discrimination result of M, the standard deviation σ obtained in each processing window SW, the maximum value σmax, and the average value σa
v, the most frequent value σpeak and the absolute value of the difference | σma
If x-σav | and | σmax-σpeak | are stored in the auxiliary storage device 72, the following effects can be obtained. Even if the shadow mask SM whose unevenness is determined to be non-defective by the above unevenness determination process is later determined to be of poor quality, if the above data at the time of the determination process is read out and displayed, it is stored as the indicated poor quality. Data can be compared. For this reason, the validity of the pass / fail determination and the validity of the map for pass / fail determination can be reexamined, and the cause of the quality defect can be easily determined by comparing the determination result with the quality defect.

Here, a modification of the shadow mask inspection apparatus 30 of the above embodiment will be described. In the above-described shadow mask inspection apparatus 30, the maximum value σma of the standard deviation σ
x, the average value σav, and the absolute value of the difference obtained from the most frequent value σpeak | σmax-σav | and | σmax-σpeak |
Is determined based on the above. However, the absolute values | σmax−σav | and | σmax−σpe
ak | In addition, the pass / fail judgment is also made for each of the whole unevenness, the partial unevenness, the vertical unevenness, and the horizontal unevenness. However, a modification of the configuration in which the processing is terminated at that time if the quality determination is determined to be defective due to any unevenness can be made. According to the configurations of these modifications, when there is a defect of unevenness, the processing time for unevenness determination can be reduced.

It is also possible to adopt a configuration in which a processing window SW having a size other than the above-mentioned size is used, or only a processing window SW having an optimum size among the above-mentioned sizes is used.

The embodiments of the present invention have been described above.
The present invention is not limited to such embodiments at all, and it goes without saying that the present invention can be implemented in various modes without departing from the gist of the present invention.

[Brief description of the drawings]

FIG. 1 is a front view showing an external configuration of a shadow mask inspection apparatus 30 according to an embodiment.

FIG. 2 is a block diagram showing an electrical configuration of the shadow mask inspection apparatus 30 shown in FIG.

FIG. 3 is a flowchart showing an entire process of inspecting unevenness of a shadow mask SM performed by a shadow mask inspection device 30;

FIG. 4 is a flowchart showing detailed contents of an initial setting process in the unevenness inspection process.

FIG. 5 is an explanatory diagram for explaining processing content in the unevenness inspection processing shown in FIG. 3;

FIG. 6 is an explanatory diagram for schematically explaining the contents of processing from obtaining gradation data of a raw image to standardized data, together with the contents of processing for displaying a non-uniform image.

FIG. 7 is a flowchart showing the contents of step S140 shown in FIG. 3;

FIG. 8 is a flowchart showing detailed contents of an entire unevenness determination process performed in step S150 shown in FIG. 7;

FIG. 9 is an explanatory diagram for describing a state of scanning of a processing window SW performed in step S154 shown in FIG. 8;

FIG. 10 is an explanatory diagram for explaining a histogram in which the value of the standard deviation σ and its frequency are plotted.

FIG. 11 is an explanatory diagram for explaining a state of a conventional unevenness inspection for a shadow mask SM.

[Explanation of symbols]

 REFERENCE SIGNS LIST 30 shadow mask inspection device 40 optical measurement device 41 surface plate 43 diffusion plate 44 light source 45 mask plate 45 a opening 46 stand support arm 47 camera holding beam 49 CCD camera 50 data processing device 52 Display 52a Display area 54 Image processing device 62 Camera control device 64 Bus line 66 CPU 68 Main storage device 70 Keyboard 72 Auxiliary storage device 74 Printer 77 Lens unit SM Shadow mask SMe0 Processing target Area SMe: Through-hole area SW: Processing window

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-6-229736 (JP, A) JP-A-7-208959 (JP, A) JP-A-3-202707 (JP, A) JP-A-9-209 68502 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01B 11/00-11/30 G01N 21/88

Claims (8)

(57) [Claims] 1. A method for inspecting a perforated plate in which a plurality of perforated plates are arranged in a substantially periodic manner, wherein the perforated plate is inspected for unevenness in light transmittance caused by a dimensional abnormality of the perforated holes. An imaging step of irradiating the perforated plate with light from one main surface side to image the perforated plate from the other main surface side, and obtaining gradation data of the captured image; A low-frequency removal step of removing low-frequency components appearing in the distribution of data and generating standardized data from the grayscale data; scanning the normalized data for each predetermined region group; A variation calculation step of calculating the degree of variation of the data for each of the region groups; calculating a maximum value and an average value of the degree of variation obtained for each of the region groups; and calculating the calculated maximum value and average value. And a presentation step for presenting the difference between Inspection method of the through-hole plate to be. 2. The method for inspecting a perforated plate according to claim 1, wherein the quality of the non-uniformity of the perforated plate to be inspected is determined based on a difference between the calculated maximum value and the average value. A method of inspecting a perforated plate having a determining step to be performed instead of or together with the presenting step. 3. A perforated plate inspection apparatus for inspecting non-uniformity in light transmittance caused by a dimensional abnormality of a perforated plate, wherein the perforated plate on which a large number of perforated holes are arranged approximately periodically is provided. An irradiating means for supporting the perforated plate and irradiating the supported perforated plate with light from one main surface side thereof, and capturing an image of the perforated plate irradiated with the light from the other main surface side; Imaging means for obtaining gradation data of a captured image; low-frequency removal means for removing low-frequency components appearing in the distribution of individual data belonging to the gradation data and generating standardized data from the gradation data; A variation calculating unit that scans the normalized data for each predetermined region group and calculates a degree of variation of individual data belonging to the region group for each of the region groups; and a degree of the variation obtained for each of the region groups. Calculate the maximum value and average value of Inspection device of the through-hole plate, characterized in that it comprises a presentation means for presenting the difference between the value and the average value. 4. The inspection apparatus for a perforated plate according to claim 3, wherein the quality of the non-uniformity of the perforated plate to be inspected is determined based on a difference between the calculated maximum value and the average value. An inspection apparatus for a perforated plate, wherein a discriminating means is provided instead of or together with the presenting means. 5. A method for inspecting a perforated plate in which a large number of perforated plates are arranged in a substantially periodic manner for inspecting unevenness in light transmittance caused by an abnormal dimensional dimension of the perforated holes. An imaging step of irradiating the perforated plate with light from one main surface side to image the perforated plate from the other main surface side, and obtaining gradation data of the captured image; A low-frequency removal step of removing low-frequency components appearing in the distribution of data and generating standardized data from the grayscale data; scanning the normalized data for each predetermined region group; A variation calculation step of calculating the degree of variation of the data for each of the region groups, and calculating the maximum value and frequency of the degree of variation obtained for each of the region groups, and calculating the degree of variation of the calculated maximum value. The difference from the maximum frequency variation Method of inspecting hole plate, characterized in that it comprises a Shimesuru presenting step. 6. The perforated plate inspection method according to claim 5, wherein the perforated plate to be inspected is determined based on a difference between the calculated degree of variation of the maximum value and the degree of variation of the maximum frequency. A determining step of determining whether the unevenness is good or bad,
A method for inspecting a perforated plate that is provided instead of or together with the presenting step. 7. A perforated plate inspection apparatus for inspecting non-uniformity of light transmittance caused by a dimensional abnormality of a perforated plate in which a plurality of perforated holes are arranged substantially periodically. An irradiating means for supporting the perforated plate and irradiating the supported perforated plate with light from one main surface side thereof, and capturing an image of the perforated plate irradiated with the light from the other main surface side; Imaging means for obtaining gradation data of a captured image; low-frequency removal means for removing low-frequency components appearing in the distribution of individual data belonging to the gradation data and generating standardized data from the gradation data; A variation calculating unit that scans the normalized data for each predetermined region group and calculates a degree of variation of individual data belonging to the region group for each of the region groups; and a degree of the variation obtained for each of the region groups. Calculate the maximum value and frequency of Inspection device of the through-hole plate, characterized in that it has a, and presenting means for presenting the difference in the degree of variation in the degree and the maximum frequency variation of. 8. The inspection apparatus for a perforated plate according to claim 7, wherein the perforated plate to be inspected is based on a difference between the calculated degree of variation of the maximum value and the degree of variation of the maximum frequency. Determining means for determining the quality of the unevenness,
An inspection apparatus for a perforated plate which is provided instead of or together with the presenting means.