KR101068356B1 - Method for inspecting defect of the Pixels in display panel device by image - Google Patents

Abstract

The present invention provides a method for acquiring image information of a display panel using a line scan camera capable of capturing each pixel of the display panel with A × B scan pixels, and analyzing brightness values of the respective scan pixels in the image information. Dividing a boundary into a pixel area formed by scan pixels above a threshold and a background area formed by scan pixels below a threshold, and searching for center points having a maximum brightness value in each row direction of scan pixels located in the pixel area, Determining a specific column direction in which the center points are arranged the most as a center position of the pixel in the X direction and a center position for the specific column direction as the center position in the Y direction to determine the center coordinates x and y of each pixel; Finally determining the rectangular area having A × B scan pixels as the pixel portion with the coordinate as the center, and the texture within the display panel. Computing individual brightness characteristic values of each determined pixel, and comparing the individual brightness characteristic values with average brightness characteristic values calculated for all pixels to determine whether each pixel is defective or not. Provide a defect inspection method.

Disclosed pixel defect inspection method of the display panel using the disclosed image processing, it is possible to effectively and quickly determine the defects of the individual pixels of the display panel using the image processing of the line scan camera has the advantage of minimizing the loss in the manufacturing process have.

Description

Method for inspecting defect of the Pixels in display panel device by image}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for inspecting pixel defects of a display panel using image processing, and more particularly, to organic light emitting diode (OLED), liquid crystal display (LCD), plasma display panel (PDP), and FED using image processing. The present invention relates to a method for inspecting pixel defects in a display panel such as a field emission display.

Inspection systems using image processing have been utilized in various fields. Among them, the display inspection field is also used as a very important tool. In general, large defects in the display screen are easily noticeable and can be easily sorted out, but in the case of fine defects associated with high quality, the size is very small and difficult to identify easily.

The study of the conventional display inspection is mainly focused on the LCD or PDP, etc. mass production is in progress. However, as OLED displays are emerging as next-generation displays, mass production is being carried out mainly by large corporations, and thus, research of inspection systems or methods for them is required.

An OLED display is a display device having a principle different from that of the LCD or PDP. Such OLEDs do not need a separate external light source and emit light by R-G-B color, so there is no need for a background light source. On the other hand, OLEDs may burn out or partially dark spots due to overcurrent, causing defects in the display.

An object of the present invention is to provide a method for inspecting pixel defects of a display panel using image processing that can easily determine whether an individual R-G-B pixel is defective in the display panel using image processing of a line scan camera.

The present invention provides a method for acquiring image information of a display panel using a line scan camera capable of capturing each pixel of the display panel with A × B scan pixels, and analyzing brightness values of the respective scan pixels in the image information. Dividing a boundary into a pixel area formed by scan pixels above a threshold and a background area formed by scan pixels below a threshold, and searching for center points having a maximum brightness value in each row direction of scan pixels located in the pixel area, Determining a specific column direction in which the center points are arranged the most as a center position in the X direction of the pixel and a center position in the specific column direction as the center position in the Y direction to determine the center coordinates x and y of each pixel; Finally determining a rectangular area having A × B scan pixels as the pixel portion with the center coordinate as the center, and in the display panel Calculating individual brightness characteristic values of each of the determined pixels and comparing the individual brightness characteristic values with average brightness characteristic values calculated for all pixels to determine whether each pixel is defective. Provides a pixel defect inspection method.

According to the pixel defect inspection method of the display panel using the image processing according to the present invention, it is possible to effectively and quickly determine whether the individual pixels of the display panel defects by using the image processing of the line scan camera to minimize the loss in the manufacturing process There is an advantage to this. In addition, since the pixels in the display panel may be searched not only for pixels having a defect in brightness but also for pixels with fine checkpoint damage, an effective pixel defect may be detected simply by a line scan camera.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for inspecting pixel defects of a display panel using image processing, that is, displays such as organic light emitting diode (OLED), liquid crystal display (LCD), plasma display panel (PDP), field emission display (FED), and the like. Applicable to pixel defect inspection in the panel. Hereinafter, for the convenience of description, the pixel defect inspection will be described for the OLED display among the various display panel types. Of course, it is obvious that the present invention is not necessarily limited to being applied only to the OLED display.

Prior to the description of the present invention, the pixel state of the OLED display will be considered. 1 and 2 are pixel photographs of normal and abnormal OLED displays.

Organic EL (Organic Electro Luminescence) or OLED manufacturing process is largely divided into pattern forming process, thin film deposition process, encapsulation process, module assembly process. Among them, the thin film deposition process using a shadow metal mask may exhibit pixel misalignment, and the encapsulation process using an organic material sensitive to oxygen or moisture may have a dark spot defect. , Defects such as pixel size expansion / contraction are very likely.

At this time, it is very important to filter the defects before the module assembly process by immediately inspecting the defects using a separate inspection device after completion of the encapsulation process. This is because when the defect is filtered to some extent through the above-described inspection process, cost reduction can be realized by minimizing a loss problem in the manufacturing process due to the completion of the defective module.

FIG. 1 is a photograph taken by magnifying an optical microscope about 100 times of each R-G-B pixel in a normal OLED display, and FIG. 2 is taken of an abnormal pixel. Comparing FIGS. 1 and 2, the abnormal red pixel of FIG. 2 has a very low brightness compared to the red pixel of FIG. It can be seen that the green pixel of FIG. 2 is slightly reduced in brightness compared to the normal green pixel of FIG. 1. The blue pixel of FIG. 2 partially shows the distribution of black spots or dark spots within the pixel.

If the inspection of such defective pixels is totally inspected by the quality inspector, it will be able to accurately identify the defects, but the results will be very poor in terms of production efficiency. In addition, when using a high magnification equipment such as an optical microscope in terms of inspection equipment, the accuracy of the defect inspection can be significantly increased, but there is a problem that the image processing time takes a long time because the inspection for a large number of pixels.

In this case, if the existing line scan type monochrome camera is used as it is, high-speed precise inspection can be performed. That is, the pixel defect inspection method of the display panel using the image processing of the present invention relates to the inspection method using the black and white camera of the line scan method.

3 is a flowchart of a method for inspecting pixel defects of a display panel using image processing according to an exemplary embodiment of the present invention, and FIG. 4 is a system configuration diagram used in FIG. 3. FIG. 5 is a configuration diagram of an OLED pixel and a scan pixel in the acquired image information, FIG. 6 is a configuration diagram of a center point for positioning an OLED pixel, and FIG. 7 is an actual setup picture of FIG. 4.

The system of FIG. 4 is available for pixel defect inspection of the OLED. The system of FIG. 4 is actually buildable through the setup of FIG. 7. Referring to FIG. 4, the OLED defect inspection system includes a line scan camera 110, an X-axis servo motor 120, a Y-axis servo motor 130, a servo drive 140, a motion control board 150, and an image. Acquisition board 160, PC 170, ON / OFF controller 180, and a power supply 190 for supplying power to each component.

OLED display panels can be installed on separate transports. The X-axis servomotor 120 and the Y-axis servomotor 130 may drive the transfer table on which the OLED display panel is placed in the X and Y directions. The line scan camera 110 has a high magnification lens to acquire image information of the OLED display panel, and the image acquisition board 160 performs image processing on the acquired image information.

The motion control board 150 issues a command to scan the next line, and the servo drive 140 sends a command of the motion control board 150 to the X-axis servomotor 120 or the Y-axis servomotor 130. Instructs the drive to the side. The controller 170 receives the information of the image acquisition board 160 to perform various processing for the inspection method. The controller 170 may also control the ON / OFF controller 180. The ON / OFF controller 180 controls on / off of the R-G-B pixel in the OLED display panel.

Of course, the pixel defect inspection method of the display panel using the image processing is not necessarily limited to being realized through the configuration of the above-described inspection system 100, it is possible to apply a system of a variety of configurations within the technical category Self-explanatory

Hereinafter, a method of inspecting pixel defects of a display panel using the image processing will be described in more detail with reference to FIGS. 3 to 6.

First, the OLED display is photographed using the line scan camera 110 to acquire image information of the OLED display (S110). The line scan camera 110 may photograph each OLED pixel in the OLED display as A × B scan pixels (8 × 14 in FIG. 5). Each one OLED pixel means one pixel of each of R, G, and B.

The image information is acquired by the image acquisition board 160 and can be processed by the controller 170. That is, the controller 170 may be regarded as an performing agent for the overall processing of the OLED pixel defect inspection.

Normally, the acquired image data type is based on the color format. However, when the image data read by the line scan camera 110 is black and white, since all of the R-G-B data are the same, only one color data may be used. In fact, for the pixel defect inspection, a line scan camera for scanning with black and white data can be used.

Then, the background area and the pixel area should be distinguished from the acquired image information, and the OLED pixel should be distinguished from the background. More specifically, after the image acquisition step (S110), the controller 170 analyzes the brightness value of each scan pixel in the image information (S120), and includes a 'pixel area' formed of scan pixels having a threshold value or more. The boundary is divided into a 'background area' formed of scan pixels below a threshold (S130).

Conventionally, a partial differential operator calculator is used to detect an edge between a background and a pixel, and the detected edges are connected to form an outline and determine the shape of an object. This method has an advantage that can be applied irrespective of a specific shape, but it is time-consuming to apply to the OLED pixel defect inspection of the present invention and has a disadvantage in that the information processing process for an errored pixel is rather difficult. In particular, when the number of pixels is large, image processing takes a long time. For example, the number of pixels in a 3 × 4 cm screen used for a mobile phone is about tens of thousands, and there are considerable difficulties in applying the above-described shape analysis method to each of these tens of thousands of pixels.

Therefore, the proposed method distinguishes the background from the OLED pixel part by analyzing scan pixels for each scan line, considering that the pixel shape is a quadrangular shape. The processing sequence and method for recognizing a range of pixels refer to FIG. 5. Here, the large rectangle represents the OLED pixel 10 as each pixel in the OLED display, and the small grid rectangle represents each scan pixel scanned by the line scan camera 110. The scan pixels corresponding to the small grid rectangles exist not only in the inside of the OLED pixel 10 but also in the outside of the background area. An example of one scan pixel on the surrounding white area is shown in FIG. 20. In the case of FIG. 5, 8 × 14 camera scan pixels consisting of eight scan pixels in the X-axis direction and 14 scan pixels in the Y-axis direction actually correspond to one OLED pixel 10. However, the ratio between this scan pixel and the OLED pixel 10 varies with the zoom lens magnification of the line scan camera 110. In FIG. 5, 1, 2, and n represent data processing for each line. In addition, X _S and X _E indicate the positions of the start and end of the scan pixel in the X-axis direction, and Y _S and Y _E denote the positions of the start and end of the scan pixel in the Y-axis direction.

In order to determine the coordinates of X _S , X _E , Y _S , and Y _E , a threshold value for distinguishing a 'background area' from a 'pixel area' must be determined. The threshold is used in Equation 1 below, and the determination of Equation 1 is limited to the case where the preconditions of Equations 2 and 3 are satisfied.

[Equation 1]

[Equation 2]

&Quot; (3) "

In Equations 1 to 3, B represents a brightness value. The brightness value is a value between D1 and D2 (D1 <D2) and may have, for example, a value between 0 and 255 which is 8-bit data.

In addition, the threshold value B _Th applied in the step S120 is an average brightness value of the lower C% range (0 <C≤45) in the brightness value distribution for each line scanned, and in the case of Equation 1, the lower 40% The average brightness value ( B _{L40, avg} ) of the range is shown. That is, in order to set the boundary between the pixel area and the background area, it means that the lower 40% of brightness values in each line are averaged and set as a threshold reference.

In addition, as a first precondition when the threshold is applied, the average brightness value of the lower E% range (0 <E≤15) in the line-level brightness value distribution should be F or more, in the case of Equation 2, the lower 10% range This is the case where the condition that the average brightness value of is 5 or more must be satisfied. This shows the condition that the average value should be 5 or more when 10% brightness value data belonging to the dark side is selected in the brightness distribution per line.

As a second precondition when the threshold is applied, the average brightness value of the upper G% range (0 <G≤15) should be at least H times (3≤H≤7) of the average brightness value of the lower E% range. In the case of Equation 3, the average brightness value in the upper 10% range should satisfy the condition that at least 5 times the average brightness value in the lower 10% range, indicating that the overall brightness of each cell should fall in a certain range. It means.

The reason for setting the conditions of Equations 1 to 3 as described above is as follows. Referring to FIG. 5, the ratio of the length occupied by the OLED pixel 10 and the background region outside thereof is about 6: 4 (36 μm: 24 μm) in the X axis. Therefore, the pixel in the 10% range of the darker side is almost certainly the brightness portion of the background area. However, if the average brightness value in the lower 10% range is too small (less than 5), it means that the camera scan itself is considered to have a problem. This is to exclude the case in which there is only a background without a normal pixel or a large number of defective areas in the cell itself.

As described above, after dividing the pixel area and the background area based on the threshold value (S130), an accurate position of the actual OLED pixel 10 must be determined in the divided pixel area (S140). To this end, referring to FIG. 6, the controller 170 searches for the center points 30 having the maximum brightness value in each row direction of the scan pixels located in the pixel area. In the case of FIG. 6, the center points having the maximum brightness values are displayed in the shaded state every 14 rows. Then, the specific column direction in which the center points are arranged the most is determined as the x-direction center position of the OLED pixel 10, and the center position with respect to the specific column direction is the y-center position in the Y-direction. cell), the center coordinates x and y of each OLED pixel 10 are determined. In addition, a rectangular region having A × B (8 × 14 in FIG. 5) scan pixels with the center coordinates (x, y) as the center is finally determined as the OLED pixel 10 portion (S140). .

The principle of the step S140 is as follows. Since all OLED pixels 10 have a two-dimensional form, it would be reasonable to position each OLED pixel 10 in terms of its center in x, y position coordinates. That is, when the x-direction occupied area of each cell is determined by comparing the thresholds using the steps S120 to S130, it is not a perfect rectangle. That is, as shown in FIG. 6, the coordinates of the maximum brightness center point, which are intermediate positions of X _S and X _E , do not all appear the same on the Y axis. Therefore, the brightest center points of each row with respect to the X-axis direction are found in advance and stored in an array, and then the position of a specific column direction in which the center points show the highest frequency is finally determined as the x center coordinate of the OLED pixel 10. do. The center coordinate in the Y direction of the OLED pixel 10 can be simply determined by the median value of the Y axis of the brightness center points for each row.

That is, in the step S140 described above, the OLED pixel 10 is formed in a quadrangular shape, and through pixel analysis for each line, the pixel region is first extracted in the X-axis direction, and the OLED is overlapped again in the Y-axis. The algorithm is configured to recognize the quadrangular range of the pixel 10.

After the position of the OLED pixel 10 is accurately determined, the defective pixel is checked to determine whether each cell is normal or abnormal for each determined OLED pixel 10 (150). To this end, the controller 170 calculates individual brightness characteristic values of the determined OLED pixels 10 in the OLED display, and calculates the individual brightness characteristic values for the entire OLED pixels 10. By comparing with the value, it is determined whether each OLED pixel is defective.

That is, for the step S150, the characteristic value of the OLED pixel 10 must be calculated first. In the proposed method, the brightness integration value and the brightness fluctuation value were selected as characteristic values for each OLED pixel 10.

More specifically, the individual brightness characteristic value, including each of the individual brightness integrals of OLED pixels (10) _(k M) and the individual brightness change value (S _k). The average brightness characteristic value includes an average brightness integration value M _{k, N} and an average brightness variation value S _{k, N} calculated for all the OLED pixels.

First, the individual brightness integration value M _k , which is defined by Equation 4, corresponds to a value obtained by integrating and averaging the brightness value B of the scan pixels in the rectangular area with respect to the OLED pixel 10.

&Quot; (4) &quot;

The average brightness integration value M _{k, N} is defined by Equation 5, and the sum of the individual brightness integration values calculated for each OLED pixel 10 is the number N of the entire OLED pixels 10. Divided by the mean.

[Equation 5]

In addition, the individual brightness change value (S _k) is, as defined by equation (6), the OLED pixel 10 brightness (B) in the individual brightness integration of the corresponding OLED pixel values for each scanned pixel in the for the rectangular area to After adding ( M _k ), it corresponds to the value integrated by adding the absolute value.

&Quot; (6) &quot;

The average brightness change value (S _{k, N)} is, as defined by Equation (7), the number a of the entire OLED pixel sum of the operation the individual brightness change value (S _k) by the respective OLED pixels (10) N The mean value divided by. That is, Equation 6 integrates the absolute value of the brightness variation value for one OLED pixel 10. When there is a checkpoint defect in the OLED pixel 10, since the value of Equation 6 will appear larger than the average value of Equation 7, it is possible to determine the bad pixel.

[Equation 7]

On the other hand, when scanning with a black and white camera, the R-G-B appears with different brightness for each color. Therefore, if it is determined that the abnormality only by the simple pixel brightness, this is not preferable because the red pixel having the low brightness among the OLED pixels 10 is always a bad pixel. However, using the fact that the same color is always repeated in the Y-axis direction and every third in the X-axis direction, it is easy to find the R-G-B array.

Next, the condition for distinguishing a bad pixel by comparing the brightness average value and the variation value of the OLED pixel 10 for each R-G-B color may be expressed by Equations 8 and 9 below.

[Equation 8]

[Equation 9]

That is not less than, the respective brightness integrated value (M _k) is the average of the brightness integrated value (M _{k, N)} or less, or if the individual brightness change value (S _k) is the average brightness change value (S _{k, N)} The OLED pixel is determined to be a bad pixel.

Here, the defective pixels of the corresponding OLED pixel 10 are distinguished by comparing the brightness integration value and the brightness fluctuation value for each corresponding RGB color of each OLED pixel 10. Referring to Equations 8 and 9, the bad pixels by applying the weight coefficients ( W _{s , color} ) for each RGB color to the average brightness integrated value ( M _{k , N} ) and the average brightness variation value ( S _{k , N} ), respectively. Judge. These W _{s and color} values are weight coefficients for each color for determining defects. The cells with defects for normal cells appear somewhat darker in brightness. Therefore, when the average brightness value of a particular OLED pixel is lower than the weighting factor ( W _{s , color} ) added to the average value M _{k , N, color} of all OLED pixels having the same color, the OLED pixel is determined to be a bad pixel. You can do it. In addition, the individual brightness change value (S _k) is larger appears pixels may therefore increase the probability of containing the black dot look poor.

Hereinafter, the configuration of the inspection apparatus used to examine the performance of the pixel defect inspection method of the display panel using the image processing and the performance results will be analyzed. In the configuration of the inspection apparatus of FIG. 4, the line scan camera 110 used Piranha HS-80-08k manufactured by DALSA. Since the pixel size of the camera 110 is 7 μm × 7 μm and the size of the OLED pixel 10 to be inspected is about 40 μm, the pixel measurement ratio is about 6: 1, which is a somewhat lower resolution for measurement. However, since the magnification of the zoom lens in front of the camera used in the inspection apparatus is twice, the ratio of the observation pixel and the target pixel could have a ratio of 10: 1 or more. 7 shows an inspection table of the actual experimental apparatus. A camera with a zoom lens is installed at the top, and the OLED display, the object to be inspected, is placed on the X-Y conveyor.

8 is an actual photograph of a portion where a defect occurs in an OLED display. Here, the portion indicated by the red square area is the portion of interest. This part is not a defect in the manufacturing process, and some OLED pixels are completely damaged or partially damaged due to deterioration due to high current or short current during display operation.

The performance result is as follows. 9 to 12 are enlarged photographs of a process of performing the image processing by enlarging a portion of FIG. 8. 9 shows the original scanned image near the defective OLED pixel. In the image of FIG. 9, the third OLED pixels are not regularly seen in the X direction because the red OLED pixels are completely lost. OLED pixels of the other two colors are also partially damaged, showing a lot of black spots.

FIG. 10 is a process of searching for a center point performed in step S140. The center point having the maximum brightness value is searched for each row of the OLED pixel 10 through line-by-line analysis in the X direction, and is displayed as a bright point. Through this process, we can check whether the algorithm is applied well.

FIG. 11 illustrates the defective pixel detection step (S150), in which a portion of the defective OLED pixels 10 is represented by a small red square box. Most of the OLED pixels that are missing and where OLED pixels are darker than average, and partially damaged OLED pixels are displayed. Based on these results, it can be seen that the algorithm proposed in this study is quite effective for detecting bad pixels.

After performing the bad pixel determination of FIG. 11 (S150), the image information scanned in black and white of FIG. 11 may be reduced to the original color through an average brightness value analysis (S160). 12 shows that the reduced image appears to be quite similar to the actual color image. This characteristic can be seen that even if you do not dare to use an expensive color scan camera can be equivalent to the black and white line scan camera 110.

As described above, the present invention relates to a pixel inspection method of an OLED type display, which accurately detects even when there is a defect in the brightness of the OLED pixel 10 or a partial fine speckle damage to the OLED pixel 10. The algorithm is presented. As a result of applying this to the actual case, the following conclusions were obtained.

First, unlike the existing edge detection method used for high-speed inspection in the mass production line, the present invention focuses on the shape of the OLED pixel 10 is a quadrangular shape, and once the pixel region is extracted in the X-axis direction Then, the algorithm is configured to recognize the range of pixels by overlapping the Y axis again. Second, the present invention sets a threshold value corresponding to 40% of the brightness distribution based on the fact that the length ratio of the background to the pixel is 6: 4 when the normal pixel is constant. As a result, an excellent result of calculating the boundary area relatively quickly was obtained. Third, through experiments using an OLED display panel in which defective pixels were actually generated, it was verified that the condition comparing the average brightness value and the variation value of the pixels proposed in the present invention can be effectively applied to the actual defect detection. In addition, considering the brightness of the original color was able to distinguish the bad pixels.

Although the present invention has been described with reference to the embodiments shown in the drawings, these are merely exemplary and those skilled in the art will understand that various modifications and equivalent other embodiments are possible. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

1 and 2 are pixel photographs of normal and abnormal OLED displays,

3 is a flowchart of a method for inspecting pixel defects of a display panel using image processing according to an embodiment of the present invention;

4 is a system configuration diagram used in FIG.

5 is a configuration diagram of an OLED pixel and a scan pixel in the acquired image information;

6 is a configuration diagram of a center point for positioning an OLED pixel;

7 is an actual setup picture of FIG. 4;

8 is a photograph taken of a defect occurs in the OLED display,

9 to 12 are enlarged photographs of a process of performing the image processing by enlarging a portion of FIG. 8.

BRIEF DESCRIPTION OF THE DRAWINGS FIG.

10: OLED Pixel 20: Background

30: center point 100: inspection device

110: line scan camera 120: X-axis servo motor

130: axis servomotor 140: servo drive

150: motion control board 160: image acquisition board

170: control unit 180: ON / OFF controller

190: power supply

Claims (4)

Acquiring image information of the display panel using a line scan camera capable of capturing each pixel in the display panel with A × B scan pixels; Analyzing a brightness value of each scan pixel in the image information, and dividing a boundary into a pixel area formed of scan pixels above a threshold and a background area formed of scan pixels below a threshold; The center points having the maximum brightness value are searched in each row direction of the scan pixels located in the pixel area, and the specific column direction in which the center points are arranged the most is determined as the center position of the pixel in the X direction and the center of the specific column direction is determined. Determining a position as a center position in the Y direction and determining the center coordinates x and y of each pixel; Finally determining a rectangular region having A × B scan pixels as the pixel portion with the central coordinate as the center; And Calculating an individual brightness characteristic value of each of the determined pixels in the display panel, and comparing the individual brightness characteristic value with an average brightness characteristic value calculated for all pixels to determine whether each pixel is defective. Method of pixel defect inspection of display panel using processing. The method according to claim 1, The threshold is, Represents the average brightness value in the lower C% range (0 <C≤45) in the line-by-line brightness value distribution The brightness value is a value between D1 (= 0) and D2 (= 2 ⁿ ) representing the brightness degree as n-bit data, When the threshold is applied, the average brightness value of the lower E% range (0 <E≤15) in the line-level brightness value distribution should be equal to or greater than F (the D1 <F <D2), and the upper G% range (0 < A method for inspecting pixel defects of a display panel using image processing, wherein an average brightness value of G ≦ 15) is equal to or more than H times (3 ≦ H ≦ 7) of the average brightness value in the lower E% range. The method of claim 1, wherein in the determining whether the pixel is defective, The individual brightness characteristic value includes an individual brightness integration value and an individual brightness variation value of each pixel, wherein the individual brightness integration value is a value obtained by integrating and averaging brightness values of scan pixels in a rectangular area for the pixel, The individual brightness fluctuation value is a value obtained by adding the absolute brightness integral value of the pixel to the brightness value of each scan pixel in the rectangular area with respect to the pixel, and then adding the absolute value to integrate the brightness value. The average brightness characteristic value includes an average brightness integration value calculated for all the pixels and an average brightness variation value, wherein the average brightness integration value is a sum of the individual brightness integration values calculated for each pixel. The average brightness change value is a value obtained by dividing the sum of the individual brightness change values calculated for each pixel by the number of all pixels. And determining the pixel as a bad pixel when the individual brightness integration value is less than or equal to the average brightness integration value or when the individual brightness variation value is more than or equal to the average brightness variation value. The method of claim 3, The defective value is determined by applying a weight coefficient for each RGB color to the average brightness integral value and the average brightness variation value so as to distinguish the defective pixels of the pixel by comparing the brightness integral value and the brightness variation value for each corresponding RGB color of each pixel. Judge the pixels, If the individual brightness integral value of a specific pixel is smaller than the average brightness integration value of all pixels having the same color multiplied by the color-specific weighting factor, the pixel is determined to be a bad pixel, If the individual brightness change value of a specific pixel is larger than the average brightness change value of all pixels having the same color multiplied by the weighting factor for each color, the pixel is determined to be a bad pixel. How to check for pixel defects.

Images