JP4610656B2 - Inspection device, inspection method, program, and recording medium - Google Patents

Description

  The present invention relates to an inspection technique for a liquid crystal panel, and more particularly to an inspection technique for performing an inspection based on a captured image of a liquid crystal panel that is an object to be inspected.

  In the inspection of a liquid crystal panel, there are various types of defects to be detected, size, brightness, and the like. In the visual determination, individual differences occur in the determination of these defects, and it is difficult to make a determination based on a certain standard. Therefore, an area sensor having an image sensor such as a CCD is used to determine a defect using a certain standard. In particular, due to the recent increase in the density of elements in liquid crystal panels, the application of pixel shifting methods and the development of super-resolution technology, etc., the liquid crystal panel to be inspected is imaged at a high resolution by an area sensor. It is common to inspect LCD panels.

  Problems in inspecting a liquid crystal panel using an area sensor include resolution deterioration due to the effects of viewing angle characteristics of the liquid crystal panel and aberrations in imaging. As a result, in the defect group to be determined to be equivalent, the measurement value serving as the defect index varies depending on the position in the image plane. In particular, in the liquid crystal panel, the influence of the viewing angle characteristic is large, and the gray value or the like greatly varies depending on the position of the image. For this reason, it is difficult to perform evaluation using a certain standard between the image center and the image edge. This is considered to be mainly due to a decrease in crossed Nicols characteristics of polarizing plates arranged on the front and back sides.

  Such a problem is a problem that occurs in a general inspection apparatus for a liquid crystal panel. In order to solve the above problem, the method of Patent Document 1 is disclosed.

  In the method described in Patent Document 1, first, a correction coefficient in each pixel is obtained before actual inspection. First, an image of the liquid crystal panel is taken by the area sensor. Next, the maximum value of the gray value in the captured image is extracted. Then, for each pixel in the image, a correction coefficient for each pixel is obtained by dividing the maximum value of the gray value by the gray value of each pixel. At the time of actual inspection, by multiplying all the pixels of the captured image by each correction coefficient, the gray value of each pixel can be corrected to compensate for the shading by the lens and the deterioration of the gray value due to the viewing angle characteristics. .

JP 2000-81368 A (published on March 21, 2000)

  However, in the defect inspection technology for liquid crystal panels, the accuracy of detecting defects is not yet sufficient. For example, with the method described in Patent Document 1, it is difficult to correctly detect a defect caused by contamination with foreign matter.

  Explaining this in detail, first, there are various defects in the inspection items of the liquid crystal panel. Within a large framework, it is mainly divided into bright spots and black spots, bright lines and black lines, and bright spots caused by TFT defects and bright spots caused by contamination of foreign substances are also mentioned. In particular, among these defects, defects caused by contamination of foreign matter are different from others. Described below is the mechanism of defects due to contamination of foreign matters, particularly the micro-defects in which bright spots are generated when fine foreign matter is mixed. The light emitted from the backlight passes through the polarizing plate disposed on the back side and reaches the foreign material, thereby causing light scattering. Light that has passed through the back-side polarizing plate becomes linearly polarized light, but the polarization direction is changed in various directions due to scattering by the foreign matter. When this scattered light passes through the front polarizing plate, it is detected by the human eye or sensor as a bright spot or black spot on the panel. In the case of such a micro defect, since the polarization state is different from that of the background portion where no defect exists, it is considered that the viewing angle characteristics are also different between the defect portion and the background portion.

  Here, in the method described in Patent Document 1, the correction coefficient is calculated using the maximum gray value for the entire image, and all pixels are corrected. Therefore, the same correction coefficient is used for both the defective portion and the background portion. It becomes. As described above, since the viewing angle characteristics of the micro defect are different between the background portion and the defect portion, it is not possible to obtain a correct result by correcting the entire surface of the image to a uniform value, and an in-plane inspection result without defects is obtained. It is not possible.

  The present invention has been made in view of the above problems, and a main object of the present invention is to provide a technique for accurately detecting defects resulting from contamination with foreign matter in a defect inspection technique for liquid crystal panels.

  In order to solve the above problems, the inspection apparatus according to the present invention is an inspection apparatus for a liquid crystal panel, in which an imaging means, a front polarizing plate, a liquid crystal panel used as a standard, a back polarizing plate, and a light source are arranged in this order. A first image acquisition control unit that acquires the first image by causing the imaging unit to capture an image of the first imaging target from the front polarizing plate side, and a second imaging target that includes the front polarizing plate and a light source. A second image acquisition control unit configured to acquire a second image by capturing the image from the surface polarizing plate side, the first image acquired by the first image acquisition control unit, and the second image acquisition control unit Viewing angle characteristic information generating means for generating viewing angle characteristic information associated with each position of the liquid crystal panel, the front polarizing plate, the liquid crystal panel to be inspected, the back polarizing plate, and the light source in this order using the second image. Side-by-side inspection An inspection image acquisition control means for acquiring an inspection image by causing the imaging means to pick up an image from the surface polarizing plate side, and an inspection image acquired by the inspection image acquisition control means. The gray value of the target pixel that is a pixel corresponding to a certain position, the gray value of the background pixel that is a pixel different from the target pixel in the inspection image, and the certain position generated by the viewing angle characteristic information generating unit Defect determining means for determining whether or not a defect has occurred at the certain position of the liquid crystal panel to be inspected based on the obtained viewing angle characteristic information is provided.

  According to said structure, since the 2nd to-be-photographed object which consists of a front polarizing plate and a light source passes the front polarizing plate, the scattered light which emitted the said light source passes a front polarizing plate, a liquid crystal panel, a back polarizing plate, and a light source. In the objects to be imaged arranged in this order, the viewing angle characteristic (viewing angle) in a state where the foreign matters mixed in the liquid crystal panel scatter light can be reproduced in a pseudo manner. Therefore, a first image obtained by imaging the first image pickup object in which the front polarizing plate, the liquid crystal panel used as a standard, the back polarizing plate, and the light source are arranged in this order, and a second image obtained by imaging the second image pickup object Is used to detect the viewing angle characteristics of a defective liquid crystal panel mixed with foreign matter and to detect a defect in the liquid crystal panel based on this by comparing the first image pickup object and the second image pickup object for each visual angle. Viewing angle characteristic information can be generated. Since the viewing angle characteristic information reflects the viewing angle characteristic of the defective liquid crystal panel, the inspection polarizing plate, the liquid crystal panel to be inspected, the back polarizing plate, and the light source for inspection are imaged. By determining whether or not a defect has occurred at a certain position (picture element) of the liquid crystal panel using the image and the viewing angle characteristic information, the defect of the liquid crystal panel can be detected with high accuracy.

  In the inspection apparatus, it is preferable that the defect is contamination of a liquid crystal panel.

  According to the above configuration, as described above, the inspection apparatus according to the present invention performs the defect determination using the viewing angle characteristic information that reflects the viewing angle characteristic of the defective liquid crystal panel in which the foreign matter is mixed. It can be suitably used to determine whether foreign matter is mixed in at a certain position of the panel.

  In the inspection apparatus, the viewing angle characteristic information associated with the certain position includes the gray value of the pixel corresponding to the certain position in the first image and the gray value of the pixel corresponding to the certain position in the second image. The determination means compares the gray value of the target pixel and the background pixel, at least one of which is corrected using the viewing angle characteristic information. Thus, it is preferable to determine whether or not a defect has occurred in the certain position of the liquid crystal panel to be inspected.

  According to said structure, it can determine suitably whether the defect has arisen in the said certain position of the said liquid crystal panel to be examined. That is, the gray value of the pixel corresponding to the certain position in the first image represents the luminance of the liquid crystal panel having no defect at the certain position. On the other hand, the gray value of the pixel corresponding to the certain position in the second image represents the luminance of the liquid crystal panel mixed with the foreign substance at the certain position. Therefore, these ratios indicate the luminance between the liquid crystal panel having no defect and the liquid crystal panel having the defect when the liquid crystal panel is viewed at a certain position, in other words, when the liquid crystal panel is viewed at a certain viewing angle. It is an indicator of how much the difference is.

  Here, the determination means of the inspection apparatus compares the gray value of the target pixel corresponding to a certain position of the liquid crystal panel to be inspected with the gray value of a background pixel that is a pixel different from the target pixel. Then, it is determined whether or not there is a defect at the certain position of the liquid crystal panel to be inspected. That is, if the density value of a pixel for determining whether or not a defect (target pixel) is an abnormal value compared to the density value of another pixel (background pixel), it is determined that there is a defect there. . At this time, as described above, when the liquid crystal panel is viewed at a certain viewing angle, the above is an indicator of how much the luminance difference between the liquid crystal panel having no defect and the liquid crystal panel having the defect is By comparing using the viewing angle characteristic information (that is, comparing one or both of the comparison targets in advance after correcting using the viewing angle characteristic information in advance), the difference between the density value of the target pixel and the density value of the background pixel is determined. Whether or not it represents a defect can be determined with high accuracy regardless of the viewing angle.

  The inspection apparatus according to the present invention is also an inspection apparatus for a liquid crystal panel, wherein an imaging means, a front polarizing plate, a liquid crystal panel used as a standard, a back polarizing plate, and a light source are arranged in this order. Using the first image obtained by imaging from the front polarizing plate side and the second image obtained by imaging the second object to be imaged comprising the front polarizing plate and the light source from the front polarizing plate side The generated viewing angle characteristic information storage means for storing the viewing angle characteristic information associated with each position of the liquid crystal panel, the front polarizing plate, the liquid crystal panel to be inspected, the back polarizing plate, and the light source are arranged in this order. In the inspection image acquisition control means for acquiring the inspection image by causing the imaging means to pick up the imaging object for inspection from the front polarizing plate side, and the inspection image acquired by the inspection image acquisition control means, Location with LCD panel Corresponding to the gray value of the target pixel as the corresponding pixel, the gray value of the background pixel as a pixel different from the target pixel in the inspection image, and the certain position stored in the viewing angle characteristic information storage means And a defect determination means for determining whether or not a defect has occurred at the certain position of the liquid crystal panel to be inspected based on the viewing angle characteristic information.

  According to said structure, the said test | inspection apparatus stores the viewing angle characteristic information which memorize | stores the viewing angle characteristic information produced | generated using the 1st image which imaged the 1st to-be-photographed object, and the 2nd image which imaged the 2nd to-be-photographed object. A storage means is provided. Since such viewing angle characteristic information reflects the viewing angle characteristic of the defective liquid crystal panel as described above, the front polarizing plate, the liquid crystal panel to be inspected, the back polarizing plate, and the light source are in this order. Whether or not a defect has occurred at a certain position (picture element) of the liquid crystal panel using the inspection image obtained by imaging the inspection target objects arranged side by side and the viewing angle characteristic information stored in the viewing angle characteristic information storage unit By determining whether or not, a defect in the liquid crystal panel caused by the contamination can be detected with high accuracy.

  An inspection method according to the present invention is an inspection method for a liquid crystal panel, in which an inspection apparatus including an imaging unit includes a first polarizing plate, a liquid crystal panel used as a standard, a back polarizing plate, and a light source arranged in this order. A first image acquisition step of capturing an imaged object from the surface polarizing plate side to acquire a first image, and a second imaged object in which the inspection device includes a surface polarizing plate and a light source, A second image acquisition step of capturing the second image and acquiring the second image; the first image acquired by the inspection apparatus in the first image acquisition step; and the second image acquired in the second image acquisition step. The viewing angle characteristic information generating step for generating the viewing angle characteristic information associated with each position of the liquid crystal panel, and the inspection device includes the front polarizing plate, the liquid crystal panel to be inspected, the back polarizing plate, and the light source in this order. Side-by-side inspection The inspection image acquisition step of acquiring an image for inspection by imaging the object to be imaged from the front polarizing plate side, and the liquid crystal panel in the inspection image acquired by the inspection device in the inspection image acquisition step Correlating the gray value of the target pixel that is a pixel corresponding to a certain position, the gray value of the background pixel that is different from the target pixel in the inspection image, and the certain position generated in the viewing angle characteristic information generating step And a defect determination step of determining whether or not a defect has occurred at the certain position of the liquid crystal panel to be inspected based on the obtained viewing angle characteristic information.

  According to said structure, there can exist an effect equivalent to the test | inspection apparatus based on this invention.

  Also included in the scope of the present invention are a program for operating the inspection apparatus according to the present invention, which causes a computer to realize the functions of each of the above apparatuses and a computer-readable recording medium on which the program is recorded. .

  According to the inspection apparatus of the present invention, viewing angle characteristic information using an image obtained by imaging an object to be reproduced that simulates the viewing angle characteristic (viewing angle) in a state where foreign matter mixed in the liquid crystal panel scatters light. And using the viewing angle characteristic information to determine whether or not a defect has occurred at a certain position (picture element) of the liquid crystal panel, it is possible to accurately detect defects in the liquid crystal panel due to foreign matter contamination. it can.

It is a mimetic diagram showing a schematic structure of an inspection device concerning one embodiment of the present invention. It is a schematic diagram explaining the process in which the test | inspection apparatus which concerns on one Embodiment of this invention images a to-be-photographed object. It is a functional block diagram showing a schematic structure of an inspection device concerning one embodiment of the present invention. It is a schematic diagram explaining the state of the light when a foreign material mixes in the liquid crystal layer of a liquid crystal panel. It is a schematic diagram explaining the viewing angle by an observation position. It is a three-dimensional graph explaining the calculation process of the viewing angle characteristic information which concerns on one Embodiment of this invention. It is a schematic diagram explaining the relationship between the attention pixel and background pixel in the image for a test | inspection which concerns on one Embodiment of this invention. It is a schematic diagram explaining the relationship between the attention pixel and background pixel in the image for a test | inspection which concerns on one Embodiment of this invention. It is a graph which shows the correction result of the feature-value by the inspection apparatus concerning one Embodiment of this invention.

  Hereinafter, an inspection apparatus 1 according to an embodiment of the present invention will be described with reference to the drawings. Note that the inspection method according to the present invention is not limited to the case where the inspection apparatus 1 is used, and one or a plurality of apparatuses that can perform equivalent processing may be used. In the following, a case where a liquid crystal panel is used as an inspection object will be described, but the use of the present invention for inspection of other articles is not excluded. The liquid crystal panel may be, for example, a liquid crystal panel applied to a display product or the like with a common front polarizing plate, back polarizing plate, and light source. The front polarizing plate and the rear polarizing plate refer to polarizing plates (polarizing films) disposed on the front surface or the back surface of the liquid crystal panel, respectively.

  FIG. 1 is a schematic diagram showing a schematic configuration of the inspection apparatus 1. As illustrated in FIG. 1, the inspection apparatus 1 includes a display unit 200, an imaging unit (imaging unit) 300, an imaging object placement unit 400, and a main control unit 100.

  The display unit 200 displays various information acquired by the inspection apparatus 1 (including inspection results), instructions to the user, and the like. The display unit 200 can be configured using a general display technology, and for example, an LCD (Liquid Crystal Display) or a CRT (Cathode Ray Tube) can be used.

  The imaging unit 300 captures an object to be imaged (inspection object) on the object to be imaged arrangement unit 400, and includes an optical system 320 and a solid-state image sensor 310. The optical system 320 is a means for forming an image of an object to be imaged (inspection object) on the light receiving unit of the solid-state image sensor 310. The optical system includes a lens and other optical elements. The solid-state image sensor 310 is means for converting an image of an object to be imaged (inspection object) formed on the light receiving unit into an image signal. As the solid-state imaging device 310, for example, an area sensor type CCD image sensor, a CMOS image sensor, or the like can be used. A lens and other optical elements constituting the optical system are attached to the inner wall of the housing of the imaging apparatus via various adjustment mechanisms.

  The imaging object arrangement unit 400 is a member for arranging an imaging object (inspection object). The imaging object placement unit 400 may be incorporated in the inspection apparatus 1, but may be configured as a separate apparatus.

  On the imaging object arrangement unit 400, for example, as shown in FIG. 1, a front polarizing plate 10, a liquid crystal panel 20 to be inspected, a back polarizing plate 30, and a backlight (light source) 40 from the side closer to the imaging unit 300. It is possible to arrange the inspection object 50 to be arranged in order. In addition to this, an object to be imaged as shown in FIGS. 2A and 2B can be arranged on the object to be imaged arrangement unit 400. In either case, the object to be imaged (front polarizing plate 10, liquid crystal panel 20, back polarizing plate 30, backlight 40, etc.) is removable.

  In addition, the imaging object placement unit 400 includes power supply means (not shown), lights the backlight 40, and emphasizes defects in the imaging object (inspection object).

  The front polarizing plate 10 and the back polarizing plate 30 are arranged in consideration of the polarization directions of each other. That is, the front polarizing plate 10 and the back polarizing plate 30 are arranged in consideration of each other's polarization direction so that it becomes dark when the liquid crystal panel is not sandwiched if it is a normally black panel and bright if it is normally white. To do. The liquid crystal panel 20 to be inspected is arranged between the back polarizing plate 30 and the front polarizing plate 10 as shown in FIG.

  Further, the imaging object placement unit 400 applies a predetermined voltage to the liquid crystal panel 20 to control the liquid crystal panel 20 to an appropriate lighting state such as lighting the entire surface black, or does not apply a voltage. As a result, the liquid crystal panel 20 is brought into a desired lighting state. In the embodiment described later, a normally black panel is assumed, so that a voltage for lighting in black is applied or no voltage is applied.

  The main control unit 100 controls the imaging unit 300 and also corrects the evaluation and viewing angle characteristics of the liquid crystal panel 20 to be inspected based on the captured image obtained by the imaging unit 300 (viewing angle characteristic information). It is a means for calculating 90. The main control unit 100 can be constituted by, for example, a personal computer or a workstation.

  FIG. 3 is a functional block diagram showing a functional schematic configuration of the inspection apparatus 1. As shown in FIG. 3, the main control unit 100 includes an inspection control unit 120 that performs an inspection of the liquid crystal panel, a correction information generation control unit 110 that calculates a correction coefficient 90 used in the inspection, and a correction that stores the correction coefficient 90. A coefficient storage unit (viewing angle characteristic information storage means) 130 is provided. The main control unit 100 also includes an imaging control unit 140 and a display control unit 150.

  The imaging control unit 140 causes the imaging unit 300 to capture an object to be captured and obtains a captured image obtained. The captured image can be in a general image data format such as BMP. That is, any format that can acquire a gray value (luminance) for each pixel of the captured image may be used, and a gray value matrix may be used. The imaging control unit 140 also controls the imaging object arrangement unit 400 to control lighting of the backlight 40 arranged on the imaging object arrangement unit 400, application of a voltage to the liquid crystal panel 20 or 20 ′, and the like. .

  The display control unit 150 receives input of data to be displayed and displays the data on the display unit 200. As the input data, well-known and commonly used data such as image data indicating an image to be displayed and a description language describing the image to be displayed can be used.

  The correction coefficient storage unit 130 stores the correction coefficient 90. The data format of the correction coefficient 90 will be described later.

  The correction information generation control unit 110 includes a first image acquisition control unit (first image acquisition control unit) 111, a first image storage unit 112, a second image acquisition control unit (second image acquisition control unit) 113, and a second image. A storage unit 114 and a correction coefficient calculation unit (viewing angle characteristic information generation means) are provided, and a first image 70 obtained by imaging the first object 51 as shown in FIG. A correction coefficient 90 is calculated based on the second image 80 obtained by imaging the second object 52 as shown in 2 (b). This will be described in detail below.

  The first image acquisition control unit 111 controls the imaging unit 300 and the imaging object placement unit 400 via the imaging control unit 140, whereby the front polarizing plate 10, the liquid crystal panel 20 ′ used as a standard, the back polarizing plate 30, The first imaging object 51 including the backlight 40 is captured by the imaging unit 300 from the front polarizing plate 10 side to acquire the first image 70, and the acquired first image 70 is stored in the first image storage unit 112. Remember me. As shown in FIG. 2A, the first object 51 includes the front polarizing plate 10, the liquid crystal panel 20 ′ used as a standard, the back polarizing plate 30, and the backlight 40 from the side closer to the imaging unit 300. They are sequentially arranged on the imaging object arrangement unit 400. The “liquid crystal panel used as a standard” may be a liquid crystal panel having the same standard (size, pixel pitch, material, etc.) as the liquid crystal panel to be inspected.

  The second image acquisition control unit 113 controls the imaging unit 300 and the imaging object placement unit 400 via the imaging control unit 140, so that the second imaging object 52 including the front polarizing plate 10 and the backlight 40 is The imaging unit 300 captures an image from the surface polarizing plate 10 side, acquires the second image 80, and stores the acquired second image 80 in the second image storage unit 114. 2B, the second object to be imaged 52 is arranged on the object to be imaged arrangement unit 400 in the order of the front polarizing plate 10 and the backlight 40 from the side closer to the imaging unit 300. .

  In a state where the first image 70 is stored in the first image storage unit 112 and the second image 80 is stored in the second image storage unit 114, the correction coefficient calculation unit 115 performs the first image 70 and the second image 80. Based on the above, the correction coefficient 90 associated with each position of the liquid crystal panel 20 is calculated and stored in the correction coefficient storage unit 130.

  The “correction coefficient associated with each position of the liquid crystal panel” can be, for example, a correction coefficient associated with each pixel of the captured image of the liquid crystal panel. That is, each pixel of the captured image of the liquid crystal panel indicates the result of detecting the light emitted from the coordinates corresponding to each pixel on the liquid crystal panel. Therefore, the correction coefficient associated with each pixel is determined for each position. The associated correction factor. The correspondence between the position on the liquid crystal panel and the pixel on the captured image can be easily obtained from the setting (resolution [μm / pix]) of the imaging unit 300.

  Although a detailed calculation method of the correction coefficient 90 will be described later, in one embodiment, the correction coefficient 90 associated with a certain position is the gray value (p) of the pixel associated with the certain position in the first image 70. And the gray value (q) of the pixel associated with the certain position in the second image 80, specifically, a value obtained by dividing q by p, or a value between p and q Can be a set of In addition, when it is set as the value set of p and q, those skilled in the art easily understand that the correction coefficient 90 may be data of the first image 70 and the second image 80 itself.

  As described above, the data format of the correction coefficient 90 is obtained by acquiring the correction coefficient (for example, the set of q / p or (q, p) described above) associated with the position (pixel) for each position (for each pixel). It can be a matrix of correction coefficients associated with each position (pixel).

  As will be described in detail later, the first image 70 simulates a captured image of a liquid crystal panel having no defect, and the second image 80 simulates a captured image of a liquid crystal panel mixed with foreign matter. . Therefore, the inspection apparatus 1 calculates the correction coefficient 90 reflecting the captured image of the liquid crystal panel mixed with foreign matter based on the first image 70 and the second image 80, and uses the correction coefficient 90 for the inspection image 60. Therefore, the defect determination can be performed in consideration of the viewing angle characteristic of the liquid crystal panel mixed with a foreign substance, which is significantly different from the viewing angle characteristic of the liquid crystal panel having no defect.

  The correction information generation control unit 110 may cause the display unit 200 to display the calculated correction coefficient 90 via the display control unit 150.

  In a state where the correction coefficient 90 is stored in the correction coefficient storage unit 130, the inspection control unit 120 starts inspection of the liquid crystal panel 20 to be inspected. The inspection control unit 120 includes an inspection image acquisition control unit (inspection image acquisition control means) 121, an inspection image storage unit 122, a feature amount calculation unit 123, a feature amount storage unit 124, a defect candidate extraction unit 125, and a defect candidate storage. Unit 126 and a defect determination unit (defect determination means) 127.

  The inspection image acquisition control unit 121 controls the imaging unit 300 and the imaging object placement unit 400 via the imaging control unit 140, whereby the front polarizing plate 10, the liquid crystal panel 20 to be inspected, the back polarizing plate 30, The imaging object 300 made of the backlight 40 and the backlight 40 is imaged by the imaging unit 300 from the surface polarizing plate 10 side to acquire the inspection image 60, and the acquired inspection image 60 is stored in the inspection image storage unit 122. Remember. As shown in FIG. 1, the inspection object 50 includes a front polarizing plate 10, a liquid crystal panel 20 ′ to be inspected, a back polarizing plate 30, and a backlight 40 in this order from the side closer to the imaging unit 300. Arranged on the imaging object arrangement unit 400. Subsequently, the inspection image acquisition control unit 121 causes the feature amount calculation unit 123 to start processing.

  Based on the inspection image 60 stored in the inspection image storage unit 122, the feature amount calculation unit 123 is associated with the feature amount associated with each position of the liquid crystal panel 20 (associated with each pixel 61 of the inspection image 60). Feature amount) 62 is calculated.

  The feature amount 62 is information for estimating whether or not a defect is present at the position associated with the feature amount 62, and the feature amount associated with each position (pixel) is stored. It can be a matrix. For example, the feature amount calculation unit 123 sets the feature amount 62 associated with a certain position of the liquid crystal panel 20, the gray value of the target pixel 61 a that is the pixel 61 of the inspection image 60 corresponding to the certain position, and the inspection image. The feature amount 62 is calculated by obtaining a value (for example, a difference between the two shade values) representing the result of comparing the shade value of the background pixel 61c, which is a pixel different from the target pixel 61a in 60.

  The background pixel 61c is a pixel that is treated as having no defect. Therefore, as described above, for example, by comparing the gray value of the background pixel 61c with the gray value of the target pixel 61a, it is determined whether or not there is a defect at a position corresponding to the target pixel 61a of the liquid crystal panel 20. Can be determined. The pixel 61 used as the background pixel 61c may be a pixel different from the pixel of interest 61a, but it is preferable to select a pixel that satisfies the conditions described later. The comparison between the gray value of the pixel of interest 61a and the gray value of the background pixel 61c may preferably be a comparison between the gray value of the pixel of interest 61a and the average of the gray values of the plurality of background pixels 61c.

  The feature amount calculation unit 123 repeats the above procedure, acquires the feature amount 62 associated with each pixel, and stores it in the feature amount storage unit 124. Subsequently, the feature amount calculation unit 123 causes the defect candidate extraction unit 125 to start processing.

  The defect candidate extraction unit 125 extracts a position (or pixel) that becomes a defect candidate, for example, by binarization processing (comparison with a predetermined threshold) based on the feature value 62 of each position (or pixel). . Extraction results (for example, the list of extracted pixel coordinates) are stored in the defect candidate storage unit 126. Subsequently, the defect candidate extraction unit 125 causes the defect determination unit 127 to start processing.

  The defect determination unit 127 performs defect determination for each position (pixel) that is a defect candidate stored in the defect candidate storage unit 126 using the correction coefficient 90 stored in the correction coefficient storage unit 130. The defect determination unit 127 determines the gray value of the target pixel 61a that is the pixel 61 in the inspection image 60 corresponding to a certain position of the liquid crystal panel 20 and the background pixel 61c that is a pixel different from the target pixel 61a in the inspection image 60. What is necessary is just to determine whether or not a defect has occurred at the certain position of the liquid crystal panel 20 based on the gray value and the correction coefficient 90. That is, as will be described later, since the correction coefficient 90 reflects the viewing angle characteristics of the liquid crystal panel mixed with foreign matter, the defect determination unit 127 compares the gray value of the target pixel 61a with the gray value of the background pixel 61c. In the determination of defects by using the correction coefficient 90, it is possible to detect the defects caused by the inclusion of foreign matter with high accuracy.

  For example, as described above, when the correction coefficient 90 is information indicating the ratio of the gray values of the corresponding pixels of the first image 70 and the second image 80, the defect determination unit 127 determines that the target pixel 61a The evaluation amount (for example, the feature amount 62) indicating the difference between the gray value and the gray value of the background pixel 61c may be evaluated more greatly as the ratio is smaller and smaller as the ratio is larger. The position where the ratio is large is a position where the difference between the gray value of the target pixel 61a and the gray value of the background pixel 61c is predicted to increase when there is a defect, and thus indicates the actually calculated difference. By evaluating the evaluation amount as small as possible, it is possible to determine the defect according to the uniform evaluation standard throughout the liquid crystal panel 20.

  For example, when the correction coefficient 90 is information representing a set of the first image 70 and the second image 80, the defect determination unit 127 determines the gray value of the target pixel 61 a as described above in the second image 80. The pixel corresponding to the position is corrected so as to be relatively smaller than the gray value of the background pixel 61c as the gray value of the pixel corresponding to the position is large, and the gray value of the background pixel 61c is changed to a pixel corresponding to the certain position of the first image 70. The gray level value of the pixel of interest 61a is corrected so as to be relatively small with respect to the gray level value of the target pixel 61a. This also makes it possible to perform defect determination based on uniform evaluation criteria throughout the liquid crystal panel 20 as described above. The grayscale values of the target pixel 61a and the background pixel 61c may be acquired based on the inspection image 60 read from the inspection image storage unit 122.

  Preferably, the defect determination unit 127 compares the gray value of the target pixel 61a and the background pixel 61c, at least one of which has been corrected using the correction coefficient 90, so that the defect is detected at the certain position of the liquid crystal panel 20. Whether or not has occurred. For example, after correcting one or both of the grayscale values of the target pixel 61a and the background pixel 61c using the correction coefficient 90, the difference may be taken. Further, if they are equivalent by mathematical transformation, the difference may be corrected using the correction coefficient 90 after calculating the difference between the gray values of the target pixel 61a and the background pixel 61c. Those skilled in the art will easily understand that this case also corresponds to comparing the gray values of the pixel of interest 61a and the background pixel 61c, at least one of which has been corrected using the correction coefficient 90. Detailed mathematical formulas will be described later.

  The defect determination unit 127 causes the display unit 200 to display information indicating the position on the liquid crystal panel 20 that has been determined to have a defect via the display control unit 150. As a method for calculating the position on the liquid crystal panel 20 corresponding to the target pixel 61a, for example, based on the setting of the resolution [μm / pix] in the optical system 320 of the imaging unit 300, the liquid crystal that appears in the inspection image 60 By calculating the distance from the edge of the panel 20 to the defect, the coordinates of the defect on the liquid crystal panel can be obtained. Further, it is possible to obtain information on how many pixels of the captured image are allocated per picture element based on the size of the picture element 22 of the liquid crystal panel 20 to be inspected.

  Note that the inspection control unit 120 according to the present embodiment temporarily does not use the correction coefficient 90, and based on the inspection image 60, positions (or pixels on the inspection image 60) that become defect candidates on the liquid crystal panel 20. 61) (the feature amount calculation unit 123 and the defect candidate extraction unit 125), and determines whether each defect candidate position is a defect using the correction coefficient 90 (defect determination unit 127). However, more simply, the defect determination unit 127 uses the correction coefficient 90 to directly determine whether a certain position on the liquid crystal panel 20 has a defect based on the inspection image 60. Also good. In that case, the correction coefficient 90 is directly applied to the feature quantity 62 without extracting the defect candidates, or at least the gray value of the target pixel 61a and the gray value of the background pixel 61c of the inspection image 60 are applied. After applying the correction coefficient 90 to either one, the feature quantity 62 may be obtained based on the corrected quantity density value, and defect determination may be performed using the result.

  In addition, the inspection apparatus 1 according to the present embodiment includes the correction information generation control unit 110. However, in another embodiment, if the correction coefficient 90 is stored in the correction coefficient storage unit 130, the inspection apparatus Even without the correction information generation control unit 110, the liquid crystal panel 20 can be inspected with high accuracy. For example, the correction information generation control unit 110 is removable (for example, software for operating the correction information generation control unit 110 in a network is temporarily supplied). The inspection apparatus in which the correction information generation control unit 110 is removed after storing the correction information, the inspection apparatus that acquires the correction coefficient 90 from the other inspection apparatus 1 and stores the correction coefficient 90 in the correction coefficient storage unit 130 included in the inspection apparatus, etc. Category.

  That is, the inspection apparatus according to another embodiment of the present invention is an inspection apparatus for the liquid crystal panel 20, and includes the imaging unit 300, the front polarizing plate 10, the liquid crystal panel 20 ′ used as a standard, the back polarizing plate 30, and the backlight. A first image 70 obtained by imaging the first imaging object 51 in which 40 is arranged in this order from the front polarizing plate 10 side, and a second imaging object 52 comprising the front polarizing plate 10 and the backlight 40. A correction coefficient storage unit 130 that stores a correction coefficient 90 associated with each position of the liquid crystal panel 20 generated using the second image 80 obtained by imaging from the surface polarizing plate 10 side, and a table The inspection object 50, in which the polarizing plate 10, the liquid crystal panel 20 to be inspected, the back polarizing plate 30, and the backlight 40 are arranged in this order, is imaged by the imaging unit 300 from the front polarizing plate 10 side and the inspection image 60. The inspection image acquisition control unit 121 to be acquired, the density value of the pixel of interest 61a that is the pixel 61 corresponding to a certain position of the liquid crystal panel 20 in the inspection image 60 acquired by the inspection image acquisition control unit 121, and the inspection Based on the gray value of the background pixel 61c, which is a pixel 61 different from the pixel of interest 61a in the image 60, and the correction coefficient 90 associated with the certain position stored in the correction coefficient storage unit 130, the liquid crystal panel 20 The inspection apparatus may include a defect determination unit 127 that determines whether or not a defect is generated at the certain position.

(Program and recording medium)
The main control unit 100 of the inspection apparatus 1 may be configured by hardware logic. Alternatively, it may be realized by software using a CPU (Central Processing Unit) as follows.

  That is, the main control unit 100 includes a CPU such as an MPU that executes instructions of a program that realizes each function, a ROM (Read Only Memory) that stores the program, and a RAM (Random Access) that expands the program into an executable format. Memory) and a storage device (recording medium) such as a memory for storing the program and various data.

  The object of the present invention is not limited to the case where the program is stored in the program memory of the main control unit 100, and the program code (executable program, intermediate code program, or source program) of the above program is recorded. This can also be achieved by supplying a recording medium to the inspection device 1 and reading and executing the program code recorded on the recording medium.

  The recording medium is not limited to a specific structure or type. That is, the recording medium includes, for example, a tape system such as a magnetic tape and a cassette tape, a magnetic disk such as a floppy (registered trademark) disk / hard disk, and an optical disk such as a CD-ROM / MO / MD / DVD / CD-R. System, a card system such as an IC card (including a memory card) / optical card, or a semiconductor memory system such as a mask ROM / EPROM / EEPROM / flash ROM.

  Further, even if the main control unit 100 (or the inspection apparatus 1) is configured to be connectable to a communication network, the object of the present invention can be achieved. In this case, the program code is supplied to the main control unit 100 via the communication network. The communication network is not limited to a specific type or form as long as it can supply the program code to the main control unit 100. For example, the Internet, intranet, extranet, LAN, ISDN, VAN, CATV communication network, virtual private network, telephone line network, mobile communication network, satellite communication network, etc. may be used.

  The transmission medium constituting the communication network may be any medium that can transmit the program code, and is not limited to a specific configuration or type. For example, even with wired lines such as IEEE 1394, USB, power line carrier, cable TV line, telephone line, ADSL (Asymmetric Digital Subscriber Line) line, infrared rays such as IrDA and remote control, Bluetooth (registered trademark), 802.11 wireless, HDR, NFC It can also be used by radio such as DLNA, mobile phone network, satellite line, and digital terrestrial network. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.

  Thus, the inspection method according to an embodiment of the present invention is a method for inspecting the liquid crystal panel 20, and includes a first image acquisition step, a second image acquisition step, a correction coefficient calculation step (viewing angle characteristic information generation step), An inspection image acquisition process and a defect determination process can be included.

  In the first image acquisition step, the inspection apparatus 1 including the imaging unit 300 includes a front polarizing plate 10, a liquid crystal panel 20 ′ used as a standard, a back polarizing plate 30, and a backlight 40 arranged in this order. The object 51 is imaged from the front polarizing plate 10 side to obtain the first image 70.

  In the second image acquisition step, the inspection apparatus 1 acquires the second image 80 by imaging the second object to be imaged 52 including the front polarizing plate 10 and the backlight 40 from the front polarizing plate 10 side.

  In the correction coefficient calculation step, the inspection apparatus 1 uses the first image 70 acquired in the first image acquisition step and the second image 80 acquired in the second image acquisition step for each position of the liquid crystal panel 20. The correction coefficient 90 associated with is calculated.

  In the inspection image acquisition process, the inspection apparatus 1 uses the front polarizing plate 10, the liquid crystal panel 20 to be inspected, the back polarizing plate 30, and the backlight 40 in this order as the inspection target object 50. An inspection image 60 is acquired by imaging from the 10 side.

  In the defect determination step, the inspection device 1 in the inspection image 60 acquired in the inspection image acquisition step, the gray value of the target pixel 61a that is the pixel 61 corresponding to a certain position of the liquid crystal panel 20, and the inspection image 60, based on the grayscale value of the background pixel 61c, which is a pixel 61 different from the target pixel 61a, and the correction coefficient 90 associated with the certain position generated in the correction information generation step. It is determined whether or not a defect has occurred in the certain position.

  In the above description, the inspection apparatus 1 performs all the processes. However, the present invention is not limited to this, and each process may be performed by separate apparatuses.

[Description of functions]
Hereinafter, the principle of the present invention and the function of each part will be described in detail.

(Changes in viewing angle characteristics of micro defects)
Here, a change in viewing angle characteristics in a foreign matter mixed type defect such as a micro defect will be described. In this case, normally black is assumed, but normally white may be used. In the following, a voltage is applied to the liquid crystal layer 21 so as to be lit black.

  First, a minute defect will be described with reference to FIG. FIG. 4 is a schematic diagram for explaining the state of light when a foreign substance is mixed in the liquid crystal layer 21 of the liquid crystal panel 20. First, the light 41 emitted from the backlight 40 reaches the back polarizing plate 30 as scattered light. The scattered light is not constant in polarization direction and is polarized in various directions, but only the linearly polarized light component (here, horizontally polarized light) is extracted by the polarization property of the back polarizing plate 30. The linearly polarized lights 42 a to 42 c normally reach the front polarizing plate 10 through the liquid crystal panel 20. Here, since the front polarizing plate 10 intentionally changes the 90 ° polarization direction from the back polarizing plate 30 (here, the polarization in the vertical direction), the light 42 a and the light 42 c are shielded by the front polarizing plate 10. . On the other hand, as shown in FIG. 4, when the foreign matter 29 is mixed in the liquid crystal layer 21 of the liquid crystal panel 20, the foreign matter 29 scatters the light 42 b that has passed through and changes its polarization direction. That is, the light 43 that has passed through the foreign material 29 is not in a linearly polarized state, so that the light 44 that is the polarization direction component of the front polarizing plate 10 is transmitted through the front polarizing plate 10. Therefore, the defect portion becomes a bright spot. In the case of normally-white, the polarization directions of the back polarizing plate 30 and the front polarizing plate 10 are the same. However, since the liquid crystal layer 21 has an optical rotation of 90 °, the defect is similar to that of normally black. The part becomes a bright spot.

  Next, viewing angle characteristics in the liquid crystal panel inspection will be described with reference to FIG. FIG. 5 is a schematic diagram for explaining the viewing angle depending on the observation position. The viewing angle characteristics of the liquid crystal panel 20 are considered to vary mainly depending on the polarizing plates 10 and 30 arranged on the front and back sides.

  First, the viewing angle characteristics of a region (background portion) where the foreign matter 29 is not mixed in the liquid crystal layer 21 of the liquid crystal panel 20 will be described. When the optical system 320 is observed in the vertical direction 321 with respect to the liquid crystal panel 20, the front polarizing plate 10, and the back polarizing plate 30, the crossed Nicols characteristics of the front polarizing plate 10 and the back polarizing plate 30 are well exhibited. Therefore, the light is completely blocked. However, when the observation angle is changed, especially when the viewing angle is directed in the oblique direction 322 as shown in FIG. 5, the crossed Nicols characteristics of the front polarizing plate 10 and the back polarizing plate 30 are deteriorated. Light leakage will occur.

  On the other hand, the viewing angle characteristics are slightly different from the above in a minute defect (a region where the foreign material 29 is mixed in the liquid crystal layer 21). As described above, since light is scattered in the foreign material 29 in the minute defect, even when observed in the vertical direction 321 with respect to the liquid crystal panel 20, the front polarizing plate 10, and the back polarizing plate 30, the front polarizing plate Light of 10 polarization direction components is transmitted. In addition, even when the crossed Nicols characteristics of the front polarizing plate 10 and the back polarizing plate 30 are deteriorated in the observation in the oblique direction 322, the scattered light state slightly changes. Is transparent.

  Thus, the viewing angle characteristics are different between the background portion and the minute defect portion, and an inspection method that takes this viewing angle property into consideration is required.

(Viewing angle correction coefficient calculation process)
Here, the step of calculating the correction coefficient of the viewing angle characteristic will be described.

  As described above, the correction information generation control unit 110 performs imaging as illustrated in FIGS. 2A and 2B and acquires the first image 70 and the second image 80.

  First, the first image 70 will be described. As shown in FIG. 2A, the first image 70 is arranged as “backlight, back polarizing plate, liquid crystal panel, front polarizing plate”, and is acquired by the imaging unit 300 capturing the image. It is an image. The first image 70 is regarded as an image of a normal part having no defect in the liquid crystal panel inspection. The first image acquisition control unit stores the first image 70 in the first image storage unit 112.

  Next, the second image 80 will be described. As shown in FIG. 2B, the second image is an image that is arranged as “backlight, front polarizing plate” and is acquired by the imaging unit 300 imaging it. The purpose of this is to allow the scattered light 41 emitted from the backlight 40 to pass through the front polarizing plate 10 so as to be polarized in only one direction. This is a pseudo representation of what is observed through the plate 10. As a result, the result when the scattered light passes only through the front polarizing plate 10 can be simulated.

  Next, the correction coefficient calculation process will be described. In the present embodiment, a method of correcting the evaluation value (feature value 62) used for the inspection by taking the ratio between the first image 70 and the second image 80 is used, but the present invention is not limited to this. For example, a method of correcting the gray values of the defective portion (target pixel 61a) and the background portion (background pixel 61c) using the two image gray values of the first image 70 and the second image 80 may be used. .

  As described above, the first image 70 is considered as a normal background portion having no defect, and the second image 80 is considered to correspond to a minute defect portion. In the present invention, the correction coefficient 90 is calculated by comparing the first image 70 having the viewing angle characteristics of the background portion and the second image 80 having the viewing angle characteristics of the defective portion.

  First, the correction coefficient calculation unit 115 reads the gray value of the first image 70 and the second image 80 stored in the first image storage unit 112 and the second image storage unit 114, and calculates the gray value of the second image 80. Divide by the gray value of the first image 70.

  FIG. 6 is a graph showing a correction coefficient calculation process. The three-dimensional graph shown in FIG. 6 represents data about a region obtained by dividing the first image 70 or the second image into four, and the right side before the graph shows the center of the image.

  FIG. 6A shows the result of dividing the gray value of the second image 80 by the gray value of the first image 70. As shown in FIG. 6A, the result of the division includes noise and the like. This is due to random noise during image capturing, fixed noise, defects in the liquid crystal panel used as a standard, and it is preferable to eliminate the influence of these noises and the like when calculating the correction coefficient. For this reason, the correction coefficient calculation unit 115 further performs processing to eliminate the influence of noise.

  As a method for eliminating the influence of noise, in the present embodiment, the correction coefficient calculation unit 115 performs fitting to a bicubic surface, but is not limited to this, and other than fitting to a bicubic surface Further, fitting to an n-order curved surface, smoothing processing using a filter, or the like may be used. These processes are preferably methods in which the average error between the result of eliminating noise and the original data is suppressed to about several percent.

  FIG. 6B shows the fitting result. The correction coefficient calculation unit 115 further normalizes all the pixels by setting the value at the center of the image to 1 for this fitting result, and this value is stored in the correction coefficient storage unit 130 as the correction coefficient 90 associated with each pixel. Remember. FIG. 6C shows a graph representing the correction coefficient 90 after the above normalization. The data format of the correction coefficient 90 can be, for example, a matrix having the correction coefficient 90 associated with each pixel as an element. The correction coefficient calculation unit 115 may also display the result on the display unit 200 so that the operator can check it.

(Correction method during inspection)
Next, a method for correcting the feature amount 62 using the correction coefficient 90 will be described.

  In the inspection process including the defect determination process, the imaging unit 300 of the inspection apparatus 1 images the liquid crystal panel 20 to be inspected. The inspection image acquisition control unit 121 stores the inspection image 60 obtained by the imaging in the inspection image storage unit 122 in the main control unit 100.

  Subsequently, the feature amount calculation unit 123 and the defect candidate extraction unit 125 perform processing for extracting defect candidates in the liquid crystal panel 20 without using the correction coefficient 90.

  First, the feature amount calculation unit 123 calculates a feature amount 62 for detecting a defect in the liquid crystal panel 20 using the inspection image 60 stored in the inspection image storage unit 122. The feature amount 62 is associated with each position of the liquid crystal panel 20 or each pixel of the inspection image 60, and is represented, for example, as a matrix having the feature amount associated with each position (pixel) as an element.

  When performing defect detection processing by image processing, in order to detect a defect, it is necessary to compare a target pixel for determining whether or not the pixel is a defective pixel with a background pixel that is regarded as not a defective pixel. Therefore, as the feature amount 62, information indicating the difference between the gray value of the background pixel regarded as a pixel having no defect and the gray value of the target pixel is used.

  Such a feature amount 62 is not particularly limited, and for example, normalized contrast, luminance, non-normalized contrast, luminance, difference value, luminance volume, and the like can be used. In the present embodiment, in order to eliminate the influence of the luminance decrease at the end of the inspection image 60 due to the shading of the lens included in the optical system 320 of the imaging unit 300, the value obtained as the difference in each pixel as the feature amount 62 However, other features such as brightness, contrast without normalization, and the like may be used.

  Here, the normalized contrast used in the present embodiment will be described with reference to the drawings. FIG. 7 shows the liquid crystal panel 20 and its picture element 22 to be inspected and the inspection image 60 and its pixel 61 in an overlapped manner. When the foreign substance 29 is mixed in the picture element 22, FIG. The case where a defect is determined is shown with the pixel 61 corresponding to the position 29 mixed as the target pixel 61a.

  As shown in FIG. 7, when a certain pixel 61 is the target pixel 61a, the background pixel 61c may be any pixel that is different from the target pixel 61a, but preferably the repetitive pattern in the liquid crystal panel 20 is defective. The difference is generated by taking the difference between the portion (the picture element 22 corresponding to the pixel of interest 61a) and the background point assigned the same. At this time, it is preferable that the comparison pixel uses several points as close as possible, and uses the average of those points. That is, as shown in FIG. 7, as the background pixel 61 c of the target pixel 61 a corresponding to approximately the center of the green (G) picture element 22, the pixel 61 corresponding to approximately the center of the green (G) picture element 22. It is preferable to use a background pixel 61c indicated by hatching in the vicinity of the target pixel 61a, and the feature value 62 is a difference between the average value of the gray values of the background pixel 61c and the gray value of the target pixel 61a. It is preferable to obtain it as information representing.

  More specifically, as shown in FIG. 8, the background pixel is not the pixel 61b that overlaps the pixel 22b with a pattern different from the overlap of the pixel of interest 61a and the pixel 22a, but the pixel 22 with the same pattern. The background pixel 61c overlapping is used.

  The feature amount calculation unit 123 overlaps the background pixel 61c as described above based on, for example, the setting of the resolution [μm / pix] in the optical system 320 of the imaging unit 300 and the size of the picture element 22 of the liquid crystal panel 20. This can be selected by deriving the pattern. For example, when the resolution is l [μm / pix] and the size (horizontal × vertical) of the picture element 22 is m × n [μm], 1 cm (l, m) / l in the horizontal direction from the target pixel 61a. A pixel 61 that is separated by 1 cm (l, n) / l pixel in the vertical direction can be suitably used as the background pixel 61c.

  Next, the difference (contrast) obtained as described above is normalized. As described above, the captured image obtained by imaging the liquid crystal panel 20 by the solid-state imaging device 310 (area sensor) of the imaging unit 300 has a luminance distribution due to the shading of the optical system lens or the like. The value will be different. In order to consider the influence of the luminance distribution at such an image position, in this embodiment, the feature amount calculation unit 123 normalizes the difference value between the target pixel and the background pixel with the background luminance. This normalized difference value is called normalized contrast and is used as one index representing the feature of the defect.

  If the gray value of the target pixel 61a of the (i, j) component is Ie (i, j), and the gray value of the background pixel 61c corresponding to the target pixel 61a of the (i, j) component is Ib (i, j), The value of the normalized contrast C can be obtained as in the following formula (1).

C = (Ie (i, j) -Ib (i, j)) / Ib (i, j) (1)
The feature amount calculation unit 123 stores the normalized contrast for each pixel calculated as described above in the feature amount storage unit 124 as the feature amount 62.

  Subsequently, the defect candidate extraction unit 125 performs a binarization process by setting a threshold value for the normalized contrast C, and obtains an area in the binarized region. “Area in binarized region” refers to the number of all connected pixels when pixels that are equal to or greater than a threshold are connected when binarization processing is performed. For example, when the pixels exceeding the threshold are not connected and exist as a single pixel, the area is 1.

  Then, the defect candidate extraction unit 125 calculates a normalized contrast volume obtained by adding the normalized contrast C in the binarized region, extracts defect candidates whose normalized contrast volume is equal to or greater than a certain threshold, Store in the candidate storage unit 126.

  Note that such a normalized contrast volume includes both normalized contrast and area. Here, in the actual defect inspection, there are cases where the above-described one having a small contrast (or luminance) even if the area is large, and conversely, the one having a small area even if the contrast (luminance) is large may be allowed. For this reason, it is difficult to determine either the area or the contrast (brightness). In this embodiment, the defect is determined using the contrast volume. Note that the determination index in defect detection is not limited to the normalized contrast volume, and brightness, a difference value, a brightness volume, or the like may be used.

  However, it is considered that the defect candidate group extracted by the defect candidate extraction unit 125 is based on the feature amount having no variation in viewing angle characteristics and having in-plane variation. Therefore, the defect determination unit 127 uses the correction coefficient 90 stored in the correction coefficient storage unit 130 to perform correction considering the viewing angle characteristics.

  In the present embodiment, as the correction coefficient 90, the background pixel is obtained by using the value obtained by dividing the gray value of each pixel of the second image 80 by the gray value of the corresponding pixel of the first image 70 in the correction coefficient calculation step. Perform the correction. When the value of the (i, j) component of the correction coefficient 90 is α (i, j), the corrected background pixel Ib ′ (i, j) of the (i, j) component is expressed by the following equation (2). Can be asking.

Ib ′ (i, j) = α (i, j) × Ib (i, j) (2)
Here, considering the above equation (1), the normalized contrast C ′ after correction can be obtained as in the following equation (3).

C ′ = (C + 1 + α) / α (3)
The defect determination unit 127 uses the feature amount 62 read from the feature amount storage unit 124 and the correction coefficient 90 read from the correction coefficient storage unit 130 to obtain a normalized contrast after correction according to the above equation (3). Then, the defect determination unit 127 corrects the normalized contrast for the defect candidate group stored in the defect candidate storage unit 126 using the obtained normalized contrast after correction, and finally calculates the normalized contrast volume. Determine the defect.

(Effects of correction of viewing angle characteristics)
FIG. 9 shows the effect of the correction considering the viewing angle characteristics according to the present invention. FIG. 9 shows a normalized contrast volume (center portion) when the same minute defect is imaged so as to be located at the center portion of the captured image, and a normalized contrast when the image is positioned so as to be positioned at the end portion of the captured image. The result (corrected rear end part) obtained by performing correction according to the present invention on the volume (corrected front end part) and the normalized contrast volume of the corrected front end part is shown.

  As shown in FIG. 9, before the viewing angle correction using the correction coefficient 90 according to the present invention, the image is picked up so that it is located at the center of the picked-up image and the edge of the picked-up image. Sometimes there is a difference in the normalized contrast volume. However, as a result of performing the viewing angle correction according to the present invention, it can be seen that the normalized contrast volume at the rear end of the correction approaches the normalized contrast volume at the center, and an equivalent result is obtained.

  As described above, it was confirmed that the effect of correcting the viewing angle characteristic according to the present invention was obtained. That is, according to the present invention, it is possible to perform the same reference inspection within the image plane.

  In the above description, all cases where the liquid crystal panel is lit black have been described. However, the inspection apparatus according to the present invention can also be applied when the liquid crystal panel is lit white. Even when the liquid crystal panel is lit white, as shown in FIG. 4, the light 43 through the foreign material 29 becomes scattered light, so that the viewing angle characteristics are different from the light 42 a and 42 c through the defect-free portion. Have When the inspection apparatus according to the present invention is used, the imaging unit 300 includes the front polarizing plate 10, the liquid crystal panel 20 ′ used as a standard, the back polarizing plate 30, and the backlight 40 arranged in this order. And information on both the viewing angle characteristics of the liquid crystal panel having no defect and the viewing angle characteristics of the liquid crystal panel having the defect are acquired by imaging the second imaging object 52 including the front polarizing plate 10 and the backlight 40. Therefore, when the inspection object 50 including the liquid crystal panel 20 to be inspected is imaged and the liquid crystal panel 20 is inspected, the inspection is performed in consideration of the viewing angle characteristics of the liquid crystal panel having a defect. Can do.

  The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope shown in the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. Is also included in the technical scope of the present invention. For example, the following first to eighth configurations are also within the scope of the present invention.

(First configuration)
In an inspection apparatus for a liquid crystal panel, a means for imaging an object composed of a backlight, a back polarizing plate, a liquid crystal panel, and a front polarizing plate and registering it as a first image, and a backlight / front polarizing plate Means for imaging a configured object and registering it as a second image; means for dividing the registered second image by the gray value of the first image; and repeating the liquid crystal panel in actual inspection processing A means for normalizing the difference value of the gray value with the background pixel to be compared with a pixel allocation equivalent to that of the defect occurrence portion with respect to the pattern, and the gray value of the background pixel for the normalization An image processing inspection apparatus for correcting a registered second image by a value obtained by dividing the registered second image by a gray value of the first image and calculating a uniform feature amount in a plane.

(Second configuration)
In an inspection apparatus for a liquid crystal panel, a means for imaging an object composed of a backlight, a back polarizing plate, a liquid crystal panel, and a front polarizing plate and registering it as a first image, and a backlight / front polarizing plate Means for imaging the configured object and registering it as a second image, and registering a background pixel to be compared with a pixel allocation equivalent to that of the defect generation part for the repeated pattern of the liquid crystal panel in the actual inspection process Means for dividing and registering the pixel gray value of the same coordinate in the first image, and dividing the pixel gray value of the pixel in the defect generating unit by the pixel gray value of the same coordinate in the second image; A difference value between the pixel gray value of the defective portion divided by the second image and the pixel gray value of the background portion divided by the first image by the pixel gray value of the background portion To do To obtain a corrected result of viewing angle characteristics I was characterized by an image processing inspection apparatus.

(Third configuration)
The first image is obtained by placing a liquid crystal panel serving as a reference on an inspection table composed of a backlight for inspection, a back polarizing plate, and a front polarizing plate, and capturing an image of the liquid crystal panel having viewing angle characteristics. And registering the image processing inspection apparatus of the first configuration.

(Fourth configuration)
The second image simulates the scattered light of the defect in which the foreign substance is mixed in the liquid crystal by the configuration of the inspection backlight and the front polarizing plate, and captures the visual angle characteristic of the foreign substance defect in the liquid crystal by imaging this. An image processing inspection apparatus having the first configuration, characterized in that an image is acquired and registered.

(Fifth configuration)
The means for dividing the registered second image by the gray value of the first image is a ratio of the first image having the viewing angle characteristic of the liquid crystal panel and the second image having the viewing angle characteristic of the foreign matter defect in the liquid crystal. The image processing inspection apparatus according to the first configuration, wherein the calculation is performed for all pixels of an image, and the influence of noise components is eliminated with respect to the result using a least square method or a smoothing filter.

(Sixth configuration)
An image processing inspection method for a liquid crystal panel, in which an image of a target composed of a backlight, a back polarizing plate, a liquid crystal panel, and a front polarizing plate is captured, and the target composed of a backlight and a front polarizing plate is captured. A correction coefficient calculation step of calculating a correction coefficient of the viewing angle characteristic by dividing the second image, and a background to be compared with a pixel allocation equivalent to the defect occurrence portion for the repeated pattern of the liquid crystal panel in the actual inspection process An inspection method comprising: an inspection step of normalizing the difference value of the gray value with the pixel by the background pixel and correcting the result by the correction coefficient obtained in the step.

(Seventh configuration)
An inspection program for causing a computer to execute the inspection method of the sixth configuration, wherein the computer causes the computer to execute each of the steps.

(Eighth configuration)
The computer-readable recording medium which recorded the test | inspection program of the said 7th structure.

  The present invention can be used in the field of manufacturing liquid crystal panels.

DESCRIPTION OF SYMBOLS 1 Inspection apparatus 10 Front polarizing plate 20 Liquid crystal panel (liquid crystal panel of test object)
20 'liquid crystal panel used as standard 21 liquid crystal layer 22 picture element 29 foreign material 30 back polarizing plate 40 backlight (light source)
41-44 Light 50 Inspection object 51 First image pickup object 52 Second image pickup object 60 Inspection image 70 First image 80 Second image 61a Pixel of interest 61c Background pixel 90 Correction coefficient (viewing angle characteristic information)
100 Main Control Unit 110 Correction Information Generation Control Unit 111 First Image Acquisition Control Unit (First Image Acquisition Control Unit)
113 2nd image acquisition control part (2nd image acquisition control means)
115 Correction coefficient calculation unit (correction information generation means)
120 Inspection Control Unit 121 Inspection Image Acquisition Control Unit (Inspection Image Acquisition Control Unit)
127 Defect determination unit (defect determination means)
130 Correction coefficient storage unit (viewing angle characteristic information storage means)
200 Display unit 300 Imaging unit (imaging means)
310 Solid Image Sensor 320 Optical System 400 Object Placement Unit

Claims (6)

An inspection device for a liquid crystal panel,
Imaging means;
A first image for acquiring a first image by causing the imaging means to take an image from the front polarizing plate side of a first object to be imaged in which a front polarizing plate, a liquid crystal panel used as a standard, a back polarizing plate, and a light source are arranged in this order. Acquisition control means;
Second image acquisition control means for acquiring a second image by causing the image pickup means to pick up an image of the second object to be picked up comprising a front polarizing plate and a light source;
Viewing angle characteristic information associated with each position of the liquid crystal panel is generated using the first image acquired by the first image acquisition control unit and the second image acquired by the second image acquisition control unit. Viewing angle characteristic information generating means;
Inspection image for acquiring an image for inspection by causing the imaging means to image from the surface polarizing plate side an object to be inspected in which a front polarizing plate, a liquid crystal panel to be inspected, a back polarizing plate and a light source are arranged in this order Acquisition control means;
In the inspection image acquired by the inspection image acquisition control means, the gray value of the target pixel which is a pixel corresponding to a certain position of the liquid crystal panel, and the background pixel which is a pixel different from the target pixel in the inspection image Whether or not there is a defect in the certain position of the liquid crystal panel to be inspected based on the gray value of the image and the viewing angle characteristic information associated with the certain position generated by the viewing angle characteristic information generating unit and a defect judgment means for,
The viewing angle characteristic information associated with the certain position represents a ratio between the gray value of the pixel corresponding to the certain position in the first image and the gray value of the pixel corresponding to the certain position in the second image. Information,
The defect determination means compares the grayscale values of the target pixel and the background pixel, at least one of which is corrected using the viewing angle characteristic information, to thereby determine the position of the liquid crystal panel to be inspected. An inspection apparatus for determining whether or not a defect has occurred .   The inspection apparatus according to claim 1, wherein the defect is contamination of a liquid crystal panel. An inspection device for a liquid crystal panel,
Imaging means;
A first image obtained by imaging from the front polarizing plate side, a first image obtained by imaging a front polarizing plate, a liquid crystal panel used as a standard, a back polarizing plate and a light source in this order, and the front polarizing plate and Stores viewing angle characteristic information associated with each position of the liquid crystal panel, which is generated using a second image obtained by imaging a second object to be imaged composed of a light source from the front polarizing plate side. Viewing angle characteristic information storage means;
Inspection image for acquiring an image for inspection by causing the imaging means to image from the surface polarizing plate side an object to be inspected in which a front polarizing plate, a liquid crystal panel to be inspected, a back polarizing plate and a light source are arranged in this order Acquisition control means;
In the inspection image acquired by the inspection image acquisition control means, the gray value of the target pixel which is a pixel corresponding to a certain position of the liquid crystal panel, and the background pixel which is a pixel different from the target pixel in the inspection image Whether or not there is a defect in the certain position of the liquid crystal panel to be inspected based on the gray value of the image and the viewing angle characteristic information associated with the certain position stored in the viewing angle characteristic information storage unit and a defect judging means for judging,
The viewing angle characteristic information associated with the certain position represents a ratio between the gray value of the pixel corresponding to the certain position in the first image and the gray value of the pixel corresponding to the certain position in the second image. Information,
The defect determination means compares the grayscale values of the target pixel and the background pixel, at least one of which is corrected using the viewing angle characteristic information, to thereby determine the position of the liquid crystal panel to be inspected. An inspection apparatus for determining whether or not a defect has occurred . An inspection method for a liquid crystal panel,
An inspection apparatus equipped with an imaging means picks up a first image to be captured from the front polarizing plate side by taking a first object to be imaged in which a front polarizing plate, a liquid crystal panel used as a standard, a back polarizing plate and a light source are arranged in this order. A first image acquisition step to acquire;
A second image acquisition step in which the inspection apparatus captures a second image pickup object including a front polarizing plate and a light source from the front polarizing plate side to acquire a second image;
Viewing angle characteristic information associated with each position of the liquid crystal panel using the first image acquired in the first image acquisition step and the second image acquired in the second image acquisition step by the inspection apparatus. Viewing angle characteristic information generating step for generating
The above inspection apparatus is for inspection to obtain an image for inspection by imaging from the front polarizing plate side an object to be inspected in which a front polarizing plate, a liquid crystal panel to be inspected, a back polarizing plate and a light source are arranged in this order. An image acquisition process;
The inspection apparatus acquired in the inspection image acquisition step in the inspection image, and the gray value of the target pixel corresponding to a certain position of the liquid crystal panel and the pixel different from the target pixel in the inspection image Whether a defect has occurred in the certain position of the liquid crystal panel to be inspected based on the gray value of the background pixel and the viewing angle characteristic information associated with the certain position generated in the viewing angle characteristic information generating step. Including a defect determination step of determining whether or not ,
The viewing angle characteristic information associated with the certain position represents a ratio between the gray value of the pixel corresponding to the certain position in the first image and the gray value of the pixel corresponding to the certain position in the second image. Information,
In the defect determination step, at least one of them is corrected using the viewing angle characteristic information, and the grayscale values of the target pixel and the background pixel are compared, so that the predetermined position of the liquid crystal panel to be inspected is obtained. An inspection method characterized by determining whether or not a defect has occurred .   A program for causing a computer to operate as the inspection apparatus according to any one of claims 1 to 3, wherein the program causes the computer to function as each means included in the inspection apparatus.   A computer-readable recording medium on which the program according to claim 5 is recorded.

Images