KR20140082334A

KR20140082334A - Method and apparatus of inspecting mura of flat display

Info

Publication number: KR20140082334A
Application number: KR1020120152184A
Authority: KR
Inventors: 히데오 나카야; 츠토무 이치카와
Original assignee: 엘지디스플레이 주식회사
Priority date: 2012-12-24
Filing date: 2012-12-24
Publication date: 2014-07-02

Abstract

A device for inspecting mura of a flat display according to the present invention comprises a camera to photograph a display panel in an emission state; and an image processing unit to detect a mura defect of the display panel by receiving and processing an image photographed by the camera, wherein the image processing unit includes a preprocessing unit which divides the input image into blocks and generates a maximum value-based reduced image and/or a minimum value-based reduced image by extracting a maximum value and/or a minimum value of the block.

Description

FIELD OF THE INVENTION [0001] The present invention relates to a method and an apparatus for inspecting a flat display,

More particularly, the present invention relates to a method and apparatus for detecting a stain on a flat panel display, which can reduce the cost of the manufacturing process and the efficiency of the inspection process by automatically detecting the stain .

2. Description of the Related Art [0002] As an information-oriented society develops, demands for a display device for displaying an image have increased in various forms. Recently, a liquid crystal display (LCD), a plasma display panel (PDP) Various flat displays such as organic light emitting diodes (OLEDs) have been utilized.

Flat panel displays can have unevenness in brightness and color due to various factors. The inspection of the stain was generally carried out visually by the inspector. In such a case, automation of the stain inspection has been required since it takes a lot of time, manpower and costs.

As a method of automating the inspection of stains, conventionally, a method has been proposed in which a display in a light emitting state is photographed by a camera and processing for the photographed image is performed to perform smear detection. Such a conventional stain inspection method comprises a preprocessing process for a photographed image, a defect candidate extraction process, and a determination process.

Smearing defects on the display are often seen as subtle changes in luminance. Incidentally, since the display has a structure in which the positions of pixels are fixed, the pixel structure is reflected in shadows in the camera image for inspection. Therefore, it becomes considerably difficult to detect a spot defect through an image in such a state.

In order to improve this, there is a method in which a pixel structure is not shown using a filter of a predetermined band in a preprocessing process, and only the contrast information displayed in each pixel is extracted. However, depending on the use of the filter, the characteristic of the spot defect itself is reduced, which makes it difficult to carry out the subsequent process. In addition, it is necessary to adjust the characteristics of the filter and the defect detection algorithm according to the types of defects and displays, thereby complicating the process.

As described above, the conventional inspection method of stains lowers the efficiency, slows the detection speed, and increases the display manufacturing cost.

A problem to be solved by the present invention is to provide a method for achieving high efficiency and high speed of stain inspection and reducing display manufacturing cost.

According to an aspect of the present invention, there is provided a display device comprising: a camera for photographing a display panel in a light emitting state; And an image processing unit for receiving and processing an image photographed by the camera and detecting an unevenness defect in the display panel, wherein the image processing unit divides the input image into blocks; And a preprocessing unit for extracting a maximum value and / or a minimum value of the block and generating a maximum value reduced image and / or a minimum value reduced image.

Here, in generating the maximally reduced image, the block may be arranged such that the center of the block coincides with the center of the pixel of the display panel.

In generating the minimum value reduced image, the block may be arranged so that the center of the block coincides with the boundary of four adjacent pixels of the display panel.

The block may be configured to have a size capable of covering pixels of the display panel.

Wherein the image processing unit comprises: a defect candidate extracting unit for binarizing the maximum value reduced image and / or the minimum value reduced image based on a set threshold value and extracting a defect candidate area based on the binarized image; And a determination unit for calculating a feature amount for the extracted defect candidate region and determining whether or not the defect candidate region is defective based on the calculated feature amount.

In another aspect, the present invention provides a method of manufacturing a display device, comprising: photographing a display panel in a light emitting state through a camera; And an image processing step of receiving and processing an image photographed by the camera to detect an unevenness defect in the display panel, the image processing step including the steps of: dividing the input image into blocks; And extracting a maximum value and / or a minimum value of the block to generate a maximum value reduced image and / or a minimum value reduced image.

Wherein the image processing step includes binarizing the maximum reduced image and / or the minimum reduced image with reference to a threshold value, and extracting a defect candidate area based on the binarized image; Calculating a feature amount for the extracted defect candidate region, and determining whether the defect candidate region is defective based on the calculated feature amount.

According to the present invention, by generating at least one of the maximum / minimum value reduced images, the pixel structure in the input image photographed by the camera can be effectively removed, and also the spot defect can be properly preserved.

Therefore, the pixel structure removal and defect candidate extraction processes can be simplified and speeded up to a considerable extent as compared with the conventional method.

As a result, the image processing process for detecting a spot defect can be simplified and speeded up, thereby reducing the manufacturing cost of the display device.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram schematically illustrating a flat panel display stain test apparatus according to an embodiment of the present invention. FIG.
2 is a diagram showing luminance waveforms of a part of an input image and images according to a processing procedure of a preprocessing unit according to an embodiment of the present invention;
Figures 3 and 4 are block diagrams for generating a minimized reduced image and a minimized reduced image in accordance with an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

1 is a block diagram schematically illustrating a flat panel display unevenness testing apparatus according to an embodiment of the present invention.

Referring to FIG. 1, a stain testing apparatus 100 according to an embodiment of the present invention may include a camera 110 and an image processing unit 120. Meanwhile, the stain testing apparatus 100 may further include a display unit 130 for visually displaying a stain test result through the image processing unit 120.

The camera 110 corresponds to a means for photographing the display panel 400 to be inspected and generating an image. The thus generated image is transferred to the image processing unit 120. [

In order to inspect the unevenness, a test signal is applied to the display panel 400 to cause the display panel 400 to emit light, and the camera 110 photographs the display panel 400 in a light emitting state.

As the display panel 400, various types of display devices can be used. For example, a liquid crystal display, an organic light emitting display, a plasma display, or the like can be used, but the present invention is not limited thereto.

The image processing unit 120 receives an image photographed through the camera 110 and processes the input image to detect a spot defect.

The image processing unit 120 according to the embodiment of the present invention may include a preprocessing unit 121, a defect candidate extracting unit 122, and a determining unit 123. [

The preprocessing unit 121 generates a maximum value reduced image and / or a minimum value reduced image with respect to the input image.

The preprocessing unit 121, which generates the maximum / minimum value reduced image, is a characteristic configuration of the embodiment of the present invention, which will be described in more detail with reference to FIGS. 2 to 4.

2 is a diagram showing images and a luminance waveform of a part of an input image according to a process of a preprocessing unit according to an embodiment of the present invention, FIG. 2B is a view showing a luminance waveform of a part of the image in FIG. 2A, FIGS. 2C and 2D are enlarged views of a part of the maximum-reduced image and the minimum-reduced image according to the embodiment of the present invention, 2e and 2f are enlarged views of a part of the minimum value reduced image and the minimum value reduced image according to the embodiment of the present invention, respectively. 3 and 4 are diagrams showing block division for generating a maximum value reduced image and a minimum value reduced image according to the embodiment of the present invention.

The preprocessing unit 121 receives the image of FIG. 2A. Referring to FIG. 2A, a grid-like pixel structure is displayed on an image photographed through the camera 110, and a white spot exists as a smudge.

In this regard, referring to FIGS. 2A and 2B, as the pixel structure of the display panel 400 is displayed in the form of a lattice in the input image, it can be seen that the luminance of the input image varies periodically along the x direction or the y direction have. In this embodiment, a case where the pixel structure of the display panel 400 is displayed on the input image at a pitch of every 7 points is taken as an example. Here, 'point' corresponds to a pixel of an input image as an image unit constituting an input image.

For such an input image, the preprocessing unit 121 divides the input image into blocks and generates a maximum value and / or a minimum value reduced image.

This will be described in more detail with reference to FIGS. 3 and 4. FIG.

3, in dividing an input image into blocks, each block B is formed to have a size enough to cover the pixels P of the organic light emitting panel 400 . In the embodiment of the present invention, for convenience of explanation, the block B has substantially the same size as the pixel P is exemplified.

In such a case, it is preferable that block division be performed so that the center of the block B substantially coincides with the center of the pixel P in generating the maximally reduced image.

For the input image divided in units of blocks as described above, the maximum value of the luminance in each block (B) is searched and extracted to be a representative value of the block (B). In this regard, for example, the block B can be divided into a plurality of sub regions, and the maximum value among the luminance values of the plurality of sub regions can be extracted.

As shown in Fig. 2C, by using the maximum value of each of the extracted blocks B as described above, it is possible to generate a maximally reduced image in units of a reduced image, that is, the block B having the maximum value of the block (B) do.

4, in dividing the input image into blocks, each of the blocks B is divided into blocks each having a size that can cover the pixels P of the organic light emitting panel 400 Size. In the embodiment of the present invention, for convenience of explanation, the block B has substantially the same size as the pixel P is exemplified.

In such a case, the center of the block B is shifted from the center of the four pixels P (that is, the boundaries of the four pixels P) and the four pixels P adjacent to each other in the x- It is preferable that the block division is performed so as to be substantially matched.

The minimum value of the luminance in each block (B) is searched and extracted for the input image divided on a block-by-block basis, and this is used as a representative value of the block (B). In this regard, for example, the block B can be divided into a plurality of sub regions, and a minimal value among the luminance values of the plurality of sub regions can be extracted.

Using the minimum value of each extracted block B as described above, as shown in Fig. 2D, it is possible to generate a minimized reduced image in units of the reduced image, that is, the block B, of the minimum value of the block (B) do.

The maximum value / minimum value reduced image can be generated through the process as described above.

Referring to FIGS. 2E and 2F in which a part of the reduced / maximum value reduced image is enlarged, it can be seen that the lattice-like pixel structure is substantially eliminated. It can also be seen that the white spot defect is preserved as an odd defect in the input image in Fig.

Here, it can be seen that the white spot defect is more apparent than in the case of the maximum reduction image. On the other hand, in the case of a black spot defect, it can be seen that the defect is more apparent than in the case of a minimal-size reduced image.

As described above, in the maximum / minimum value reduced image according to the embodiment of the present invention, the pixel structure of the display panel 400 is effectively removed, and the luminance change of the defect is also maintained.

The maximum / minimum value reduced image having the pixel structure removed through the preprocessing unit 121 is input to the defect candidate extracting unit 122, so that the defect candidate region can be extracted.

As described above, in the maximum / minimum value reduced image, spot defects are well preserved. Therefore, the defect candidate extracting unit 122 binarizes the inputted image based on the threshold value set, for example, to generate a binarized image. Based on the binarized image, labeling is performed to assign an identification number to the defect candidate area, thereby extracting the defect candidate area.

The determining section 123 finally determines whether or not the defect candidate region extracted by the defect candidate extracting section 122 is a spot defect. In this regard, for example, the determining section 123 can calculate a feature quantity (e.g., a parameter such as a luminance value, a contrast ratio, and a SEMU value) for a defect candidate region to which an identification number is assigned. Here, SEMU is an abbreviation of Semiconductor Equipment and Materials International (SEMI) Mura. Then, it is possible to finally determine whether or not the defect is based on the calculated feature amount, and such determination can be made based on the accumulated defect determination information or the like.

The above-described defect candidate extraction process and determination process are performed on a reduced image having a reduced size as compared with the input image, so that the processing time of the process can be reduced to a considerable extent.

The defect determination result of the image processing unit 120 through the above-described process can be transmitted to the display unit 130 and displayed as a screen.

On the other hand, in the embodiment of the present invention described above, when noise is emphasized on the image photographed by the camera 110, in order to suppress noise, for example, a low- pass filter can be configured.

It is also possible to perform a tilt correction on an image by detecting a relative tilt between the camera 110 and the display panel 400 for an image in which noise is suppressed using an image or a filter photographed by the camera 110 have. As described above, for the image on which the tilt correction is performed, the above-described image processing process can be performed.

As described above, according to the embodiment of the present invention, by generating at least one of the maximum / minimum value reduced images, the pixel structure in the input image photographed by the camera is effectively removed, and also, .

As a result, according to the embodiment of the present invention, the image processing process for detecting a spot defect can be simplified and speeded up, thereby reducing the manufacturing cost of the display device.

The embodiment of the present invention described above is an example of the present invention, and variations are possible within the spirit of the present invention. Accordingly, the invention includes modifications of the invention within the scope of the appended claims and equivalents thereof.

100: stain inspection device 110: camera
120: image processing unit 121: preprocessing unit
122: smear candidate detection unit 123:
130: Display section 400: Display panel

Claims

A camera for photographing a display panel in a light emitting state;
And an image processing unit for receiving and processing an image photographed by the camera and detecting a spot defect of the display panel,
Wherein the image processing unit comprises:
Dividing the input image into blocks; And a preprocessing unit for extracting the maximum value and / or the minimum value of the block and generating the maximum value reduced image and / or the minimum value reduced image
Flat panel display stain inspection device.

The method according to claim 1,
In generating the maximally reduced image, the block is arranged such that the center of the block coincides with the center of the pixel of the display panel
Flat panel display stain inspection device.

The method according to claim 1,
In generating the minimum value reduced image, the block is arranged so that the center of the block coincides with the boundary of four pixels adjacent to each other in the display panel
Flat panel display stain inspection device.

The method according to claim 1,
The block is configured to have a size capable of covering pixels of the display panel
Flat panel display stain inspection device.

The method according to claim 1,
Wherein the image processing unit comprises:
A defect candidate extracting unit for binarizing the maximum value reduced image and / or the minimum value reduced image based on the set threshold value and extracting a defect candidate area based on the binarized image;
And a judgment unit for calculating a feature amount for the extracted defect candidate area and judging whether or not the defect candidate area is defective based on the calculated feature amount
Flat panel display stain inspection device.

Photographing a display panel in a light emitting state through a camera;
And an image processing step of receiving and processing an image photographed by the camera to detect a spot defect of the display panel,
Wherein the image processing step comprises:
Dividing the input image into blocks;
And extracting a maximum value and / or a minimum value of the block to generate a maximum value reduced image and / or a minimum value reduced image
Flat panel display stain test method.

The method according to claim 6,
In generating the maximally reduced image, the block is arranged such that the center of the block coincides with the center of the pixel of the display panel
Flat panel display stain test method.

The method according to claim 6,
In generating the minimum value reduced image, the block is arranged so that the center of the block coincides with the boundary of four pixels adjacent to each other in the display panel
Flat panel display stain test method.

The method according to claim 6,
The block is configured to have a size capable of covering pixels of the display panel
Flat panel display stain test method.

The method according to claim 6,
Wherein the image processing step comprises:
Binarizing the maximum value reduced image and / or the minimum value reduced image based on a set threshold value, and extracting a defect candidate area based on the binarized image;
Calculating a feature amount for the extracted defect candidate region, and determining whether or not the defect candidate region is defective based on the calculated feature amount
Flat panel display stain test method.