KR101261016B1 - method and apparatus for automatic optical inspection of flat panel substrate - Google Patents

Abstract

An automatic optical inspection method for a flat panel substrate according to the present invention includes photographing a partial region of the substrate and generating a reference image based on unit images which are cell layout patterns that are periodically and repeatedly present in the photographed image. Comprising the step of comparing each unit image with the reference image to determine the presence of a defect, and if there is a defect in the substrate, obtaining the position information of the defect and outputting the position information of the obtained defect.

Description

 Method and apparatus for automatic optical inspection of flat panel substrate

The present invention relates to a manufacturing technology of a flat panel, and more particularly, to an automatic optical inspection method and apparatus for a flat panel substrate.

A flat panel includes a plurality of pixels and expresses an image by adjusting the intensity of light in each pixel. The flat panel includes a liquid crystal display, a plasma display panel, and an organic light emitting display. ), For example.

The flat panel is formed by forming a plurality of pixels on one substrate and encapsulating the upper part with another substrate. In particular, in the liquid crystal display or the organic light emitting display of the flat panel, a plurality of pixel circuits are formed on the substrate by repeating the process of depositing, etching, and cleaning the thin film on the substrate. In the process of performing the above deposition, etching, and cleaning process, there is a defect in which the wiring for forming the pixel circuit formed of the thin film is shorted or opened. However, when the defective substrate is discarded, there is a problem that the yield is low and the manufacturing cost is high, thereby repairing the wiring of the flat panel substrate. In this case, in order to repair the flat panel substrate, it is common for the inspector to determine whether the defect is present through visual inspection.

However, when the inspector visually inspects, the inspection time is determined according to the condition of the inspector, and there is a problem in that the quality of the flat panel is not maintained uniformly because the inspection criteria are different for each inspector. In addition, the visual inspection has a problem that the test takes a long time.

SUMMARY OF THE INVENTION An object of the present invention is to provide an automatic optical inspection method for a flat panel substrate and an apparatus thereof for automatically detecting a defect of the flat panel substrate.

According to a first aspect of the present invention, there is provided a method for automatic optical inspection of a flat panel substrate, the method comprising: unit images corresponding to a cell layout pattern which is photographed on a portion of the substrate and periodically repeated in each of the photographed images; Generating a reference image based on the reference image, determining whether there is a defect by comparing each unit image with the reference image, acquiring position information of the defect if a defect exists on the substrate, and outputting position information of the acquired defect It is to provide an automatic optical inspection method of a flat panel substrate comprising a step.

Additionally, the generating of the reference image may include: a pixel value of one point of the unit image and pixel values corresponding to the point of the point of the other unit images adjacent to the unit image of the corresponding point of the reference image. An automatic optical inspection method of a flat panel substrate which generates a reference image by obtaining a pixel value is provided.

In addition, in the step of generating the reference image, the reference image determining whether a defect is present has a size of M × N, and the pixel value J (i, j) at one point is J (i, j) = MED { Automatic Optical Inspection of Flat Panel Substrate Determined by I (i, j), I (i, jN), I (i, j + N), I (iM, j), I (i + M, j)} To provide a way.

In addition, the present invention provides an automatic optical inspection method of a flat panel substrate in which pixel values of respective points of a reference image compared to a unit image located at an edge of an image are calculated except for pixel values of a unit image missing from an adjacent unit image. It is.

In addition, the reference image provides an automatic optical inspection method of a flat panel substrate having an image having the same size as a unit image.

In addition, the area of the substrate photographed by the camera in the step of acquiring the unit image is to provide an automatic optical inspection method of the flat panel substrate is determined based on the rough position information of the defect input from the outside.

A second aspect of the present invention is to provide a recording medium which can be read by a computer on which a program in which the automatic optical inspection method for the flat panel substrate is implemented is recorded.

According to a third aspect of the present invention, in an automatic optical inspection apparatus for a flat panel substrate, the apparatus includes: a memory, an image acquisition unit for acquiring an image of a portion of the substrate, and periodically in each of the images The reference image generator generates a reference image based on the unit images, which are repeatedly present in the cell layout pattern, and compares the reference image with the unit images if the defect exists in the substrate. An automatic optical inspection apparatus for a flat panel substrate including a defect determining unit stored in a memory and an output unit for outputting position information of a defect of a substrate obtained by the defect determining unit to the outside.

In addition, the reference image generating unit may generate a pixel value of a corresponding point of the reference image from pixel values of a point of one unit image and pixel values of a point corresponding to the point of another unit image adjacent to the unit image. It is to provide an automatic optical inspection device for a flat panel substrate to obtain a reference image.

Additionally, the image acquisition unit may include: an input unit for receiving schematic position information of a defect from the outside, a camera module for capturing an image and storing the image in the memory, and the camera module based on the schematic position information of the defect input from the input unit. It is to provide an automatic optical inspection apparatus for a flat panel substrate further comprising a camera transfer unit for transferring to the position.

According to the automatic optical inspection method and apparatus of the flat panel according to the present invention, it is possible to automatically detect a defect located on the substrate and to determine its position. Therefore, the defect of the substrate may not be determined with the naked eye, so that the inspection of the substrate may be quickly performed, and the quality of the flat panel manufactured using the substrate may be maintained uniformly.

1 is a structural diagram showing an embodiment of an automatic optical inspection apparatus for a flat panel substrate according to the present invention.
FIG. 2 is a structural diagram illustrating an embodiment of an image acquisition unit illustrated in FIG. 1.
3 is a flow chart showing an embodiment of a method for detecting a defect of a flat panel substrate using an automatic optical inspection device for a flat panel substrate according to the present invention.
4A and 4B are flowcharts illustrating another embodiment of a method for detecting a defect of a flat panel substrate using an automatic optical inspection apparatus for the flat panel substrate according to the present invention.
FIG. 5 is a conceptual diagram illustrating an embodiment of a method of selecting a reference image by the reference image generator illustrated in FIG. 1.

Hereinafter, the present invention will be described in detail so that those skilled in the art can easily understand and reproduce the present invention through these embodiments.

1 is a structural diagram showing an embodiment of an automatic optical inspection apparatus for a flat panel substrate according to the present invention. Referring to FIG. 1, an automatic optical inspection apparatus for a flat panel substrate includes a memory 110, an image acquisition unit 120, a reference image generation unit 130, a defect determination unit 140, and an output unit 150. .

The memory 110 stores an image obtained by photographing a substrate of one flat panel and generates and stores a reference image based on unit images and unit images acquired by analyzing an image of the substrate of the flat panel. . In addition, the memory 110 may store the location information of the image acquisition unit 120 and provide the location information to the defect determination unit 140 to map the location information of the image acquisition unit 120 to unit images. .

The image acquisition unit 120 captures a portion of the flat panel substrate, acquires an image of the substrate, and stores the image in the memory 110. In addition, the image acquisition unit 120 is photographed while moving on the flat panel substrate, and stores the position information on the photographing location in the memory (110). The image acquisition unit 120 may be an example of a high resolution camera such as a charge-coupled device (CCD) camera.

The reference image generator 130 generates a unit image by analyzing a layout formed on a substrate displayed in the photographed image stored in the memory 110. In the layout of the substrate, a pattern that is periodically repeated is formed in the process of patterning a thin film. The layout is analyzed and one repeated pattern is set as one unit image. In this case, a plurality of unit images are set to one image acquired by the image acquisition unit 120. The reference image generator 130 recognizes the unit image based on the repeated periodic value of the pattern. The periodic value in the horizontal / vertical direction may be input by the operator or may be automatically recognized. In addition, the size of the unit image is expressed by the number of pixels of the image included in one period of the pattern. Since the image is photographed by the camera, the period value is not necessarily an integer value and may have a decimal point such as 127.5.

In addition, the reference image generator 130 generates a reference image used to identify the defect of the substrate compared to the unit image, and stores the generated reference image in the memory 110. The reference image is generated based on pixel values of respective points corresponding to the pixels constituting the unit images, and the reference image generator 130 selects one unit image and unit images neighboring the unit images. Then, pixel values of positions corresponding to each point of each unit image are read from each unit image. The intermediate value is selected from these pixel values, and the selected intermediate value is selected as a pixel value of one point of the reference image corresponding to one point of the unit image. That is, the pixel value of the reference image is selected as an intermediate value of the pixel values of the plurality of unit images.

The defect determiner 140 sequentially compares the unit images acquired by photographing the flat panel substrate by the image acquisition unit 120 with the reference images generated by the reference image generator 130 to determine whether a defect exists in the substrate. do. In addition, the defect determiner 140 obtains the location information in which the defect exists on the flat panel using the location information of the image acquisition unit 120 mapped to the unit image and stores the location information in the memory 110. In this case, since the reference image has the same size as the unit image, it is possible to determine the location of the defect by comparing simple pixel values.

The output unit 150 outputs the position information of the defect of the flat panel substrate obtained by the defect determination unit 140 to the outside so as to recognize the information on the position where the defect of the flat panel exists.

FIG. 2 is a structural diagram illustrating an embodiment of an image acquisition unit illustrated in FIG. 1.

2, the image acquisition unit 120 includes an input unit 121, a camera transfer unit 122, and a camera module 123.

The input unit 121 receives schematic position information of a substrate defect from the outside.

The camera transfer unit 122 allows the camera module 123 to move to a schematic position of a substrate defect received from the input unit 121. For example, the camera transfer unit 122 may be an X-Y axis robot that moves the camera module 123 in the X axis or Y axis direction on the X-Y axis. The camera transfer unit 122 allows the camera module 123 to be moved horizontally and vertically on a plane on the substrate so that the camera module 123 can capture the entire area of the substrate. In addition, the camera transfer unit 122 transmits the position information of the camera module 130 on the substrate to the memory 110 so that the position information of the camera module 123 and the unit image may be mapped.

The camera module 123 photographs the substrate to obtain a high resolution image of the substrate. The camera module 123 may store the acquired image in the memory 110.

3 is a flow chart showing an embodiment of a method for detecting a defect of a flat panel substrate using an automatic optical inspection device for a flat panel substrate according to the present invention.

Referring to FIG. 3, after performing a patterning process of depositing, etching, and cleaning a thin film on a substrate of a flat panel, an image obtained by photographing a substrate using an image acquisition unit 120 is obtained. After dividing the substrate into a plurality of regions according to the size of the substrate, one or more image acquisition units 120 acquires an image of the substrate while moving the respective regions. The thin film may be formed of a silicon layer, a metal layer, an insulating layer, or the like.

Then, by analyzing the image photographing the substrate, the layout of the thin film formed on the substrate displayed on the photographed image is analyzed and the pattern repeated on the substrate is identified. In addition, a plurality of unit images are included in the photographed image by generating a unit image corresponding to one pattern. (210) A repeating pattern on the substrate may include, for example, a pixel circuit, and one unit image is included in the unit image. Pixel circuits may correspond, and pixel circuits composed of red, green and blue subpixels may correspond.

Next, a reference image is generated by analyzing a plurality of unit images. In more detail, a reference image is generated by measuring a pixel value of a point corresponding to each unit image, and setting a median value of the measured pixel values as the pixel value of the reference image.

When the reference image is generated, pixel values of one point of one unit image and pixel values of corresponding points of the reference image are compared. As a result of the comparison, if there are more than a predetermined number of points in the predetermined area that are different from each other, it is determined that a defect exists in the substrate area corresponding to the unit image.

Then, the position of the defect existing on the substrate is determined using the position information of the image acquisition unit 120 mapped to the unit image, and the position information of the region where the defect is located is output to the outside.

4A and 4B are flowcharts illustrating another embodiment of a method of detecting a defect of a flat panel using an auto-optical inspection apparatus of a flat panel according to the present invention.

Referring to FIGS. 4A and 4B, after performing a process of depositing, etching, and cleaning a thin film on a substrate of a flat panel, a pattern is photographed using the image acquisition unit 120 to obtain an image. 300) In this case, the entire substrate or the substrate is divided into a plurality of regions according to the size of the substrate, and one or more image acquisition units 120 capture each region while moving to obtain an image of the substrate.

Then, the image is analyzed and it is determined roughly whether or not there is a defect on the substrate (310). In this case, the method of roughly determining whether or not there is a part having a particularly high or low luminance in the image by observing the image. For example, to judge. When it is determined that a defect of the substrate exists on the substrate, an image obtained by photographing the substrate is obtained by photographing a region in which the defect of the substrate is determined using the image acquisition unit 120. , The position where the defect is present can be determined through the position information of the image acquisition unit 120.

Then, by analyzing the image photographing the substrate, the layout of the thin film formed on the substrate displayed on the photographed image is analyzed and the pattern repeated on the substrate is identified. In addition, a plurality of unit images are included in the photographed image by generating a unit image corresponding to one pattern. In operation 330, a pattern repeated on a substrate may include a pixel circuit. Pixels having pixel circuits may correspond, and pixels having pixel circuits of red and green blue subpixels may correspond.

Next, the pixel values of the plurality of unit images are measured by analyzing the plurality of unit images, and the reference image is generated by setting an intermediate value of the measured pixel values as the pixel value of the reference image.

When the reference image is generated, pixel values of one unit image and corresponding points of the reference unit image are compared. As a result of the comparison, when there are more than a predetermined number of pixels having a difference greater than or equal to the reference value, it is determined that a defect exists in the substrate region corresponding to the unit image.

Then, the position of the defect existing on the substrate is determined using the position information of the image acquisition unit 120 mapped to the unit image, and the position information of the region where the defect exists is output to the outside.

FIG. 5 is a conceptual diagram illustrating an embodiment of a method of selecting a reference image by the reference image generator illustrated in FIG. 1. In the image photographed by the image acquisition unit 120, it is assumed that the 4 × 4 unit image is included in the image by analyzing the layout pattern of the substrate.

Referring to FIG. 5, the reference image selects a first unit image 410 that determines whether a defect exists in a substrate of the first unit image M, N among 4 x 4 unit images, and then selects the first unit image 410. ) And the unit image adjacent to the first unit image 410 are selected. Here, the resolution of each unit image is assumed to be M × N. In addition, the unit images adjacent to the first unit image 410 may include a unit image 410a located above the first unit image 410, a unit image 410b positioned below, and a unit image 410c located on the left side. ) May be selected as the unit image 410d existing on the right side. If the resolution of each unit image is M × N, the number of one point of each unit image is M × N.

If the coordinates of one point of the first unit image 410 are assumed to be (i, j), neighboring unit images 410a corresponding to the coordinates of one point of the first unit image M, N may be used. One points of 410b, 410c, and 410d may be set to (iM, j), (i + M, j), (i, jN), and (i, j + N), respectively. The pixel value at one point of the first unit image 410 and the pixel values at one point of the neighboring unit images 410a, 410b, 410c, and 410d are measured, and an intermediate value of the measured pixel values is selected. The pixel value of one point of the reference image is set. Then, the process repeats the intermediate values of pixel values corresponding to all points of the first unit image 410. In this case, when the resolution of the reference image is set to be the same as the resolution of the unit image, the pixel values of all points of the reference image may be set using the median of the pixel values of the unit images. Therefore, pixel value setting of all points of the reference image may be expressed as in Equation 1 below.

J (i, j) = MED {I (i, j), I (i, jN), I (i, j + N), I (iM, j), I (i + M, j)} One)

Where J (i, j) is a pixel value of a reference image, MED {} is a function of outputting a median value among a plurality of arguments, I is a pixel value of a unit image, i, j is a coordinate of a point in the unit image, M is the horizontal resolution of the unit image, N is the vertical resolution of the unit image.

When generating the reference image, the reason for using the median value of the pixel value of each point of the unit images, in general, the number of defects on the substrate is very small, one unit image and one point of the unit image neighboring the unit image There is a very small probability that there will be a defect. In addition, the defect of the substrate is caused by the short or open of the wiring generated by the thin film during the process. Therefore, since unnecessary wiring is photographed or necessary wiring is not photographed in the image in which the defect of the image is captured, the point where the defect exists in the unit image has a different pixel value from the normal point. Therefore, when the pixel value is selected as an intermediate value when the reference image is generated, the pixel value of the image in which the defect is captured may be removed.

In the case of the second unit image 420 located at the corner of the unit image, only the unit image 420a positioned on the upper side and the unit image 420b positioned on the right side may be set as adjacent unit images. Therefore, the generation of the reference image is calculated by excluding pixel values of the unit images missing from the adjacent unit images.

It is to be understood that the technical spirit of the present invention has been specifically described in accordance with the preferred embodiments thereof, but it is to be understood that the above-described embodiments are intended to be illustrative and not restrictive. In addition, it will be understood by those of ordinary skill in the art that various embodiments are possible within the scope of the technical idea of the present invention.

110: memory 120: image acquisition unit
130: reference image generation unit 140: defect determination
150: output unit

Claims (11)

In the automatic optical inspection method of a flat panel substrate,
Photographing a portion of the substrate and generating a reference image based on unit images corresponding to a cell layout pattern that is periodically and repeatedly present in each captured image;
Comparing each unit image with the reference image to determine whether there is a defect, and obtaining position information of the defect when a defect exists in the substrate; And
Outputting position information of the acquired defect; Including;
The reference image has a size of M × N, and the pixel values J (i, j) of one point are MED {I (i, j), I (i, jN), I (i, j + N), I (iM, j), I (i + M, j)}, MED {} is a function for outputting an intermediate value among a plurality of arguments, I is a pixel value of a unit image, and i, j is a point in the unit image. Automatic optical inspection method of a flat panel substrate, which means the coordinates of. The method of claim 1, wherein the generating of the reference image comprises:
A flat panel substrate for generating a reference image by obtaining a pixel value of a corresponding point of a reference image from pixel values of one point of a unit image and pixel values corresponding to the one point of a point of another unit image adjacent to the unit image. Automated Optical Inspection Method. delete The method of claim 1,
And a pixel value of each point of the reference image compared to the unit image located at the corner of the image is calculated by excluding the pixel value of the unit image missing from the adjacent unit image. The method of claim 1,
And the reference image is an image having the same size as a unit image. The method of claim 1,
And an area of the substrate to be photographed in the step of generating the reference image is determined based on rough position information of a defect input from the outside. delete delete delete delete delete

Images