JP2005195514A - Defect inspection device and its method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a defect inspection device for easily performing inspection even on a luminescent spot defect with a low luminance value. <P>SOLUTION: A spatial filter having an emphasis coefficient is applied to an inspected pixel and to each of eight circumferential pixels existing on the circumference thereof to emphasize the luminance value of the inspected pixel further than the circumferential pixels, thereby determining whether the inspected pixel is a defective pixel or not. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a defect inspection apparatus for inspecting a defect pixel of a display panel composed of a plurality of pixels such as a liquid crystal display device and an organic EL device.

  Conventionally, various defect inspection apparatuses for photographing a display panel with a CCD camera and determining whether or not the pixel is a defective pixel on the basis of the luminance value of the pixel constituting the display panel are various. Proposed.

  For example, a method for detecting a bright spot defect in a general defect inspection apparatus will be described.

  First, the luminance value for each pixel of the display panel is input as image data.

  Next, in the input image data, the luminance value of the inspection pixel to be inspected and the luminance values of surrounding pixels around it are averaged to obtain an average luminance.

  Next, a difference between the obtained average luminance and the luminance value of the inspection pixel is obtained.

  Next, if the difference value is equal to or greater than the reference value, it is determined that the defect is a bright spot defect.

  In such a defective pixel inspection method, when the luminance value at the bright spot defect is high, the difference from the luminance value with the surrounding pixels becomes large and can be easily found.

  However, a dark luminescent spot defect or unevenness has a problem in that the luminance value is not so high as compared to the surrounding pixels, so that the difference is small and it is difficult to determine that it is a luminescent spot defect.

For example, in the technique described in Patent Document 1, the amount of change in luminance in the horizontal direction and the vertical direction is detected using a corresponding spatial filter, and unevenness and the like can be detected based on the result. Yes.
JP-A-8-145907

However, even in the defective pixel inspection method as described above, it is still difficult to ensure sufficient sensitivity with sufficient accuracy compared with visual inspection. In addition, if spatial filters having different directions are used, inconveniences such as time-consuming processing occur when the number of pixels increases with high-definition display.
Therefore, in view of the above problems, the present invention provides a defect inspection apparatus and method capable of ensuring sufficient sensitivity and accuracy.

  According to a first aspect of the present invention, in the defect inspection apparatus for inspecting a defective pixel of a display panel composed of a plurality of pixels, an image data input means for inputting a luminance value for each pixel of the display panel as image data; The value A of the product of the luminance value of the inspection pixel to be inspected in the image data and the enhancement coefficient β is calculated, and the total luminance value B of the respective luminance values of the n surrounding pixels around the inspection pixel. And the enhancement coefficient α (where β = −α × n), and the calculated luminance value of the inspection pixel is obtained by subtracting the calculated C value from the calculated A value. A defect inspection apparatus comprising: an emphasized luminance value calculation unit for calculating; and a determination unit that determines whether the pixel is a defective pixel by comparing the emphasized luminance value of the inspection pixel with a reference value.

  In the invention according to claim 2, when the defective pixel to be inspected is a bright spot defect, the value of the enhancement coefficient β in the enhancement luminance calculation means is a positive value, and the value of the enhancement coefficient α is negative. 2. The defect pixel device according to claim 1, wherein the determination means determines that the defect is a bright spot defect when the emphasized luminance value of the inspection pixel is equal to or greater than the reference value.

  In the invention according to claim 3, when the defective pixel to be inspected is a dark spot defect, the value of the enhancement coefficient β in the enhancement luminance calculation means is a negative value, and the value of the enhancement coefficient α is positive. 2. The defect inspection apparatus according to claim 1, wherein the determination means determines that the defect is a dark spot defect when the emphasized luminance value of the inspection pixel is equal to or less than the reference value.

  The invention according to claim 4 is the defect inspection apparatus according to claim 1, wherein the surrounding pixels are arranged in a lattice shape adjacent to the inspection pixel in the center, and n = 8.

  The invention according to claim 5 is the defect inspection apparatus according to claim 1, wherein one color pixel is used as the inspection pixel, and each adjacent color pixel is used as the surrounding pixel.

  6. The defect inspection according to claim 1, wherein a single color pixel is used as the inspection pixel, and a color pixel of the same color among adjacent pixels is used as the surrounding pixel. Device.

  According to a seventh aspect of the present invention, in the defect inspection apparatus for inspecting a defective pixel of a display panel composed of a plurality of pixels arranged in a grid in the xy direction, a luminance value for each pixel of the display panel is used as image data. Image data input means for inputting; initial position inspection means for inspecting one defective pixel serving as an initial reference defect pixel based on the image data; and registering the coordinate position of the initial reference defect pixel; If a pixel adjacent to and surrounding the reference defective pixel is determined to be a defective pixel, and it is determined that the adjacent pixel is a defective pixel, one defect among the determined defective pixels A first search means for registering a coordinate position and a defect pixel search range of all the determined defect pixels and a next reference defect pixel as a next reference defect pixel based on a predetermined condition, When it is determined whether or not there is a defective pixel among pixels that are adjacent to the heart and are outside the registered search range, and the adjacent pixel is determined to be a defective pixel. One defective pixel in the determined defective pixels is set as the next reference defective pixel based on the predetermined condition, and the coordinate positions of all the determined defective pixels and the defective pixel search range are registered. And the second search means for repeating this search until no defective pixel is found in the adjacent pixel or until it is determined that the adjacent pixel to be searched is a registered defective pixel. A range determining unit that sets all the pixels included in the range of the minimum coordinate value and the maximum coordinate value in the x direction and the minimum coordinate value and the maximum coordinate value in the y direction of the plurality of defective pixels as defective pixels. With features That is a defect inspection apparatus.

  In the invention according to claim 8, the predetermined condition in the first search means and the second search means is to select a defect pixel having a minimum x-coordinate value among a plurality of searched defect pixels, and 8. The defect inspection apparatus according to claim 7, wherein when there are a plurality of selected defect pixels, the defect pixel having the minimum y coordinate value is set as the next reference defect pixel.

  In the invention according to claim 9, the predetermined condition in the first search means and the second search means is to select a defect pixel having a maximum x-coordinate value among a plurality of searched defect pixels, and 8. The defect inspection apparatus according to claim 7, wherein when there are a plurality of selected defect pixels, the defect pixel having the minimum y coordinate value is set as the next reference defect pixel.

  According to a tenth aspect of the present invention, in the defect inspection method for inspecting a defective pixel of a display panel composed of a plurality of pixels, an image data input step of inputting a luminance value for each pixel of the display panel as image data; The value A of the product of the luminance value of the inspection pixel to be inspected in the image data and the enhancement coefficient β is calculated, and the total luminance value B of the respective luminance values of the n surrounding pixels around the inspection pixel. And the enhancement coefficient α (where β = −α × n), and the calculated luminance value of the inspection pixel is obtained by subtracting the calculated C value from the calculated A value. A defect inspection method, comprising: an enhanced luminance value calculation step for calculating; and a determination step for determining whether the pixel is a defective pixel by comparing the enhanced luminance value of the inspection pixel with a reference value.

  According to an eleventh aspect of the present invention, in the defect inspection method for inspecting a defective pixel of a display panel composed of a plurality of pixels arranged in a grid in the xy direction, a luminance value for each pixel of the display panel is used as image data. An input image data input step, an initial position inspection step of inspecting one defective pixel which becomes an initial reference defective pixel based on the image data, and registering a coordinate position of the initial reference defective pixel; If a pixel adjacent to and surrounding the reference defective pixel is determined to be a defective pixel, and it is determined that the adjacent pixel is a defective pixel, one defect among the determined defective pixels A first search step for registering a coordinate position of all the determined defective pixels and a search range of the defective pixels, and setting the pixel as a next reference defective pixel based on a predetermined condition; It is searched whether there is a defective pixel among pixels that are adjacent to the quasi-defective pixel in the vicinity and are outside the registered search range, and that the adjacent pixel is a defective pixel. If determined, one defective pixel in the determined defective pixel is set as the next reference defective pixel based on the predetermined condition, and the coordinate positions of all the determined defective pixels and the defective pixels are determined. A second search step of registering a search range and repeating this search until no defective pixel is found in the adjacent pixel or until it is determined that the adjacent pixel to be searched for is a registered defective pixel. And a range determining step in which all the pixels included in the range of the minimum coordinate value and maximum coordinate value in the x direction and the minimum coordinate value and maximum coordinate value in the y direction of the plurality of registered defective pixels are defective pixels; Which is a defect inspection method characterized by having.

  According to a twelfth aspect of the present invention, in a program for realizing, by a computer, a defect inspection method for inspecting a defect pixel of a display panel composed of a plurality of pixels, a luminance value for each pixel of the display panel is input as image data. Calculates the data input function and the value A of the product of the luminance value of the inspection pixel to be inspected and the enhancement coefficient β in the image data, and each luminance value of n surrounding pixels around the inspection pixel Is calculated by subtracting the calculated C value from the calculated A value and calculating the value C of the product of the total luminance value B and the enhancement coefficient α (where β = −α × n). An enhancement luminance value calculation function for calculating an enhancement luminance value of a pixel and a determination function for determining whether or not the pixel is a defective pixel by comparing the enhancement luminance value of the inspection pixel with a reference value are realized. Defect inspection method program Lamb.

  According to a thirteenth aspect of the present invention, in a program for realizing a defect inspection method by a computer for inspecting a defect pixel of a display panel composed of a plurality of pixels arranged in a grid in the xy direction, An image data input function for inputting luminance values as image data, and an initial position for inspecting one defective pixel serving as an initial reference defect pixel based on the image data and registering the coordinate position of the initial reference defect pixel An inspection function and whether or not adjacent pixels around the initial reference defective pixel are defective pixels are searched, and if it is determined that the adjacent pixels are defective pixels, the determined defective pixels One defect pixel in is set as the next reference defect pixel based on a predetermined condition, and the coordinate positions and defect pixel search ranges of all the determined defect pixels are registered. A first search function and a pixel adjacent to and surrounding the next reference defective pixel, and whether there is a defective pixel among pixels outside the registered search range, If it is determined that the adjacent pixel is a defective pixel, one defective pixel in the determined defective pixel is set as the next reference defective pixel based on the predetermined condition, and all the determined pixels are determined. The coordinate position of the defective pixel and the search range of the defective pixel are registered, and until the defective pixel is not found in the adjacent pixel, or the adjacent pixel to be searched is a registered defective pixel. A second search function that repeats until it is determined, and all the pixels included in the range of the minimum coordinate value and maximum coordinate value in the x direction and the minimum coordinate value and maximum coordinate value in the y direction of the plurality of registered defective pixels The defective pixels A program defect inspection method characterized by realizing range determining function and a of.

  The invention according to claims 1 to 6, 10, and 12 will be described.

  First, the luminance value for each pixel of the display panel is input as image data.

  Next, a product value A of the luminance value of the inspection pixel in the image data and the enhancement coefficient β is calculated. Also, a value C of the product of the total luminance value B of the n luminance values around the inspection pixel and the enhancement coefficient α is calculated. However, β = −α × n. When the defective pixel to be inspected is a bright spot defect, the value of the enhancement coefficient β is a positive value and the value of the enhancement coefficient α is a negative value. When the defective pixel to be inspected is a dark spot defect, the value of the enhancement coefficient β is a negative value, and the value of the enhancement coefficient α is a positive value. Further, for example, surrounding pixels are arranged in a lattice shape adjacent to the center of the inspection pixel, n = 8, and inspection is performed in a 3 × 3 lattice-shaped range vertically and horizontally.

  The emphasized luminance value of the inspection pixel is calculated by subtracting the calculated C value from the calculated A value.

  The calculated emphasized luminance value of the inspection pixel is compared with a reference value to determine whether or not the pixel is a defective pixel.

  In the present invention, the luminance values of the inspection pixels to be inspected are spatially enhanced by multiplying the luminance values of the surrounding pixels and the inspection pixels by the enhancement coefficients α and β having a relationship of β = −α × n, respectively. The defect inspection can be performed accurately and easily with sufficient sensitivity.

  The invention according to claims 9, 11 and 13 will be described.

  First, the luminance value for each pixel of the display panel is input as image data.

  Next, one defective pixel which becomes an initial reference defect pixel is inspected based on the image data, and the coordinate position of the initial reference defect pixel is registered.

  Next, whether or not adjacent pixels around the initial reference defective pixel are the defective pixels is searched, and if it is determined that the adjacent pixel is the defective pixel, among the determined defective pixels. One defective pixel is set as the next reference defective pixel based on a predetermined condition. Also, the coordinate positions of all the determined defective pixels and the search range of the defective pixels are registered.

  Then, a search is performed to determine whether or not a neighboring pixel around the next reference defective pixel is a defective pixel. If it is determined that the adjacent pixel is a defective pixel, the determined defective pixel is The reference defective pixel is used, and this search is repeated until no defective pixel is found in the adjacent pixels, or until the adjacent pixel to be searched for is a registered reference defective pixel.

  Then, all the pixels included in the range of the minimum coordinate value and the maximum coordinate value in the x direction and the minimum coordinate value and the maximum coordinate value in the y direction of the plurality of registered reference defect pixels are defined as the defect pixels.

  As a result, even if the pixel is surrounded by the defective pixels and is erroneously determined as a normal pixel by comparison with the luminance value of the defective pixel, the pixels surrounded by the defective pixels are All become defective pixels. Accordingly, the defective pixel region can be accurately determined.

(First embodiment)
Hereinafter, the defect inspection apparatus 10 according to the first embodiment of the present invention will be described.

  The defect inspection apparatus 10 of the present embodiment is an apparatus for inspecting the bright spot defect and the dark spot defect of the display pixels constituting the display panel of the liquid crystal display device 12.

(1) Configuration of Defect Inspection Device 10 The configuration of the defect inspection device 10 will be described with reference to FIG.

  The defect inspection apparatus 10 includes a control device 14, an imaging device 16, a display device 18 as a monitor, and an input device 20 such as a keyboard.

  A liquid crystal display device 12 to be inspected is placed on an inspection table, and a display panel is displayed by a control signal such as an image signal from the control device 14.

  The displayed display panel of the liquid crystal display device 12 is photographed by the imaging device 16. The imaging device 16 is a CCD camera and is not provided with a color identification filter or the like. In this example, the CCD camera is not provided with a color identification filter or the like, but it may be provided. In addition, the image pickup pixel of the image pickup element of the CCD camera has a low resolution corresponding to one of the pixels of the liquid crystal display device 12. However, the pixel in this specification means a state in which pixels of RGB colors are combined into one, and a pixel for each color is referred to as a “color pixel”.

  The control device 14 checks whether or not each pixel is a defective pixel using the luminance value b (x, y) for each pixel photographed by the imaging device 16. Here, the x direction is the horizontal direction on the display panel, the y direction means the vertical direction, and the pixel position can be specified by coordinates (x, y). In the control device 14, a central processing unit CPU 22 and a plurality of image memories 26 are provided, and processing described below can be executed by a program stored in the control device 14.

(2) Outline of Defect Inspection Method An outline of the defect inspection method of the defect inspection apparatus 10 having the above configuration will be described based on the flowchart of FIG.

  In step 1, the luminance value b (x, y) for each pixel of the display panel of the liquid crystal display device 12 is input using the imaging device 16.

  In step 2, for each pixel, it is examined whether the pixel is a defective pixel. In this case, the process is performed for all the pixels. This process will be described in detail later.

  In step 3, using the data of the inspected defective pixels, the range of the defective pixels is inspected. This process will also be described in detail later.

  In step 4, the data determined to be defective pixels is displayed on the display device 18.

(3) Defect Pixel Inspection Method The defect pixel inspection method in step 2 will be described with reference to FIGS.

(3-1) Case of a luminescent spot defect A case where a defective pixel to be inspected is a luminescent spot defect will be described.

(3-1-1) Theory of inspection method One inspection pixel to be inspected is selected. In this selection method, for example, the pixels from the upper left of the display panel are sequentially reached to the right end pixel, and then the lower pixels are inspected from the left end to the right end and finally selected to the lower right.

  A 3 × 3 spatial differential filter is applied around the selected inspection pixel (i, j). This spatial differential filter multiplies the central inspection pixel by an enhancement coefficient of β (where β> 1), and the surrounding eight pixels (hereinafter referred to as surrounding pixels) have α (where α Multiply by the enhancement factor <-1. In addition, the surrounding pixels refer to pixels above, below, left, right, and upper right, lower right, upper left, and lower left in FIG. 1, that is, spatial surrounding pixels. Also, β = 8 × −α. The reason for this relationship is that the value of the enhancement coefficient β to be applied to the inspection pixel is equal to the number of surrounding pixels 8 × −α, so that the difference between the inspection pixel and the surrounding pixel is canceled out and only the enhancement pixel is spatially Because it tries to emphasize. That is, instead of emphasizing using information on pixels in one direction in the vertical direction or the horizontal direction, an attempt is made to emphasize using spatial information on pixels in at least two directions in the vertical direction and the horizontal direction. In the case of the present embodiment, in addition to the two directions described above, diagonal information on the upper left and lower right and diagonal information on the upper right and lower left are also used for spatial emphasis.

  By applying this spatial filter, an enhanced luminance value c (i, j) obtained by emphasizing the luminance value b (i, j) of the inspection pixel (i, j) is obtained by equations (1) and (2).

c (i, j) = β × b (i, j) −α × B (1)

B = b (i−1, j−1) + b (i, j−1) + b (i + 1, j−1) + b (i−1, j) + b (i + 1, j) + b (i−1, j + 1) + B (i, j + 1) + b (i + 1, j + 1) (2)

If the emphasized luminance value c (i, j) is higher than the reference value, it is determined as a bright spot defect, and if it is lower, it is determined as a dark spot defect at a time.

(3-1-2) Specific Example FIG. 2 is an example of a spatial filter in the case where specific values are entered for α and β, where β = 8 and α = −1. Hereinafter, the reason why the defect inspection method of this embodiment is superior to the conventional defect inspection method will be described with reference to FIGS.

  FIGS. 5A and 6A show the pixels of the photographed display panel, and the pixel in the center is the inspection pixel. It is assumed that the green G color pixel of the inspection pixel is a bright spot defect, a bright spot defect in FIG. 5A, and a dark spot defect in FIG. 6A.

  FIGS. 5B and 6B show the nine pixels in FIGS. 5A and 6A converted into luminance values b (i, j), respectively. . The luminance value of the inspection pixel in FIG. 5B is 70, and the average luminance of nine pixels is 34. The luminance value of the inspection pixel in FIG. 6B is 45, and the average luminance of nine pixels is 31.

  A specific example of the conventional defect inspection method will be described with reference to FIGS.

  The case of the bright bright spot defect in FIG. 5 will be described.

  Since the luminance value of the inspection pixel is 70 and the average luminance is 34, a value 36 obtained by subtracting the average luminance 34 from the luminance value 70 is registered in the coordinate position (i, j) of the inspection pixel.

  Also for the surrounding pixels, a value obtained by subtracting the average luminance value from the luminance value of the surrounding pixels is registered. In the case of a negative luminance value, 0 is registered. This registration is shown in FIG. In this case, the value obtained by subtracting the inspection pixel is 36, which is higher than the reference value 20, and it is determined that the inspection pixel has a bright spot defect.

  The case of the dark bright spot defect in FIG. 6 will be described.

  Similarly, FIG. 6D shows a state in which the average luminance 31 is subtracted from the luminance value 45 of the inspection pixel and the luminance values of the surrounding pixels. In this case, the value obtained by subtracting the inspection pixel is 14, which is lower than the reference value 20, and it is not determined that the inspection pixel is a defective pixel even though the inspection pixel is a bright spot defect.

  A specific example of the defect inspection method of this embodiment will be described with reference to FIGS.

  The case of the bright bright spot defect in FIG. 5 will be described.

  First, as shown in FIG. 5B, the luminance values of the inspection pixel and the surrounding pixels are obtained as in the conventional case.

  Next, when the calculated inspection pixel and surrounding pixels are calculated based on the above-described spatial filters (1) and (2), FIG. 5C is obtained. That is, the emphasized luminance value c (i, j) of the inspection pixel is 320, and the emphasized luminance value of the surrounding pixels is zero. Then, when compared with the reference value 20 as in the previous case, it can be determined that this inspection pixel is a bright spot defect.

  The case of the dark bright spot defect in FIG. 6 will be described.

  Similarly in the case of FIG. 6, the luminance values of the inspection pixel and the surrounding pixels are obtained as shown in FIG.

  When the spatial filter described above is applied, the emphasized luminance value is as shown in FIG. Since the emphasized luminance value c (i, j) of the inspection pixel to which this spatial filter is applied is 120 and is larger than the reference value 20, it can be determined that this inspection pixel is a bright spot defect.

  As described above, in the conventional defect inspection method, bright bright spot defects can be easily inspected. However, in the case of a dark bright spot defect, although it is almost the same as the surrounding luminance value, it cannot be inspected. However, in the present embodiment, by applying a spatial filter having an enhancement coefficient, it is possible to easily inspect even a luminescent spot defect that hardly changes from the luminance value of surrounding pixels. This is because the value of the enhancement coefficient β to be applied to the inspection pixel is equal to the number of surrounding pixels 8 × −α, so that the difference between the inspection pixel and the surrounding pixel is canceled and only the enhancement pixel is enhanced. .

(3-1-3) Modified Example In the above specific example, α = −1 and β = 8 have been described, but instead of this, if α = −2 and β = 16 as shown in FIG. The luminance value of the pixel is emphasized.

(3-2) In the case of a dark spot defect The case where the defective pixel to be inspected is a dark spot defect will be described.

  In order to inspect the dark spot defect, a 3 × 3 spatial filter as shown in FIG. 4 is applied. That is, the enhancement coefficient β applied to the inspection pixel is a negative value, and the enhancement coefficient α applied to the surrounding pixels is a positive value. As a result, the luminance value of the inspection pixel, which is a dark spot defect, is emphasized so as to be darker, and the inspection can be facilitated.

  A specific example thereof will be described with reference to FIG.

  The pixel at the center in FIG. 7A is the inspection pixel, which is a dark spot defect.

  As shown in FIG. 7B, each luminance value of nine display pixels is registered, and nine average luminance values 65 are obtained.

  When the conventional defect inspection method is used, a value obtained by subtracting the average luminance 65 from the luminance value 32 of the inspection pixel is registered as shown in FIG. This value is -38.

  When the defect inspection method of the present embodiment is used, the emphasized luminance value of the inspection pixel becomes 304 as shown in FIG.

  Therefore, it is emphasized by about 10 times compared to −38 of the conventional method, and can easily be determined as a dark spot defect.

(3-3) Modified Example In the above embodiment, a 3 × 3 spatial differential filter is used, but a 5 × 5 spatial differential filter may be used. In this case, the relationship β = 24 × α is necessary. That is, also in this case, it is important to include spatially surrounding pixels with respect to the inspection pixel.

(4) Defect Pixel Range Determination Method Next, the defect pixel range determination described in step 3 above will be described with reference to FIGS.

(4-1) Reason for Determining the Range of Defect Pixels FIG. 8 shows the luminance value of each pixel by inputting an image, and the portion indicated by hatching may be a bright spot defect. It is a certain pixel. However, it is not determined which pixel has the bright spot defect in the state of FIG.

  Using the defect inspection method described above, it is determined whether or not all pixels are bright spot defects. The determined state is shown in FIG.

  Here, a problem is whether or not a pixel surrounded by a bright spot defect is a defect pixel. In the above method, since it is determined whether or not the inspection pixel at the center is a bright spot defect based on the brightness value of the surrounding pixel, the surrounding pixel is a bright spot defect and has a higher brightness than a normal pixel. In the case of a value, even if the inspection pixel at the center has a bright spot defect, it may be erroneously determined to be a normal pixel.

  On the other hand, when a large number of pixels have a bright spot defect, the probability that the middle pixel is a bright spot defect is high. In such a case, it is necessary to determine that the middle pixel is a bright spot defect. There is. Therefore, in order to search for the range of the bright spot defect, the following determination method is adopted.

(4-2) Contents of Range Determination Method First, in order to simplify the following description, coordinates are defined in FIG. In FIG. 9, the coordinate value (x, y) of the pixel on the upper left is (0, 0), and the coordinate value (x, y) of the pixel on the upper right is (5, 0). Similarly, the lower left coordinate value is (0, 5), and the lower right coordinate value (x, y) = (5, 5).

  Then, in order from the upper left, it is determined whether each pixel is a defective pixel as indicated by an arrow shown in FIG. 9. If it is a normal pixel, 0 is registered. Registers a value of 1. The registered state is shown in FIG. Here, the problem is the pixels at the coordinate values (2, 3) and (3, 3). As described above, the pixel here has a high probability of being a defective pixel, but it is determined to be a normal pixel because its luminance value is low when compared with surrounding pixels. Therefore, since it is necessary to specify the range of the defective pixel, this determination method will be described based on the flowcharts of FIGS. 11 and 15.

  In step 31, as shown in FIG. 11A, it is determined whether the pixel is a defective pixel in order from the upper left to the lower right, and an initial reference defective pixel is searched. Here, the initial reference defective pixel is a defective pixel that is first found by searching in order from the upper left to the lower right, and the coordinate value is (2, 1).

  In step 32, it is determined whether or not there are defective pixels in eight neighboring pixels centering on the initial reference defective pixel of (2, 1). Here, the pixels (1, 2) (2, 2) (2, 3) on the lower left, lower left, and lower right of the reference defective pixel (2, 1) are defective pixels.

  In step 33, the found (1, 2) (2, 2) (2, 3) registers the position of the defective pixel and the searched range.

  In step 34, the next reference defective pixel is selected. The selection condition is that the defect pixel (1, 2) having the smallest x coordinate value among the plurality of defect pixels (1, 2) (2, 2) (2, 3) searched is the next reference defect. Select as a pixel. When a plurality of defective pixels exist at the same x coordinate, the defective pixel having the minimum y coordinate value is set as the next reference defective pixel.

  In step 35, it is determined whether the surrounding pixels are defective pixels as shown in FIG. 11B with the selected next reference defective pixel (1,2) as the center. However, in this case, the range (1, 1) (2, 1) (2, 2) that has already been searched is not searched, and other unsearched ranges are searched. By this search, it is possible to determine that the pixel (1, 3) under the reference defective pixel (1, 2) is a defective pixel.

  In this way, as shown in FIGS. 11C to 11F and steps 34 to 38, until the following search end condition is satisfied, one discovered defective pixel is set as the next reference defective pixel, Search one after another. When the search end condition is satisfied, the routine proceeds to step 39.

  The search end condition is whether a defective pixel is not found in the adjacent pixel as shown in step 36, or whether the adjacent pixel to be searched for is a registered defective pixel as shown in step 37. It is. That is, if a defective pixel is not found in an adjacent pixel, all the defective pixels in the vicinity are searched, and when a defective pixel that has already been registered is reached, as shown in FIG. This is because the defective pixel is searched by making a round in the counterclockwise direction from the reference defective pixel.

  In step 39, the minimum coordinate value and the maximum coordinate value in the x direction are extracted from the coordinate values of the registered defective pixels. Similarly, the minimum coordinate value and the maximum coordinate value in the y direction are extracted. The display pixels from the minimum coordinate value in the x direction to the maximum coordinate value and the display pixels from the minimum coordinate value in the y direction to the maximum coordinate value are all defective pixels. Thereby, as shown in FIG. 12, even a pixel surrounded by a bright spot defect can be determined as a defect pixel.

  In step 40, it is determined whether or not it is a defective pixel in order from the upper left to the lower right of the display panel, and the next initial reference defective pixel is searched. In the case of this search, the already registered defect ranges are excluded. Then, the process returns to step 32.

  As described above, even if the defective pixel surrounded by the defective pixels is erroneously determined to be a normal pixel, by searching the range of the defective pixels, the erroneously determined display pixel is also accurately the defective pixel. I can judge.

(4-3) Modification 1
In the above embodiment, the case where all the display pixels exist around the inspection pixel has been described. However, for the pixels existing in the corner, upper side, right side, left side, and lower side of the display panel, some of the surrounding pixels exist do not do. Therefore, in this case, it is determined whether the inspection pixel is a defective pixel by using the overall average luminance value of the entire display panel instead of the luminance value of the surrounding pixels.

(4-4) Modification 2
In the range determination method, the defective pixel having the minimum x coordinate value is selected from the plurality of searched defective pixels, and when there are a plurality of the selected defective pixels, the defective pixel having the minimum y coordinate value is selected. The pixel was the next reference defect pixel. Alternatively, a defective pixel having the maximum x coordinate value is selected from the plurality of searched defective pixels, and when there are a plurality of the selected defective pixels, the defective pixel having the minimum y coordinate value is selected. May be the next reference defect pixel.

(4-5) Modification 3
The range determination method was performed after the defect pixel inspection method (3) was used. However, the range determination method of the present embodiment may be used to determine the range after using the defective pixel inspection method and other defective pixel inspection methods described in the related art.

(Second Embodiment)
A second embodiment using high resolution resolution pixels corresponding to the image pickup pixels of the image pickup device for each RGB color pixel will be described.

  In the first embodiment, the image pickup pixel of the image pickup element of the CCD camera uses a low-resolution resolution corresponding to one of the pixels of the liquid crystal display device 12. In addition, a high resolution resolution corresponding to the imaging pixel of the imaging element is used.

  As shown in FIG. 16, one color pixel (for example, G color pixel) is selected as the inspection pixel (in FIG. 16, it is a color pixel indicated by double hatching), and adjacent to the left as surrounding pixels. R color pixel, B color pixel adjacent in the right direction, G color pixel adjacent in the upward direction, G color pixel adjacent in the downward direction, R color pixel adjacent to the diagonal, and B color pixel Use.

(Third embodiment)
A third embodiment using a high resolution resolution image sensor corresponding to each RGB color pixel will be described. In this embodiment, the CCD camera is provided with a color identification filter.

  As shown in FIG. 17, one color pixel (for example, a G color pixel) is selected as the inspection pixel (in FIG. 17, it is a color pixel indicated by double hatching), and adjacent to the left as surrounding pixels. G color pixel of the pixel to be adjacent, G color pixel of the pixel adjacent to the right direction, G color pixel of the pixel adjacent to the upper direction, G color pixel of the pixel adjacent to the lower direction, G color pixels are used. That is, in this method, a color pixel of the same color is compared among pixels adjacent to the inspection pixel.

  The present invention is suitable for testing defects of display pixels in flat display devices such as liquid crystal display devices, organic EL devices, and plasma display devices.

It is explanatory drawing of the spatial filter in the case of discovering the luminescent spot defect which shows one Embodiment of this invention. It is explanatory drawing of the spatial filter of (alpha) =-1 when discovering a bright spot fault similarly. It is explanatory drawing of the spatial filter of (alpha) =-2 similarly. It is explanatory drawing of the spatial filter of a dark spot fault. It is explanatory drawing in the case of discovering a bright luminescent spot defect. It is explanatory drawing in the case of discovering a dark luminescent spot defect. It is explanatory drawing in the case of discovering a dark spot fault. It is a figure showing the luminance value of each display pixel in the case of defining the range of a defective pixel. It is a figure showing the emphasis luminance value similarly. It is the figure which distinguished the defective pixel and normal pixel before performing determination. It is a figure which searches the range of a fault pixel. It is the figure by which the range of the fault pixel was determined. It is a block diagram of a defect inspection apparatus. It is a flowchart of a defect inspection apparatus. It is a flowchart which searches for a defective pixel. FIG. 6 is a layout diagram of color pixels according to a second embodiment. FIG. 10 is a layout diagram of color pixels according to a third embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Defect inspection apparatus 12 Liquid crystal display device 14 Control apparatus 16 Imaging device 18 Display apparatus 20 Input device

Claims (13)

In a defect inspection apparatus for inspecting a defect pixel of a display panel composed of a plurality of pixels,
Image data input means for inputting a luminance value for each pixel of the display panel as image data;
A value A of the product of the luminance value of the inspection pixel to be inspected in the image data and the enhancement coefficient β is calculated, and the total luminance value of the luminance values of n surrounding pixels around the inspection pixel A value C of a product of B and the enhancement coefficient α (where β = −α × n) is calculated;
Enhanced brightness value calculating means for subtracting the calculated C value from the calculated A value to calculate an enhanced brightness value of the inspection pixel;
A determination unit that determines whether the pixel is a defective pixel by comparing the emphasized luminance value of the inspection pixel with a reference value;
A defect inspection apparatus characterized by comprising: If the defective pixel to be inspected is a bright spot defect,
The value of the enhancement coefficient β in the enhancement luminance calculation means is a positive value, and the value of the enhancement coefficient α is a negative value,
The defect pixel device according to claim 1, wherein the determination unit determines that the defect is a bright spot defect when an emphasized luminance value of the inspection pixel is equal to or greater than the reference value. When the defective pixel to be inspected is a dark spot defect,
The value of the enhancement coefficient β in the enhancement luminance calculation means is a negative value, and the value of the enhancement coefficient α is a positive value,
The defect inspection apparatus according to claim 1, wherein the determination unit determines that the defect is a dark spot defect when an emphasized luminance value of the inspection pixel is equal to or less than the reference value. The defect inspection apparatus according to claim 1, wherein the surrounding pixels are arranged in a lattice shape adjacent to the inspection pixel in the center, and n = 8. The defect inspection apparatus according to claim 1, wherein one color pixel is used as the inspection pixel and each adjacent color pixel is used as the surrounding pixel. The defect inspection apparatus according to claim 1, wherein a single color pixel is used as the inspection pixel, and a color pixel of the same color among adjacent pixels is used as the surrounding pixel. In a defect inspection apparatus for inspecting a defect pixel of a display panel composed of a plurality of pixels arranged in a grid in the xy direction,
Image data input means for inputting a luminance value for each pixel of the display panel as image data;
Initial position inspection means for inspecting one defective pixel serving as an initial reference defect pixel based on the image data and registering the coordinate position of the initial reference defect pixel;
A search is performed as to whether or not a neighboring pixel around the initial reference defective pixel is a defective pixel. If it is determined that the adjacent pixel is a defective pixel, one of the determined defective pixels is selected. A first search means for registering the coordinate positions of all the determined defective pixels and the search range of the defective pixels, as a next reference defective pixel based on a predetermined condition,
It is searched whether there is a defective pixel among the pixels adjacent to and surrounding the next reference defective pixel and out of the registered search range, and the adjacent pixel is the defective pixel. Is determined to be one defective pixel in the determined defective pixel as a next reference defective pixel based on the predetermined condition, and the coordinate positions of all the determined defective pixels and First, a defective pixel search range is registered, and this search is repeated until no defective pixel is found in the adjacent pixel or until it is determined that the adjacent pixel to be searched is already registered. Two search means;
A range determination unit that sets all the pixels included in the range of the minimum coordinate value and the maximum coordinate value in the x direction and the minimum coordinate value and the maximum coordinate value in the y direction of the plurality of registered defective pixels as defective pixels;
A defect inspection apparatus characterized by comprising: The predetermined conditions in the first search means and the second search means are:
Selecting a defective pixel having a minimum x-coordinate value among a plurality of searched defective pixels;
The defect inspection apparatus according to claim 7, wherein when there are a plurality of the selected defect pixels, the defect pixel having the minimum y coordinate value is set as the next reference defect pixel. The predetermined conditions in the first search means and the second search means are:
Selecting a defective pixel having the maximum x-coordinate value among the plurality of searched defective pixels;
The defect inspection apparatus according to claim 7, wherein when there are a plurality of the selected defect pixels, the defect pixel having the minimum y coordinate value is set as the next reference defect pixel. In a defect inspection method for inspecting a defective pixel of a display panel composed of a plurality of pixels,
An image data input step of inputting a luminance value for each pixel of the display panel as image data;
A value A of the product of the luminance value of the inspection pixel to be inspected in the image data and the enhancement coefficient β is calculated, and the total luminance value of each luminance value of n surrounding pixels around the inspection pixel A value C of a product of B and the enhancement coefficient α (where β = −α × n) is calculated;
An enhanced luminance value calculating step of subtracting the calculated C value from the calculated A value to calculate an enhanced luminance value of the inspection pixel;
A determination step of determining whether the pixel is a defective pixel by comparing the emphasized luminance value of the inspection pixel with a reference value;
A defect inspection method characterized by comprising: In a defect inspection method for inspecting a defective pixel of a display panel composed of a plurality of pixels arranged in a grid in the xy direction,
An image data input step of inputting a luminance value for each pixel of the display panel as image data;
An initial position inspection step of inspecting one defective pixel which is an initial reference defect pixel based on the image data and registering a coordinate position of the initial reference defect pixel;
A search is performed as to whether or not a neighboring pixel around the initial reference defective pixel is a defective pixel. If it is determined that the adjacent pixel is a defective pixel, one of the determined defective pixels is selected. A first search step of registering the determined coordinate positions of all the defective pixels and the search range of the defective pixels, as a next reference defective pixel based on a predetermined condition,
It is searched whether there is a defective pixel among the pixels adjacent to and surrounding the next reference defective pixel and out of the registered search range, and the adjacent pixel is the defective pixel. Is determined to be one defective pixel among the determined defective pixels as a next reference defective pixel based on the predetermined condition, and the coordinate positions of all the determined defective pixels and First, a defective pixel search range is registered, and this search is repeated until no defective pixel is found in the adjacent pixel or until it is determined that the adjacent pixel to be searched is already registered. Two search steps;
A range determining step in which all the pixels included in the range of the minimum coordinate value and maximum coordinate value in the x direction and the minimum coordinate value and maximum coordinate value in the y direction of the plurality of registered defective pixels are defective pixels;
A defect inspection method characterized by comprising: In a program for realizing by a computer a defect inspection method for inspecting a defect pixel of a display panel composed of a plurality of pixels,
An image data input function for inputting a luminance value for each pixel of the display panel as image data;
A value A of the product of the luminance value of the inspection pixel to be inspected in the image data and the enhancement coefficient β is calculated, and the total luminance value of each luminance value of n surrounding pixels around the inspection pixel A value C of a product of B and the enhancement coefficient α (where β = −α × n) is calculated;
An enhanced luminance value calculation function for subtracting the calculated C value from the calculated A value to calculate the enhanced luminance value of the inspection pixel;
A determination function for determining whether the pixel is a defective pixel by comparing the emphasized luminance value of the inspection pixel with a reference value;
A defect inspection method program characterized by realizing the above. In a program for realizing, by a computer, a defect inspection method for inspecting a defect pixel of a display panel composed of a plurality of pixels arranged in a grid in the xy direction,
An image data input function for inputting a luminance value for each pixel of the display panel as image data;
An initial position inspection function for inspecting one defective pixel serving as an initial reference defect pixel based on the image data and registering the coordinate position of the initial reference defect pixel;
A search is performed as to whether or not a neighboring pixel around the initial reference defective pixel is a defective pixel. If it is determined that the adjacent pixel is a defective pixel, one of the determined defective pixels is selected. A first search function for registering all the determined defect pixel coordinate positions and defect pixel search ranges as a next reference defect pixel based on a predetermined condition;
It is searched whether there is a defective pixel among the pixels adjacent to and surrounding the next reference defective pixel and out of the registered search range, and the adjacent pixel is the defective pixel. Is determined to be one defective pixel in the determined defective pixel as a next reference defective pixel based on the predetermined condition, and the coordinate positions of all the determined defective pixels and First, a defective pixel search range is registered, and this search is repeated until no defective pixel is found in the adjacent pixel or until it is determined that the adjacent pixel to be searched is already registered. 2 search function,
A range determination function in which all the pixels included in the range of the minimum coordinate value and maximum coordinate value in the x direction and the minimum coordinate value and maximum coordinate value in the y direction of the plurality of registered defective pixels are defective pixels;
A defect inspection method program characterized by realizing the above.

Images