JPH06208017A - Method and device for inspecting defect of color filter - Google Patents

Abstract

PURPOSE:To provide the method and device for inspecting the defect of the color filter of an LCD, etc., which can automate the defect inspection of the color filter. CONSTITUTION:The LCD color filter 5 is irradiated by a transparent lighting device 8 and its transmitted light is converted by a camera 2 into an electric signal, which is inputted to an image processor 1. A dark visual field lighting device 9 irradiates the LCD color filter 6 and its reflected light is converted by a camera 3 into an electric signal, which is inputted to the image processor 1. A dark visual field lighting device 10 irradiates the LCD color filter 7 and the reflected light is converted by a camera 4 into an electric signal, which is inputted to the image processor 1. The image processor 1 detects a white void defect by performing arithmetic processing after S conversion and I conversion of the transmitted light and also detects a Cr, defect by performing image arithmetic processing after the I conversion. Further, a flaw and a projection defect are detected with dark visual field light and an ITO defect is also detected similarly with light visual field light.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color filter defect inspection method and defect inspection apparatus, and more particularly to an LCD (Li
The present invention relates to a color filter defect inspection method and a defect inspection device.

[0002]

2. Description of the Related Art Liquid crystal display (hereinafter referred to as LCD)
In addition to its features such as thinness, light weight, and low power consumption, it is also widely used for flat panel displays because of its advantages such as low price and the ability to support full color.

A color filter used in an LCD has external dimensions of 18 mm × 15 mm × as shown in FIG. 5, for example.
Chromium (Cr) 102 is vapor-deposited on a surface of a glass substrate 101 having a thickness of about 2 mm with a small window for the number of pixels (for example, 480 × 268) opened. This small window is R (Re
d), G (Green) and B (Blue) color filters 103 are color-coated so that light is transmitted therethrough. An electrode film (not shown) is formed by coating on Cr 102 and RGB.

The defects of this LCD color filter are classified into four types as follows. FIG. 6 is a diagram for explaining a defect of the color filter, and the defect will be described with reference to this figure.

(1) Whiteout defect It means that the color coating is missing in a part of the area of the small window color-coded in RGB as shown in FIG.

(2) Cr defect This is a defective patterning of Cr, and as shown in FIG.
It means that r is excessively formed in a region of each small window of RGB, or is excessively small in amount.

(3) Scratch / protrusion defect It means that dust or the like is attached to the glass substrate as shown in FIG. 6 so that the RGB region or the Cr region is scratched or a protrusion is formed.

(4) ITO (Indium Tin Oxide) Defects It means that the film thickness of the electrode film is uneven, and this unevenness spreads to some extent as shown in FIG.

Conventionally, the defects of these LCD color filters have been detected visually by using a microscope under illumination suitable for detecting the defects. Hereinafter, the illumination system used for detecting each defect and the detection status thereof will be described.

(1) Whiteout defect The whiteout defect portion appears to be white in transmitted illumination.

(2) Cr Defects In transillumination, since Cr does not transmit light, a difference in brightness occurs between the light transmitting portion of the RGB colored portion and the Cr defect portion which does not transmit light due to Cr. C due to the difference
r Detect defect part.

(3) Defects and Defects of Defects Defects of defects of scratches and protrusions appear to be white in dark field illumination.

(4) ITO (Indium Tin Oxide) Defects The ITO defect portion appears to be hazy even when the hue is disturbed in bright field illumination.

Table 1 describes the contents of each defect and the suitability and unsuitability depending on the illumination method.

[0015]

[Table 1]

[0016]

In the defect inspection of the LCD color filter described above, some defects may be overlooked by visual observation using a microscope, which causes a problem in quality of the defect inspection. Further, the pain of the worker for detecting the minute defects exceeds the limit, which is a problem.

Therefore, an object of the present invention is to provide a color filter defect inspection method and a defect inspection apparatus capable of automating the defect inspection of a color filter such as an LCD.

[0018]

According to the present invention, there is provided a defect inspection method for a color filter formed on a transparent substrate, wherein a plurality of illuminations having different illuminances or transmittances are sequentially applied to the color filter and A defect inspection method of a color filter is characterized in that the defect of the color filter is inspected by irradiating the peripheral portion of the color filter and image-processing each image information obtained by the reflected light or the transmitted light. .

According to the present invention, there is provided a method for inspecting a defect of a color filter formed on a transparent substrate, the method comprising:
The defect of the characteristic of the color filter is detected by irradiating a specific illumination to the color filter and a peripheral portion of the color filter, and image-processing the image information obtained by the reflected light or the transmitted light. It is solved by the color filter defect inspection method.

Further, according to the present invention, the above-mentioned problem is that the light transmitted to the color filter is good and the light not transmitted to the peripheral portion of the color filter or having poor transmittance is changed to the color filter. And S conversion processing for illuminating the peripheral part of the color filter and performing classification by saturation on the image information obtained by the transmitted light and I conversion processing for performing classification by luminance, and the S conversion processing 3. The color filter defect inspection method according to claim 2, wherein the color painting defect of the color filter is detected by performing an arithmetic process on the output of the I and I conversion processes.

[0021] According to the present invention, the above-mentioned problem is obtained by transmitting light having good transmittance to the color filter and not transmitting to the peripheral portion of the color filter or having poor transmittance to the color filter and the color filter. Irradiate the peripheral part of the filter, and perform I conversion processing for performing classification by brightness on image information obtained by the transmitted light,
This can be solved by a defect inspection method for a color filter, which is characterized by detecting an over-forming defect or an under-forming defect in the peripheral portion of the color filter by performing image comparison of the images obtained by the I conversion process.

According to the present invention, the above-mentioned problem is S conversion for categorizing image information obtained by irradiating the color filter and the peripheral portion of the color filter with the reflected light and classifying the image information by the saturation degree. Processing and I conversion processing for performing classification according to brightness, and performing arithmetic processing on the output of the S conversion processing and I conversion processing to detect defects due to scratches or protrusions of the color filter. It is solved by the color filter defect inspection method.

According to the present invention, there is provided a method for inspecting a color filter, comprising: a color filter formed on a transparent substrate; and an electrode film formed on at least the color filter. The image information obtained by irradiating the filter and the peripheral part of the color filter, and H-conversion processing for performing classification by hue value is performed on the image information obtained by the reflected light, and one hue value image obtained by the H conversion processing is obtained. The above-mentioned predetermined hue value ranges are used to classify each of the classified images, and the classified images are compared with each other. By performing arithmetic processing on the output of the image comparison, the unevenness of the electrode film is obtained. This is solved by a defect inspection method for a color filter, which is characterized by detecting defects due to

According to the present invention, the above object is to detect an illumination means for irradiating a color filter formed on a transparent substrate and a reflected light or a transmitted light of the light from the illumination means on the color filter. Light detection means for obtaining image information, and an image processing device for image-processing the image information to detect defects in the color filter,
It is solved by a color filter defect inspection device characterized by being provided.

According to the present invention, the above-mentioned problem is solved by a plurality of illuminating means having different illuminances or transmittances for irradiating a color filter formed on a transparent substrate, and reflection of light by the illuminating means on the color filter. A detection means for detecting light or transmitted light to obtain image information, and an image processing device for image-processing the image information to detect a defect of the color filter. Solved by color filter defect inspection equipment.

[0026]

According to the present invention, the defective portion of the color filter is irradiated with illumination having different illuminance or transmittance so that the hue (hue) and the saturation (color) which are the three attributes of the color of the reflected light or the transmitted light are irradiated. Degree) and brightness, the hue, saturation and brightness are obtained by image processing for the image information obtained by this reflected light or transmitted light, and the presence or absence of color filter defects is calculated by calculating these. And the position can be obtained. Therefore, the specific defect can be detected by performing the specific illumination and the image processing suitable for detecting the defect of the color filter.

Since the defective coloring of the color filter shines white in transmitted illumination, the S conversion causes the defective portion to be identified as a non-bright portion, and the I conversion causes the defective portion to be identified as a bright portion, so that the S conversion and the I conversion are performed. By calculating the converted result, only the defective portion can be extracted.

When the light transmittance of the color filter is good and the peripheral portion of the color filter is not transmitted or the light of low light transmittance is radiated, the image information based on the light transmittance is classified by brightness. Since the peripheral part is classified as dark and the normal color filter part is classified as bright, it is possible to detect an excessive or under-formed defect part in the peripheral part of the color filter by comparing the patterns of the areas of the bright part.

Since the defects due to the scratches and protrusions of the color filter shine white in the dark field light, it is possible to extract only the defective portion by calculating the result of S conversion and I conversion as in the case of defective coloring.

Further, the unevenness of the electrode film, for example, the ITO defect is
Since the hue is disturbed in bright field illumination, the image information is
When the color distribution is obtained by performing conversion and the pattern is compared with this hue distribution, the ITO defect portion can be identified because the pattern is different between the ITO defect portion and other normal portions.

[0031]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of an LCD color filter defect inspection apparatus according to the present invention. Figure 1
As shown in FIG. 3, a transmission illumination device 8 for performing transmission illumination on the LCD color filter 5 is arranged in A, and LC
A camera 2 for arranging the light transmitted through the D color filter 5 to be converted into an RGB electric signal is arranged. B
Is provided with a dark-field illumination device 9 for performing dark-field illumination on the LCD color filter 6, and a camera 3 for converting light reflected by the surface of the LCD color filter 6 into RGB electrical signals. Is provided. A bright-field illumination device 10 for performing bright-field illumination on the LCD color filter 7 is provided in C, and a camera 4 for entering the light reflected on the surface of the LCD color filter 7 and converting it into RGB electrical signals. Is provided. Camera 2,
3 and 4 are connected to the image processing apparatus 1.

FIG. 2 is a circuit diagram of the image processing apparatus 1. As shown in FIG. 2, the output from the camera 2 is A / D converted (not shown), and the transmitted light converted into a digital signal is an S conversion processing unit 21, an I conversion processing unit 22, and an I conversion processing unit 2.
It has been entered in 3. The S conversion processing unit 21 converts the transmitted light into S
(Saturation) conversion, then binarization is performed by a predetermined threshold value, and inversion processing is performed to convert vivid portions into "0" and unvivid portions into "1". Further, the I conversion processing units 22 and 23 perform I (Intensity) conversion of the transmitted light and binarize the light according to a predetermined threshold value to “1” for a bright portion and “0” for a dark portion.
Convert to. The S conversion processing unit 21 and the I conversion processing unit 22 are connected to the AND circuit 31. AND circuit 3
A blank defect is output from 1. The I conversion processing unit 23 is connected to the pattern comparison unit 32. The pattern comparison section 32 compares the patterns of the adjacent RGB color painting sections and C
Output r defect.

The output from the camera 3 is A / D converted (not shown), and the dark-field reflected light converted into a digital signal is input to the S conversion processing section 24 and the I conversion processing section 25.
The S conversion processing unit 24 S-converts the dark-field reflected light to convert a vivid portion into “0” and an unvivid portion into “1”.
The I conversion processing unit 25 binarizes the dark field reflected light in luminance. The S conversion processing unit 24 and the I conversion processing unit 25 are the AN
It is connected to the D circuit 33. The AND circuit 33 outputs a flaw / protrusion defect.

The output from the camera 4 is A / D converted (not shown), and the bright field reflected light converted into a digital signal is input to the H conversion processing unit 26. The H conversion processing unit 26 performs H (Hue) conversion of the bright field reflected light to obtain a hue distribution of 0 to 120. The H conversion processing unit 26 is connected to the hue distribution processing unit 27. The hue distribution processing unit 27 calculates the hue distribution from 0 to 4
Divided into 3 bands of 0, 40-80, 80-120 and 2
Quantify. The hue distribution processing unit 27 includes a pattern comparison unit 28,
It is connected to 29 and 30. Pattern comparison unit 28
Performs a pattern comparison of adjacent repeating patterns of "1" or "0" for a hue distribution of 0 to 40, and detects an ITO defect in this band. Similarly, the pattern comparison unit 29 in the band of 40 to 80
0 detects ITO defects in the 80-120 band. The pattern comparison units 28 to 30 are connected to the OR circuit 34. The OR circuit 34 outputs the ITO defect in the entire band of 0 to 120.

Next, a defect inspection method using the LCD color filter defect inspection apparatus described above will be described in detail.

White light defects and Cr defects are inspected by using transmitted illumination in the illumination system. The transmission illumination device 8 shown in FIG. 1 illuminates the LCD color filter 5, and the transmitted light is converted by the camera 2 into R, G, B analog signals. After converting the output from this camera 2 into a digital signal,
It is input to the S conversion processing unit 21, the I conversion processing units 22 and 23. S = (MAX (R, G,
B) -MIN (R, G, B)) / MAX (R, G, B)
Is converted into a binary signal by a predetermined threshold value and inverted to obtain an inverted signal of the S signal. The S signal represents the vividness, and white spots, defective portions and Cr portions are "1",
The R, G, and B color filter units are output as "0". By the I conversion processing unit 22, I = MAX (R, G,
B) and is binarized by a predetermined threshold. The I signal represents luminance, and the white defect portion and the R, G, and B color-painted portions are output as "1", and the Cr portion is output as "0". Therefore, as shown in Table 2, since the whiteout defect is output from the AND circuit 31 as "1", the whiteout defect is discriminated.

[0038]

[Table 2]

Next, the I conversion processing unit 23 is converted into I = MAX (R, G, B) in the same manner as the I conversion processing unit 22, and binarized to a predetermined threshold value. The I signal represents luminance, and the R, G, and B color-painted portions are output as "1" and the Cr portion is output as "0". FIG. 3 shows a pattern comparison part 32 for chromium defect inspection.
It is a figure for explaining. Since the shaded portion is a color filter (R, G, B color coating portion), "1" is output from the I conversion processing unit 23 and "0" is output otherwise. Since the color filters are formed with the same size and the same pitch, the patterns of the adjacent "1" areas, that is, the areas of the color filter are compared, and the difference is taken to obtain Cr.
Defects can be detected.

The dark field illumination is used for the illumination system to inspect for scratches and protrusion defects. LC by the dark field illumination device 9 shown in FIG.
The D color filter 6 is irradiated with the reflected light and the camera 3
Are converted into R, G, B analog signals. The output from the camera 3 is converted into a digital signal and then input to the S conversion processing unit 24 and the I conversion processing unit 25. With dark field illumination, the scratches and protrusion defects appear to glow white. Therefore S
The conversion processing unit 24 performs S conversion similarly to the S conversion processing unit 21, outputs the scratch / projection defect portion and the Cr portion as “1”, and outputs the color filter portion as “0”, and the I conversion processing unit 25 performs I conversion. I conversion is performed similarly to the processing unit 22, and the scratch / projection defect portion and the color filter portion are output as “1” and the Cr portion is output as “0”. Therefore, as shown in Table 3, AN
A scratch / protrusion defect is output from the D circuit 33.

[0041]

[Table 3]

Bright field illumination is used in the illumination system to inspect for ITO defects. The bright field illumination device 10 shown in FIG.
The D color filter 7 is irradiated and the reflected light is reflected by the camera 4.
To convert to R, G, B analog. After converting the output from the camera 4 into a digital signal, the H conversion processing unit 2
Enter in 6. The H conversion processing unit 26 performs H conversion according to the following equation.

When R, G ≧ B, H = (GB) / (R
+ G-2B) When G, B ≧ R H = (BR) / (G + B-2R) When R, B ≧ G H = (RG) / (B + R-2G) FIG. It is a figure for explaining a hue distribution, and an annular hue distribution can be obtained by H conversion with red as 0 and 120, green as 40, and blue as 80. The hue distribution processing unit 27 first classifies the hue distribution obtained by H conversion into the following bands according to the value of H, and binarizes each band.

In the case of 0≤H <40 (I) In the case of 40≤H <80 (II) In the case of 80≤H <120 (III) In each of the cases of (I) to (III), the following 3 Images of types (I ') to (III') are output.

(I ') image ... Pixel in the range of (I) is "1", other pixels are in "0"(II') image ... (II) "1" is the pixel that is present, "0" is the other pixel (III ') image ... "1" is the pixel that falls within the range of (III), and "0" is the other pixel. Is disturbed, so (I ') ~ (II
For each image of I ′), the adjacent repeating pattern is C
Comparison is performed in the same manner as the pattern comparison unit 32 in the r defect detection to detect the ITO defect part in the bands (I) to (III) corresponding to each image.

The pattern comparison unit 28 performs pattern comparison on the image of (I '), and the ITO in the band of (I).
Detect defective parts. The pattern comparison unit 29 performs pattern comparison on the image of (II ′) and detects the ITO defect portion in the band of (II). The pattern comparison unit 30 (II
Pattern comparison is performed on the image of I ′) to detect the ITO defect portion in the band of (III). Pattern comparison unit 2
By taking the OR of the outputs of 8 to 30 by the OR circuit 34, the ITO defect in the entire band of 0 to 120 can be detected.

The OR circuit 35 detects a blank defect, a flaw / protrusion defect, and an ITO defect.

[0048]

As described above, according to the present invention, the defect inspection of the color filter is automated by using image processing, so that a certain level of defect inspection can be performed and the inspection quality and efficiency can be improved. In addition, the quality and yield of color filters can be improved. Further, it is possible to eliminate the visual work of the microscope, and the pain of the operator can be eliminated.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an LCD color filter inspection device according to an embodiment.

FIG. 2 is a configuration diagram of an image processing apparatus according to an embodiment.

FIG. 3 is a diagram for explaining a pattern comparison of Cr defects.

FIG. 4 is a hue distribution by H conversion.

FIG. 5 is a perspective view of an LCD color filter.

FIG. 6 is a diagram for explaining a defect of an LCD color filter.

[Explanation of symbols]

 1 Image Processing Device 2, 3, 4 Camera 5, 6, 7 LCD Color Filter 8 Transmission Illumination Device 9 Dark Field Illumination Device 10 Bright Field Illumination Device 21, 24 S Conversion Processing Unit 22, 23, 25 I Conversion Processing Unit 26 H Conversion processing unit 27 Hue distribution processing unit 28, 29, 30, 32 Pattern comparison unit 31, 33 AND circuit 34, 35 OR circuit 100 Glass substrate 101 Chrome 102 Color filter

Claims (8)

[Claims] 1. A method for inspecting a defect of a color filter formed on a transparent substrate, wherein a plurality of illuminations having different illuminances or transmittances are sequentially applied to the color filter and the peripheral portion of the color filter, and the reflected light or A defect inspection method for a color filter, which comprises inspecting a defect of the color filter by image-processing each image information obtained by transmitted light. 2. A method for inspecting a defect of a color filter formed on a transparent substrate, wherein image information obtained by irradiating the color filter and a peripheral portion of the color filter with specific illumination, and reflected light or transmitted light thereof. A method for inspecting a defect of a color filter, wherein the defect of the characteristic of the color filter is detected by image processing. 3. The color filter and the peripheral portion of the color filter are irradiated with light that is good in transmitted light to the color filter and is not transmitted or has poor transmittance in the peripheral portion of the color filter. , S conversion processing for performing classification by saturation and I conversion processing for performing classification by luminance with respect to image information obtained by the transmitted light, and for output by the S conversion processing and I conversion processing. The defect inspection method for a color filter according to claim 2, wherein a defective color painting of the color filter is detected by performing a calculation process according to the above. 4. The color filter and the peripheral portion of the color filter are irradiated with light having good transmittance for the color filter and not transmitted or poor transmittance for the peripheral portion of the color filter, The image information obtained by the transmitted light is subjected to an I conversion process for performing classification according to brightness, and images obtained by this I conversion process are subjected to image comparison, whereby an excessive formation defect or underformation of the peripheral portion of the color filter is performed. The defect inspection method for a color filter according to claim 2, wherein a defect is detected. 5. The dark field light is applied to the color filter and the peripheral portion of the color filter, and image information obtained by the reflected light is subjected to S conversion processing for performing classification by saturation and classification by brightness. 3. The I-conversion processing for performing the I-conversion processing and the arithmetic processing on the output of the S-conversion processing and the I-conversion processing to detect a defect due to a scratch or a protrusion of the color filter. Color filter defect inspection method. 6. A method for inspecting a defect of a color filter, wherein a color filter is formed on a transparent substrate, and an electrode film is formed on at least the color filter, wherein bright-field light is applied to the color filter and a peripheral portion of the color filter. The image information obtained by irradiating the reflected light is subjected to H conversion processing for performing classification by hue value, and the hue value image obtained by the H conversion processing is provided with one or more predetermined hue values. It is possible to detect defects due to the unevenness of the electrode film by performing classification processing according to the range of 1 to create classification images, performing image comparison on the classification images, and performing arithmetic processing on the output by the image comparison. Characteristic color filter defect inspection method. 7. Illuminating means for irradiating a color filter formed on a transparent substrate, and light for detecting reflected light or transmitted light of the illuminating means by the color filter to obtain image information. A defect inspection apparatus for a color filter, comprising: detection means; and an image processing apparatus for performing image processing on the image information to detect a defect in the color filter. 8. A plurality of illuminating means having different illuminances or transmittances for irradiating a color filter formed on a transparent substrate, and detecting reflected light or transmitted light of the light from the illuminating means on the color filter. A defect inspection apparatus for a color filter, comprising: a detection unit for obtaining image information by means of an image processing apparatus; and an image processing apparatus for image-processing the image information to detect a defect in the color filter.