JP7635245B2 - Method for inspecting optical laminate - Google Patents

Description

The present invention relates to an inspection method for an optical laminate in which a polarizer and an optical film are laminated, and a release film is further laminated on at least one outermost surface side in the thickness direction. In particular, the present invention relates to an inspection method for an optical laminate that can suppress overdetection of defects present on the surface of the release film and accurately detect defects present between the polarizer and the optical film.

2. Description of the Related Art Conventionally, there has been known an inspection method for optically inspecting an optical laminate including a polarizer for defects and judging the acceptability of the optical laminate.
Defects in the optical laminate include foreign matter (referred to as "bonding foreign matter" as appropriate in this specification) present between layers of the optical laminate (specifically, between the polarizer and the optical film that constitute the optical laminate) and defects (foreign matter, stains, scratches, etc.) present on the surface of the optical laminate.

The optical conditions under which defects are easily detected vary depending on the type of defect, and therefore various inspection methods that combine a number of optical conditions have been proposed.
For example, Patent Documents 1 and 2 propose inspection methods for detecting defects in an optical film based on a transmitted image of the optical film generated by light passing through the optical film and a reflected image of the optical film generated by light reflected by the optical film (paragraphs 0023 to 0026 of Patent Document 1, claim 2 of Patent Document 2, etc.).
Furthermore, Patent Document 3 proposes an inspection method for detecting defects in an optical stack based on a reflected image of the optical stack generated by light reflected from an optical stack including a polarizer, and a crossed Nicol image of the optical stack generated by light passing through an inspection polarizing filter and the optical stack arranged in a crossed Nicol state with respect to the polarization axis of the polarizer (e.g., claim 1 of Patent Document 3).

Here, when the inspection object is an optical laminate in which a polarizer and an optical film (e.g., a retardation film) are laminated, and further a release film (e.g., a separator or a surface protection film) is laminated on at least one outermost surface side in the thickness direction, defects (foreign matter, dirt, scratches, etc.) present on the surface of the release film are harmless and do not pose a problem. This is because when the optical laminate is used (e.g., when the optical laminate is bonded to a liquid crystal cell), the release film is peeled off and does not remain.

The inventors of the present invention have investigated an inspection method for detecting defects based on a crossed Nicol image of an optical laminate containing a polarizer, and found that although it is possible to detect bonding foreign matter present between the layers of the optical laminate, it also detects (overdetects) defects present on the surface of the harmless release film. As a result, optical laminates that only have defects on the surface of the release film (which is not a problem because there are no defects between the layers) may be treated as defective, which may result in a poor product yield for the optical laminate.

Patent Document 3 describes that in order to suppress overdetection as described above, if the position of a defect candidate detected based on a reflected image of the optical stack is identical to the position of a defect candidate detected based on a crossed Nicol image of the optical stack, the defect candidate is not treated as a defect (see, e.g., Claim 1, paragraph 0083 of Patent Document 3).
However, according to the inventors' research, it was found that even the inspection method that combines reflected images and crossed Nicol images as described above may not be able to sufficiently suppress overdetection of harmless defects.

The inspection method described in Patent Document 1 aims to accurately count the number of defects (paragraph 0007 of Patent Document 1), and is not a method for preventing overdetection of harmless defects.
Furthermore, the inspection method described in Patent Document 2 aims to accurately determine the type of defect (paragraph 0018 of Patent Document 2), and is not a method for preventing overdetection of harmless defects.

JP 2003-329601 A JP 2012-167975 A Patent No. 4960161

The present invention has been made to solve the problems of the conventional technology as described above, and aims to provide an inspection method for an optical laminate that can suppress overdetection of defects present on the surface of the release film and accurately detect defects that exist between the polarizer and the optical film.

In order to solve the above-mentioned problems, the inventors conducted extensive research and discovered that defect candidates that were detected in both the transmitted image and the crossed Nicol image but not in the reflected image were highly likely to be defects (bonding foreign matter) that existed between the polarizer and the optical film, and thus completed the present invention.

That is, in order to solve the above-mentioned problems, the present invention provides a method for inspecting an optical laminate in which a polarizer and an optical film are laminated, and a release film is further laminated on at least one outermost surface side in the thickness direction, the method comprising: a transmission inspection process for generating a transmission image of the optical laminate by light transmitted through the optical laminate, and detecting defect candidates present in the optical laminate based on the transmission image; a crossed Nicol inspection process for generating a crossed Nicol image of the optical laminate by an inspection polarizing filter arranged so as to be in a crossed Nicol state with respect to the polarization axis of the polarizer and light transmitted through the optical laminate, and detecting defect candidates present in the optical laminate based on the crossed Nicol image; The present invention provides an inspection method for an optical stack, the method comprising: a reflection inspection process for generating a reflected image of the optical stack by light reflected from the optical stack, and detecting defect candidates present in the optical stack based on the reflected image; and a calculation process for determining defects present between the polarizer and the optical film based on the defect candidates detected in the transmission inspection process, the crossed Nicols inspection process, and the defect candidates detected in the reflection inspection process, wherein in the calculation process, a defect candidate detected in both the transmission inspection process and the crossed Nicols inspection process but not detected in the reflection inspection process is determined to be a defect present between the polarizer and the optical film.

According to the present invention, in the transmission inspection process, defect candidates present in the optical stack are detected based on a transmission image of the optical stack. The transmission image is generated, for example, by arranging a light source on one surface side of the optical stack and an imaging means on the other surface side, and receiving light emitted from the light source and transmitted through the optical stack by the imaging means to form an image (capture). Defect candidates in the transmission image are detected, for example, by applying known image processing such as binarization to the transmission image, which extracts pixel regions having a different luminance value (pixel value) from other pixel regions.
According to the present invention, in the crossed Nicol inspection process, defect candidates present in the optical stack are detected based on the crossed Nicol image of the optical stack. The crossed Nicol image is generated, for example, by arranging a light source and an inspection polarizing filter on one side of the optical stack, arranging an imaging means on the other side, and capturing and receiving the light emitted from the light source and transmitted through the inspection polarizing filter and the optical stack by the imaging means to form an image (image). In this case, the crossed Nicol state is disrupted by a defect present between the inspection polarizing filter and the polarizer of the optical stack, so that in the crossed Nicol image, the pixel area corresponding to the defect present between the inspection polarizing filter and the polarizer becomes bright (the luminance value becomes large). Alternatively, the crossed Nicol image can also be generated by arranging a light source on one side of the optical stack, arranging an inspection polarizing filter and an imaging means on the other side, and receiving and forming an image (image) with the imaging means the light emitted from the light source and transmitted through the optical stack and the inspection polarizing filter. In this case, too, the crossed Nicol state is disrupted by a defect existing between the polarizer of the optical stack and the polarizing filter for inspection, so that in the crossed Nicol image, the pixel region corresponding to the defect existing between the polarizer and the polarizing filter for inspection becomes brighter (brighter). Defect candidates in the crossed Nicol image are detected, for example, by applying known image processing such as binarization to the crossed Nicol image, which extracts pixel regions having a different brightness value (pixel value) from other pixel regions (specifically, pixel regions having a higher brightness value than other pixel regions).
Furthermore, according to the present invention, in the reflection inspection process, defect candidates present in the optical stack are detected based on a reflection image of the optical stack. The reflection image is generated, for example, by arranging a light source and an imaging means on one side of the optical stack, and receiving light emitted from the light source and reflected by the optical stack with the imaging means to form an image (capture). Defect candidates in the reflection image are detected, for example, by applying known image processing such as binarization to the reflection image, which extracts pixel regions having a different luminance value (pixel value) from other pixel regions.

In the present invention, the term "optical film" refers to an optical film that cannot be peeled off from a polarizer.
In addition, in the present invention, the term "inspection polarizing filter arranged so as to be in crossed Nicols with respect to the polarization axis of the polarizer" is not limited to the case where the angle between the polarization axis of the polarizer and the polarization axis of the inspection polarizing filter is exactly 90°, but includes the case where the angle is within the range of 90°±10°.
Furthermore, in the present invention, the transmission inspection process, the crossed Nicols inspection process, and the reflection inspection process do not necessarily have to be performed in this order, and they can be performed in any order (including the case where multiple inspection processes are performed with some overlap).

Furthermore, according to the present invention, in the calculation process, a defect candidate that is detected in both the transmission inspection process and the crossed Nicol inspection process but not detected in the reflection inspection process is determined to be a defect (bonding foreign matter) present between the polarizer and the optical film.
Whether or not a defect candidate is detected in both the transmission inspection process and the crossed Nicols inspection process is determined, for example, by whether or not a defect candidate detected in the crossed Nicols inspection process exists at a position equivalent (the same or nearby) to the position of a certain defect candidate detected in the transmission inspection process. If a defect candidate detected in the crossed Nicols inspection process exists at the equivalent position, it is determined that this defect candidate has been detected in both the transmission inspection process and the crossed Nicols inspection process. On the other hand, if no defect candidate detected in the crossed Nicols inspection process exists at the equivalent position, it is determined that this defect candidate has not been detected in both the transmission inspection process and the crossed Nicols inspection process.
Whether or not a defect candidate detected in both the transmission inspection process and the crossed Nicols inspection process is detected in the reflection inspection process is determined, for example, by whether or not a defect candidate detected in the reflection inspection process exists at a position equivalent (the same or nearby) to the position of a defect candidate detected in both the transmission inspection process and the crossed Nicols inspection process. If a defect candidate detected in the reflection inspection process exists at the equivalent position, it is determined that this defect candidate was detected in the reflection inspection process. On the other hand, if no defect candidate detected in the reflection inspection process exists at the equivalent position, it is determined that this defect candidate was not detected in the reflection inspection process.

As described above, according to the findings of the inventors, defect candidates that are detected in both the transmitted image and the crossed Nicol image but not in the reflected image are likely to be defects (bonding foreign matter) that exist between the polarizer and the optical film. According to the present invention, in the calculation process, defect candidates that are detected in both the transmitted inspection process and the crossed Nicol inspection process (i.e., detected in both the transmitted image and the crossed Nicol image) but not in the reflected inspection process (i.e., not detected in the reflected image) are determined to be defects that exist between the polarizer and the optical film, so that overdetection of defects that exist on the surface of the release film is suppressed, and defects that exist between the polarizer and the optical film can be detected with high accuracy.

In the present invention, when the release film is a separator and the optical film is located between the separator and the polarizer, it is preferable to position the inspection polarizing filter on the separator side in the crossed Nicol inspection process.

When the optical laminate is attached to a liquid crystal cell of an image display device, the separator side is attached to the liquid crystal cell (the separator is peeled off and then attached). When the separator side of the optical laminate is attached to the liquid crystal cell, defects existing between the polarizer of the optical laminate and the liquid crystal cell (specifically, defects existing between the polarizer and the optical film located closer to the liquid crystal cell than the polarizer) are displayed as bright spots in the image display device when the liquid crystal cell is driven, which is a quality problem.
According to the above-mentioned preferred method, in the crossed Nicol inspection process, the inspection polarizing filter is placed on the same side of the separator as the liquid crystal cell, so that defects that become a problem when driving the liquid crystal cell (defects that exist between the liquid crystal cell and the optical film) can be detected as defect candidates.

In the present invention, if there is a portion of the optical laminate in which the orientation direction of the release film is significantly misaligned with the direction of the polarization axis of the polarizer, or if there is a portion of the optical laminate in which the orientation direction of the release film is significantly misaligned with the direction of the polarization axis of the inspection polarizing filter, even if there is no defect between the inspection polarizing filter and the polarizer of the optical laminate, the crossed Nicol state will be disrupted at the above-mentioned portion, and the detection accuracy of defect candidates in the crossed Nicol inspection process will be reduced.
Therefore, the present invention is preferably used when the orientation direction of the release film is within ±6° (more preferably within ±3.5°) with respect to the predetermined specified orientation direction. In this way, if the release film has a uniform orientation direction (within ±6° with respect to the default orientation direction), the release film and the polarizer are laminated so that the default orientation direction of the release film and the direction of the polarization axis of the polarizer coincide, or the default orientation direction of the release film and the direction of the polarization axis of the inspection polarizing filter coincide (in other words, the default orientation direction of the release film and the direction of the polarization axis of the polarizer are perpendicular to each other), it is possible to prevent a decrease in the detection accuracy of defect candidates in the crossed Nicol inspection process for an optical laminate in which the release film and the polarizer are laminated so that the default orientation direction of the release film and the direction of the polarization axis of the polarizer coincide (in other words, the default orientation direction of the release film and the direction of the polarization axis of the polarizer are perpendicular to each other).

Preferably, the imaging means for generating the transmitted image in the transmission inspection process and the imaging means for generating the crossed Nicol image in the crossed Nicol inspection process are the same, and the timing for performing imaging by the imaging means in the transmission inspection process and the timing for performing imaging by the imaging means in the crossed Nicol inspection process are switched.

According to the above-mentioned preferred method, since the imaging means for generating the transmission image and the imaging means for generating the crossed Nicol image are the same, the coordinates of the transmission image and the coordinates of the crossed Nicol image can be matched with high precision. Therefore, in the calculation process, it is possible to accurately determine whether or not a defect candidate is detected in both the transmission inspection process and the crossed Nicol inspection process (for example, it is possible to accurately determine whether or not a defect candidate detected in the crossed Nicol inspection process exists at a position equivalent to the position of a certain defect candidate detected in the transmission inspection process).

Preferably, the transmission inspection process and/or the crossed Nicols inspection process includes a noise removal procedure for excluding from the detected defect candidates those defect candidates whose dimensions are greater than a predetermined threshold value.

Defects (lamination foreign matter) present between the polarizer and the optical film of the optical laminate are often smaller in size than defects present on the surface of the release film.
According to the above-mentioned preferred method, among the defect candidates detected in the transmission inspection step and/or the crossed Nicols inspection step, defect candidates having dimensions larger than a predetermined threshold value are excluded from the defect candidates, so that the number of defect candidates to be determined whether or not they have been detected in both the transmission inspection step and the crossed Nicols inspection step can be reduced in the calculation step, which has the advantage of being able to shorten the time required for the calculation step.

According to the present invention, it is possible to suppress overdetection of defects present on the surface of the release film and accurately detect defects present between the polarizer and the optical film.

1 is a diagram for illustrating a schematic configuration of an inspection device for performing an inspection method for an optical laminate according to an embodiment of the present invention. 1 is a flow chart showing schematic steps of an inspection method for an optical laminate according to an embodiment of the present invention. 3 is a diagram for explaining an example of a defect candidate detected in the transmission inspection step S1 shown in FIG. 2; FIG. 3 is a diagram for explaining an example of a defect candidate detected in the crossed Nicols inspection process S2 shown in FIG. 2; FIG. 2 is a diagram for explaining the details of switching control executed by a control calculation means 9 shown in FIG. 1 . FIG. 3 is a diagram for explaining an example of a defect candidate detected in the reflection inspection step S3 shown in FIG. 2; FIG. FIG. 3 is a diagram for explaining the content of a calculation step S4 shown in FIG. 2 .

Hereinafter, a method for inspecting an optical laminate according to one embodiment of the present invention will be described with reference to the accompanying drawings as appropriate.
Fig. 1 is a diagram for explaining a schematic configuration of an inspection device for carrying out an inspection method for an optical laminate according to the present embodiment. Fig. 1(a) is a side view showing the schematic configuration of the inspection device. Fig. 1(b) is a cross-sectional view showing the schematic configuration of the optical laminate. In Fig. 1, X indicates a horizontal direction parallel to the transport direction of the optical laminate S, Y indicates a horizontal direction perpendicular to the X direction, and Z indicates a vertical direction.

<Optical laminate S>
First, the configuration of the optical laminate S that is the object of inspection by the inspection device 100 of this embodiment will be described.
1B, the optical laminate S of this embodiment is cut into chips according to the application, and a polarizer 10 and optical films 20 and 30 are laminated, and release films 40 and 50 are further laminated on the outermost surface side in the thickness direction (Z direction). In this embodiment, one optical film 20 located below the polarizer 10 is a retardation film, and the other optical film 30 located above the polarizer 10 is a protective film. In this embodiment, one release film 40 located below the polarizer 10 is a separator, and the other release film 50 located above the polarizer 10 is a surface protective film.
Hereinafter, each component of the optical laminate S will be described.

[Polarizer 10]
The polarizer 10 is typically made of a resin film containing a dichroic material.
As the resin film, any appropriate resin film that can be used as a polarizer can be adopted, and the resin film is typically a polyvinyl alcohol-based resin (hereinafter referred to as "PVA-based resin") film.

Any suitable resin can be used as the PVA-based resin that forms the PVA-based resin film. Examples include polyvinyl alcohol and ethylene-vinyl alcohol copolymer. Polyvinyl alcohol is obtained by saponifying polyvinyl acetate. Ethylene-vinyl alcohol copolymer is obtained by saponifying ethylene-vinyl acetate copolymer.

The average degree of polymerization of the PVA-based resin can be appropriately selected depending on the purpose. The average degree of polymerization is usually 1000 to 10000, preferably 1200 to 4500, and more preferably 1500 to 4300. The average degree of polymerization can be determined in accordance with JIS K 6726-1994.

Examples of dichroic substances contained in the resin film include iodine and organic dyes. These can be used alone or in combination of two or more. Iodine is preferably used.

The resin film may be a single layer resin film or a laminate of two or more layers.

A specific example of a polarizer composed of a single-layer resin film is a PVA-based resin film that has been subjected to a dyeing treatment with iodine and a stretching treatment (typically, a uniaxial stretching treatment). The dyeing treatment with iodine is performed, for example, by immersing the PVA-based film in an aqueous iodine solution. The stretching ratio for the uniaxial stretching is preferably 3 to 7 times. The stretching may be performed after dyeing or while dyeing. Alternatively, dyeing may be performed after stretching. If necessary, the PVA-based resin film is subjected to a swelling treatment, a crosslinking treatment, a washing treatment, a drying treatment, etc.

Specific examples of polarizers made of a laminate include a laminate of a resin substrate and a PVA-based resin layer (PVA-based resin film) laminated on the resin substrate, or a polarizer made of a laminate of a resin substrate and a PVA-based resin layer applied to the resin substrate. A polarizer made of a laminate of a resin substrate and a PVA-based resin layer applied to the resin substrate can be produced, for example, by applying a PVA-based resin solution to the resin substrate, drying the resin substrate to form a PVA-based resin layer on the resin substrate, obtaining a laminate of the resin substrate and the PVA-based resin layer, and then stretching and dyeing the laminate to make the PVA-based resin layer into a polarizer. In this embodiment, stretching typically includes immersing the laminate in an aqueous boric acid solution and stretching it. Furthermore, stretching may include, as necessary, stretching the laminate in the air at a high temperature (e.g., 95°C or higher) before stretching in the aqueous boric acid solution. The obtained laminate of resin substrate/polarizer may be used as it is (i.e., the resin substrate may be used as a protective layer for the polarizer), or the resin substrate may be peeled off from the laminate of resin substrate/polarizer, and any suitable protective layer depending on the purpose may be laminated on the peeled surface. Details of such a method for producing a polarizer are described, for example, in JP2012-73580A. The entire disclosure of this publication is incorporated herein by reference.

The thickness of the polarizer 10 is preferably 15 μm or less, more preferably 1 μm to 12 μm, even more preferably 3 μm to 10 μm, and particularly preferably 3 μm to 8 μm.

The polarizer 10 preferably exhibits absorption dichroism at any wavelength within the range of 380 nm to 780 nm. The single transmittance of the polarizer 10 is preferably 40.0% to 45.0%, and more preferably 41.5% to 43.5%. The degree of polarization of the polarizer 10 is preferably 97.0% or more, more preferably 99.0% or more, and even more preferably 99.9% or more.

[Retardation film 20]
The retardation film 20 may be, for example, a compensation plate that provides a wide viewing angle, or a retardation plate (circular polarizing plate) that is used together with a polarizing film to generate circularly polarized light. The thickness of the retardation film 20 is, for example, 1 to 200 μm. Note that, instead of the retardation film 20, other films such as a protective film or a reflective polarizer, which will be described later, may be used.

The retardation film 20 is typically formed of any suitable resin capable of realizing the above characteristics. Examples of resins forming the retardation film 20 include polyarylate, polyamide, polyimide, polyester, polyaryletherketone, polyamideimide, polyesterimide, polyvinyl alcohol, polyfumaric acid ester, polyethersulfone, polysulfone, norbornene resin, polycarbonate resin, cellulose resin, and polyurethane. These resins may be used alone or in combination. Cycloolefin-based norbornene resin is preferred.

[Protective film 30]
Any suitable resin film is used as the protective film 30. Examples of materials for forming the resin film include (meth)acrylic resins, cellulose resins such as diacetyl cellulose and triacetyl cellulose, cycloolefin resins such as norbornene resins, olefin resins such as polypropylene, ester resins such as polyethylene terephthalate resins, polyamide resins, polycarbonate resins, and copolymer resins thereof. Note that "(meth)acrylic resin" means acrylic resin and/or methacrylic resin.

The thickness of the protective film 30 is typically 10 μm to 100 μm, preferably 20 μm to 40 μm.

The surface of the protective film 30 opposite the polarizer 10 may be subjected to a surface treatment such as a hard coat treatment, an anti-reflection treatment, an anti-sticking treatment, an anti-glare treatment, etc., as necessary. Furthermore, the surface of the protective film 30 opposite the polarizer 10 may be subjected to a treatment to improve visibility when viewed through polarized sunglasses (typically, a treatment to impart (elliptical) polarizing function, a treatment to impart ultra-high phase difference) as necessary. Note that when a surface treatment is performed to form a surface treatment layer, the thickness of the protective film 30 includes the thickness of the surface treatment layer.

The retardation film 20 and the protective film 30 are laminated to the polarizer 10 via any suitable adhesive layer (not shown). Typical examples of the adhesive constituting the adhesive layer include a PVA-based adhesive and an activated energy ray curing adhesive.

[Separator 40]
Any appropriate separator can be used as the separator 40. Specific examples include plastic films, nonwoven fabrics, or papers whose surfaces are coated with a release agent. Specific examples of the release agent include silicone-based release agents, fluorine-based release agents, and long-chain alkyl acrylate-based release agents. Specific examples of the plastic film include polyethylene terephthalate (PET) films, polyethylene films, and polypropylene films. The thickness of the separator can be, for example, 10 μm to 100 μm.

The separator 40 is laminated on the retardation film 20 via any suitable adhesive layer (not shown). Specific examples of adhesives constituting the adhesive layer include acrylic adhesives, rubber adhesives, silicone adhesives, polyester adhesives, urethane adhesives, epoxy adhesives, and polyether adhesives. By adjusting the type, number, combination, and compounding ratio of monomers forming the base resin of the adhesive, as well as the compounding amount of the crosslinking agent, reaction temperature, reaction time, and the like, an adhesive having desired properties according to the purpose can be prepared. The base resin of the adhesive may be used alone or in combination of two or more types. From the viewpoints of transparency, processability, durability, and the like, an acrylic adhesive is preferred. Details of the adhesive constituting the adhesive layer are described in, for example, JP 2014-115468 A, and the description of the publication is incorporated herein by reference. The thickness of the adhesive layer can be, for example, 10 μm to 100 μm. The pressure-sensitive adhesive layer can have a storage modulus G' at 25° C. of, for example, 1.0×10 ⁴ [Pa] to 1.0×10 ⁶ [Pa]. The storage modulus can be determined, for example, by dynamic viscoelasticity measurement.

In this embodiment, the separator 40 has an orientation direction within ±6° of a predetermined specified orientation direction. For example, if the polarization axis direction of the polarizer 10 in this embodiment is the X direction, the default orientation direction of the separator 40 is the Y direction, and the separator 40 is stacked so that the orientation direction of any part of the separator 40 forms an angle of ±6° with the Y direction.

[Surface protection film 50]
The surface protection film 50 typically has a substrate and a pressure-sensitive adhesive layer. In this embodiment, the thickness of the surface protection film 50 is, for example, 30 μm or more. The upper limit of the thickness of the surface protection film 50 is, for example, 150 μm. In this specification, the "thickness of the surface protection film" refers to the total thickness of the substrate and the pressure-sensitive adhesive layer.

The substrate can be made of any suitable resin film. Examples of materials for forming the resin film include ester-based resins such as polyethylene terephthalate-based resins, cycloolefin-based resins such as norbornene-based resins, olefin-based resins such as polypropylene, polyamide-based resins, polycarbonate-based resins, and copolymer resins of these. Ester-based resins (particularly polyethylene terephthalate-based resins) are preferred.

Any suitable adhesive can be used as the adhesive that forms the adhesive layer. Examples of the base resin of the adhesive include acrylic resins, styrene resins, silicone resins, urethane resins, and rubber resins.

<Inspection device 100>
Next, the configuration of the inspection device 100 of this embodiment will be described.
The inspection device 100 of this embodiment is a device for inspecting the optical laminate S having the configuration described above.
As shown in FIG. 1A, the inspection device 100 of this embodiment includes a plurality of belt conveyors 1 for transporting the optical laminate S in the X direction, and a clean roller 2 for adsorbing and removing foreign matter attached to the outermost surface (uppermost surface and lowermost surface) of the optical laminate S. The inspection device 100 of this embodiment also includes a light source 3 and an imaging means 4 for performing a transmission inspection process S1 described later. The inspection device 100 of this embodiment also includes a pair of light sources 5a, 5b and a pair of inspection polarizing filters 6a, 6b for performing a crossed Nicol inspection process S2 described later. The imaging means 4 is also used as an imaging means for performing the crossed Nicol inspection process S2. The inspection device 100 of this embodiment also includes a light source 7 and an imaging means 8 for performing a reflection inspection process S3 described later. Furthermore, the inspection device 100 of this embodiment is equipped with a control and calculation means 9 that is electrically connected to the light source 3, the imaging means 4, the light sources 5a, 5b, the light source 7, and the imaging means 8, and controls the operation of these means, as well as processes the imaging signals output from the imaging means 4 and the imaging means 8 to determine defects.
Each component of the inspection device 100 will be described below.

[Belt conveyor 1]
The belt conveyor 1 is configured such that an annular belt stretched over rollers at both ends moves with the rotation of the rollers, thereby conveying the optical laminate S placed on the belt. After being cut into chips, the optical laminate S is placed on the belt conveyor 1 shown at the left end of FIG. 1(a), and is sequentially conveyed in the X direction by each belt conveyor 1 toward the right side of FIG. 1(a). In this embodiment, as shown in FIG. 1(b), the optical laminate S is placed on the belt conveyor so that the separator 40 side is downward and conveyed. The conveying speed V of the optical laminate S by the belt conveyor 1 is set to, for example, 50 mm/sec to 750 mm/sec.

[Cleaning roller 2]
The clean roller 2 includes a pair of upper and lower rollers through which the optical laminate S passes, and a roll-shaped adhesive tape (not shown) that rotates in contact with each roller. When the pair of upper and lower rollers come into contact with the optical laminate S, foreign matter adhering to the outermost surface (the uppermost surface and the lowermost surface, i.e., the lower surface of the separator 40 and the upper surface of the surface protection film 50) of the optical laminate S is adsorbed to the rollers, and the foreign matter adsorbed to the rollers is transferred to the adhesive tape and removed.
Before performing the transmission inspection process S1 described below, foreign matter present on the outermost surface of the optical laminate S can be removed to a certain extent using the clean roller 2, which makes it possible to further suppress overdetection of defects present on the surface of the release film (separator 40, surface protection film 50).

[Light source 3]
The light source 3 is a light source for performing a transmission inspection step S1 described later, and in this embodiment, is disposed on the lower surface side (separator 40 side) of the optical stack S. The optical axis of the light source 3 (shown by a dashed line in FIG. 1( a)) is oriented in a vertical direction (Z direction) parallel to the thickness direction of the optical stack S, and the light source 3 emits light vertically upward toward the optical stack S in accordance with a control signal output from the control calculation means 9.
The light source 3 is not limited as long as it can emit light of a wavelength that can be transmitted through the optical laminate S, but for example, an LED or a halogen lamp can be used.

[Imaging means 4]
The imaging means 4 is an imaging means for performing a transmission inspection step S1 and a crossed Nicol inspection step S2 described later, and in this embodiment, is arranged on the upper surface side (surface protection film 50 side) of the optical laminate S. The optical axis of the imaging means 4 (shown by a dashed line in FIG. 1A) is directed in a vertical direction (Z direction) parallel to the thickness direction of the optical laminate S, and the imaging means 4 receives light emitted from the light source 3 and transmitted through the optical laminate S in accordance with a control signal output from the control calculation means 9, forms an image, and outputs an electrical signal corresponding to the amount of light to the control calculation means 9 as an imaging signal. In addition, the imaging means 4 receives light emitted from the light sources 5a and 5b and transmitted through the inspection polarizing filters 6a and 6b and the optical laminate S in accordance with a control signal output from the control calculation means 9, forms an image, and outputs an electrical signal corresponding to the amount of light to the control calculation means 9 as an imaging signal. The focus of the imaging means 4 is set on the upper surface of the optical laminate S (upper surface of the surface protection film 50).

In this embodiment, a line sensor is used as the imaging means 4, in which a plurality of imaging elements (CCD or CMOS) are arranged in a straight line in a direction (Y direction) perpendicular to the transport direction (X direction) of the optical stack S, and which outputs an imaging signal at a constant scanning period (for example, 7 μsec to 14 μsec). By using a line sensor as the imaging means 4, the field of view of the imaging means 4 in the X direction is reduced, so that the required irradiation range in the X direction of the light emitted from the light sources 3, 5a, and 5b can be narrowed, and the constraints on the arrangement of the light sources 3, 5a, and 5b are relaxed. As the optical stack S is transported in the X direction, the imaging elements of the line sensor are scanned in the Y direction, so that a two-dimensional transmission image is generated in a transmission inspection process S1 described later, and a two-dimensional crossed Nicol image is generated in a crossed Nicol inspection process S2 described later.
However, the imaging means 4 is not necessarily limited to a line sensor, and for example, a two-dimensional camera with a high-speed shutter can also be used as the imaging means 4.

[Light source 5a, 5b]
The light sources 5a and 5b are light sources for performing a crossed Nicol inspection process S2 described later, and in this embodiment, are arranged on the lower surface side (separator 40 side) of the optical stack S. The optical axes of the light sources 5a and 5b (shown by dashed lines in FIG. 1A) are directed in a direction inclined with respect to the vertical direction (Z direction) parallel to the thickness direction of the optical stack S. Specifically, the optical axis of the light source 5a is directed in a direction inclined with respect to the vertical direction toward the downstream side in the conveying direction of the optical stack S, and the optical axis of the light source 5b is directed in a direction inclined with respect to the vertical direction toward the upstream side in the conveying direction of the optical stack S. The light sources 5a and 5b emit light upward toward the optical stack S according to a control signal output from the control calculation means 9.
The light sources 5a and 5b are not limited as long as they can emit light of a wavelength that can be transmitted through the optical laminate S, but for example, an LED or a halogen lamp can be used.
In this embodiment, the imaging means 4 is used (shared) as an imaging means for performing both the transmission inspection process S1 and the crossed Nicol inspection process S2. That is, since the light emitted from the light source 3 and the light sources 5a and 5b is received by the same imaging means 4, the direction of the optical axis of the light source 5a and 5b is different from the direction of the optical axis of the light source 3. In addition, the pair of light sources 5a and 5b are arranged so that a sufficient amount of light emitted from an inclined direction can be secured. However, for example, by adopting a coaxial epi-illumination optical system composed of a half mirror or the like, the direction of the optical axis of the light source 5a and 5b can be directed vertically like the direction of the optical axis of the light source 3, and it is also possible to configure using a single light source instead of a pair of light sources 5a and 5b.

[Inspection polarizing filters 6a, 6b]
The inspection polarizing filters 6a and 6b are arranged so as to be in a crossed Nicol state with respect to the polarization axis of the polarizer 10 of the optical laminate S. For example, if the direction of the polarization axis of the polarizer 10 is the X direction, the inspection polarizing filters 6a and 6b are arranged so as to be in the Y direction perpendicular to the X direction. However, the angle between the polarization axis of the polarizer 10 and the polarization axis of the inspection polarizing filters 6a and 6b does not necessarily have to be exactly 90°, and may be within the range of 90°±10°.
The configuration and manufacturing method of the inspection polarizing filters 6a and 6b are similar to those of the polarizer 10, and therefore a detailed description thereof will be omitted here.

In this embodiment, the inspection polarizing filters 6a and 6b are disposed on the lower surface side (separator 40 side) of the optical stack S. Specifically, the inspection polarizing filters 6a and 6b are disposed between the optical stack S and the light sources 5a and 5b, respectively, and the light emitted from the light sources 5a and 5b passes through the inspection polarizing filters 6a and 6b, respectively, and is irradiated onto the optical stack S. In the case of this embodiment, the crossed Nicol state is disrupted by a defect existing between the inspection polarizing filters 6a and 6b and the polarizer 10, so that in the crossed Nicol image of the optical stack S generated in the crossed Nicol inspection process S2 described later, the pixel region corresponding to the defect existing between the inspection polarizing filters 6a and 6b and the polarizer 10 becomes bright (the luminance value becomes large), and this defect can be detected as a defect candidate.
However, the present invention is not necessarily limited to this, and the inspection polarizing filters 6a and 6b can also be arranged on the upper surface side (surface protection film 50 side) of the optical stack S. Specifically, a configuration can be adopted in which one inspection polarizing filter is arranged between the optical stack S and the imaging means, and light emitted from the light sources 5a and 5b and transmitted through the optical stack S is transmitted through this inspection polarizing filter and received by the imaging means 4. In this case, since the crossed Nicol state is destroyed by a defect existing between the inspection polarizing filter and the polarizer 10, in the crossed Nicol image of the optical stack S generated in the crossed Nicol inspection process S2 described later, the pixel area corresponding to the defect existing between the inspection polarizing filter and the polarizer 10 becomes bright (the luminance value becomes large), and this defect can be detected as a defect candidate.

[Light source 7]
The light source 7 is a light source for performing a reflection inspection step S3 described later, and in this embodiment, is arranged on the lower surface side (separator 40 side) of the optical stack S. The optical axis of the light source 7 (shown by a dashed line in FIG. 1A) is directed in a direction inclined with respect to the vertical direction (Z direction) parallel to the thickness direction of the optical stack S. In the example shown in FIG. 1A, the optical axis of the light source 7 is directed in a direction inclined toward the upstream side of the conveying direction of the optical stack S with respect to the vertical direction. However, this is not limited to this, and the optical axis of the light source 7 can also be directed in a direction inclined toward the downstream side of the conveying direction of the optical stack S with respect to the vertical direction. In addition, for example, by adopting a coaxial epi-illumination optical system composed of a half mirror or the like, the optical axis of the light source 7 can also be directed in the vertical direction. The light source 7 emits light upward toward the optical stack S according to a control signal output from the control calculation means 9.
The light source 7 is not limited as long as it can emit light of a wavelength that can be reflected by the optical laminate S, but for example, an LED or a halogen lamp can be used.

[Imaging means 8]
The imaging means 8 is an imaging means for performing a reflection inspection step S3 described later, and in this embodiment, is disposed on the lower surface side (separator 40 side) of the optical stack S. The optical axis of the imaging means 8 (shown by a dashed line in FIG. 1A) is directed in a vertical direction (Z direction) parallel to the thickness direction of the optical stack S, and the imaging means 8 receives light emitted from the light source 7 and reflected by the optical stack S in accordance with a control signal output from the control and calculation means 9, forms an image, and outputs an electrical signal according to the amount of light to the control and calculation means 9 as an imaging signal. The focal point of the imaging means 8 is set on the lower surface of the optical stack S (the lower surface of the separator 40).

In this embodiment, as in the imaging means 4, a line sensor is used as the imaging means 8, in which a plurality of imaging elements (CCD or CMOS) are arranged in a straight line in a direction (Y direction) perpendicular to the transport direction (X direction) of the optical stack S, and which outputs an imaging signal at a constant scanning period (for example, 7 μsec to 14 μsec). By using a line sensor as the imaging means 8, the field of view of the imaging means 8 in the X direction is reduced, so that the required irradiation range in the X direction of the light emitted from the light source 7 is narrow, and the constraints on the arrangement of the light source 7, etc. are relaxed. As the optical stack S is transported in the X direction, the imaging elements of the line sensor are scanned in the Y direction, and a two-dimensional reflection image is generated in the reflection inspection process S3 described later.
However, the imaging means 8 is not necessarily limited to a line sensor, and for example, a two-dimensional camera with a high-speed shutter can also be used as the imaging means 8.

[Control and calculation means 9]
The control and calculation means 9 is constituted by, for example, a personal computer or a programmable logic controller (PLC) in which a program for executing the control processing and calculation processing described below is installed.

<Inspection method according to this embodiment>
Hereinafter, a method for inspecting the optical laminate S according to this embodiment using the inspection device 100 described above will be described.
FIG. 2 is a flow diagram showing an outline of the steps of the method for inspecting the optical laminate S according to this embodiment.
As shown in FIG. 2, the inspection method according to this embodiment includes a transmission inspection step S1, a crossed Nicols inspection step S2, a reflection inspection step S3, and a calculation step S4.
Each of steps S1 to S4 will be described below.

[Transmission inspection step S1]
In the transmission inspection step S1, a transmission image of the optical stack S is generated by light passing through the optical stack S, and defect candidates present in the optical stack S are detected based on this transmission image (S11 in FIG. 2).
Specifically, the light source 3 and the imaging means 4 are driven by a control signal output from the control and calculation means 9 just before the optical stack S reaches directly below the imaging means 4. The imaging means 4 receives the light emitted from the light source 3 and transmitted through the optical stack S, forms an image, and outputs an electrical signal according to the amount of light as an imaging signal to the control and calculation means 9. The control and calculation means 9 generates a two-dimensional transmission image based on this input imaging signal. The control and calculation means 9 then applies known image processing, such as binarization, to the generated transmission image to extract pixel regions whose luminance values (pixel values) are different from other pixel regions, thereby detecting defect candidates.

Fig. 3 is a diagram for explaining an example of a defect candidate detected in the transmission inspection process S1. Fig. 3(a) is a cross-sectional view for explaining an example of a defect present in the optical laminate S. Fig. 3(b) is a diagram for explaining an example of a defect candidate detected before executing the noise removal procedure S12 in the transmission inspection process S1. Fig. 3(c) is a diagram for explaining an example of a defect candidate remaining after executing the noise removal procedure S12 in the transmission inspection process S1.
In Fig. 3(a), the symbol F1 indicates a harmless foreign object attached to the surface of the separator 40, which is a release film. The symbol F2 indicates a harmless scratch present on the surface of the separator 40. The symbol F3 indicates a harmful attached foreign object present between the polarizer 10 and the retardation film 20. The symbol F4 indicates a harmless foreign object attached to the surface of the surface protection film 50, which is a release film. Figs. 3(b) and (c) show the transmission images after binarization, and in Fig. 3(b), three foreign objects F1 (F1a to F1c), two scratches F2 (F2a, F2b), one attached foreign object F3, and two foreign objects F4 (F4a, F4b) are each detected as defect candidates. In FIG. 3C, one foreign matter F1 (F1c), one scratch F2 (F2a), one bonding foreign matter F3, and two foreign matters F4 (F4a, F4b) are detected as defect candidates.

The transmission inspection process S1 of this embodiment includes a noise removal procedure (S12 in FIG. 2) that removes from the detected defect candidates those with dimensions (e.g., area) larger than a predetermined threshold value. Therefore, from the detected defect candidates shown in FIG. 3(b), foreign bodies F1a, F1b, and scratches F2b, which are defect candidates with relatively large dimensions, are removed, resulting in the state shown in FIG. 3(c).

[Crossed Nicols inspection process S2]
In the crossed Nicol inspection process S2, a crossed Nicol image of the optical stack S is generated using light passing through the inspection polarizing filters 6a, 6b and the optical stack S, and potential defects present in the optical stack S are detected based on this crossed Nicol image (S2 in Figure 2).
Specifically, the light sources 5a, 5b and the imaging means 4 are driven by a control signal output from the control calculation means 9 at a timing immediately before the optical stack S reaches directly under the imaging means 4. Then, the imaging means 4 receives the light emitted from the light sources 5a, 5b, respectively, and transmitted through the inspection polarizing filters 6a, 6b and the optical stack S, respectively, forms an image, and outputs an electric signal according to the amount of light as an imaging signal to the control calculation means 9. The control calculation means 9 generates a two-dimensional crossed Nicol image based on this input imaging signal. Then, the control calculation means 9 detects defect candidates by applying known image processing such as binarization to the generated crossed Nicol image to extract pixel areas whose luminance values (pixel values) are different from other pixel areas (luminance values are large).

Figure 4 is a diagram illustrating an example of defect candidates detected in the crossed Nicols inspection process S2. Figure 4 shows the crossed Nicols image after binarization, in which three foreign bodies F1 (F1a-F1c), three scratches F2 (F2b-F2d), and one bonding foreign body F3 have been detected as defect candidates. Foreign body F4 attached to the surface of the surface protection film 50 is not located between the polarizer 10 and the inspection polarizing filters 6a, 6b, and therefore is not detected in the crossed Nicols image, unlike the transmitted image.

In this embodiment, since the imaging means 4 for generating the transmitted image and the imaging means 4 for generating the crossed Nicols image are the same, the control and calculation means 9 controls switching between the timing for performing imaging by the imaging means 4 in the transmission inspection process S1 and the timing for performing imaging by the imaging means 4 in the crossed Nicols inspection process S2.
Specifically, the control calculation means 9 performs control to switch the timing of emitting light from the light source 3 used to generate a transmission image and the timing of emitting light from the light sources 5a, 5b used to generate a cross image for each scanning period of the imaging means 4. That is, the control calculation means 9 outputs a control signal to the light source 3 to cause the light source 3 to emit light in one scanning period, and then outputs a control signal to the light sources 5a, 5b to cause the light sources 5a, 5b to emit light in the next scanning period. The control calculation means 9 further outputs a control signal to the light source 3 to cause the light source 3 to emit light in the next scanning period. The control calculation means 9 repeats the above operations until one optical laminate S has passed directly under the imaging means 4.

Fig. 5 is a diagram for explaining the switching control executed by the control calculation means 9. As described above, the control calculation means 9 switches the timing of emitting light from the light source 3 and the timing of emitting light from the light sources 5a and 5b for each scanning period of the imaging means 4, so that, as shown in Fig. 5(a), the imaging means 4 alternately forms images of the light emitted from the light source 3 and transmitted through the optical stack S (the region shown by the white outline in Fig. 5(a)) and the light emitted from the light sources 5a and 5b and transmitted through the inspection polarizing filters 6a and 6b and the optical stack S (the region hatched in dots in Fig. 5(a)) in the X direction at a pitch corresponding to the scanning period.
The control and calculation means 9 generates a transmission image as shown in Fig. 5(b) by extracting only the regions shown in white in Fig. 5(a) and combining them in the X direction in accordance with the scanning cycle of the imaging means 4. The control and calculation means 9 also generates a crossed Nicol image as shown in Fig. 5(c) by extracting only the regions hatched in dots in Fig. 5(a) and combining them in the X direction in accordance with the scanning cycle of the imaging means 4.
In this manner, the control and calculation means 9 is capable of generating a transmission image and a crossed Nicol image separately even if the imaging means 4 for generating a transmission image and the imaging means 4 for generating a crossed Nicol image are the same.

[Reflection inspection step S3]
In the reflection inspection step S3, a reflection image of the optical stack S is generated by light reflected by the optical stack S, and defect candidates present in the optical stack S are detected based on this reflection image (S3 in FIG. 2).
Specifically, the light source 7 and the imaging means 8 are driven by a control signal output from the control and calculation means 9 just before the optical stack S reaches directly below the imaging means 8. The imaging means 8 receives the light emitted from the light source 7 and reflected by the optical stack S, forms an image, and outputs an electrical signal according to the amount of light to the control and calculation means 9 as an imaging signal. The control and calculation means 9 generates a two-dimensional reflected image based on this input imaging signal. The control and calculation means 9 then applies known image processing, such as binarization, to the generated reflected image to extract pixel areas that have a different luminance value (pixel value) from other pixel areas, thereby detecting defect candidates.

Figure 6 is a diagram illustrating an example of a defect candidate detected in the reflection inspection process S3. Figure 6 shows the reflected image after binarization, in which three foreign bodies F1 (F1a to F1c) have been detected as defect candidates. Because the reflected image is generated by light reflected by the optical laminate S, only the foreign body F1 attached to the surface of the separator 40 on the side where the light source 7 is located is detected. In the example shown in Figure 6, only the foreign body F1 is detected, but there are also cases where a scratch F2 present on the surface of the separator 40 is detected.

[Calculation step S4]
In the calculation process S4, defects present between the polarizer 10 and the optical film (in this embodiment, the phase difference film 20) are determined based on the defect candidates detected in the transmission inspection process S1, the defect candidates detected in the crossed Nicol inspection process S2, and the defect candidates detected in the reflection inspection process S3 (S4 in Figure 2).
Specifically, in the calculation step S4, the calculation control means 9 first judges whether a certain defect candidate is detected in both the transmission inspection step S1 and the crossed Nicols inspection step S2 (S41 in FIG. 2). Whether a defect candidate is detected in both the transmission inspection step S1 and the crossed Nicols inspection step S2 is judged based on whether a defect candidate detected in the crossed Nicols inspection step S2 exists at a position equivalent (the same or close) to the position of a certain defect candidate detected in the transmission inspection step S1. Specifically, for example, it is judged based on whether a center of gravity of a defect candidate detected in the crossed Nicols inspection step S2 exists at a position equivalent to the position of the center of gravity of a defect candidate detected in the transmission inspection step S1 (for example, a position ±2 mm from the position of the center of gravity). In order to judge whether a defect candidate detected in the crossed Nicols inspection step S2 exists at a position equivalent to the position of a defect candidate detected in the transmission inspection step S1, the coordinates of the transmission image and the coordinates of the crossed Nicols image must match. In this embodiment, as described above, the imaging means 4 for generating the transmission image and the imaging means 4 for generating the crossed Nicol image are the same, so the coordinates of the transmission image and the coordinates of the crossed Nicol image almost match, and there is little need to make the coordinates of both images match strictly. However, strictly speaking, as can be seen from the contents described with reference to Fig. 5, the coordinates of the transmission image and the coordinates of the crossed Nicol image are shifted in the X direction by a pitch corresponding to the scanning period of the imaging means 4, so it is preferable that the calculation control means 9 corrects this shift to make the coordinates of both images match.

FIG. 7 is a diagram for explaining the content of the calculation step S4.
7A is a diagram for explaining an example of defect candidates determined to be detected in both the transmission inspection step S1 and the crossed Nicols inspection step S2 in the calculation step S4 (specifically, S41). As shown in FIG. 3C, in the transmission inspection step S1, the foreign matter F1c, the scratch F2a, the bonding foreign matter F3, and the foreign matters F4a and F4b are detected as defect candidates, and as shown in FIG. 4, in the crossed Nicols inspection step S2, the foreign matter F1a to F1c, the scratches F2b to F2d, and the bonding foreign matter F3 are detected as defect candidates. For example, the foreign matter F1c detected in the transmission inspection step S1 is also detected in the crossed Nicols inspection step S2, so the calculation control means 9 determines that the foreign matter F1c is detected in both steps (S41 in FIG. 2 is "Yes"). On the other hand, for example, the scratch F2a detected in the transmission inspection process S1 is not detected in the crossed Nicols inspection process S2, so the calculation control means 9 determines that the scratch F2a is not detected in both processes ("No" in S41 in FIG. 2), and determines that this defect candidate (scratch F2a) is not the bonding foreign matter F3 present between the polarizer 10 and the retardation film 20 (S44 in FIG. 2). In this embodiment, by performing the same calculation for all the defect candidates detected in the transmission inspection process S1, it is determined that the defect candidates (foreign matter F1c, bonding foreign matter F3) shown in FIG. 7(a) are defect candidates detected in both the transmission inspection process S1 and the crossed Nicols inspection process S2.

Next, in the calculation step S4, the calculation control means 9 judges whether the defect candidate detected in both the transmission inspection step S1 and the crossed Nicols inspection step S2 was detected in the reflection inspection step S3 (S42 in FIG. 2). Whether the defect candidate detected in both the transmission inspection step S1 and the crossed Nicols inspection step S2 was detected in the reflection inspection step S3 is judged based on whether the defect candidate detected in the reflection inspection step S3 exists at a position equivalent (the same or close) to the position of a defect candidate detected in both the transmission inspection step S1 and the crossed Nicols inspection step S2. Specifically, for example, it is judged based on whether the center of gravity of the defect candidate detected in the reflection inspection step S3 exists at a position equivalent (for example, a position ±2 mm from the center of gravity) to the position of the center of gravity of the defect candidate detected in both the transmission inspection step S1 and the crossed Nicols inspection step S2. In order to determine whether or not a defect candidate detected in the reflection inspection process S3 exists at a position equivalent to the position of the defect candidate detected in both the transmission inspection process S1 and the crossed Nicols inspection process S2, the coordinates of the transmission image and the crossed Nicols image must match the coordinates of the reflection image. The coordinates of the transmission image and the crossed Nicols image and the coordinates of the reflection image are shifted in the X direction by the time obtained by dividing the separation distance L (see FIG. 1(a)) in the X direction between the imaging means 4 and the imaging means 8 by the conveying speed V of the optical laminate S, so the arithmetic control means 9 needs to correct this shift to make the coordinates of the transmission image and the crossed Nicols image match the coordinates of the reflection image. In addition, since there is a possibility that the coordinates of the transmitted image and the crossed Nicol image and the coordinates of the reflected image may be misaligned in the Y direction, it is preferable that the calculation control means 9 applies known image processing to detect the edges in the Y direction (edges of the optical stack S) in the transmitted image and the crossed Nicol image and the edges in the Y direction (edges of the optical stack S) in the reflected image, and match the coordinates of the transmitted image and the crossed Nicol image with the coordinates of the reflected image so that the positions of these edges match.

Figure 7(b) is a diagram illustrating an example of a defect candidate determined in the calculation step S4 (specifically, S42) not to have been detected in the reflection inspection step S3. As shown in the above-mentioned Figure 7(a), of the foreign matter F1c and the bonding foreign matter F3 detected in both the transmission inspection step S1 and the crossed Nicol inspection step S2, as shown in the above-mentioned Figure 6, the foreign matter F1c is also detected in the reflection inspection step S3, so the calculation control means 9 determines that the foreign matter F1c was detected in the reflection inspection step S3 (S42 in Figure 2 is "Yes") and determines that this defect candidate (foreign matter F1c) is not the bonding foreign matter F3 present between the polarizer 10 and the phase difference film 20 (S44 in Figure 2). On the other hand, of the foreign matter F1c and the bonding foreign matter F3 detected in both the transmission inspection process S1 and the crossed Nicols inspection process S2, as shown in FIG. 6 described above, the bonding foreign matter F3 is not detected in the reflection inspection process S3. Therefore, the arithmetic control means 9 determines that the bonding foreign matter F3 is not detected in the reflection inspection process S3 ("No" in S42 in FIG. 2), and determines that this defect candidate (bonding foreign matter F3) is the bonding foreign matter F3 present between the polarizer 10 and the retardation film 20 (S43 in FIG. 2).

According to the findings of the inventors, a defect candidate that is detected in both the transmission image and the crossed Nicol image and not detected in the reflection image is likely to be a defect (bonding foreign matter) that exists between the polarizer 10 and the phase difference film 20. According to the inspection method of this embodiment, as described above, in the calculation step S4, a defect candidate that is detected in both the transmission inspection step S1 and the crossed Nicol inspection step S2 (i.e., detected in both the transmission image and the crossed Nicol image) and not detected in the reflection inspection step S3 (i.e., not detected in the reflection image) is determined to be a defect that exists between the polarizer 10 and the phase difference film 20, so that overdetection of defects that exist on the surface of the release film (separator 40, surface protection film 50) is suppressed, and defects that exist between the polarizer 10 and the phase difference film 20 can be detected with high accuracy.

In this embodiment, the inspection object is an optical laminate S in which optical films 20, 30 (phase difference film 20, protective film 30) are laminated on both sides of the polarizer 10 in the thickness direction, and release films 40, 50 (separator 40, surface protective film 50) are laminated on both outermost surfaces in the thickness direction. However, the present invention is not limited to this. The present invention is applicable to various optical laminates as long as the optical laminate is an optical laminate in which at least one optical film (e.g., only phase difference film 20) is laminated on the polarizer 10, and a release film (e.g., only separator 40) is laminated on at least one outermost surface.

In addition, in this embodiment, the case where the inspection object is an optical laminate S cut into chips has been described as an example, but the present invention is not limited to this. It is also possible to adopt a configuration in which the inspection is performed while transporting a long optical laminate by roll-to-roll, similar to the inspection methods described in Patent Documents 1 to 3.

In addition, in this embodiment, the light sources 3, 5a, 5b, and 7 are arranged on the underside (separator 40 side) of the optical stack S, but the present invention is not limited to this, and it is also possible to adopt a configuration in which the light sources 3, 5a, 5b, and 7 are arranged on the upper side (surface protection film 50 side) of the optical stack S (the inspection polarizing filters 6a and 6b are also arranged on the upper side of the optical stack S). In this case, the imaging means 4 is arranged on the lower side of the optical stack S, and the imaging means 8 is arranged on the upper side of the optical stack S. In this case, in the crossed Nicol inspection process S2, any bonding foreign matter present between the polarizer 10 and the protective film 30 is detected.

In addition, in this embodiment, an example has been described in which the imaging means 4 for generating a transmission image in the transmission inspection process S1 and the imaging means 4 for generating a crossed Nicol image in the crossed Nicol inspection process S2 are the same, but the present invention is not limited to this, and it is also possible to provide an imaging means for generating a transmission image and an imaging means for generating a crossed Nicol image separately.

Furthermore, in this embodiment, the case where the transmission inspection process S1, the crossed Nicols inspection process S2, and the reflection inspection process S3 are performed in this order (however, the timing of performing imaging in the transmission inspection process S1 and the crossed Nicols inspection process S2 overlaps) has been described as an example, but the present invention is not limited to this, and they can be performed in any order. It is also possible to adopt a procedure in which, after performing the transmission inspection process S1 and the crossed Nicols inspection process S2, only S41 of the calculation process S4 is performed first, and then the reflection inspection process S3 is performed, and then S42 of the calculation process S4 is performed.

REFERENCE SIGNS LIST 1: Belt conveyor 2: Clean roller 3, 5a, 5b, 7: Light source 4, 8: Imaging means 6a, 6b: Inspection polarizing filter 10: Polarizer 20: Retardation film (optical film)
30: Protective film (optical film)
40: Separator (release film)
50: Surface protection film (release film)
100: Inspection device S: Optical laminate S1: Transmission inspection process S2: Cross Nicol inspection process S3: Reflection inspection process S4: Calculation process

Claims (5)

A method for inspecting an optical laminate in which a polarizer and an optical film are laminated together, and a release film is further laminated on at least one outermost surface side in a thickness direction, comprising the steps of:
A transmission inspection process of generating a transmission image of the optical stack by light transmitted through the optical stack, and detecting defect candidates present in the optical stack based on the transmission image;
a crossed Nicol inspection process for generating a crossed Nicol image of the optical stack using an inspection polarizing filter arranged to be crossed Nicol with respect to the polarization axis of the polarizer and light passing through the optical stack, and detecting defect candidates present in the optical stack based on the crossed Nicol image;
A reflection inspection process of generating a reflection image of the optical stack by light reflected by the optical stack, and detecting defect candidates present in the optical stack based on the reflection image;
a calculation step of determining a defect present between the polarizer and the optical film based on the defect candidates detected in the transmission inspection step, the defect candidates detected in the crossed Nicols inspection step, and the defect candidates detected in the reflection inspection step,
In the calculation step, a defect candidate that is detected in both the transmission inspection step and the crossed Nicol inspection step and not detected in the reflection inspection step is determined to be a defect that exists between the polarizer and the optical film.
A method for inspecting an optical laminate. the release film is a separator,
the optical film is located between the separator and the polarizer;
In the crossed Nicol inspection step, the inspection polarizing filter is disposed on the separator side.
A method for inspecting the optical laminate according to claim 1 . The orientation direction of the release film is within ±6° of a predetermined specified orientation direction;
A method for inspecting the optical laminate according to claim 1 or 2. an imaging means for generating the transmission image in the transmission inspection step and an imaging means for generating the crossed Nicol image in the crossed Nicol inspection step are identical;
switching a timing for performing imaging by the imaging means in the transmission inspection step and a timing for performing imaging by the imaging means in the crossed Nicols inspection step;
A method for inspecting the optical laminate according to claim 1 . the transmission inspection step and/or the crossed Nicols inspection step includes a noise removal step of excluding defect candidates having dimensions larger than a predetermined threshold value from the detected defect candidates;
A method for inspecting the optical laminate according to claim 1 .

Images