KR101076478B1 - Method of detecting the picture defect of flat display panel using stretching technique and recording medium - Google Patents

Abstract

An object of the present invention is to provide a method for inspecting an image defect of a flat panel display panel and a computer readable recording medium having recorded thereon a program for realizing the method. A method of detecting an image defect by using a level difference between the image data signals, wherein the stretching range is set to increase the level difference between the image data signals to stretch the captured image data signal to the stretching range. It is done. According to the present invention, since the level difference between the captured image data signals is increased, the discriminating power of image defect detection is increased, and the setting width of the threshold value for determining the image defect is also widened, so that the effect of detecting more precise image defects can be obtained. have.

Flat Panel Display, Burn Defect, Stretch, Pattern Image, Histogram

Description

Image defect inspection method and recording medium of flat panel display panel using stretching technique {METHOD OF DETECTING THE PICTURE DEFECT OF FLAT DISPLAY PANEL USING STRETCHING TECHNIQUE AND RECORDING MEDIUM}

1 is a view schematically showing a conventional automatic image defect inspection system,

2 is a flowchart for explaining a method of inspecting an image defect of a flat panel display by a conventional automatic image defect inspection system;

3 is a flowchart for explaining a method for inspecting an image defect of a flat panel display panel according to the present invention;

4 is a flow chart of a stretching process according to an embodiment of the present invention;

5 is an exemplary diagram of a histogram for explaining a stretching process according to FIG. 4;

6 is a flow chart of a stretching process according to another embodiment of the present invention;

7 is an exemplary diagram illustrating a histogram for explaining setting of a non-stretching range according to the stretching process of FIG. 6.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for inspecting an image defect of a flat panel display panel such as a liquid crystal panel, and more particularly, to a panel of a TFT-LCD, a TN / STN, a PDP, an organic EL panel (hereinafter referred to as a “flat panel display”). The present invention relates to an image defect inspection method of a flat panel display that uses a stretching technique to detect an image using a line scan camera or an area camera and analyzes / processes the image to determine defects. .

In general, a flat panel display is a flat panel display based on a technology different from that of a cathode ray tube display, which is thinner than a cathode ray tube display. In particular, such a flat panel display is gradually reducing the size of display pixels in accordance with the advancement of consumer preferences. It is a trend.

As a result, since the display manufacturing process requires considerable precision, the probability of product defects is increasing. However, in the display device, the display flat panel occupies most of the total price, and the more defective the product, the greater the loss on the manufacturer side. Therefore, in order to reduce the additional loss by immediately correcting or discarding the place where the defect has occurred in the flat panel display, it is important to detect image defects of the flat panel display accurately and at high speed.

Conventionally, defect inspection of a flat panel display is performed by an automatic image defect inspection system that inspects a visual image or a pattern image of a flat panel display which an inspector inspects directly by an eye, and processes the captured image image to automatically check for defective pixels. It is done by inspection.

1 is a view schematically showing a conventional automatic image defect inspection system. Referring to FIG. 1, an image defect inspection system includes a stage 100, a camera 120, an operation unit 110, and an inspection unit 200.

The stage 100 is a portion where the flat panel display panel 10 for inspection is placed, and a camera 120 for photographing the flat panel display panel placed on the stage is installed above the stage 100. The operation unit 110 is a part that operates the system as a whole, such as positioning of a camera or a stage. The inspection unit 200 includes an image unit 210 which receives images captured by a camera, converts the image into digital image values, and stores the inspection position of the flat panel display panel image. In addition, the inspection unit is based on the comparison value extraction unit 230 for extracting a difference between the signal level and the reference level of the portion corresponding to the inspection position in the flat panel display panel image transmitted to the image unit and the difference value extracted by the comparison value extraction unit. In addition, the apparatus further includes a determination unit 240 storing a threshold value for determining that the image is defective.

2 is a flowchart for explaining a method for inspecting an image defect of a flat panel display by the automatic image defect inspection system. First, a repetitive test pattern having a predetermined period is generated by a pattern generator (not shown) to display the test pattern on the flat panel display (S20). In general, an image display device having a liquid crystal display panel requires 8 bits to represent one image signal as a '0' or '1' digital signal. As in Equation 1, it can be expressed as a signal level of up to 256 from 0 (black) to 255 (white). The signal level of the repetitive test pattern having a predetermined period is, for example, (50, 50, 90, 90, 120, 120), (50, 50, 90, 90, 120, 120), ... It may be a repetitive signal level having the same signal level at positions periodically separated by 6 pixels.

2 ⁸ = 256

Next, an imaging camera, for example, a line scan camera photographs with a predetermined scan direction on a line-by-line basis to detect a pattern image corresponding to the drawn line (S30). The digital image value is extracted from the pattern image detected in the step (S40). The extracted digital image values are compared with positions where the same signal level should exist (that is, the same signal level exists for each period) (S50), and it is determined whether the pixel is an error according to the difference of the signal level. (S60). For example, the digital image values of a captured image of a flat panel display panel displaying the test pattern illustrated above are (50, 55, 90, 90, 120, 130), (55, 50, 60, 90, 120, 100). , (56, 55, 90, 90, 120, 130), (50, 55, 90, 90, 120, 130), ..., both sides separated by a period (6 pixels) The sum of the absolute values of the differences from the values is (5, 5, 30, 0, 0, 30), (6, 10, 60, 0, 0, 60), (6, 5, 30, 0, 10 , 30), ... Here, if the threshold value for determining that the pixel is an error of 40 is a pixel having a signal level of 60 and 100 of the second group, the error is determined.

However, in the conventional method for inspecting bad pixels due to the difference in signal level as described above, since the original raw image data received from the imaging camera is applied to the inspection algorithm as it is, it is not easy to secure sufficient detection capability. There is a problem. Therefore, there is a need for a method for detecting a more accurate bad pixel, which can clearly show a difference in signal level between a normal pixel and a bad pixel.

The present invention has been invented to solve the above problems, and in the method of inspecting a bad pixel by the difference in signal level between the obtained image data, the difference of the signal level between the normal pixel and the bad pixel by applying a stretching technique It is an object of the present invention to provide a method which can appear clearly and a computer-readable recording medium having recorded thereon a program for realizing the method.

The above and other objects and advantages of the present invention will become more apparent from the following description of preferred embodiments with reference to the accompanying drawings.

According to an aspect of the present invention, an image defect inspection method of a flat panel display panel includes: acquiring a digital image value by capturing a repetitive pattern image having a predetermined period of the flat panel display panel, and the range of the digital image value Determining a variable maximum value and a variable minimum value, stretching the digital image value to a range between the variable maximum value and the variable minimum value, and comparing the difference between the stretched digital image values and inspecting an image defect. Characterized in that it comprises a.

In addition, the inspecting of the image defect may include detecting a difference between the stretched digital image value at the inspection position and the stretched digital image values at positions separated by one period from the inspection position and the absolute value of the difference value. And comparing the sum of the values with a defect determination threshold.

According to the above invention, since the signal level difference between the image data obtained by applying the stretching technique is increased, the signal level difference between the pixel outputting the normal value and the defective pixel is increased to increase the discrimination ability of the defect detection. Has

According to another aspect of the present invention, the present invention comprises the steps of acquiring a digital image value by photographing a repetitive pattern image having a predetermined period of the flat panel display panel, creating a histogram of the digital image value, the digital image from the histogram Determining a range of values to set a variable maximum value and a variable minimum value, setting a non-stretching range from the histogram, and extracting the digital image value of the variable maximum value and the variable minimum value except for the non-stretching range set in the step. Selectively stretching to a range, and comparing the difference between the selectively stretched digital image values to check for image defects.

The checking of the image defect may include detecting a difference between the selectively stretched digital image value at the inspection position and the stretched digital image values at positions separated by one period from the inspection position and the difference value. And comparing the sum of the absolute values of with the defect determination threshold.

According to the present invention, in addition to the effect of increasing the signal level difference between the acquired image data, the stretching portion and the non-stretching portion is selectively stretched so as not to stretch for the image value of the pixel outputting the normal value, so that the image defect detection can be performed. There is also the effect of giving more discrimination.

Hereinafter, the present invention will be described in detail with reference to embodiments and drawings of the present invention. These examples are only presented by way of example only to more specifically describe the present invention, it will be apparent to those skilled in the art that the scope of the present invention is not limited by these examples. . In addition, in describing the embodiments of the present invention, extremely detailed details are omitted so as not to unnecessarily obscure the present invention.

3 is a flowchart illustrating a method of inspecting an image defect of a flat panel display panel according to the present invention.

By generating a repetitive inspection pattern having a predetermined period by a pattern generator (not shown) to display the inspection pattern on a flat panel display (S320), and detecting a pattern image with an imaging camera (S330). Is the same as Then, the digital image value is extracted from the pattern image detected in the step (S340). The extracted digital image value is stretched according to the present invention (S350).

With reference to Fig. 4 showing a stretching process flow chart according to the present invention, step S350 will be described in more detail. Based on the extracted image value, a histogram indicating the number of pixels according to the signal level is generated as shown in FIG. 5A (S420). FIG. 5A shows an example of a histogram when the signal level range is within the range of 50 to 180. FIG.

After detecting a range of digital image values from the histogram (S430), a stretching range, that is, a variable maximum value and a variable minimum value, is determined from the range of the digital image values (S440). In determining the variable maximum value and the variable minimum value, it is preferable not to set the full range of the signal level, that is, 255 or 0. Because the signal level of a probabilistic defective pixel is in an extremely high or low range, this is to prevent normal values from approaching this range due to stretching. For example, if the signal level of the input image value is within the range of 50 to 180, as shown in the histogram shown in FIG. 5A, the value is not stretched to 0 to 255, but to about 20 to 220. will be. That is, a value less than 20 or greater than 220 may be maintained as it is. FIG. 5B illustrates a histogram after stretching the digital image value shown in FIG. 5A in the stretching range 20 to 220. FIG. As shown in the figure, the difference in signal levels between the digital image values after stretching is increased by the stretching range.

In this way, by stretching the digital image value to the maximum utilization signal level range, the signal level difference between the obtained image data can be increased, and the discriminating power of defect detection can be increased. Also, in setting the threshold value for determining defect in defect detection, it is possible to set more widely than when using the original image data as it is, and thus more accurate pixel defect can be detected.

Of course, a variable maximum value and a variable minimum value are previously determined from a range of the test pattern generated from the pattern generator without determining a variable maximum value and a variable minimum value from the histogram by creating a histogram of the digital image value extracted in the above embodiment. It is also possible to determine. In addition, the histogram may create a histogram of an image of one line or a histogram of an entire image.

The stretched digital image values are compared with positions (that is, the same signal level exists for each cycle) where the same signal level should exist (S360), and it is determined whether the pixel is an error according to the difference in the signal level. (S370).

6 is a block diagram illustrating a stretching method according to another embodiment of the present invention. This embodiment is exactly the same as the previous embodiment except that a signal level with a high distribution is specified so that values corresponding to the signal level are not stretched. After generating a histogram of the digital image value (S630), a signal level having the highest distribution of signal levels is detected (S660). For the histogram of FIG. 5A, signal level 100 will be the signal level with the highest distribution. After detecting the signal level having the highest signal level distribution, the non-stretching range is appropriately set in consideration of the error range (S670). For example, the signal level may set the range from 95 to 105 as the non-stretching range.

As in the above embodiment, the non-stretching range is determined and selectively stretched except the non-stretching range, since the signal level range having the highest distribution is a pixel that outputs a normal value, so that the existing value is used without stretching. To give more discrimination.

In the above embodiment, in setting the non-stretching range, since a repetitive pattern image having a certain period is used as the test pattern, the a, b, and c points corresponding to each peak in the histogram shown in FIG. It is preferable to set to. The point corresponding to each peak is a test pattern using a repetitive pattern image having a certain period, and may be regarded as a pixel that outputs a normal value. Therefore, the pixel that outputs the normal value is intended to be further distinguished from the signal value indicating the defect by using the existing value as it is without stretching the value.

As described above, the method of the present invention may be implemented as a program and stored in a recording medium (CD-ROM, RAM, floppy disk, hard disk, magneto-optical disk, etc.) in a computer-readable form. Since this process can be easily implemented by those skilled in the art will not be described in more detail.

Although the present invention has been described in terms of preferred embodiments with reference to the accompanying drawings, it is apparent to those skilled in the art that various modifications and changes are possible. Therefore, the technical scope of the present invention will be determined according to the claims to be described later.

According to the image defect inspection method according to the present invention, since the signal level difference between the image data obtained by applying the stretching technique is increased, the signal level difference between the pixel outputting the normal value and the defective pixel is increased. Increase detection discrimination. Thereby, it is possible to detect the image defect of a flat panel display panel more accurately.

Further, according to the present invention, there is also an advantage of further providing discrimination power for image defect detection by selectively stretching the stretching portion and the non-stretching portion so as not to stretch the image value of the pixel outputting the normal value.

Claims (13)

In the image defect inspection method of the flat panel display panel, a) capturing a pattern image of the flat panel display panel to obtain a digital image value; b) determining a range of the digital image values and setting a variable maximum value and a variable minimum value; c) stretching the digital image value to a range of a variable maximum value and a variable minimum value; d) comparing the difference of the stretched digital image values. The method of claim 1, And the pattern image of the flat panel display is a repetitive pattern image having a certain period. The method of claim 2, The step d) includes detecting a difference between the stretched digital image value at the test position and the stretched digital image values at positions spaced one cycle apart from the test position, and the sum of the absolute values of the difference values is defective. And comparing with a determination threshold value.  In the method of inspecting a pixel defect of a flat panel display panel, a) capturing a pattern image of the flat panel display panel to obtain a digital image value; b) creating a histogram of the digital image values c) determining a range of digital image values from the histogram and setting a variable maximum value and a variable minimum value; d) setting a non-stretching range from the histogram; e) selectively stretching the digital image value to a range of a variable maximum value and a variable minimum value except for the non-stretching range set in step d); and f) comparing the differences of the selectively stretched digital image values. The method of claim 4, wherein And d) includes setting a range corresponding to a peak of the histogram as a non-stretching range. The method according to claim 4 or 5, And the pattern image of the flat panel display is a repetitive pattern image having a certain period. The method of claim 6, The step f) includes detecting a difference between the selectively stretched digital image value at the test position and the stretched digital image values at positions spaced one cycle apart from the test position and the sum of the absolute values of the difference values. Comparing the to a defect determination threshold value. A first step of extracting a digital image value according to the pattern image by photographing a pattern image line by line on a flat panel display having a predetermined pattern; A second step of stretching the digital image value extracted in the first step to a value within a predetermined range; And a third step of comparing the difference between the digital image values stretched in the second step. A recording medium having recorded thereon a program for inspecting an image defect from a pattern image of a flat panel display panel picked up by an imaging camera, a) acquiring a digital image value by photographing a repetitive pattern image having a predetermined period of the flat panel display panel; b) determining a range of the digital image values and setting a variable maximum value and a variable minimum value; c) stretching the digital image value to a range of a variable maximum value and a variable minimum value; d) a computer-readable recording medium having recorded thereon a program for executing the step of inspecting an image defect by comparing the difference of the stretched digital image values. The method of claim 9, The step d) includes detecting a difference between the stretched digital image value at the test position and the stretched digital image values at positions spaced one cycle apart from the test position, and the sum of the absolute values of the difference values is defective. And a step of comparing with a decision threshold value. A recording medium having recorded thereon a program for inspecting an image defect from a pattern image of a flat panel display panel picked up by an imaging camera, a) acquiring a digital image value by photographing a repetitive pattern image having a predetermined period of the flat panel display panel; b) creating a histogram of the digital image values c) determining a range of digital image values from the histogram and setting a variable maximum value and a variable minimum value; d) setting a non-stretching range from the histogram; e) selectively stretching the digital image value to a range of a variable maximum value and a variable minimum value except for the non-stretching range set in step d); f) A computer readable recording medium having recorded thereon a program for executing the step of checking for image defects by comparing the difference between the selectively stretched digital image values. The method of claim 11, The step d) includes setting a range corresponding to the peak of the histogram as a non-stretching range. 13. The method according to claim 11 or 12, The step f) includes detecting a difference between the selectively stretched digital image value at the test position and the stretched digital image values at positions spaced one cycle apart from the test position and the sum of the absolute values of the difference values. And comparing the defects with a defect determination threshold value. The computer-readable recording medium having recorded thereon a program comprising: checking an image defect.

Images