JP2003262843A - Method and device for inspecting liquid crystal panel - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect unevenness in a turned off state and a surface foreign matter in a liquid crystal panel. <P>SOLUTION: Light transmitted by the liquid crystal panel 1 irradiated with illumination light from a backlight 3 is subjected to image pickup on the upper side of the liquid crystal panel surface with a CCD (charge coupled device) camera 5 disposed on a position in a viewing angle of the liquid crystal panel. An image processing device 7 takes in a picture picked up with the CCD camera 5 and binarizes the picture. Based on the presence or absence of a white picture appearing in the binarized picture, the unevenness in the turned off state such as liquid crystal gap unevenness, an alignment defect or the like and the foreign matter adhering to the liquid crystal panel are detected. Furthermore, based on the presence or absence of a black picture appearing in the binarized picture, the unevenness in the turned off state and the foreign matter adhering to the liquid crystal panel are detected. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a liquid crystal panel inspection method and apparatus capable of performing a visual inspection for detecting non-lighting unevenness of a liquid crystal panel in a non-lighting state and foreign matter adhering to the liquid crystal panel.

[0002]

2. Description of the Related Art In the conventional image processing technique, when inspecting a liquid crystal panel, a CCD camera is installed right above the liquid crystal panel, an image captured by the CCD is captured by an image processing device, and the captured image is analyzed. I made a pass / fail judgment.
However, since the image captured by the CCD camera installed right above the liquid crystal panel is not located in the viewing angle direction where defects and the like can be seen well, there is a problem that the defect contrast is small and detection by image processing is difficult. Further, in the conventional panel surface inspection device disclosed in Japanese Patent Laid-Open No. 11-101712, the image pickup element surface and the lens of the image pickup unit are parallel to the panel surface, and the center of the image pickup element surface of the image pickup unit and the center of the lens are While keeping the state that the straight line connecting the points passes through a predetermined point on the panel surface, the imaging unit and the lens are moved in parallel to capture an image of the panel surface, and the image processing device captures the image captured by the imaging unit and captures it. The image having different visual fields is analyzed to detect various defects such as display unevenness, stains, line defects and point defects.

[0003]

In the conventional panel surface inspection apparatus as described above, the mechanism for capturing images from two directions in order to obtain images with different fields of view is a method of moving the image pickup section and the lens. There is a problem that the configuration of the device becomes complicated. Further, since the viewing angle of the liquid crystal panel is not taken into consideration, the image captured by the image capturing unit is not an image from a position where defects and the like can be seen well, and the defect contrast is small and detection by image processing is difficult. Was what remained.

[0004]

In a liquid crystal panel inspection method and apparatus according to the present invention, a backlight is arranged on the back surface of the liquid crystal panel, and an image pickup means is arranged at the position of the liquid crystal panel viewing angle above the liquid crystal panel surface. Then, the transmitted light of the liquid crystal panel illuminated by the illumination light of the backlight is imaged by the imaging means,
The image processing device takes in the image picked up by the image pickup means, binarizes it, and determines whether or not there is a white image appearing in the binarized image, and it adheres to the liquid crystal gap unevenness, non-lighting unevenness such as alignment failure, and the liquid crystal panel. It detects foreign matter that exists in the binarized image, and detects non-lighting unevenness such as liquid crystal gap unevenness and alignment failure and foreign matter that adheres to the liquid crystal panel based on the presence or absence of a black image that appears in the binarized image. is there.

In this way, the transmitted light of the liquid crystal panel illuminated by the illumination light of the backlight is picked up by the image pickup means arranged at the position of the liquid crystal panel viewing angle above the liquid crystal panel surface, and the image processing device picks up the image. The image picked up by the means is taken in, binarized, and the presence or absence of a white image appearing in the binarized image is used to detect non-lighting unevenness such as liquid crystal gap unevenness, alignment failure and foreign matter adhering to the liquid crystal panel. In addition, liquid crystal gap unevenness, non-lighting unevenness such as alignment failure and foreign matter adhering to the liquid crystal panel are detected based on the presence or absence of a black image appearing in the binarized image. The contrast of non-lighting irregularities such as defects and defects of foreign matter adhering to the liquid crystal panel is large, and non-lighting irregularity and defects of foreign matter adhering to the liquid crystal panel are easy to see,
Moreover, the presence or absence of a white image in the binarized image clearly showing a defect and the contrast of the defect of non-lighting unevenness and the foreign matter adhering to the liquid crystal panel are large, and the non-lighting unevenness and the liquid crystal panel adhere to the liquid crystal panel. Foreign matter defects are easy to see and can be easily detected by the presence or absence of a black image in the binarized image that clearly shows defects. Non-lighting unevenness and liquid crystal gap unevenness, non-lighting unevenness such as alignment defects, and liquid crystal. It has become possible to automate the detection of foreign matter adhering to the panel.

Further, the image processing apparatus performs the white image enhancement filter processing on the image before binarization, so that the white image appearing in the binarized image appears more clearly. Further, the image processing apparatus performs black image enhancement filter processing on the image before binarization, binarizes the black enhanced image, and performs inversion processing so that a black image appearing in the binarized image is generated. It will be a white image and will appear more clearly. It is also possible to determine the size of the non-lighting unevenness or the size of the foreign matter adhering to the liquid crystal panel from the size of the white image.

The liquid crystal panel inspection method and apparatus according to the present invention has a backlight disposed on the back surface of the liquid crystal panel,
The first image pickup means is arranged above the surface of the liquid crystal panel at the viewing angle of the liquid crystal panel, and the second image pickup means is arranged above the surface of the liquid crystal panel at a position opposite to the viewing angle of the liquid crystal panel. The transmitted light of the liquid crystal panel illuminated by the illumination light of the light is arranged above the surface of the liquid crystal panel, the first image pickup means arranged at the position of the liquid crystal panel viewing angle, and the first image pickup means arranged at the position opposite to the liquid crystal panel viewing angle. The image processing device captures the images captured by the first and second image capturing devices, respectively, and reverses the captured image captured by the second image capturing device, and the inverted image. And an image captured by the first image capturing means are subjected to image addition processing, the image subjected to the image addition processing is binarized, and liquid crystal gap unevenness is determined based on the presence or absence of a white image appearing in the binarized image. , Non-lighting unevenness such as defective orientation and foreign matter adhering to the liquid crystal panel are detected, and further non-lighting unevenness such as liquid crystal gap unevenness and alignment defect and liquid crystal panel are detected from the presence or absence of a black image appearing in the binarized image. The foreign matter adhering to the is detected.

As described above, the transmitted light of the liquid crystal panel illuminated by the illumination light of the backlight is opposite to the first image pickup means and the liquid crystal panel viewing angle arranged at the liquid crystal panel viewing angle position above the liquid crystal panel surface. The second image pickup means arranged at a position in the direction picks up the respective images, the image processing apparatus fetches the images picked up by the first and second image pickup means, respectively, and reverses the fetched image picked up by the second image pickup means. The processed and inverted image and the captured image captured by the first imaging means are subjected to image addition processing, the image addition processed image is binarized, and appears in the binarized image. Detects liquid crystal gap unevenness, non-lighting unevenness such as defective alignment, and foreign matter adhering to the liquid crystal panel from the presence or absence of a white image,
Furthermore, the presence or absence of the black image appearing in the binarized image is used to detect liquid crystal gap unevenness, non-lighting unevenness such as alignment failure, and foreign matter adhering to the liquid crystal panel. The image unevenness of the backlight can be removed by adding the inverted image of the image taken by the second image pickup means to the image taken by the second image pickup means, and the liquid crystal gap unevenness, the non-lighting unevenness such as the alignment failure and the liquid crystal panel adhere The defect of the foreign matter present is easy to see, and can be easily detected by the presence or absence of a white image and a black image appearing in the binarized image that more clearly shows the defect. Furthermore, it has become possible to automate the detection of non-lighting unevenness and foreign matter adhering to the liquid crystal panel.

Further, the image processing apparatus performs the white image enhancement filter process on the image before binarization, so that the white image appearing in the binarized image appears more clearly. Further, the image processing apparatus performs black image enhancement filter processing on the image before binarization, binarizes the black enhanced image, and performs inversion processing so that a black image appearing in the binarized image is generated. It will be a white image and will appear more clearly. It is also possible to determine the non-lighting unevenness and the size of the foreign matter adhering to the liquid crystal panel from the size of the white image.

[0010]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiment 1. 1 is a block diagram showing a configuration of a liquid crystal panel inspection apparatus according to Embodiment 1 of the present invention, FIG. 2 is a flowchart showing an operation of the liquid crystal panel inspection apparatus, and FIG. 3 is a non-lighting unevenness of the liquid crystal panel inspection apparatus. It is explanatory drawing of the image processing to detect. In FIG. 1, 1
Is a liquid crystal panel installed on the panel setting jig 2, 3 is a backlight for illuminating the backlight of the liquid crystal panel 1, and 4 is a backlight.
Is a backlight lighting device for lighting the backlight 3, 5
Is a first CCD camera which is an image pickup means and is arranged above the liquid crystal panel 1 at a liquid crystal panel viewing angle α, and 6 is above the liquid crystal panel 1 and in a direction opposite to the position of the liquid crystal panel viewing angle α. It is a second CCD camera which is a second image pickup means arranged at a position. The liquid crystal panel viewing angle α of the liquid crystal panel 1 is substantially the same for liquid crystal panels of the same product, but different for liquid crystal panels of different products. 7
Is an image processing device that performs image processing described later to determine whether the screen of the liquid crystal panel 1 is good or bad, and drives and controls the CCD camera 5, and 8 is a monitor that displays the image processed by the image processing device 7.

Next, the operation of the liquid crystal panel inspection apparatus according to the first embodiment of the present invention will be described with reference to the flowcharts of FIGS. First, the liquid crystal panel 1 to be inspected is set on the panel setting jig 2 and the inspection is started (step S1). Next, the backlight lighting device 4 lights the backlight 3 arranged on the back side of the liquid crystal panel 1 (step S2). After that, the image processing device 7 is connected to the CC arranged at the position of the liquid crystal panel viewing angle α.
The non-illuminated image of the liquid crystal panel 1 with only backlight illumination captured by the camera 5 is captured (step S3). The non-lighted image is taken in while the liquid crystal panel 1 is backlit, but the drive voltage is not applied to the liquid crystal panel 1.

Next, the image processing device 7 which has taken in the non-lighted image picked up by the CCD camera 5 detects the non-lighting unevenness of the liquid crystal panel 1 and the foreign matter adhering to the liquid crystal panel by the image processing described below. (Step S
4). Here, non-lighting unevenness refers to liquid crystal gap unevenness, alignment failure, and the like. First, in the image processing for detecting the non-lighting unevenness by the image processing device 7, the captured non-lighting image P1 shown in (a) of FIG. 3 is subjected to white enhancement filter processing, and the white emphasis image P2 shown in (b) of FIG. Then, the white-enhanced image P2 is binarized to obtain a binarized image P3 shown in (c) of FIG.

As described above, the white image P1a which appears thin and white in the non-lighted image P1 shown in FIG. 3A captured by the image processing device 7 is the non-lighting unevenness due to the liquid crystal gap unevenness and the alignment failure and the liquid crystal panel. This indicates that the adhered foreign matter is defective. The non-illuminated image P1 shown in FIG. 3A is subjected to the white enhancement filter process on the non-illuminated image P1.
Image P1a that looks thin and white due to the backlight illumination of
This is to emphasize more. Further, binarizing the white-enhanced image P2 that has been subjected to the white-enhancement filtering process shown in FIG. 3B is performed between the defective portion P2a of the white-enhanced image P2 and other normal portions of the black-saturated image. This is to clarify the difference.

Therefore, it is determined whether or not the binarized image P3 has a white image P3a. If there is a white image P3a, the number of white pixels is measured. Note that the binarized image P
3 is displayed on the monitor 8 to see the inspection progress. The number of pixels of the white image P3a in the binarized image P3 is 4 to
If there are 5 pixels, it is determined that the liquid crystal panel 1 has a non-lighting unevenness defect and a foreign matter defect adhering to the liquid crystal panel. If it is 4 pixels or less, it is due to noise or the like, and the liquid crystal panel 1 has a non-lighting unevenness defect. Also, it is determined that there is no foreign matter defect adhering to the liquid crystal panel. Then, the size of the non-lighting unevenness defect and the defect of the foreign matter adhering to the liquid crystal panel is determined based on the number of pixels of the white image. Here, this liquid crystal panel 1
The total number of pixels that have been imaged is about 1 million pixels, and it depends on the size of the liquid crystal panel 1, but since one pixel is about 40 microns, the non-lighting unevenness and the size of foreign matter adhering to the liquid crystal panel are That is 160 to 200 microns.

Further, the image processing device 7 which has taken in the non-lighted image P1 picked up by the CCD camera 5 detects foreign matter adhering to the liquid crystal panel 1 by the image processing described below (step S4). In the image processing for detecting the foreign matter adhering to the liquid crystal panel 1 by the image processing device 7, the captured non-lighted image P1 shown in (a) of FIG. 3 is subjected to black enhancement filter processing, and shown in (d) of FIG. A black-enhanced image P4, and the black-enhanced image P4 is binarized and inverted, and the binarized and inverted image P shown in FIG.
Set to 5. As described above, the black image P1b that appears light black in the non-lighted image P1 shown in FIG. 3A captured by the image processing device 7 indicates that the foreign matter is attached to the liquid crystal panel 1.

The reason why the non-illuminated image P1 is subjected to the black enhancement filter process is to further enhance the black image P1b which looks light black due to the backlight illumination of the non-illuminated image P1 shown in FIG. Further, binarization of the black-enhanced image P4 that has been subjected to the black-enhancement filtering process shown in FIG. 3D is performed by the difference between the defective portion P4b of the black-enhanced image P4 and the normal portion of the other blackish image. This is to clarify. Further, the inversion process is performed so that the black image P4b in the black emphasized image P4 becomes a white image for easy viewing.

Therefore, it is determined whether or not the binarized and inverted image P5 has a white image P5b. White image P5b
If there is, the number of white pixels is measured. The binarized image P5 is displayed on the monitor 8 to see the progress of the inspection. The white image P5b in the image P5 that has been binarized and inverted.
4 to 5 pixels, it is determined that the liquid crystal panel 1 has a non-lighting non-uniformity defect and a foreign matter defect adhering to the liquid crystal panel 1, and if it is 4 pixels or less, it is considered to be due to noise or the like. It is determined that the panel 1 has no non-lighting non-uniformity defect and no defect of foreign matter adhering to the liquid crystal panel 1. Then, the size of the non-lighting unevenness defect and the defect of the foreign matter adhering to the liquid crystal panel 1 is determined based on the number of pixels of the white image. Here, the total number of pixels of the liquid crystal panel 1 is about 100.
Although it is reasonable and depending on the size of the liquid crystal panel 1, one pixel is about 40 microns, the size of the non-lighting unevenness and the size of the foreign matter adhering to the liquid crystal panel 1 is 160 to 200 microns. Become.

When the detection processing of the lighting unevenness of the liquid crystal panel 1 and the foreign matter adhering to the liquid crystal panel 1 is completed in this way, the image processing device 7 checks the non-lighting unevenness and the liquid crystal panel 1.
The pass / fail judgment result which is the detection result of the foreign matter adhering to is output and displayed on the monitor 8 (step S5). When the display of the pass / fail judgment result on the monitor 8 is completed, the image processing device 7 outputs an inspection completion signal (step S6), and the inspection completion is displayed on the monitor 8. Thus, one liquid crystal panel 1
When the inspection of 1 is completed, the liquid crystal panel 1 is removed from the panel setting jig 2 and the next liquid crystal panel 1 to be inspected is set on the panel setting lamp jig 2 to perform the next inspection. .

In the first embodiment, the non-illuminated image captured by the CCD camera 5 arranged at the viewing angle α of the liquid crystal panel is captured from the viewing angle α direction. It is easy to see the defects of foreign matter. Furthermore, since the non-illuminated image captured by the CCD camera 5 is subjected to white enhancement filter processing and binarization processing, non-illumination unevenness and foreign matter adhering to the liquid crystal panel 1 appear clearly in the binarized image. Become. In addition, the non-illuminated image captured by the CCD camera 5 is subjected to black enhancement filter processing, and 2
The binarized image clearly shows non-lighting nonuniformity and foreign matter adhering to the liquid crystal panel 1 because the binarized image is inverted.

Embodiment 2. FIG. 4 is a block diagram showing the configuration of the liquid crystal panel inspection device according to the first embodiment of the present invention,
FIG. 5 is a flow chart showing the operation of the liquid crystal panel inspection apparatus, and FIG. 6 is an explanatory diagram of image processing for removing non-illumination unevenness by removing illumination unevenness of the liquid crystal panel inspection apparatus. In FIG. 4, 1 is a liquid crystal panel installed on the panel setting jig 2, 3 is a backlight for illuminating the backlight of the liquid crystal panel 1, 4 is a backlight lighting device for lighting the backlight 3, and 5 is the liquid crystal panel 1. The first CCD camera, which is the image pickup means, is arranged above the liquid crystal panel at the viewing angle α, and is arranged above the liquid crystal panel 1 at a position opposite to the viewing angle α of the liquid crystal panel. The second CCD camera, which is the second image pickup means, performs image processing described later,
An image processing device that determines the quality of the screen of the liquid crystal panel 1 and drives and controls the first and second CCD cameras 5 and 6, and 8 is a monitor that displays an image processed by the image processing device 7. In the second embodiment, unlike the first embodiment,
A second CCD camera 6 is provided in addition to the first CCD camera 5 for removing illumination unevenness, and image addition processing is added in image processing.

Next, the operation of the liquid crystal panel inspection device according to the second embodiment of the present invention will be described with reference to the flowcharts of FIGS. First, the liquid crystal panel 1 to be inspected is set on the panel setting jig 2 and the inspection is started (step S11). Next, the backlight lighting device 4 lights the backlight 3 arranged on the back side of the liquid crystal panel 1 (step S12). After that, the image processing device 7 captures the non-illuminated image P11 of the liquid crystal panel 1 only for backlight illumination, which is captured by the first CC camera 5 arranged at the viewing angle of the liquid crystal panel (step S
13) Further, the non-lighted image P12 of the liquid crystal panel 1 only for backlight illumination captured by the second CC camera 6 arranged at the position opposite to the position of the liquid crystal panel viewing angle α is captured (step S14). These non-illuminated images P1
1 and P12 are taken in while the liquid crystal panel 1 is backlit and the drive voltage is not applied to the liquid crystal panel 1.

Next, CCD camera 5 and CCD camera 6
Non-illuminated images P11 and P12 respectively captured by
In the image processing device 7 that has taken in, the illumination unevenness is first removed by the image processing described below (step S15).
The image processing for removing the uneven illumination by the image processing device 7 is
The non-lighted image P12 captured from the second CCD camera 6 shown in FIG. 6 (a) is subjected to the reversal processing to form the reversed image P13 shown in FIG. 6 (b), and the reversed image P13 and FIG. The non-illuminated image P11 captured from the first CCD camera 5 shown in a) is subjected to image addition processing to obtain an added image P14 shown in (d) of FIG. In this way, the second CCD
The non-illuminated image P12 captured from the camera 6 is inverted, and the inverted image P13 and the non-illuminated image P11 captured from the first CCD camera 5 are subjected to image addition processing for the following reason.

Since the non-lighted image P11 captured from the first CCD camera 5 is an image captured from the viewing angle α direction of the liquid crystal panel, defects are easy to see and uneven illumination is also visible. A to non-illuminated image P11 shown in the lower part of FIG.
The profile between b shows this. Further, the non-lighted image P12 captured from the second CCD camera 6 is an image captured from the direction opposite to the liquid crystal panel viewing angle α direction of the liquid crystal panel viewing angle α of the first CCD camera 5, so that a defect is hard to see. , Uneven illumination is visible. The profile between a and b of the non-illuminated image P12 shown in the lower part of FIG. 6A shows this.

Then, the non-lighted image P1 captured from the second CCD camera 6 in which defects are difficult to see and uneven illumination appears
2 is reversed, a to b of the reversed non-lighted image P13
Illumination unevenness of the profile in the interval has a tendency of brightness opposite to that shown in the lower part of FIG. Therefore, if the inverted non-illuminated image P13 and the non-illuminated image P11 captured by the first CCD camera 5 are subjected to the image addition processing, the illumination unevenness is canceled and only the defect is extracted. A to non-illuminated image P14 shown in the lower part of FIG.
The profile between b shows this.

In the image processing device 7 for the added image P14 in which only the defects are extracted in this way, the non-lighting unevenness of the liquid crystal panel 1 and the defect of the foreign matter adhering to the liquid crystal panel 1 by the image processing described below. Is detected (step S
16). First, the image processing for detecting non-lighting unevenness and foreign matter adhering to the liquid crystal panel 1 by the image processing device 7 is as follows.
The added image P14 shown in (d) of FIG. 6 is subjected to a white enhancement filter process to obtain a white enhanced image P15 shown in (e) of FIG. 6, and the white enhanced image P15 is binarized to obtain a white enhanced image P15 of FIG. ), Which is a binarized image P16. As described above, the white image P14a that looks thin and white in the added image P14 shown in FIG. 6D is the non-lighting unevenness defect due to the liquid crystal gap unevenness and the alignment defect and the foreign matter defect adhering to the liquid crystal panel 1. Is shown. The reason why the added image P14 is subjected to the white enhancement filter process is to further enhance the white image P14a that looks pale white in the added image P14 shown in FIG. 6D. Further, binarizing the white-enhanced image P15 that has been subjected to the white-enhancement filtering process shown in (e) of FIG. 6 is based on the difference between the defective portion P15a of the white-enhanced image P15 and the normal portion of the other blackish image. This is to clarify.

Therefore, it is determined whether or not the binarized image P16 has a white image P16a. White image P16a
If there is, the number of white pixels is measured. The binarized image P16 is displayed on the monitor 8 to see the progress of the inspection. If the number of pixels of the white image P16a in the binarized image P16 is 4 to 5 pixels, it is determined that the liquid crystal panel 1 has the non-lighting unevenness defect and the foreign matter defect adhering to the liquid crystal panel 1, and 4 pixels are determined. If it is below, it is determined that there is no non-lighting non-uniformity defect in the liquid crystal panel 1 and a defect of foreign matter adhering to the liquid crystal panel 1 as a result of noise or the like. And
The size of the non-lighting unevenness defect and the defect of the foreign matter adhering to the liquid crystal panel 1 is determined by the number of pixels of the white image. Here, the total number of pixels of the liquid crystal panel 1 captured is about 1 million pixels, and although it depends on the size of the liquid crystal panel 1, one pixel is about 40 μm.
The size of the foreign matter adhering to is about 160 to 200 microns.

Further, with respect to the added image P14 in which only the defects have been extracted in this way, the image processing device 7 performs the image processing described below to perform non-lighting unevenness of the liquid crystal panel 1 and foreign matter attached to the liquid crystal panel 1. The defect is detected (step S16). The image processing for detecting the non-lighting unevenness of the liquid crystal panel 1 and the foreign matter adhering to the liquid crystal panel 1 by the image processing device 7 is performed by black enhancement filter processing of the added image P14 shown in FIG. A black-enhanced image P17 shown in (g) is obtained, and the black-enhanced image P17 is further binarized and subjected to an inversion process to obtain an binarized and inverted image P18 shown in (h) of FIG.

As described above, the black image P14b that appears light black in the added image P14 shown in FIG. 6D indicates that the liquid crystal panel 1 is not illuminated and the foreign matter is attached to the liquid crystal panel 1. ing. The addition image P14 is subjected to the black enhancement filter processing in order to further enhance the black image P14b that looks light black in the addition image P14 shown in FIG. 5D. Further, binarizing the black-enhanced image P17 subjected to the black-enhancement filtering process shown in (g) of FIG. 6 is performed between the defective portion P17b of the black-enhanced image P17 and the normal portion of the other black-saturated image. This is to clarify the difference. Further, the inversion processing is performed so that the black image in the black-enhanced image P17 is changed to a white image for easy viewing.

Then, the binarized and inverted image P1
It is determined whether or not there is a white image P18b in 8. If there is a white image P18b, the number of white pixels is measured. 2
The binarized image P18 is displayed on the monitor 8 to see the progress of the examination. This binarized and inverted image P1
If the number of pixels of the white image P18b in 8 is 4 to 5 pixels, it is determined that the liquid crystal panel 1 has a non-lighting unevenness defect and a defect of foreign matter adhering to the liquid crystal panel 1, and if 4 pixels or less,
It is determined that there is no non-lighting non-uniformity defect in the liquid crystal panel 1 and a foreign matter defect adhering to the liquid crystal panel 1 due to noise or the like. Then, the size of the non-lighting unevenness defect and the defect of the foreign matter adhering to the liquid crystal panel 1 is determined based on the number of pixels of the white image. Here, the total number of pixels of the liquid crystal panel 1 captured is about 1 million pixels, and although it depends on the size of the liquid crystal panel 1, since one pixel is about 40 microns, non-lighting unevenness and adhesion to the liquid crystal panel 1 occur. The size of the foreign matter is 160 to 200 microns.

When the detection processing of the lighting irregularity of the liquid crystal panel 1 and the surface foreign matter is completed in this way, the image processing device 7 outputs the pass / fail judgment result which is the detection result of the non-lighting irregularity and the foreign matter adhering to the liquid crystal panel 1. It is displayed on the monitor 8 (step S17). When the display of the quality determination result on the monitor 8 is completed, the image processing device 7 outputs an inspection completion signal (step S18) and causes the monitor 8 to display the inspection completion. When the inspection of one liquid crystal panel 1 is completed in this way, the liquid crystal panel 1 is removed from the panel setting jig 2, and the next liquid crystal panel 1 to be inspected is set on the panel setting lamp jig 2 to perform the next inspection. Will be performed.

In the second embodiment, the non-illuminated image P12 captured from the second CCD camera 6 is inverted, and the inverted image P13 and the non-illuminated image P11 captured from the first CCD camera 5 are imaged. The added image that has been subjected to the addition processing is one in which illumination unevenness is removed, and it is easy to see defects such as astigmatism and unevenness and surface foreign matter. Further, since the non-illuminated image captured by the CCD camera 5 is subjected to white enhancement filter processing and binarization processing, non-illumination unevenness and foreign matter adhering to the liquid crystal panel appear clearly in the binarized image. . Further, since the non-lighted image captured by the CCD camera 5 is subjected to the black enhancement filter processing, binarized and inverted, the non-lighted unevenness and the foreign matter adhering to the liquid crystal panel 1 clearly appear in the binarized image. It will be.

In the above-mentioned embodiment, the liquid crystal panel 1 is the object of inspection, but it goes without saying that other than the liquid crystal panel 1, a display having a response such as electric luminescence can also be the object of inspection.

[0033]

As described above, according to the liquid crystal panel inspection method and apparatus of the present invention, the transmitted light of the liquid crystal panel irradiated by the illumination light of the backlight is above the surface of the liquid crystal panel.
The liquid crystal gap is detected by the image pickup device arranged at the liquid crystal panel viewing angle, the image processing device takes in the image picked up by the image pickup device, binarizes the image, and determines whether or not a white image appears in the binarized image. Non-lighting unevenness such as unevenness and poor orientation, and foreign matter adhering to the liquid crystal panel are detected, and the non-lighting unevenness of the liquid crystal panel and adhesion to the liquid crystal panel are detected based on the presence or absence of a black image appearing in the binarized image. Since it is designed to detect foreign matter that is present, the image captured from the viewing angle direction makes it easier to see the non-lighting unevenness such as liquid crystal gap unevenness and alignment failure, and the defects of foreign matter adhering to the liquid crystal panel, and more clearly. A white image in the binarized image showing a defect, and
It is possible to easily detect the presence or absence of a black image in the binarized image that more clearly shows a defect, and it is possible to automate the detection of non-lighting unevenness and foreign matter adhering to the liquid crystal panel.

Further, the image processing apparatus performs the white image enhancement filter process on the image before binarization, so that the white image appearing in the binarized image appears more clearly. Further, the image processing apparatus performs black image enhancement filter processing on the image before binarization, binarizes the black enhanced image, and performs inversion processing so that a black image appearing in the binarized image is generated. It will be a white image and will appear more clearly. It is also possible to determine the non-lighting unevenness and the size of the foreign matter adhering to the liquid crystal panel from the size of the white image.

According to another method and apparatus for inspecting a liquid crystal panel of the present invention, the transmitted light of the liquid crystal panel irradiated by the illumination light of the backlight is arranged above the surface of the liquid crystal panel at the position of the liquid crystal panel viewing angle. The first image pickup means and the second image pickup means arranged at a position opposite to the viewing angle of the liquid crystal panel respectively pick up images, and the image processing device fetches the images picked up by the first and second image pickup means, respectively. The captured image captured by the second image capturing means is subjected to inversion processing, the image subjected to the inversion processing and the captured image captured by the first image capturing means are subjected to image addition processing, and the image subjected to the image addition processing is binarized. Processing, the presence or absence of a white image appearing in the binarized image, liquid crystal gap unevenness, non-lighting unevenness such as alignment failure, and foreign matter adhering to the liquid crystal panel are detected, and the binarized image is further detected. Since non-lighting unevenness and foreign matter adhering to the liquid crystal panel are detected based on the presence or absence of the black image appearing on the screen, the image captured from the viewing angle direction and the reverse image of the image captured from the opposite viewing angle are added. Since it is processed, uneven illumination can be removed, unevenness of liquid crystal gap, non-illumination unevenness such as defective alignment, and defects of foreign matter adhering to the liquid crystal panel are easy to see, and more clearly the white image in the binarized image And it can be easily detected by the presence or absence of a black image in the binarized image that more clearly shows the defect,
This has the effect of automating the detection of non-lighting unevenness and foreign matter adhering to the liquid crystal panel.

Further, the image processing apparatus performs the white image enhancement filter processing on the image before binarization, so that the white image appearing in the binarized image appears more clearly. Further, the image processing apparatus performs black image enhancement filter processing on the image before binarization, binarizes the black enhanced image, and performs inversion processing so that a black image appearing in the binarized image is generated. It will be a white image and will appear more clearly. It is also possible to determine the non-lighting unevenness and the size of the foreign matter adhering to the liquid crystal panel from the size of the white image.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a liquid crystal panel inspection device according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing an operation of the liquid crystal panel inspection device.

FIG. 3 is an explanatory diagram of image processing for detecting non-lighting unevenness and foreign matter adhering to the liquid crystal panel of the liquid crystal panel inspection device.

FIG. 4 is a block diagram showing a configuration of a liquid crystal panel inspection device according to a second embodiment of the present invention.

FIG. 5 is a flowchart showing an operation of the liquid crystal panel inspection device.

FIG. 6 is an explanatory diagram of image processing for removing non-lighting unevenness and foreign matter adhering to the liquid crystal panel by removing uneven lighting of the liquid crystal panel inspection apparatus.

[Explanation of symbols]

1 liquid crystal panel, 2 panel setting jig, 3 backlight, 4 backlight lighting device, 5 CCD camera (imaging means), 7 image processing device

Continued front page    F term (reference) 2F065 AA49 BB02 CC25 DD09 FF02                       GG18 HH13 HH15 JJ03 JJ05                       JJ08 JJ26 NN01 PP11 QQ00                       QQ04 QQ24 QQ27 QQ33 RR06                       SS02 SS04 SS13                 2G051 AA73 AB01 AB02 AB06 AB07                       CA04 CA07 CA08 CB02 EA08                       EA11 EA16 EC05 ED14 ED21                 2G086 EE10                 2H088 FA12 HA28 MA20                 2H091 FA41Z FC30 LA30

Claims (19)

[Claims] 1. A step of arranging a backlight on a back surface of a liquid crystal panel, a step of arranging an image pickup means at a position of a liquid crystal panel viewing angle above a surface of the liquid crystal panel, and a liquid crystal illuminated by illumination light of the backlight. The step of capturing the transmitted light of the panel with the image capturing means, and the image processing apparatus takes in the image captured by the image capturing means, binarizes it, and determines whether or not there is a white image appearing in the binarized image. Detects non-lighting unevenness such as defects and foreign substances adhering to the liquid crystal panel, and determines whether there is a black image appearing in the binarized image, and determines the liquid crystal gap unevenness, non-lighting unevenness such as alignment failure, and the liquid crystal panel. A method for inspecting a liquid crystal panel, comprising: a step of detecting foreign matter adhering thereto. 2. The liquid crystal panel inspection method according to claim 1, wherein the image processing device performs white image enhancement filter processing on an image before binarization. 3. The image processing apparatus performs black image enhancement filter processing on an image before binarization, binarizes the black enhanced image, and performs inversion processing. Liquid crystal panel inspection method described. 4. The liquid crystal panel inspection method according to claim 1, wherein the size of the non-lighting unevenness or the size of a foreign substance adhering to the liquid crystal panel is determined from the size of the white image. 5. A step of arranging a backlight on the back surface of the liquid crystal panel, a step of arranging the first imaging means at a position of a liquid crystal panel viewing angle above the surface of the liquid crystal panel, and a step above the surface of the liquid crystal panel. The step of disposing the second image pickup means at a position opposite to the position of the liquid crystal panel viewing angle, and the transmitted light of the liquid crystal panel illuminated by the illumination light of the backlight is imaged by the first and second image pickup means. A process, a process in which the image processing device captures images captured by the first and second image capturing devices, respectively, and a process in which the image captured by the second image capturing device captured by the image processing device is subjected to inversion processing and the inverted image And an image captured by the first image capturing means are subjected to image addition processing, the image subjected to the image addition processing is binarized, and liquid crystal gap unevenness is determined based on the presence or absence of a white image appearing in the binarized image. , Distribution Detects non-lighting unevenness such as defects and foreign matter adhering to the liquid crystal panel, and further determines whether or not there is a black image appearing in the binarized image, liquid crystal gap unevenness, non-lighting unevenness such as alignment failure, and liquid crystal panel. And a step of detecting foreign matter adhering to the liquid crystal panel. 6. The liquid crystal panel inspection method according to claim 5, wherein the image processing device performs white image enhancement filter processing on an image before binarization. 7. The liquid crystal panel according to claim 5, wherein the image processing device performs a black image enhancement filter process on an image before binarization, and performs an inversion process on the black enhanced image. Inspection method. 8. The liquid crystal panel inspecting method according to claim 5, wherein the size of the non-lighting unevenness and the size of the foreign matter adhering to the liquid crystal panel are determined from the size of the white image. 9. The liquid crystal panel inspection method according to claim 1, wherein the image pickup means is a CCD camera. 10. A backlight for illuminating the back surface of a liquid crystal panel, an image pickup means arranged at a position of a liquid crystal panel viewing angle above the surface of the liquid crystal panel, and an image picked up by the image pickup means is captured and binarized. ,
From the presence or absence of white image appearing in the binarized image, liquid crystal gap irregularity, non-lighting irregularity such as alignment defect, and foreign matter adhering to the liquid crystal panel are detected, and the black appearing in the binarized image is detected. A liquid crystal panel inspection device, comprising: an image processing device that detects liquid crystal gap unevenness, non-lighting unevenness such as defective alignment, and foreign matter adhering to the liquid crystal panel based on the presence or absence of an image. 11. The liquid crystal panel inspection device according to claim 10, wherein the image processing device has a white image enhancement filter for performing a white image enhancement filter process on an image before binarization. 12. The image processing device comprises a black image enhancement filter for performing a black image enhancement filter process on an image before binarization, and an inverting means for inverting the black enhanced image. The liquid crystal panel inspection device according to claim 10. 13. The liquid crystal panel inspection apparatus according to claim 10, wherein the size of the non-lighting unevenness or the size of a foreign substance on the surface of the liquid crystal panel is determined from the size of the white image. 14. A backlight for illuminating the back surface of a liquid crystal panel, a first image pickup means arranged at a position of a liquid crystal panel viewing angle above the surface of the liquid crystal panel, and a liquid crystal panel viewing field above the surface of the liquid crystal panel. The second image pickup means arranged at a position opposite to the corner, and the transmitted light of the liquid crystal panel illuminated by the illumination light of the backlight are captured and captured by the first and second image pickup means, respectively. The image taken by the second image pickup means is subjected to inversion processing, and the image subjected to the inversion processing and the image taken by the first image pickup means are subjected to image addition processing, and the image subjected to the image addition processing is binarized. After processing, the presence or absence of a white image that appears in the binarized image is used to detect liquid crystal gap irregularities, non-lighting irregularities such as alignment defects, and foreign matter adhering to the liquid crystal panel. Appearance A liquid crystal panel inspection device, comprising: an image processing device that detects non-lighting unevenness such as liquid crystal gap unevenness, alignment failure and the like, and foreign matter adhering to the liquid crystal panel based on the presence or absence of a black image. 15. The liquid crystal panel according to claim 14, wherein the image processing device has a white image enhancement filter for performing a white image enhancement filter process on an image subjected to the image addition process before binarization. Inspection device. 16. The image processing apparatus has a black image enhancement filter for performing a black image enhancement filter process on an image that has been subjected to image addition processing before binarization, and an inversion means for performing an inversion process on the black enhanced image. 15. The method according to claim 14, wherein
The liquid crystal panel inspection device described. 17. The liquid crystal panel inspection apparatus according to claim 14, wherein the size of the non-lighting unevenness or the size of a foreign substance adhering to the liquid crystal panel is determined from the size of the white image. 18. The first and second image pickup means are CCDs.
The liquid crystal panel inspection device according to claim 10 or 14, which is a camera. 19. A monitor is provided, and the image processing device has an image captured by the imaging means on the monitor, an image subjected to a white image enhancement filter process, an image subjected to a black image enhancement filter process, a binarized image, or a binary image. 11. The image and the inspection result which have been converted into an inverted image are displayed.
The liquid crystal panel inspection device according to any one of 11, 12, 14, 15 and 16.