KR20180030310A - Defect detecting system of polarization unit and detecting method thereof - Google Patents

Abstract

The present invention relates to a polarizing unit test device which is composed of a light source unit, a first reference polarizing member positioned in an upper part of the light source unit, a second reference polarizing member positioned in an upper part of the first reference polarizing member, a photographing unit positioned in an upper part of the second reference polarizing member, and a test controller receiving a plurality of images from the photographing unit, synthesizing the plurality of images into a test image, and analyzing the test image. Therefore, a defect state of a polarizing unit can be determined and analyzed through the polarizing unit test device.

Description

[0001] DEFECT DETECTING SYSTEM OF POLARIZATION UNIT AND DETECTING METHOD THEREOF [0002]

The present invention relates to a system for automatically inspecting and analyzing defects in a polarizing unit, and more particularly, to a system for detecting defects occurring in a polarizing unit at a high speed and constructing and quantifying a defect map, To the polarization unit defect detection apparatus.

A polarizing unit used in a display device transmits only light vibrating in the same direction as the polarization axis in the light, and absorbs or reflects light vibrating in the other direction using a suitable medium to oscillate in a specific direction It is a film that plays a role of making.

Since the light emitted from the backlight of the liquid crystal display device propagates in the same direction in all directions and polarizing units are attached to both sides of the liquid crystal cell of the LCD module such that the polarization axes are orthogonal or parallel to each other, The intensity of the transmitted light is adjusted according to the degree of rotation of the light source. The liquid crystal display can express the gray level by adjusting the transmitted light intensity.

Recently, a display device having a structure in which a polarizing unit is built in a display device has been developed, and a polarizing unit built in a display device is previously inspected. In order to check whether a polarizing unit has foreign matters or holes, It is important to shorten the time required for the inspection for the whole inspection.

Accordingly, the present invention provides a method of inspecting a polarizing unit capable of inspecting a polarizing unit at a high speed to judge whether or not the polarizing unit is defective and sorting the types of defects, and an inspection apparatus using the same.

The apparatus for inspecting a polarizing unit according to an embodiment of the present invention includes a light source unit, a first reference polarizing member having a polarization axis in a first direction located on the light source unit, a second reference polarizing member positioned on an upper portion of the first reference polarizing member, A plurality of images are received from the photographing unit and the photographing unit located at the upper portion of the reference polarizing member, and the plurality of images are synthesized with the inspection image to obtain the to-be-inspected polarized light unit positioned between the first reference polarizing member and the second reference polarizing member And a check controller for judging whether or not the defect is defective.

The second reference polarizing member of the apparatus for inspecting a polarizing unit according to the embodiment of the present invention has a polarization axis in either one direction parallel or perpendicular to the first direction.

A plurality of images of the apparatus for inspecting a polarizing unit according to an embodiment of the present invention includes a first image obtained by photographing output light of the first reference polarizing member in a state that the light source unit is turned on, The third image obtained by photographing the output light of the second reference polarization member having the polarization axis parallel to the polarization axis of the reference polarization member and the output light of the third reference polarization member having the polarization axis perpendicular to the polarization axis of the first reference polarization member And includes at least three or more images out of one fourth image.

An inspection controller of an apparatus for inspecting a polarizing unit according to an exemplary embodiment of the present invention includes an image storage for storing a plurality of images, an image synthesizer for synthesizing a plurality of stored images to generate an inspection image, A binarization unit for binarizing the pixel data of the screen block, and an image combining unit for combining the plurality of binarized image blocks to generate an output image.

The inspection controller of the apparatus for inspecting a polarizing unit according to an embodiment of the present invention further includes a defect determination unit for automatically determining a position of a defect in an output image.

The defect determination unit of the apparatus for inspecting a polarized light unit according to the embodiment of the present invention classifies the causes of defects and displays them.

The inspection controller of the apparatus for inspecting a polarizing unit according to the embodiment of the present invention further includes a display device for superimposing and displaying the inspection image and the defect.

A polarizing unit inspection method according to an embodiment of the present invention includes a photographing step of photographing a plurality of images, an image synthesizing step of synthesizing a plurality of images into an inspection image, an image dividing step of dividing the inspection image into a plurality of screen blocks, A binarization step of binarizing a plurality of picture blocks to generate a binarized picture block, a block unit combining step of combining binarized picture blocks to generate a binarized defect image, and a defect discrimination step of discriminating a defect by analyzing the combined defect images do.

In the polarimetric unit inspection method according to the embodiment of the present invention, the photographing step of photographing a plurality of images includes disposing a light source unit and a first reference polarizing member having a polarization axis in a first direction and irradiating the first reference polarizing member A first image capturing step of capturing an output light of the object to be inspected, a first image capturing step of capturing an output light, an inspection object having a polarization axis perpendicular to the first direction on an upper portion of the first reference polarization element, A third image capturing step of disposing a second reference polarizing element having a polarization axis in a second direction on an upper portion of the object to be inspected and capturing the output light of the second reference polarizing element in a lighted state of the light source section, A third reference polarization member having a polarization axis perpendicular to the second direction is disposed on an upper portion of the first reference polarization member, And a step.

In the polarizing unit inspection method according to the embodiment of the present invention, the second direction is either a direction parallel to the first direction or a direction perpendicular to the first direction.

The third polarizing member of the polarizing unit inspection method according to the embodiment of the present invention is arranged by rotating the second polarizing member by 90 degrees.

In the image synthesizing step of the method for inspecting a polarizing unit according to an embodiment of the present invention, a plurality of images are synthesized into inspection images by using different arithmetic expressions according to kinds of defects.

The dividing step of the polarizing unit inspection method according to the embodiment of the present invention divides the inspection image into a plurality of pixel blocks arranged in a matrix, and the pixel block includes at least 128 X 128 pixels.

The defect determination step of the polarizing unit inspection method according to the embodiment of the present invention merges adjacent plural defects to construct a defect map.

The polarizing unit inspection method according to an embodiment of the present invention further includes a defect display step of superimposing and displaying the defect map on the inspection image.

The defective marking step of the polarizing unit inspection method according to the embodiment of the present invention generates defect intensity based on the brightness and area of each defect in the defect map and classifies the defect grade based on the defect intensity.

The polarizing unit inspection method and the inspection apparatus of the present invention can detect the defects of the polarizing unit built in the display device and track the type of defective and the history of the defects, thereby easily grasping the cause of defects in the production process.

1 is a configuration diagram of an inspection apparatus of a polarizing unit according to an embodiment of the present invention.
2A is a schematic diagram of a single polarization inspection mode.
2B is a schematic diagram of a cross-checking mode.
2C is a schematic diagram of the double test mode.
2D is a schematic diagram of a parallel inspection mode.
Fig. 3 is a screen for measuring the image and the degree of polarization of the subject polarized light unit.
4 is a graph showing the polarization degree of the polarization unit shown in Fig.
5 is an operational block diagram of a polarization unit inspection apparatus according to an embodiment of the present invention.
6A is a graph of a single polarization inspection image and image data.
6B is a cross-sectional inspection image and image data graph.
6C is a double inspection image and image data graph.
6D is a parallel inspection image and image data graph.
7A is a graph of a hall inspection image and an image data.
FIG. 7B is a graph of foreign substance dispersion inspection image data and image data.
8A is a conceptual diagram of image block division according to an embodiment of the present invention.
FIG. 8B is a diagram illustrating an example of image block division according to an embodiment of the present invention.
9A is a hole defect inspection image and a binarization processed image according to an embodiment of the present invention.
FIG. 9B is an image of foreign matter dispersion defect inspection and binarization processing according to an embodiment of the present invention.
10 is an operational block diagram of a polarization unit inspection apparatus according to another embodiment of the present invention.
11 is a view comparing a profile of the inspection image with an offset removal profile.
12 is a display screen of a polarizing unit inspection apparatus according to an embodiment of the present invention.
13 is an image of a defect map according to an embodiment of the present invention.
14A to 14G are explanatory diagrams of a defective pixel search method.
15A and 15B are explanatory diagrams of a defective pixel search method.
16 is a comparison screen of a high-resolution optical image and a defect map of the present invention.
17 is a graph showing the defect intensity and the number of defects.
18 is a graph of defect occurrence status for each sample.
19 is a graph showing the analysis of the cause of failure of the polarization unit.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

While the present invention has been described in connection with certain embodiments, it is obvious to those skilled in the art that various changes and modifications may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. It is to be understood, however, that the scope of the present invention is not limited to the specific embodiments described above, and all changes, equivalents, or alternatives included in the spirit and technical scope of the present invention are included in the scope of the present invention.

The terms first, second, third, etc. in this specification may be used to describe various components, but such components are not limited by these terms. The terms are used only for the purpose of distinguishing one element from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second or third component, and similarly, the second or third component may be alternately named.

In order to clearly illustrate the present invention, parts not related to the description are omitted, and the same or similar components are denoted by the same reference numerals throughout the specification.

The polarizing member used in the present invention includes both a polarizing film and a polarizing plate polarizing substrate, and the polarizing unit includes a polarizing film that can be attached to the display device, a built-in polarizer embedded in the substrate of the display device as well as a polarizing substrate, And the like.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1 is a configuration diagram of an inspection apparatus of a polarizing unit according to an embodiment of the present invention.

1, the apparatus for inspecting a polarized light unit 10 includes a light source unit 100, a first reference polarizing member 110, a second reference polarizing member 130, a photographing unit 300, and an inspection controller 400 .

The apparatus for inspecting a polarized light unit 10 further includes an inspection jig 200 for fixing the first reference polarizing element 110, the second reference polarizing element 130 and the inspection polarized light unit 120. The subject polarized light unit 120 may be positioned between the first reference polarized light member 110 and the second reference polarized light member 130 fixed by the inspection jig 200.

The polarization unit inspection apparatus 10 may further include an exchange unit (not shown) capable of exchanging the inspection target polarization unit 120 and the second reference polarization member 130.

The light source unit 100 positioned below the inspection jig 200 outputs the uniformly diffused light in the direction of the first reference polarization member 110. The light source unit 100 includes a plurality of light sources (not shown) positioned on the lower surface and a diffusion film (not shown) that uniformly diffuses the output light of the light source. The light source unit 100 may include a light source located at a peripheral portion and a light guide plate and a diffusion film for guiding output light of the light source.

The first reference polarizing member 110 is disposed on the light source 100. The first reference polarizing member 110 has a polarization axis in the first direction. The inspection target polarizing unit 120 is positioned above the first reference polarizing member 110. The subject polarized light unit 120 is arranged so as to have a polarization axis perpendicular to the polarization axis of the first reference polarizing member 110. [ The polarization axis of the first reference polarizing member 110 and the polarization axis of the subject polarized light unit 120 are perpendicular to each other so that no light is output in the normal region of the subject polarized light unit 120.

 The second reference polarizing member 130 may be disposed on the inspection target polarizing unit 120 in accordance with the inspection mode. The second reference polarizing member 130 is arranged to have a polarization axis perpendicular or parallel to the polarization axis of the inspection target polarization unit 120 according to the inspection mode.

The photographing unit 300 is disposed above the second reference polarization member 130 and measures output light output from the light source unit 100 according to each inspection mode. The image photographed by the photographing unit 300 is transmitted to the inspection controller 400. The photographing unit 300 generally includes a CCD camera, converts the photographed image into digital image data, and transmits the digital image data to the inspection controller 400.

The inspection controller 400 can determine the defect of the inspection subject polarization unit 120 using an internal operation program. Also, it is possible to analyze the types and causes of defects in the subject polarized light unit 120 by analyzing a test image in which images of a plurality of inspection modes are synthesized.

2A to 2D are schematic diagrams of an inspection mode according to an embodiment of the present invention.

2A is a schematic diagram of a single polarization inspection mode. The single polarization inspection mode is an inspection method in which the output light of the light source unit 100 is output in the direction of the first reference polarization member 110 and the output light transmitted from the front surface of the first reference polarization member 110 is photographed. In the single polarization inspection mode, the output light of the light source unit 100 is polarized through the first reference polarizing member 110 having the polarization axis of the first direction. The image photographed by the photographing unit 300 is transmitted to the inspection controller 400 and is stored as a first image.

2B is a schematic diagram of a cross-checking mode. The cross inspection mode is a mode in which the first reference polarizing member 110 having the polarization axis in the first direction and the output light having passed through the to-be-inspected polarization unit 120 having the polarization axis perpendicular to the first direction among the output light of the light source unit 100 It is a test method to take. In theory, if the two polarization axes are stacked vertically, there should be no light passing through and thus the output light should not be measured. However, in practice, there is a case where a hole is formed in the inspection target polarizing unit and passes through the polarizing unit, or the polarized state of the incident light is dispersed by the foreign matter attached to the polarizing unit.

The output light of the first reference polarizing member 110 may pass through the subject polarized light unit 120 when the hole 121 or the foreign substance 122 is attached to the subject polarized light unit 120. [ The light output through the hole 121 of the subject polarized light unit 120 is light polarized in the same axis as the light that has passed through the first reference polarized member 110. On the other hand, the polarized light of the incident light is dispersed by the foreign object 122 of the subject polarized light unit 120, and the light passing through the subject polarized light unit 120 may have a polarization axis different from that of the first reference polarized light member 110.

That is, the output light passing through the subject polarized light unit 120 may have different polarization axes depending on the type of the defect. In the cross checking mode, the image output from the photographing unit 300 is transmitted to the inspection controller 400 and is stored as a second image.

 2C is a schematic diagram of the double test mode. The double inspection mode is an inspection method in which the second reference polarizing member 130 is disposed on the upper portion of the inspection target polarizing unit 120 in FIG. 2B such that the polarization axis is in the first direction and the output light is captured. In the double inspection mode, the output light passing through the hole 121 of the inspection target polarizing unit 120 is polarized in the first direction and can be transmitted through the second reference polarizing member 130. On the other hand, the output light due to the dispersion of the foreign object 122 of the subject polarized light unit 120 is shaded or attenuated by the second reference polarized member 130 as the polarization axis is different from the first direction. In the double inspection mode, the image output from the photographing unit 300 is transmitted to the inspection controller 400 and is stored as a third image.

2D is a schematic diagram of a parallel inspection mode. The parallel inspection mode is an inspection method for photographing output light passing through a third reference polarization member 140 having a polarization axis perpendicular to the first direction, which is located on the upper side of the inspection target polarization unit 120 of FIG. 2B. The output light passing through the hole 121 of the inspection target polarizing unit 120 in the parallel inspection mode is blocked or attenuated because it can not pass through the third reference polarizing member 140 having a polarization axis perpendicular to the first direction. On the other hand, the output light passed by the dispersion of the foreign object 122 of the inspection target polarizing unit 120 can be diffracted by the polarization axis and transmitted through the third reference polarizing member 140. In the parallel inspection mode, the image output from the photographing unit 300 is transmitted to the inspection controller 400 and is stored as a fourth image.

In the parallel inspection mode, the third reference polarizing member 140 may be disposed by rotating the second reference polarizing member 130, or a separate polarizing member other than the second reference polarizing member 130 may be used.

2A to 2D, the apparatus for inspecting a polarized light unit may include a plurality of inspection modes in which different reference polarizing members are combined to inspect defects of the inspection target polarizing unit 120, Can be obtained. The inspection controller 400 analyzes a plurality of photographed images to determine whether or not the inspection subject polarizing unit 120 is defective and analyze the types of defects.

Fig. 3 is a screen for measuring the image and the degree of polarization of the subject polarized light unit.

4 is a graph showing the polarization degree of the polarization unit shown in Fig.

3 is an image obtained by arranging a light source on the back surface of the subject polarized light unit 120 and measuring the light transmitted from the front surface of the subject polarized light unit 120. FIG. The subject polarized light unit 120 is a polarized light unit having a structure integrally formed inside the substrate. A bright point is confirmed in a part of the inspection subject polarizing unit 120, and a striped unevenness is formed in the vertical direction.

3 is an image showing the degree of polarization (DOP, Degree of Polarization) of the subject polarized light unit 120 calculated based on the optical image of the left drawing. The polarization degree (DOP), which is a characteristic reference of the subject polarized light unit 120, can be expressed by the following equation (1).

[Equation 1]

Tmax and Tmin are the maximum and minimum optical output intensities outputted through the subject polarized light unit 120, respectively. Since the ideal polarizing unit produces a perfect polarization state, the degree of polarization (DOP) must be exactly one.

The polarization degree image is expressed by converting the degree of polarization of each of the inspection subject polarization unit 120 into a percentage. The portion visually recognized as vertical blur in the image is identified as the region where the deviation of the polarization degree occurs.

Referring to FIGS. 3 and 4, the polarization degree of the polarization unit is excellent in the central region of the inspection subject polarization unit 120 and low in the peripheral region. In the portion where the sharpness is lowered in the polarization degree graph, the unevenness of the polarization degree is abruptly generated, and the brightness unevenness may be caused by the difference of the polarization degree.

5 is an operational block diagram of a polarization unit inspection apparatus according to an embodiment of the present invention.

5, the inspection controller includes an image storage memory A, B, C and D, a hole image synthesizer 410, a scatter image synthesizer 411, a binarization processor 430, a block unit combiner 440, A hole defect discrimination unit 450, and a scatter defect discrimination unit 451. [

The inspection controller 400 receives a first image photographed in a single polarization inspection mode, a second image photographed in an intersection inspection mode, a third image photographed in a double inspection mode, and a fourth image photographed in a parallel inspection mode, And stores them in memory A, B, C, and D, respectively. The first to fourth image information are transmitted to the hole image combining unit 410 and the scatter image combining unit 411, respectively.

The Hole image synthesizing unit 410 synthesizes the first to fourth images to synthesize a first inspection image suitable for hole defect inspection, and the Scatter image synthesizing unit 411 synthesizes the first to fourth images, A second inspection image suitable for inspection is synthesized. The synthesized first and second inspection images are transmitted to the image block dividing unit 420, respectively. The detailed synthesis method will be described in more detail in the description of FIG.

The image block dividing unit 420 transmits the image block to the binarization processing unit 430. [ The binarization processor 430 binarizes each image block to generate a binarized image, and transmits the binarized image to the block unit combiner 440. The block unit combining unit 440 combines the binarized image blocks to generate a defect image. The generated defect image is transferred to the hole defect determination unit 450 or the scatter defect determination unit 451 according to the type of the defect. The defect discriminators 450 and 451 discriminate a defect, quantify the position of the defect and the strength of the defect, and output it.

5 illustrates that the Hole image synthesis unit 410 and the Scatter image synthesis unit 411 are composed of different blocks. However, the Hole image synthesis unit 410 and the Scatter image synthesis unit 411 are composed of one image synthesis unit . Also, the Hole defect determining unit 450 and the Scatter defect determining unit 451 may be formed of one block.

6A is a graph of a single polarization inspection image and image data.

The single polarization test is a first image obtained by photographing a state in which the output light of the light source unit 100 is transmitted through the first reference polarization member 110. The single polarization inspection is a measurement of the light polarized in the first direction among the light sources of the light source unit 100. The first image has a high image data value over the entire area.

6B is a cross-sectional inspection image and image data graph.

The cross inspection image is a second image obtained by photographing output light transmitted through the first reference polarization member 110 and the inspection target polarization unit 120 among the output light of the light source unit 100. Referring to the second image, since the polarization axes of the first reference polarizing member 110 and the inspection polarizing unit 120 are vertically arranged, light is not detected in most areas. The peak points displayed on the graph of the image data are the output light transmitted by the defects of the inspection target polarizing unit 120.

By analyzing the second image, it is possible to detect the presence or absence of a defect in the inspection subject polarizing unit 120 and the position of the defect. However, it can not be discriminated from the second image whether the defects of the inspection target polarizing unit 120 are due to a hole defect or a scatter defect.

6C is a double inspection image and image data graph.

The double inspection image is a third image obtained by photographing light transmitted through the first reference polarizing member 110, the inspection target polarizing unit 120 and the second reference polarizing member 130 among the output light of the light source unit 100. In the double inspection mode, the polarization axis of the first reference polarizing member 110 is perpendicular to the polarization axis of the subject polarized light unit 120 and is arranged parallel to the polarization axis of the second reference polarizing member 130. The output light due to the hole defect generated in the inspected polarizing unit 120 has the same polarization axis as the polarizing axis of the second reference polarizing member 130 and is transmitted without being blocked by the second reference polarizing member 130. [ Therefore, the output light due to the hole defect is detected as substantially the same image data in the third image and the second image.

6D is a parallel inspection image and image data graph.

The parallel inspection image is a fourth image of the output light of the light source unit 100, the light source unit light transmitted through the first reference polarization member 110, the inspection target polarization unit 120, and the third reference polarization member 140. In the parallel inspection mode, the polarization axis of the first reference polarizing member 110 is disposed perpendicular to the polarization axis of the subject polarizing unit 120 and the third reference polarizing member 140. The output light due to the hole defect generated in the inspected polarizing unit 120 is blocked by the third reference polarizing member 140 and is not transmitted perpendicularly to the same polarization axis as the polarizing axis of the third reference polarizing member 140. Therefore, when the peak due to the defect of the second image is greatly attenuated in the fourth image, it can be judged as a defect due to the hole.

7A is a graph of a hall inspection image and an image data.

FIG. 7B is a graph of foreign substance dispersion inspection image data and image data.

7A is a test image (left side) and a test image data graph (right side) in which the first through fourth images are combined into one image in order to detect hole defects occurring in the subject polarized light unit 120. FIG.

The inspection image of FIG. 7A is generated by combining the image data having the lightness and darkness information of the first to fourth images described in FIGS. 6A to 6D with the following equation (2) into one image data.

 &Quot; (2) "

Hall inspection image = first image * second image * third image * INV (fourth image)

Referring to the image data graph of FIG. 7A, the defective portion generated by the holes in the hole inspection image appears as a graph having a peak above the plane of the three-dimensional graph.

7B is a test image (left side) and a test image data graph (right side) in which the first through fourth images are synthesized to detect a foreign particle dispersion defect in the subject polarized light unit 120. FIG. The inspection image of FIG. 7B is generated by combining the image data having the lightness and darkness information of the first to fourth images described in FIGS. 6A to 6D with the following equation (3) into one image data.

 &Quot; (3) "

Foreign particle dispersion test image = first image * second image * third image * fourth image

Referring to the image data graph of FIG. 7B, a defect site generated by foreign scatter in the foreign particle scattering inspection image appears in the form of a peak on a plane. [Equation 2] and [Equation 3] are multiplications of light and dark information of the image. Pixels having information of 0 in any one of the first to fourth images are displayed as dark points.

The fourth image of Equation (2) is an image in which the polarization axes of the first reference polarizing member 110 and the second reference polarizing member 130 are vertically aligned, and the image to be inspected by the hole polarizing unit 120, Since the output light of the second reference polarization member 130 can not pass through the second reference polarization member 130, the fourth image is reversed to generate a hole inspection image.

The fourth image of Equation (3) is an image in which the polarization axes of the first reference polarizing member 110 and the second reference polarizing member 130 are vertically aligned, and the image to be inspected polarizing unit 120 Is transmitted through the second reference polarization member 130, so that a foreign particle dispersion inspection image is generated by multiplying the fourth image.

8A is a conceptual diagram of image block division according to an embodiment of the present invention.

FIG. 8B is a diagram illustrating an example of image block division according to an embodiment of the present invention.

Referring to FIGS. 5 and 8A, the image block dividing unit 420 receives an inspection image and divides the image into blocks arranged in a matrix. Referring to FIG. 8A, the image block dividing unit 420 divides the input test image into i regions along the horizontal direction and into j regions in the vertical direction. That is, the input image screen is divided into image blocks arranged in the form of i X j matrix. Each image block preferably includes at least 128 x 128 pixels therein. This is to improve the accuracy in binarizing the image block in the binarization processing unit described later.

The image blocks divided by the image block division unit 420 are transmitted to the binarization processing unit 430. [ The binarizing process of the image is to classify the information of all the pixels included in the block on the basis of two classes representing black or white. The binarization processing algorithm classifies pixels in each class that are lighter than a given threshold as white, and changes pixels that are not in the class to black. The OTSU binarization scheme can be applied with a more precise binarization algorithm. The Otsu binarization method is a method of classifying input images by using distributed information in each class. Otsu's binarization method computes a threshold value Th that minimizes intra-class variance between two classes when image pixels are classified into two classes based on the threshold value Th.

When the ratio of pixels darker than the threshold Th in the input image is α, the variance is σ12, the ratio of pixels lighter than the threshold Th is β, and the variance is σ22 (α + β = 1) Is calculated by the following equation (4).

(4) intra-class variance =? *? 12 +? *? 22

Minimizing the variance within the class of [Equation 4] means that the threshold with the least variance of both classes is applied.

9A is a hole defect inspection image and a binarization processed image according to an embodiment of the present invention. The binarized image of FIG. 9A is a result of binarizing the image by applying a threshold value (Th) 904. The binarized image of FIG. 9A shows that four hole defects are generated in the image block of the subject polarized light unit 120. FIG.

FIG. 9B is an image of foreign matter dispersion defect inspection and binarization processing according to an embodiment of the present invention. The binarized image of FIG. 9B is a result of binarizing by applying a threshold value (Th) 1422. The binarized image of FIG. 9B shows that one foreign particle dispersion defect occurs in the image block of the subject polarized light unit 120. FIG.

10 is an operational block diagram of a polarization unit inspection apparatus according to another embodiment of the present invention.

Referring to FIG. 10, the inspection controller 400 includes a second image B photographed in a cross-examination mode photographed in a single polarization inspection mode, a third image C photographed in a double-inspection mode, And stores the received fourth image D in the frame memory, respectively. The image data of the second to fourth images are transmitted to the hole image synthesizer 410 and the scatter image synthesizer 411, respectively.

The Hole image synthesis unit 410 combines the second through fourth images to synthesize a first inspection image suitable for hole defect inspection, and the Scatter image synthesis unit 411 synthesizes the second through fourth images, A second inspection image suitable for inspection is synthesized. The synthesized first and second inspection images are transmitted to the image block dividing unit 420, respectively. The image combining units 410 and 411 and the image block dividing unit 420 are the same as those in FIGS. 5 to 7, and a detailed description thereof will be omitted.

The image blocks divided by the image block dividing unit 420 are input to the offset processing unit 425. The input data of the input image block has a very low value, and there is an offset of the base level caused by the external light. In order to improve the precision of the binarization processing unit 430, the offset processing unit 425 removes the offset of the input data.

5 illustrates that the Hole image combining unit 410 and the Scatter image combining unit 411 are composed of different blocks. However, the Hole image combining unit 410 and the Scatter image combining unit 411 are the same as those of FIG. Function block.

 11 is a view comparing a profile of the inspection image with an offset removal profile.

The input image profile before the offset is removed shows that the dark region of the image rises to a certain level and is unevenly distributed with the peak included in the input image. The unevenness of the input image may be caused by unevenness of the output light of the light source unit 100 or may be caused by the angle of view of the camera of the photographing unit 300. If the levels of the dark portions are substantially different from each other, the defect determination period can not be maintained constant depending on the region.

The offset processing unit 425 constantly corrects the dark level of the inspection video signal to a value of 0 with respect to the entire screen. The inspection image having a constant arm level can be discriminated on the basis of the defect without deviation according to the position of the screen.

12 is a display screen of a polarizing unit inspection apparatus according to an embodiment of the present invention.

Referring to FIG. 12, the apparatus for inspecting a polarized light unit 10 may display an input image for each of a plurality of inspection modes photographed by a photographing unit on a display device of the inspection controller 400.

The inspection controller 400 analyzes the binarized inspection image to determine a hole defect and a scatter defect, and displays the result on a table device.

The inspection controller 400 may generate a defect map based on the distribution position of the defect to quantify the degree of the defect. It is also possible to automatically judge whether or not the to-be-inspected polarized light unit 120 is correct based on the number of displayed defects (Number) and the area ratio of defects (Area).

13 is an image of a defect map according to an embodiment of the present invention.

Referring to FIG. 13, the inspection controller 400 processes a test image to generate a defect image, and generates a defect map based on the image information of the defect image.

The Defect Map can be generated by analyzing the Defect Image and quantitatively quantifying the size and the degree of defect of the defect. The inspection controller 400 merges the defective pixels adjacent in the vertical and horizontal directions based on one defective pixel determined as a defect to generate a defect map by constructing adjacent defective pixels as one defect.

The code below is an algorithm block diagram for searching for defects of each pixel in a defective image and merging adjacent defective pixels to generate a defect map.

for y_position = 1: max

    for x_position = 1: max

        if (pixel is bright)

            if (pixel is included defect map)

                update previous defect map

            else

                update new defect map

            end

        end

    end

end

14A to 14G are explanatory diagrams of a defective pixel search method.

Referring to FIG. 14A, a black pixel in a pixel block arranged in a 10 X 10 matrix structure means a normal region of the subject polarized light unit 120, and a white pixel corresponds to a defect occurrence region. The number recorded in the pixel of FIG. 14A means luminance information, and the luminance information of the pixel is proportional to the intensity of the defect.

The defect search for constructing the defect map searches pixels arranged in the (1, 1) position along the first direction and the (1, 10) position in the pixel matrix, The entire pixel matrix is searched by the scanning method to be searched. The pixel information determined as a defect in the search process is recorded in a defect table. The Searching Defect Map Area searches for pixels that are already identified as defective in adjacent pixels up to the 3 X 3 region centering on defective pixels.

14B is an explanatory diagram of a defective pixel search method.

14B shows that a defect is detected in the pixels 2 and 3 of the pixel matrix during the defect search. Information of the pixels 2 and 3 is described in a defect table shown in FIG. 14B.

The defects of the pixels 2 and 3 are newly registered as the first defects in the defect table since there is no existing defect in the Searching Defect Map Area of the pixels 2 and 3.

The defect table includes defect start points (Start y_pos, Start x_pos), end points (End y_pos, End x_pos), cumulative luminance (Lumi), start frame, end frame, The frame # 1 and the frame # 2 indicating the number of included pixels include information of the frame # 2 indicating the sequence number of the image information to be inputted and analyzed.

The first defect started with the pixels 2 and 3 is generated with cumulative luminance 5, start frame 1, end frame 1, and frame # 1 as 1 in the area from (2,3) to (2,3).

14C is an explanatory diagram of a defective pixel search method.

Referring to FIG. 14C, the pixels 3 and 3 are adjacent to the pixels 2 and 3 included in the first defect determined as a defect. The information of the pixel 3, 3 is merged into the first defect of the defect table.

The defect table reflects the information of the pixels 3 and 3 to update the information of the first defect. The end point (End y_pos, End x_pos) of the first defect is updated from the existing (2,3) position to the (3,3) position and the accumulated luminance (Lumi) And the luminance value 15 is updated to the result value 20 added. Frame # 1, which means the number of pixels included in a region designated by one defect, is updated to 2 by adding a second defect.

 14D is an explanatory diagram of a defective pixel search method.

Referring to FIG. 14D, the pixels 3 and 4 are merged into the first defect as pixels adjacent to the (3, 3) pixel of the first defect.

The defect table reflects the information of the pixels 3 and 4 to update the first defect. The end point (End y_pos, End x_pos) of the first defect is updated from the (3,3) position to the (3,4) position and the cumulative brightness Lumi is updated to the defect brightness value 30 is added and the result is updated to 50. Frame # 1 is updated to 3.

14E is an explanatory diagram of a defective pixel search method.

Referring to FIG. 14E, pixels 5,5 are merged into the first defect as pixels adjacent to pixels 5,4 of the first defect.

The defect table reflects the information of the pixels 5, 5 to update the first defect. The end point (End y_pos, End x_pos) of the first defect is updated to the (5,7) position, and the accumulated luminance (Lumi) is updated to 135. [ Frame # 1 is updated to 12.

14F is an explanatory diagram of a defective pixel search method.

Referring to FIG. 14F, the pixels 8 and 4 are not adjacent to the defective pixel of the detected first defect, and are not merged into the first defect, but are newly registered as the second defect. The second defect is generated with a cumulative luminance of 5, a start frame 1, an end frame 1, and a frame # 1 at 1 in the area of the pixels 8 and 4.

14G is an explanatory diagram of a defective pixel search method.

Referring to Fig. 14G, pixels 8,5 are merged into the second defect adjacent to the second defective pixel 8,4.

The defect table reflects the information of the pixels 8 and 5 to update the information of the second defect. The end point (End y_pos, End x_pos) of the second defect is updated from the existing position (8,4) to the position (8,5), and the accumulated luminance (Lumi) The luminance value 5 is added and updated to 10. Frame # 1 is updated to 2.

In this way, defects are searched for all the pixels of the pixel matrix, and defects are blocked and stored.

15A is an explanatory diagram of a defective pixel search method.

FIG. 15A shows a method of successively detecting a defect by adding a second frame image to a result of the defect search described in FIGS. 14A to 14G.

When a defect is searched for the 10 x 10 pixel area shown in Fig. 15A, a defect is detected in the pixel (2,2). The pixel 2, 2 is the first defect detected in the second frame, but can be merged into the first defect adjacent to the position of the pixel included in the first defect of the first frame described in Figs. 14A to 14G.

The starting point (Start_y, Start_x) is updated to (2,2) and the cumulative luminance is updated from 135 to 140, reflecting the information of the pixels (2,2) of the second frame. The end frame is updated to 2, and the updated number of pixels is recorded as 1 in Frame # 2.

Pixels overlapping or adjacent to the first defect shown in Fig. 15A are merged and processed in the first defect.

15B is an explanatory diagram of a defective pixel search method.

FIG. 15B illustrates a case where a defect is detected in the pixels 6 and 9. The third defect is registered as a start point of the pixels 6 and 9 as positions overlapping or not adjacent to the areas of the first defect and the second defect detected before the pixels 6 and 9.

The third defect started from the pixels 6 and 9 is generated as the cumulative luminance 5, the start frame 2, the end frame 2, and the frame # 2 in the area from (6,9) to (6,9)

The defects adjacent to the pixels 6 and 9 are not further searched, and the third defect is composed only of the pixels 6 and 9.

The inspection controller 400 may construct a defect map based on a defect table including information on the first to third defects. The defect map is formed in the shape of a rectangle including the coordinates of the start point and the end point of the defect, and may include a cumulative luminance (Lumi) value corresponding to the strength of the defect.

16 is a comparison screen of a high-resolution optical image and a defect map of the present invention.

The optical image of FIG. 15 is an image photographed using an optical tester that optically analyzes defects of a polarization unit using an ultra-high resolution camera. High-resolution images for optical inspection of a polarizing unit are shot by a scanning method in order to inspect the entire surface, so that it takes several hours only for shooting and analysis.

On the other hand, the apparatus for inspecting a polarizing unit according to an embodiment of the present invention can be performed within a few minutes, from imaging an image through combination of reference polarizing members to generating an inspection image and analyzing defects. The results of the analysis may be displayed as a defect map so that it can be used as data for quantitative analysis to increase visibility.

The defect map according to the embodiment of the present invention shown on the right side of FIG. 16 shows the same defective portion as the optical image analysis image shown on the left side.

17 is a graph showing the defect intensity and the number of defects.

The inspection controller 400 can analyze the intensity of defects occurring in the inspection subject polarizing unit 120 through the defect map construction.

The defect intensity can be calculated through the following equation (5).

&Quot; (5) "

Defect intensity = luminance X area

Each defect classified through defect analysis can be classified into a plurality of defect classes according to defect intensity. For example, if a defect has a defect intensity of 2500 or more, a defect of a defect class has a defect intensity of less than 2500 and a defect intensity of 600 or more, the defect is classified into Large, Mid, and Small. , And Small (Small can be classified into cases where the defect intensity is less than 600. Generally, the larger the defect intensity, the smaller the number of defects is shown.

The defect grade can be selectively determined according to the nature of the defect and the measurement of the inspection system.

The polarization unit inspection apparatus 10 can store and compare the number of defects according to the defect intensity of the plurality of inspection target polarization units 120 through a series of inspections. It is possible to analyze the defect positions and the intensities of the plurality of inspection subject polarizing units 120 to determine whether or not the polarizing unit is abnormal in the production process and to analyze the cause of the defects.

18 is a graph of defect occurrence status for each sample.

Referring to FIG. 18, a defect map may be constructed and defects may be analyzed and displayed according to an embodiment of the present invention. The graph shows the classification of the grades and the number of defects according to the defect intensity that occurred in all 9 experimental samples.

It is possible to numerically manage the intensity of defects and to discriminate between the parts in real time during the manufacturing process of the polarizing unit. By accumulating and comparing numerical values for each defect intensity, raw data for analyzing the cause of defects can be obtained .

19 is a graph showing the analysis of the cause of failure of the polarization unit.

The subject polarized light unit of the present invention is constituted by a wire grid polarizer, and the wire grid polarizer forms a grid pattern using a stamp in the manufacturing process. Although the stamp can be used repeatedly, there is a high possibility that defects are generated in the polarization unit according to the number of times of use.

FIG. 19 is a diagram for analyzing a defect occurrence status of a polarization unit according to the number of times of reuse of a single stamp, by using the defect guidance method of the present invention. FIG.

Referring to FIG. 19, if the number of all defects is 51 in the first use of the stamp, or when the number of reuses is increased by 13, the number of defects is increased to 252. In particular, the defect intensity shows a small increase of 9 times from 24 to 224 defects.

10 Polarization unit inspection device
100 Light source part
110 first reference polarizing member
130 second reference polarization member
200 jig
300 shooting section
400 inspection controller
410 Hole image synthesis unit
411 Scatter image synthesis unit
420 video block division
425 offset processor
430 binarization processor
440 block unit coupling portion
450 Hole defect discrimination unit
451 Scatter defect determination unit

Claims (16)

A light source;
A first reference polarizing member having a polarization axis in a first direction located above the light source unit;
A second reference polarizing member positioned above the first reference polarizing member;
A photographing unit positioned above the second reference polarizing member; And
An inspection unit which receives a plurality of images from the photographing unit and synthesizes the plurality of images into an inspection image to determine whether the inspection target polarizing unit positioned between the first reference polarizing member and the second reference polarizing member is defective A polarizing unit inspection apparatus comprising a controller. The method according to claim 1,
And the second reference polarizing member has a polarization axis in any one of directions parallel or perpendicular to the first direction. 3. The method of claim 2,
The plurality of images
A first image obtained by photographing output light of the first reference polarizing member in a state that the light source unit is turned on;
A second image obtained by photographing the output light of the inspection object;
A third image of the output light of the second reference polarization member having a polarization axis parallel to the polarization axis of the first reference polarization member; And
And a fourth image obtained by photographing output light of the third reference polarization member having a polarization axis perpendicular to the polarization axis of the first reference polarization member. The method of claim 3,
Wherein the inspection controller comprises: an image storage unit for storing the plurality of images;
An image synthesizer for synthesizing the stored images to generate a test image;
A screen dividing unit dividing the inspection image into a plurality of screen blocks;
A binarization unit for binarizing the pixel data of the picture block; And
And an image combining unit that combines the plurality of binarized image blocks to generate an output image. 5. The method of claim 4,
Wherein the inspection controller further comprises a defect determination unit for automatically determining a position of a defect in the output image. 6. The method of claim 5,
And the defect discriminating section classifies the cause of the defect and displays the result. 6. The method of claim 5,
Wherein the inspection controller further comprises a display device for superimposing and displaying the inspection image and the defect. A photographing step of photographing a plurality of images with respect to the subject polarizing unit;
An image synthesizing step of synthesizing the plurality of images into an inspection image;
An image dividing step of dividing the inspection image into a plurality of image blocks;
A binarization step of binarizing the divided plurality of image blocks to generate a binarized image block;
A block unit combining step of combining the binarized picture blocks to generate a binarized defect image; And
And a defect determination step of analyzing the combined defect image to determine a defect. The method of claim 8, wherein
The photographing step of photographing the plurality of images
A first image capturing step of disposing a first reference polarizing member having a polarization axis in a first direction and a light source unit and photographing output light of the first reference polarizing member in a state that the light source unit is turned on;
A second image capturing step of disposing an inspection object having a polarization axis perpendicular to the first direction on the first reference polarization element and photographing the output light of the inspection object in a state in which the light source section is turned on;
A third image capturing step of disposing a second reference polarizing member having a polarization axis in a second direction on the inspection object and photographing the output light of the second reference polarizing member in a state in which the light source unit is turned on; And
The fourth reference image-taking step in which the third reference polarizing member having a polarization axis perpendicular to the second direction is disposed on the top of the to-be-inspected polarizing unit and the output light of the third reference polarizing member is photographed in a state that the light source unit is on ≪ / RTI > 10. The method of claim 9,
Wherein the second direction is any one of a direction parallel to the first direction and a direction perpendicular to the first direction. 10. The method of claim 9,
And the third polarizing member is arranged by rotating the second polarizing member by 90 degrees. 9. The method of claim 8,
Wherein the image synthesizing step synthesizes the plurality of images into a test image using different calculation equations according to kinds of defects. 9. The method of claim 8,
Wherein the dividing step divides the inspection image into a plurality of pixel blocks arranged in a matrix, and the pixel block includes at least 128 X 128 pixels. 9. The method of claim 8,
Wherein the defect determination step comprises combining a plurality of adjacent defects to constitute a defect map. 15. The method of claim 14,
And a defect display step of displaying the defect map superimposed on the inspection image. 15. The method of claim 14,
Wherein the defect display step generates a defect intensity based on the brightness and area of each defect of the defect map, and classifies the defect class based on the defect intensity.

Images