Feature point extraction methods, control programs, and information processing systems

Abstract

[Problem] To efficiently collect information that will help improve processes and set shipping specifications during the manufacturing of strip materials and in subsequent processing using this material. [Solution] A method for extracting feature points includes: step (c), which compares the first feature point information of the strip material in the first inspection data with the second feature point information of the strip material in the second inspection data; and step (d), which, based on the comparison result of step (c), extracts a third type of feature point existing in both the first and second inspection data, and / or a first type of feature point existing in the first inspection data but not in the second inspection data, or a second type of feature point existing in the second inspection data but not in the first inspection data; wherein, the inspection data of one of the first and second inspection data is inspection data obtained using one or more inspection methods, and the inspection data of the other party is inspection data obtained using one or more inspection methods that include inspection methods different from those of the inspection data of one party.

Description

Technical Field

[0001] This invention relates to a method for extracting feature points, a control program, and an information processing system. Prior Technology

[0002] Liquid crystal display (LCD) devices are increasingly used in large-screen televisions and large displays, which in turn requires wider films for the display surfaces of LCD devices. For example, films with a width of 2000mm or more are required. In addition, in order to anticipate substrate loss (film loss) or to reduce transportation costs, roll lengths of 1000m or more are required, and further, long film rolls of 3000m or more are needed.

[0003] In the post-processing stages of manufacturing products based on web-like materials such as films, when quality issues arise, it's necessary to categorize them as either defects that originated in the pre-processing stages (i.e., during the manufacturing of the web-like material) or those that arose during the post-processing stages. When it's impossible to clearly categorize defects as arising in post-processing, improvements to the pre-processing stages are sometimes required. To meet these improvement requests, the shipping specifications of the pre-processing stages must be excessively tightened, resulting in overefficiency. Furthermore, although the exact impact of the post-processing stages cannot be clearly determined, to prevent quality problems before they occur, sometimes the following approach is adopted: the shipping specifications during web-like material manufacturing are excessively tightened as expected. This situation also results in overefficiency. Overefficiency leads to poor yield rates, increased costs, and is environmentally unfriendly; moreover, it is detrimental to both the companies performing the pre-processing and post-processing stages.

[0004] Although it is necessary to distinguish between defects that originally existed on the strip or those generated in subsequent processing, in this case, it is necessary to establish a correspondence between the coordinate system on the strip in the previous process and the coordinate system in the subsequent process.

[0005] Patent Document 1 below discloses a technology in an optical film inspection system that synchronizes the inspection results of a first inspection device and a second inspection device during inspection. In this inspection system, the synchronization of the two inspection results is achieved by utilizing the distance between the first inspection device and the second inspection device arranged in the subsequent stage in the transfer direction, and by utilizing an encoder signal that displays the amount of film movement. [Previous Technical Documents] [Patent Literature]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2016-161576 Summary of the Invention

[0007] [The problem that the invention aims to solve] The technology in Patent Document 1, which achieves synchronization of inspection results by utilizing the mutual distance and movement distance between inspection devices, cannot achieve synchronization when the membrane stretches or contracts. Furthermore, synchronization cannot be achieved if there is an offline process where the membrane is wound up and fed out between the two inspection devices.

[0008] The present invention was created in view of the above circumstances, with the aim of efficiently collecting information that is helpful for process improvement and shipping specification setting in both the manufacturing process of strip material and the subsequent processing using it. [Methods used to solve problems]

[0009] The above-mentioned objectives of the present invention are achieved by the following means.

[0010] (1) A method for extracting feature points of a strip material. have: Step (a) is to obtain the first inspection data in the first manufacturing process of manufacturing strip or performing post-processing on the manufactured strip; Step (b) involves obtaining the second inspection data in the second manufacturing process, which is performed after the first manufacturing process and involves post-processing when using the aforementioned strip material; Step (c) involves comparing the first feature point information of the aforementioned strip material in the first inspection data with the second feature point information of the aforementioned strip material in the second inspection data; and Step (d) involves extracting, based on the comparison results of step (c), the third type of feature points existing in both the first and second inspection data, and / or the first type of feature points existing in the first inspection data but not in the second inspection data, or the second type of feature points existing in the second inspection data but not in the first inspection data; The inspection data of one of the aforementioned first inspection data and second inspection data is inspection data obtained by using one or more inspection methods, and the inspection data of the other party is inspection data obtained by using one or more inspection methods that include inspection methods different from the inspection methods of the aforementioned first inspection data.

[0011] (2) The feature point extraction method as described in (1) above, wherein, The aforementioned first and second inspection data are inspection data obtained by photographing the strip material using an inspection device. The photographic method of the inspection data of the other party in the aforementioned inspection device is the inspection data obtained by using one or more photographic methods that are different from the photographic method of the inspection data of the other party in the inspection device.

[0012] (3) The feature point extraction method as described in (2) above, wherein, The photographic method used in obtaining the inspection data of the aforementioned party, depending on the positional relationship between the lighting and the photographic unit, includes at least one of the following: a first photographic method for photographing transmitted light traveling straight within the strip; a second photographic method for photographing transmitted light reflected within the strip; a third photographic method for photographing reflected light orthogonally reflected from the surface of the strip; and a fourth photographic method for photographing reflected light scattered from the surface of the strip. The photographic method used for the examination materials of the other party mentioned above, including the photographic methods mentioned in the first to fourth paragraphs, is different from the photographic method used for the examination materials of the other party mentioned above.

[0013] (4) The feature point extraction method as described in (2) above, wherein, The aforementioned photographic methods include photographic methods that use polarized light to detect the polarization state of light.

[0014] (5) The feature point extraction method as described in (1) above, wherein, The inspection data of at least one of the aforementioned first inspection data or second inspection data is inspection data that integrates feature point information extracted from multiple primary data obtained using multiple types of inspection methods in the same inspection area by aligning their positions on the strip surface.

[0015] (6) The feature point extraction method as described in (5) above, wherein, Furthermore, it includes step (e), which involves extracting the third type of feature points using the aforementioned aggregated inspection data in step (d), and then classifying the feature points of the aforementioned aggregated inspection data into feature points of each of the plurality of primary data.

[0016] (7) The feature point extraction method as described in (6) above, wherein, Include: Step (f) is the selection of the aforementioned inspection method for the inspection materials received by the aforementioned party; and Step (g) is to display the feature point information of the primary data classified in step (e) corresponding to the selected inspection method, based on the selection in step (f).

[0017] (8) A method for extracting feature points of a strip material. have: Step (a) is to obtain the first inspection data in the first manufacturing process of manufacturing strip or performing post-processing on the manufactured strip; Step (b) involves obtaining the second inspection data in the second manufacturing process, which is performed after the first manufacturing process and involves post-processing when using the aforementioned strip material; Step (c) involves comparing the first feature point information of the aforementioned strip material in the first inspection data with the second feature point information of the aforementioned strip material in the second inspection data; and Step (d) involves extracting the third type of feature points existing in both the first and second examined data, based on the comparison results of step (c) above. At least one of the aforementioned first inspection data and second inspection data is inspection data that integrates feature point information extracted from multiple primary data obtained using multiple types of inspection methods in the same inspection area, and aligns their positions on the strip surface. Furthermore, it includes step (e), which involves using the aforementioned consolidated inspection data, after extracting the aforementioned third type of feature points in step (d), classifying the feature points of the aforementioned consolidated inspection data into feature points of each of the plurality of primary data.

[0018] (9) The feature point extraction method as described in (8) above, wherein, The aforementioned first and second inspection data are inspection data obtained by photographing the strip material using an inspection device. The examination device for the aforementioned examination data is the examination data obtained using multiple photographic methods.

[0019] (10) The feature point extraction method as described in (8) above, wherein, Include: Step (f) is the process of selecting the aforementioned inspection method for the inspection materials received by the aforementioned party; and Step (g) is to display the feature point information of the primary data classified in step (e) corresponding to the selected inspection method, based on the selection in step (f).

[0020] (11) A control program for causing a computer to perform any one of the extraction methods described in (1) to (10) above.

[0021] (12) An information processing system, have: The acquisition unit acquires first inspection data in the first manufacturing process of manufacturing strip or performing post-processing on the manufactured strip, and acquires second inspection data in the second manufacturing process of performing post-processing treatment when using the aforementioned strip after the first manufacturing process; The comparison unit is one that compares the first feature point information of the aforementioned strip material in the first inspection data with the second feature point information of the aforementioned strip material in the second inspection data; and The extraction unit extracts, based on the comparison results of the comparison unit, third-type feature points existing in both the first and second inspection data, and / or first-type feature points existing in the first inspection data but not in the second inspection data, or second-type feature points existing in the second inspection data but not in the first inspection data; The inspection data of one of the aforementioned first inspection data and second inspection data is inspection data obtained by using one or more inspection methods, and the inspection data of the other party is inspection data obtained by using one or more inspection methods that include inspection methods different from the inspection methods of the aforementioned first inspection data.

[0022] (13) An information processing system, have: The acquisition unit acquires first inspection data in the first manufacturing process of manufacturing strip or performing post-processing on the manufactured strip, and acquires second inspection data in the second manufacturing process of performing post-processing treatment when using the aforementioned strip after the first manufacturing process; The comparison unit is one that compares the first feature point information of the aforementioned strip material in the first inspection data with the second feature point information of the aforementioned strip material in the second inspection data; and The extraction unit extracts a third type of feature point existing in both the first and second inspection data, based on the comparison results of the aforementioned comparison unit. At least one of the aforementioned first inspection data and second inspection data is inspection data that integrates feature point information extracted from multiple primary data obtained using multiple types of inspection methods in the same inspection area, and aligns their positions on the strip surface. The aforementioned extraction unit, after extracting the aforementioned third type of feature points from the aggregated inspection data, classifies the feature points of the aggregated inspection data into feature points for each of the plurality of primary data. [Comparison with the effectiveness of previous technologies]

[0023] The method for extracting feature points of strip materials associated with this invention, have: Step (a) is to obtain the first inspection data in the first manufacturing process of manufacturing strip or performing post-processing on the manufactured strip; Step (b) involves obtaining the second inspection data in the second manufacturing process, which is performed after the first manufacturing process and involves post-processing when using the aforementioned strip material; Step (c) involves comparing the first feature point information of the aforementioned strip material in the first inspection data with the second feature point information of the aforementioned strip material in the second inspection data; and Step (d) involves extracting, based on the comparison results of step (c), the third type of feature points existing in both the first and second inspection data, and / or the first type of feature points existing in the first inspection data but not in the second inspection data, or the second type of feature points existing in the second inspection data but not in the first inspection data; The inspection data of one of the aforementioned first and second inspection data refers to inspection data obtained using one or more inspection methods, while the inspection data of the other refers to inspection data obtained using one or more inspection methods that include inspection methods different from those of the aforementioned first inspection data. Accordingly, information that will help improve processes and set shipping specifications can be collected efficiently by both the manufacturing process of the strip material and the manufacturing process that uses it for subsequent processing. Simple Explanation of the Diagram

[0024] The advantages and features provided by one or more embodiments of the present invention will be more fully understood from the following detailed description and drawings. However, these are for illustrative purposes only and are not intended to limit the invention. [Figure 1] is a schematic diagram illustrating an application example of the information processing system associated with this embodiment. [Figure 2] is a table illustrating the characteristics related to various photographic methods. [Figure 3] is a schematic diagram illustrating the relationship between lighting and camera position for photography modes 1 to 5. [Figure 4] is a table illustrating the correspondence between the photography units 1 to 6 and the photography methods in this embodiment. [Figure 5A] is a schematic diagram illustrating an example of the configuration of the inspection device. [Figure 5B] is a schematic diagram illustrating an example of the configuration of the inspection device. [Figure 6] is a table illustrating the list of inspection devices. [Figure 7] is a block diagram illustrating the schematic structure of an information processing system. [Figure 8] shows examples of various materials memorized in the memory section. [Figure 9] shows an example of the database (DB) stored in the memory department. [Figure 10] is a flowchart illustrating the generation process of the first inspection data performed in the first manufacturing process. [Figure 11] shows an example of the check data DB stored in the memory department. [Figure 12] is a flowchart illustrating the generation process of the second inspection data performed in the second manufacturing process. [Figure 13] shows an example of the check data DB stored in the memory department. [Figure 14] is a flowchart illustrating the feature point extraction process performed in the information processing system. [Figure 15] shows an example of an operation screen displayed on a terminal device. [Figure 16] is a schematic diagram illustrating the feature point extraction process. [Figure 17] is a subroutine flowchart illustrating the comparison process 1 in step S34. [Figure 18] shows an example of the probability density function for the position and intensity of display feature points calculated using kernel density estimation. [Figure 19] is an example of a list of corresponding points. [Figure 20] is a table for illustrating the first to third type of feature points. [Figure 21A] shows an example of the feature point extraction results. [Figure 21B] is an example of the check data DB in the memory department corresponding to Figure 13. [Figure 22] is a subroutine flowchart illustrating the display process for step S38. [Figure 23] shows an example of the operation screen for accepting one data selection. [Figure 24A] shows an example of a display on the preview screen. [Figure 24B] shows an example of a display on the preview screen. Implementation

[0025] Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the scope of the present invention is not limited to the disclosed embodiments. Furthermore, the same symbols are used to denote the same elements in the drawings, and repeated descriptions are omitted. Additionally, the dimensions in the drawings are exaggerated for ease of explanation and sometimes differ from the actual dimensions.

[0026] (Strip material) In this embodiment, strip material refers to sheet-like material, including resin film and metal film. Furthermore, strip material includes laminated bodies. Hereinafter, strip material refers to a long resin film, and the process will be described as a film roll. The processing includes coating with a coating liquid, and also includes a lamination process for forming a laminate by stacking other film-like materials.

[0027] (Feature Point Information) In this embodiment, feature points refer to defects on the membrane, and feature points are generated by analyzing image data. Image analysis can be performed using known techniques, where pixels whose pixel values ​​deviate from the surrounding average value by more than a predetermined amount (the difference is more than a predetermined amount) from image data taken of the membrane surface are extracted as feature points. Alternatively, they can be calculated using the "image processing for feature point generation" method described later. From one or more image data taken of a membrane roll 80 (several km in length), in most cases, tens to thousands of feature points are generated. Defects include both anomalies that reach the level of product defects and minor anomalies that do not reach the level of product defects. Feature points include defects related to poor adhesion, axial unevenness, etc., when the membranes are bonded together (e.g., ultrasonic fusion). Feature point information includes size and position (xy coordinates). Furthermore, feature point information can be obtained by grouping multiple close feature points into one (grouping). Image processing related to feature point generation will be discussed later.

[0028] Figure 1 is a schematic diagram illustrating an application example of the information processing system 50 associated with this embodiment. As shown in Figure 1, the information processing system 50 and terminal devices 70 in factories A and B are interconnected via a network. The network is a communication line such as a data communication network. In some cases, wired LANs, wireless LANs, etc. (e.g., LANs conforming to the IEEE 802.11 standard) may also be used. Details regarding the information processing system 50 will be described later.

[0029] Terminal device 70, for example, is a PC (personal computer). For example, terminal device 70 is a PC used by employees of a manufacturing company that operates factory A and factory B.

[0030] The aforementioned film roll manufacturing apparatus 1000 is provided at factory A. Factory A may be operated or managed by a film manufacturer, for example. At factory A, a first manufacturing process for manufacturing film roll 80 is carried out. In the first manufacturing process, a plurality of sub-processes such as drying, stretching, and winding are performed. Before and after the plurality of sub-processes, inspection devices 90a1 and a2 (hereinafter, these are also collectively referred to as inspection devices 90) are arranged. The film surface of film roll 80 is inspected using the plurality of inspection devices 90a1 and 90a2. Although two inspection devices 90a1 and 90a2 are shown in Figure 1, the number of inspection devices 90 is not limited to this; it may be one or more than three.

[0031] Factory B is equipped with a product manufacturing facility 2000. Factory B is operated or managed, for example, by a coating manufacturer (hereinafter also referred to as a user company or user). There are multiple user companies operating each of Factory B. In Factory B, products are manufactured using film rolls 80 shipped and conveyed from Factory A. Factory B includes multiple sub-processes for subsequent processing, which include coating using film (film F8 described below) fed from film roll 80, or laminating or bonding with other films. Sub-processes are equivalent to the first or second manufacturing process. Multiple inspection devices 90b1 to 90b4 (hereinafter collectively referred to as inspection devices 90) are arranged before and after the multiple sub-processes. Although four inspection devices 90b1, 90b2, 90b3, and 90b4 are shown in Figure 1, the number of inspection devices 90 is not limited to this and may be three or less or five or more.

[0032] In this embodiment, in a typical example, the process of manufacturing the film (strip) is referred to as the first manufacturing process, and the post-processing performed on the film manufactured in this process at factory B is referred to as the second manufacturing process. For example, in the post-processing process, a coating process to impart a functional layer to the surface is performed, and a lamination process is performed to stack other strips such as films. In the second manufacturing process, the film surface of the film roll 80 is also inspected using the inspection device 90.

[0033] Regarding the material of the manufactured membrane F8 shown in Figure 1, while not particularly limited, generally, examples include polycarbonate resin, polyurethane resin, acrylic resin, polyolefin resin, cycloolefin resin, polyether resin, polyester resin, polyamide resin, polysulfide resin, unsaturated polyester resin, epoxy resin, melamine resin, phenolic resin, diallyl phthalate resin, polyimide resin, carbamate resin, polyvinyl acetate resin, polyvinyl alcohol resin, styrene resin, cellulose acetate resin, and polyvinyl chloride resin. Furthermore, for example, considering productivity and quality, a width of 1000 mm to 3200 mm is preferred for the membrane F8. Considering quality and processing, a thickness of 15 μm to 500 μm is preferred for the membrane F8.

[0034] In a typical example other than those of this embodiment, when the film manufactured by factory B undergoes multiple post-processing steps, the earlier post-processing step is sometimes referred to as the first manufacturing process, and the subsequent post-processing step is sometimes referred to as the second manufacturing process. Furthermore, sometimes the film roll 80 manufactured by factory A is sent out for post-processing in a later process. In a typical example, the sub-processes performed by factory B include a lamination process (also called an adhesion process) that laminates and adheres other films. The lamination process forms a laminate 80B (see the enlarged cross-sectional view of the dialog box effect in FIG1). Hereinafter, using several sub-processes of factory B, layers 1 to 3 are adhered or coated onto the original roll film (F8). Also, in the enlarged cross-sectional view of the dialog box effect, the markings of the adhesion layers between each layer are omitted. For example, film F8 is a PVA (polyvinyl alcohol resin) layer, layer 1 is a TAC (cellulose triacetate) layer, layer 2 is an optical functional film layer such as a polarizer, and layer 3 is a protective device. In addition, the third layer can also be a separation layer or an anti-glare film, etc. The third layer is an opaque layer.

[0035] The inspection device 90 photographs the membrane and other components to generate image data. As shown in Figure 1, multiple inspection devices 90 are configured before and after sub-processes such as the extension process and lamination process in the membrane production line. Hereinafter, the inspection device 90 will be described with reference to Figures 2 to 6.

[0036] (Inspection device 90) For inspection devices that use transparent materials such as film F8 as the subject of inspection and detect surface irregularities, bubbles, cracks, and internal structural strain within the transparent material, these include imaging units with both transmissive and reflective imaging methods. Similarly, for inspection devices that use opaque materials such as laminate 80B as the subject of inspection and detect surface irregularities such as scratches, these include imaging units with reflective imaging methods. Furthermore, regarding both transmissive and reflective types, depending on the relationship between the camera's optical axis, the light source, and the subject's position, these include bright-field inspection devices that receive non-scattered light from the surface and dark-field inspection devices that receive scattered light. Bright-field inspection devices correspond to the first, third, and fifth imaging methods described later and receive non-scattered light. Dark-field inspection devices correspond to the second, fourth, and fifth imaging methods described later and receive scattered light. In a bright-field inspection device, when there are no defects, there is no light scattering, so light from the light source is not blocked and enters the light detection device. When a defect exists, the defect blocks the light, preventing it from entering the light detection device. Therefore, the defect is observed as a dark spot or stripe against a bright background. In contrast, in a dark-field inspection device, when there are no defects, light is not scattered and therefore does not enter the light detection device. However, when a defect exists, light is scattered due to the defect and enters the light detection device. Therefore, the defect is observed as a bright spot or stripe against a dark background.

[0037] The inspection apparatus 90 in this embodiment can be any type of inspection apparatus. One of the first and second inspection data is inspection data obtained using one or more inspection methods, while the other is inspection data obtained using one or more inspection methods that include inspection methods different from those used for the first inspection data. Alternatively, one of the inspection data can be inspection data obtained using multiple inspection methods (hereinafter referred to as integrated inspection data). While inspection methods with high detection sensitivity are preferred, sometimes the available inspection methods are limited by the state of the object being inspected (the object being photographed). For example, inspection of films with optical functions such as polarizers, phase retardation films, optical compensation films, and liquid crystal layers requires photographing using a polarizing camera or similar polarizing photographic method. Furthermore, in the case of opaque or low-transparency films, sometimes through-photography methods cannot be used, or even if used, the sensitivity is low. Therefore, the first and second inspection data may use different inspection methods.

[0038] In this embodiment, each inspection device 90 has a photographic unit when any of the following first to fifth photographic methods are applied. Figure 2 is a table explaining features related to various photographic methods. Figure 3 is a schematic diagram illustrating the positional relationship of the photographic unit 95 for various photographic methods. The photographic unit 95 is configured with an illumination 91 and a camera 92. An explanation related to the configuration examples of the illumination 91 and camera 92 of the photographic unit 95 will be given later.

[0039] (First photography method) As shown in Figures 3(A) and 2, the first photographic method involves photographing light transmitted vertically within a transparent film F8. The camera 92 and illumination 91 are positioned facing each other, spanning the object being photographed (film F8). Furthermore, the illumination direction of illumination 91 is aligned with the optical axis of the camera 92. As shown in Figure 2, in the first photographic method, when the object being photographed has a defect, the transmitted light is obstructed by the defect, causing a change in the amount of light received. The object being inspected is a transparent film. Hereinafter, the first photographic method will also be labeled "Transmission 1".

[0040] (Second photography method) As shown in Figures 3(B) and 2, the second photographic method involves photographing transmitted light reflected or scattered within the transparent film F8. The camera 92 and illumination 91 are positioned facing each other, spanning the subject (referring to film F8, hereinafter the same). Furthermore, the illumination direction of illumination 91 is offset from the optical axis of camera 92. In the second photographic method, when a foreign object is present within the subject, the scattered light caused by that foreign object is received. The subject being inspected is a transparent film. Hereinafter, the second photographic method will also be labeled "Transmission 2".

[0041] (Third photography method) As shown in Figures 3(C) and 2, the third photographic method involves photographing reflected light that undergoes positive reflection on the surface of the transparent film F8 or the opaque laminate 80B. The camera 92 and illumination 91 are positioned on the same side of the object being photographed. Furthermore, the angle of incidence of the illumination direction of the illumination 91 relative to the surface of the object being photographed is the same as the angle of incidence of the optical axis of the camera 92. The third photographic method utilizes the phase difference of the arrival distance, which depends on the surface shape of the object being photographed, for detection. The objects being inspected are the transparent film and the opaque film (laminate 80B). Hereinafter, the third photographic method will also be referred to as "Reflection 1".

[0042] (Fourth photography method) As shown in Figures 3(D) and 2, the fourth photographic method involves photographing reflected light scattered on the surface of the transparent film F8 or the surface of the opaque laminate 80B. The camera 92 and the illumination 91 are positioned on the same side of the object being photographed. Furthermore, the angle of incidence of the illumination 91 relative to the surface of the object and the angle of incidence of the optical axis of the camera 92 are offset and inconsistently configured. The fourth photographic method receives scattered light caused by foreign objects on the surface of the object. The objects being inspected are transparent and opaque films. Hereinafter, the fourth photographic method will also be referred to as "Reflection 2".

[0043] (Fifth photography method) As shown in Figures 3(E) and (F), the fifth photographic method is photography using polarized light. Hereinafter, the fifth photographic method will also be referred to as "polarized light". The configuration of illumination 91 and camera 92 can also be applied to any of the first to fourth methods. Photography using polarized light is used to check the polarization state. Photography using polarized light can use a polarizing camera that photographs linearly polarized light, or illumination that illuminates linearly polarized light (called a polarizing light source). Alternatively, in the fifth photographic method using polarized light, as shown, a polarizing plate can be used, or this can be combined with a polarizing camera and a polarizing light source. The fifth photographic method is used, for example, for checking a film for a polarizing plate. A film for a polarizing plate includes a film (polarizer) having polarizing properties, and a film used to manufacture this polarizer (hereinafter also referred to as a raw material film). The raw material film is required to have non-polarizing properties.

[0044] Figure 3(E) shows a photographic method using polarized light for the raw material film or protective film of a polarizer. The raw material film is a film that imparts polarization properties in a post-processing step, and the protective film is a film used to laminate onto the polarizer (polarizing film) in a post-processing step. In the photographic method shown in Figure 3(E), unpolarized light is output from illumination 91 (the same applies to Figure 3(F)). A polarizing plate 99a is directly positioned between illumination 91 and the film F8 (raw material film) being inspected, and a polarizing plate 99b is positioned linearly between film F8 and camera 92. The linear polarizing plates 99a and 99b are arranged in a cross-niche configuration. That is, when viewed from the Z direction (the direction perpendicular to the film surface), the absorption axes of the two linear polarizing plates 99a and 99b intersect at a predetermined angle that is approximately orthogonal to each other. For example, the absorption axes intersect approximately orthogonally around 90° (89~91°). With this configuration, when film F8 is normal, the linearly polarized light irradiated by linear polarizer 99a and transmitted through film F8 is difficult to transmit through linear polarizer 99b. On the other hand, when film F8 has defects (feature points), the light transmitted through the defective portion passes through linear polarizer 99b and is therefore observed as a bright spot.

[0045] Figure 3(F) shows a photographic method for using polarized light on a polarizer (also known as a polarizing film) that has been given polarization properties. In the photographic method shown in Figure 3(F), the linear polarizer 99b is configured such that the absorption axis of the film F8 and the absorption axis of the linear polarizer 99b intersect approximately orthogonally. Unpolarized light is output from the illumination 91 to the film F8 (polarizer). Light that has passed through the normal area of ​​the film F8 has difficulty passing through the linear polarizer 99b, but light that has passed through the defective areas passes through the linear polarizer 99b and is therefore observed as a bright spot.

[0046] (A variation of the first photographic method) As shown in Figure 3(G), a variation of the first photographic method (hereinafter also referred to as the first b photographic method) is a photographic method that uses a blade to block a portion of the illumination light. In the first b photographic method, transmitted light that travels directly within the transparent film F8 and is reflected or scattered is photographed. The configuration of the camera 92 and the illumination 91 is the same as in the first photographic method, with the illumination direction of the illumination 91 aligned with the optical axis of the camera 92. Furthermore, in the first b photographic method, a light-blocking member 915 with a blade is positioned near the subject on the illumination 91 side. The light-blocking member 915 is a flat plate, configured so that its front end is on the optical axis, blocking illumination light from one side of the optical axis. In the example shown in Figure 3(G), half of the light on the paper surface, which is to the left of the optical axis, is blocked. Due to variations in the thickness of the subject, the refractive index changes, and the angle at which the light arrives changes. The subject being inspected is a transparent film. The first b photographic method can be modified as appropriate to use the first photographic method.

[0047] Figure 4 is a table illustrating the correspondence between photographic units 1 to 6 and photographic modes in this embodiment. Photographic units 1 to 6 correspond to photographic modes 1 to 5, respectively.

[0048] Figures 5A and 5B are schematic diagrams illustrating examples of the configuration of the inspection device 90. In Figures 5A and 5B, among the first to fifth imaging methods, the fourth imaging method (reflection 2) is shown as a representative example. Furthermore, although Figure 5A shows an example of the inspection device 90 configured with a single imaging unit 95, as will be described later, the inspection device 90 may sometimes include multiple imaging units 95 in cases where multiple imaging methods are applied.

[0049] Figure 5A is a schematic diagram illustrating the configuration of the reflective inspection device 90 when viewed from the width direction (X direction). Figure 5B is a schematic diagram illustrating the configuration of the inspection device 90 when viewed from the transport direction (Y direction). The inspection device 90 includes an illumination 91, a camera 92 as an optical sensor, an image analysis unit 93 as a data processing unit, and a memory unit 94. The inspection device 90 optically inspects feature points (hereinafter referred to as defects) generated on the film F8 during transport. The camera 92 of the inspection device 90 optically inspects the film F8 of the film roll 80 and generates image data as inspection data. The image data includes not only still images but also animation data composed of time-series continuous still images. Furthermore, the inspection data can also be configured to directly extract feature points from signal data without image data digitization. Camera 92 is configured such that the entire width of the film F8 is the inspection area (photographic range), with the number of cameras, the field of view, and the distance from the film surface set. The number of cameras is chosen to allow for multiple cameras to be arranged in the width direction when it is not possible to adequately photograph the entire width of the film with a single camera. Figure 5B shows an example of two cameras 92 arranged in the width direction (X direction). The image resolution unit 93 can also combine multiple images obtained by continuous photography from one camera 92 to generate a single image containing the entire film surface of the film roll 80. Alternatively, the image resolution unit 93 can establish a correspondence with the photography time and store multiple image data in the memory unit 94. Furthermore, the same image data obtained by multiple cameras 92 arranged in the width direction can also be combined. The image analysis unit 93, by referring to the memorized transport speed (winding speed or delivery speed) and the corresponding photographic time established with the image data, determines the position of the film F8 along its long side. Hereinafter, it will be explained as follows: For each film roll 80, multiple image data obtained through continuous photography are established and memorized with corresponding photographic times. The image analysis unit 93 analyzes the image data to generate defect information. The inspection device 90 inspects for defects generated during manufacturing processes such as the winding of the long film F8.

[0050] Illumination 91 illuminates the inspection area of ​​membrane F8. Illumination 91 is the illumination of light uniformly along the width direction of the rolled membrane F8 (a direction orthogonal to the long side of membrane F8 and parallel to the membrane surface). Here, uniform means that the illuminance of membrane F8 is approximately the same along the width direction of membrane F8 (the difference between the maximum and minimum values ​​is less than a predetermined value, etc.).

[0051] Camera 92 is an optical sensor that optically reads the inspection area of ​​film F8. Camera 92 includes an imaging element such as a CCD (Charge Coupled Device) or CMOS (Complementary Metal Oxide Semiconductor), and a lens. Camera 92 is an area sensor that generates 2D image data from the output signals of each imaging element. Camera 92 detects diffused light reflected from the light illuminated by illumination 91 and reflected from the inspection area of ​​film F8. Here, camera 92 can be either a color camera or a monochrome camera.

[0052] One or more cameras 92 have a total imaging range across the width direction of the membrane F8, and can simultaneously read the entire width direction of the membrane F8 in a single photograph. The camera 92 can be used to detect light in the visible light region or light in the infrared region. Furthermore, in the fifth imaging mode, a polarizing camera (a general-purpose camera) with a polarizing filter is used.

[0053] Furthermore, regarding the output signal of the camera 92, the contrast between the signal value corresponding to the irradiated portion of the film F8 illuminated by the light source 91 and the signal value corresponding to the non-irradiated portion illuminated by the light source 91 is expected to be above a predetermined value. That is, it is expected that only the areas of the film F8 illuminated by the light source 91 (irradiated portions) appear bright.

[0054] In terms of contrast, it is expressed as the difference or ratio of two values ​​of the processed object (here, the signal value corresponding to the illuminated part and the signal value corresponding to the unilluminated part). The greater the difference between the two values, the greater the contrast. To increase the contrast between the illuminated part and the unilluminated part, it is desirable to use a powerful and highly linear illumination 91.

[0055] Here, "powerful" means that the illuminance E50 at an illumination distance of 50mm is 50,000 lx or more. Furthermore, "high straightness" means that when the illuminance is E50 at an illumination distance of 50mm and E100 at an illumination distance of 100mm, (E50-E100) / E50 < 0.5.

[0056] The image analysis unit 93, which consists of a CPU, RAM, etc., reads various processing programs stored in the memory unit 94 and expands them into RAM, and performs various processing in cooperation with the programs.

[0057] The memory unit 94, composed of HDD, SSD (Solid State Drive), etc., stores various processing programs and the data required for their execution. Furthermore, the memory unit 94 associates the captured image data (inspection data) with the time of capture. The memory unit 94 stores the winding speed of the film roll manufacturing apparatus 1000 (e.g., 100 m / min) or the delivery conditions of the film F8 in the product manufacturing apparatus 2000 (e.g., 30 m / min). These winding speeds and delivery conditions can also be included in the inspection checklist for the DB (refer to Table T13 in Figure 9).

[0058] The image analysis unit 93 processes the output signal from the camera 92 (optical sensor) to detect feature points (position, intensity) such as defects in the film F8. The data processing includes: image processing, which processes image data obtained from the output signal from the camera 92; defect determination processing, which determines defects based on the processed image data; and quantitative evaluation processing, which quantitatively evaluates defects based on the processed image data.

[0059] (Inspection Equipment List) Figure 6 is a table illustrating the list of inspection devices. As shown in Table T01 of Inspection Device List 1, inspection devices a to f each have one or more inspectors. For example, inspection device b has photographic unit 2 and photographic unit 3, and inspection device c has photographic unit 1 and photographic unit 4. Photographic units 1 to 6 are shown in Figure 4 above. For example, photographic unit 1 performs photography using the first photographic method shown in Figure 3(A). Inspection devices a to f correspond to any one of the inspection devices 90 (90a1, 90a2, 90b1, etc.) shown in Figure 1. Furthermore, the combination of photographic units in the inspection devices (inspection devices 90) shown in Table T01 is exemplified, but other combinations of photographic units may also be present.

[0060] Furthermore, as shown in Table T01, when an inspection device includes multiple imaging units, the inspection data (first-order data) obtained from the multiple imaging units is combined into a single inspection data set with added positional information between the imaging units. Hereinafter, the inspection data before combination will also be referred to as first-order data, and the inspection data combining multiple first-order data sets will be referred to as combined inspection data. Combined inspection data is equivalent to the first inspection data or the second inspection data. The positional information between these multiple imaging units, as shown in Table T01, is recorded as "positional information between inspectors." Although the multiple imaging units are set to photograph the same inspection area of ​​the object such as film F8 in the wide-side direction (X direction), they may sometimes be slightly offset in the long-side direction (Y direction). This slight offset in the Y direction is aligned using this "positional information between inspectors."

[0061] Table T02 of the inspection device list 2 shows the combinations of inspection devices corresponding to the first inspection data and the second inspection data. The selection of the inspection device for the first inspection data and the second inspection data can be made by the user, as described later (button b1 in Figure 15, described later).

[0062] In this embodiment, the inspection data of one of the first inspection data and the second inspection data is inspection data obtained using one or more inspection methods. The inspection data of the other party is inspection data obtained using one or more inspection methods that include inspection methods different from those of the inspection data of one party. Alternatively, the inspection data of one party is inspection data obtained using multiple inspection methods. For example, in Example 1 of Table T02, the first inspection data uses inspection data obtained from inspection device a, and the second inspection data uses inspection data obtained from inspection device c. In this case, the first inspection data of one party is inspection data obtained using one inspection method (first inspection method), and the second inspection data of the other party is inspection data obtained using two inspection methods (first and fourth inspection methods). Examples 2 and 3 are the same.

[0063] Up to this point, the inspection device 90 has been described. The following description focuses on the inspector who, as a combination of Example 1 in Table T02, obtains the first and second inspection data (within the dashed box of Table T02).

[0064] (Information Processing System 50) The information processing system 50 will now be described with reference to Figures 7 to 9. Figure 7 is a block diagram illustrating the schematic configuration of the information processing system 50. The information processing system 50 is, for example, a server. As shown in Figure 7, the information processing system 50 includes a control unit 51, a memory unit 52, and a communication unit 53.

[0065] (Control Department 51) The control unit 51 includes a CPU and memory such as RAM and ROM. The CPU is a control circuit composed of a multi-core processor that executes the control and various arithmetic operations of the above-mentioned parts according to a program. The various functions of the information processing system 50 are performed by the CPU and its corresponding program.

[0066] The control unit 51, in cooperation with the communication unit 53, functions as an acquisition unit 511 and a receiving unit 512. Furthermore, the control unit 51 functions as a comparison unit 513, a parsing unit 514, an extraction unit 515, and a display output unit 516. The acquisition unit 511 acquires the first and second inspection data obtained from inspections in the first and second manufacturing processes. The receiving unit 512 receives the user's selection of the inspection method or the inspection device (photographic unit). The comparison unit 513 extracts feature points from each of the first and second inspection data. The comparison unit 513 uses comparison processing to explore corresponding feature points between the first and second inspection data. In the comparison processing, the comparison unit 513 generates descriptors for feature points (descriptor 2 described later) for the two inspection data (image data), and uses these feature point descriptors to perform feature point matching processing between the inspection data, outputting the comparison results (shown in the corresponding point list in Figure 19 described later). The analysis unit 514 calculates indicators showing the relationship (hereinafter referred to as relationship information) using a list of corresponding points. The relationship information includes a scatter plot and statistical information such as mean, regression line, dispersion, standard deviation, and correlation coefficient. The extraction unit 515 extracts (classifies) the first to third type of feature points using the comparison results. The display output unit 516 sends the feature point extraction results and the display data of feature points based on the data in one pass to the terminal device 70, or displays them on a display unit (not shown), depending on the requirements of the terminal device 70.

[0067] (Memory Department 52) The memory unit 52 is a large-capacity auxiliary memory device for storing various programs and data of the operating system. Storage devices include, for example, hard disks, solid-state drives, flash memory, and ROM. The memory unit 52 stores user lists, batch lists, inspection data databases (DBs), and inspection device lists. The management and registration of the user lists and batch lists are recorded by the terminal device 70, depending on the access of the administrator. For example, this administrator might be the person in charge of the department responsible for operating factory A. Furthermore, the inspection device list is as shown in Figure 6 above.

[0068] (User List) Figure 8 shows an example of various data stored in the memory unit 52. Table T11 shown in Figure 8 is an example of a user list. The user list stores user ID, username, contact information, etc. In addition, each user is assigned access rights to the database (DB) and is granted access rights to various data (examination data, extraction data, etc.) related to the membrane roll 80 (identified by batch ID) associated with the user.

[0069] (Batch List) Table T12 in Figure 8 is an example of a batch list. The batch list records the batch ID assigned to each roll of film, the product name (also known as the type), the target user ID (the client), and multiple manufacturing conditions, dimensions (width, length, thickness), manufacturing date, etc.

[0070] (Check data DB) Figure 9 shows an example of the inspection database (DB) stored in the memory unit 52. The inspection database stores information related to the inspection of various film rolls 80, such as the first and second inspection data shown in Figure 9, and the results of feature point extraction. As described above, the first inspection data is the data obtained during the inspection in the first manufacturing process. The second inspection data is the data obtained during the inspection in the second manufacturing process. The feature point extraction results are data generated using the information processing system 50 using these first and second inspection data.

[0071] Table T13 in Figure 9 is an example of an inspection checklist logged in the inspection data DB. The inspection checklist stores the inspection ID, batch ID, inspection device ID, inspection data, inspection date and time, etc. The inspection device ID corresponds to the inspection device name in Table T01 of Figure 6. An example of the inspection data included in the inspection checklist will be described later.

[0072] (Communications Department 53) The communications section 53 is also an interface for network connection with external devices such as PCs.

[0073] (The generation and processing of the data for the first and second checks) Hereinafter, referring to Figures 10 to 13, the generation process of the first and second inspection data performed in the first and second manufacturing processes will be described. Figure 10 is a flowchart illustrating the generation process of the first inspection data performed in the first manufacturing process.

[0074] As described above, the following explanation will take the case of Example 1 of Table T02 in Figure 6 as an example. That is, the first inspection data is obtained using inspection device a, and the second inspection data is obtained using inspection device c. In this case, the first inspection data is the inspection data obtained using the first inspection method (first photographic method). The second inspection data is the inspection data obtained using two of the first and fourth inspection methods (first and second photographic methods).

[0075] (First inspection data generation and processing) (Step S11) In the typical example described above, in the first manufacturing process, a film roll 80 is manufactured using a film roll manufacturing apparatus 1000.

[0076] (Step S12) The inspection device 90 photographs the membrane and saves the image data. The inspection device 90 is as described in Figures 3 to 6.

[0077] (Step S13) When the inspection device 90 has multiple imaging units, step S15 is performed; when there is only one imaging unit, step S14 is performed. Here, the inspection device a has one imaging unit 1 (through 1), so step S14 is performed.

[0078] (Step S14) The image analysis unit 93 performs image processing on the image data as described below, generating a plurality of feature points.

[0079] (Image processing for feature point generation) The image analysis unit 93 acquires the 2D image data generated by the camera 92 and stored in the memory unit 94.

[0080] The image analysis unit 93 processes the image data (inspection data) acquired from the camera 92.

[0081] The image analysis unit 93 divides the image data into a plurality of regions. For example, the image analysis unit 93 divides the image data into n (e.g., several to dozens) regions in the width direction (hereinafter referred to as region a1 to region an).

[0082] Next, the image analysis unit 93 acquires image data of region a1 and performs mathematical processing on the image data of region a1. Depending on the type of defect of the object being inspected (strip-shaped protrusion, longitudinal wrinkle, oblique wrinkle, etc.), appropriate mathematical processing is prepared.

[0083] In terms of mathematical processing, it includes preprocessing, emphasis processing, signal processing, and image feature extraction.

[0084] Regarding preprocessing, the following are examples. Image cropping, • Low-pass filter, high-pass filter, Gaussian filter, median filter, bilateral filter, morphology transformation, color transformation (L*a*b*, sRGB, HSV, HSL), contrast adjustment, noise removal, restoration of blurred images, masking, Hough transform, projective transformation, etc.

[0085] The text emphasizes processing techniques, citing examples such as the Sobel filter, Scharr filter, Laplacian filter, Gabor filter, and Canny method.

[0086] Regarding signal processing, the following are examples. • Calculate basic statistics (maximum, minimum, mean, median, standard deviation, dispersion, quartiles), sum and square root, difference, sum, product, ratio, distance matrix processing, differential and integral calculus, threshold processing (binarization, adaptive binarization, etc.), Fourier transform, wavelet transform, peak detection (peak value, peak count, half-width, etc.), etc.

[0087] Regarding image feature extraction, template matching and SIFT features are listed.

[0088] Next, the image analysis unit 93 performs thresholding on the values ​​(feature quantities) obtained through mathematical processing of the image data for region a1. The thresholding process involves determining whether a defect is a target for detection based on a predetermined threshold, and further determining the level (intensity) of the defect.

[0089] In threshold processing, determining the existence and type of a defect is equivalent to "defect determination processing." Furthermore, in threshold processing, classifying defects into multiple ranks based on thresholds is equivalent to "quantitative evaluation processing."

[0090] For example, for parameters (features) with values ​​from 1 to 100, defects can be classified into multiple levels. For instance, levels can be classified according to defect size (diameter, area, etc.). Furthermore, levels classified by parameter value can be further subdivided.

[0091] The image analysis unit 93 processes areas other than region a1 in the same way.

[0092] After processing each region a1 to an, the image analysis unit 93 integrates the results for each region a1 to an, and the data processing ends. Specifically, the image analysis unit 93 generates corresponding data based on the level and location (xy coordinates) of the detected defects according to the region (based on the position in the width direction of the membrane F8).

[0093] After data processing, the image analysis unit 93 stores the processing results in the memory unit 94. The image analysis unit 93 performs data processing on each of the plurality of image data obtained from the examination of one film roll 80 to obtain processing results. These processing results are then combined to generate the examination data as shown in Figure 11.

[0094] Figure 11 illustrates an example of the contents of the inspection data (inspection ID: i0101) in the inspection checklist. The inspection data automatically assigns feature point IDs with consecutive numbers to each feature point, and feature point descriptors 1 and 2 (hereinafter referred to only as descriptor 1, etc.) based on the feature point IDs. Furthermore, for each feature point in the inspection data, information is provided indicating which camera unit was used to obtain the data (the line for "camera unit"). In the example of Figure 11, one set of data obtained using one camera unit 1 was used; therefore, camera unit 1 is described for all feature points.

[0095] Descriptor 1 is individual information for a feature point, recording its XY coordinate position and intensity. Intensity refers to the level of the feature point, as described later. Furthermore, as intensity information, it may also include information on the size (diameter, area) and brightness of the feature point. The XY coordinate position is the XY coordinate relative to the starting point of the membrane surface (e.g., the left end of the front end). X is the coordinate in the width direction of the membrane, which can be in the range of 0 to 3000 mm, depending on the membrane size (refer to Table T12). Y is the coordinate in the long side direction of the membrane, which can be in the range of 0 to 10000 μm, depending on the membrane size. Descriptor 1 is generated by the image resolution unit 93 of the inspection device 90.

[0096] Descriptor 2 is the surrounding information, which includes vectors, arrangement information, etc., representing information about the surrounding environment such as relationships with other feature points. For example, SIFT features can be used as descriptors, or the probability density function of feature points calculated by kernel density estimation can be used as a descriptor. The generation of this descriptor 2 is mainly performed using the comparison unit 513.

[0097] (Steps S15~S16) When there are multiple photography units, proceed with steps S15 and S16. This process is the same as steps S25 and S26 in Figure 12, which will be described later. The details of this process will be explained in the following description of steps S25 and S26.

[0098] (Step S17) In the first manufacturing process, the terminal device 70 sends inspection data, including information on a plurality of feature points obtained from the processing up to step S14 or S16, to the information processing system 50. The acquisition unit 511 of the information processing system 50 stores the acquired inspection data as first inspection data in the inspection data DB of the memory unit 52. The inspection data shown in Figure 11 is an example of such first inspection data.

[0099] (Second inspection data generation and processing) Figure 12 is a flowchart illustrating the generation process of the second inspection data performed in the second manufacturing process.

[0100] (Step S21) In the second manufacturing process, for example after the first manufacturing process, the product manufacturing apparatus 2000 performs post-processing in the case of using membrane roll 80 to manufacture the product in the case of using membrane F8.

[0101] (Step S22) The inspection device 90 photographs the surface of the membrane F8 before post-processing, or the membrane F8 or laminate 80B during or after post-processing, and saves the image data.

[0102] (Step S23) When the inspection device 90 has multiple imaging units, step S25 is performed; when there is only one imaging unit, step S24 is performed. Here, the inspection device c, as an example, has two imaging units 1 (transmitting 1) and 4 (reflecting 2), so step S15 is performed.

[0103] (Step S24) If the inspection device is configured with a single imaging unit (imaging unit 25), this step S24 is performed. The process here is the same as step S14, and the explanation is omitted.

[0104] (Step S25) The image analysis unit 93, following the same process as step 14 or S24, generates a plurality of feature points using the inspection data (image data) obtained from the individual imaging units, thus generating one set of inspection data. In this case, corresponding to imaging units 1 and 4 (2), two sets of one set of data are generated. The structure of each set of one set of data is the same as that of the inspection data shown in FIG11, consisting of feature point ID and feature point descriptor 1 (XY coordinates, size).

[0105] (Step S26) The image analysis unit 93 combines multiple single-data points into a single inspection data point based on position information and inter-inspector position information (refer to Table T01). The processing in steps S25 and S26 is the same as that in steps S15 and S16.

[0106] (Step S27) In the second manufacturing process, the terminal device 70 sends inspection data containing information on a plurality of feature points obtained from the processing up to step S24 or step S26 to the information processing system 50. The acquisition unit 511 of the information processing system 50 stores the acquired inspection data as the second inspection data in the inspection data DB of the memory unit 52.

[0107] Figure 13 illustrates an example of the contents of the second inspection data (inspection ID: i0102) generated and saved in this manner. In the example shown in Figure 13, the "Photography Section" row of the inspection data describes which photography section (photography method) was used in Photography Section 1 and Photography Section 4.

[0108] (Feature point extraction process) The feature point extraction process performed in the information processing system 50 will be described below with reference to Figures 14 to 24C. Figure 14 is a flowchart illustrating the feature point extraction process. Figure 15 is an example of an operation screen 701 displayed on the terminal device 70. Figure 16 is a schematic diagram illustrating the feature point extraction process.

[0109] Information processing system 50 initiates the processing from step S31 onwards based on a start instruction from the user (via the operation screen in the case of terminal device 70) or when the second inspection data is registered in the inspection data DB of memory unit 52 and both the first and second inspection data are complete. On operation screen 701 of Figure 15, the user, after selecting a batch, selects the first and second inspection data from a plurality of inspection data associated with this batch using button b1. The second inspection data is the inspection data obtained in a process downstream of the first inspection data. In the example of Figure 15, it is shown that, depending on the selection, the first inspection data is obtained using inspection device a, and the second inspection data is obtained using inspection device c. The user operates the parsing start button b0. Accordingly, control unit 51 begins feature point extraction processing.

[0110] (Step S31) Section 511 obtains the same batch from inspection data DB, that is, obtains a pair of first and second inspection data.

[0111] (Steps S32, S33) The comparison unit 513 performs preprocessing on the first inspection data under the first condition and performs preprocessing on the second inspection data under the second condition.

[0112] As shown in FIG16, the comparison unit 513 performs preprocessing for the second condition. If there is a difference between the winding (first manufacturing process) and the output (second manufacturing process) of the second inspection data, the Y-coordinate (vertical) is reversed to ensure consistency. Furthermore, in the second manufacturing process, based on information set regarding whether the imaging area of ​​the camera 92 is the surface or back of the film F8, the comparison unit 513 performs preprocessing to reverse the X-coordinate (left-right) for the second inspection data (or the first inspection data).

[0113] Furthermore, the comparison unit 513 performs at least one of the following as noise removal processing included in the first and second conditions for the first and second inspection data. (1) Removal of low-level feature points. (2) Remove extremely small feature points. (3) Remove consecutive dots. (4) Remove concentrated dots along the wide side. This is especially common at the very front and rear ends of membrane F8.

[0114] (Step S34) The comparison unit 513 uses comparison processing to explore feature points that are the same as or correspond to feature points in the inspection data of one side from the inspection data of the other side, and establish correspondence (matching) between feature points. Figure 17 is a subroutine flowchart illustrating the processing of this step S34.

[0115] (Step S401) The comparison unit 513 generates descriptors 2 for each feature point of the first and second examination data. The comparison unit 513 may use SIFT feature quantities as descriptors, or use the probability density function of the feature points calculated by kernel density estimation as descriptors.

[0116] (Steps S402, S403) The comparison unit 513 compares the feature points of the first and second inspection data with each other and retrieves the most corresponding point. The comparison unit 513 evaluates the similarity between feature points using descriptor 1 and descriptor 2, and regards the most similar point as the corresponding feature point.

[0117] For example, when comparing feature points of the first and second inspection data, the comparison unit 513 explores feature points corresponding to the feature points of the object in the second inspection data from among the feature points of the first inspection data. In this case, the comparison unit 513 determines the feature points of the first inspection data that are consistent with or have the closest intensity to the X and Y coordinates of the feature points of the second inspection data as identical points. Alternatively, the feature points whose distance (Euclidean distance) between the X and Y coordinates of the descriptor 1 is closest to each other are determined as identical points (corresponding points). The comparison unit 513 excludes feature points that are separated by a distance greater than the set threshold from the determination of corresponding points. In addition, in this embodiment, the threshold is set to a value that is two to three digits larger than the threshold in the X direction and the threshold in the Y direction. For example, the threshold in the X direction is several millimeters, and the threshold in the Y direction is several meters. The reason for setting the threshold in the Y direction to be two or three digits larger than the threshold in the X direction is that the amount of variation is greater.

[0118] Furthermore, the comparison unit 513 can also be configured to use descriptor 2 together with descriptor 1 to extract the most similar feature points from each other based on the distance between vectors in the high-dimensional vector space. In this case, the comparison unit 513, as descriptor 2, can also use the probability density function calculated using kernel density estimation as described above. Figure 18 shows an example of the probability density function that displays the position and intensity (density) of feature points calculated using kernel density estimation. In Figure 18, the horizontal and vertical axes are XY coordinates, and it is shown that the higher the concentration, the higher the density.

[0119] The comparison unit 513 determines that each feature point corresponds to only one feature point. For example, for feature points in the second inspection data, the feature points whose descriptor vectors in the first inspection data are closest to each other are recorded in the corresponding point list.

[0120] Figure 19 shows an example of a list of corresponding points stored in the memory unit 52. The list of corresponding points is created by establishing correspondences between the most similar feature points in the first inspection data for each feature point in the second inspection data. Furthermore, the list of corresponding points describes the X and Y coordinates of the feature points in the second inspection data, the Euclidean distance between the corresponding feature points, the difference dx on the X coordinate, and the difference dy on the Y coordinate.

[0121] (Step S404) The analysis unit 514 generates correlation information between the feature points existing in both the first and second checked data. That is, the analysis unit 514, corresponding to the feature points in the point list that have corresponding feature points, generates a scatter plot and statistical information with the horizontal axis as X or Y and the vertical axis as dx or dy as the data displaying correlation information. The statistical information includes a regression line and a correlation coefficient. Following this, the control unit 51 terminates the processing of Figure 17 and returns to the processing of Figure 14, proceeding to step S35 and below.

[0122] (Step S35) The extraction unit 515 categorizes feature points from the corresponding list, the first inspection data, and the second inspection data into three types of feature points, from the first type to the third type. The first to third types of feature points will be described later. Specifically, the extraction unit 515 categorizes feature points in the corresponding list and the second inspection data that do not have a corresponding feature point in the first inspection data into the second type of feature points. Furthermore, the extraction unit 515 categorizes feature points in the first inspection data that are not included in the corresponding list (feature points that do not have a corresponding feature point in the second inspection data) into the first type of feature points. Additionally, the extraction unit 515 categorizes feature points in the integrated inspection data that are classified into the third type of feature points into feature points for each of the multiple primary data sets.

[0123] (Feature points 1 through 3) Figure 20 is a table used to illustrate the first to third types of feature points.

[0124] (Type 1 feature point) The first type of feature point is a feature point that exists in the first inspection data but does not exist in the second inspection data. The second type of feature point is a feature point that disappears in the second manufacturing process (e.g., coating process, lamination process, etc.). This first type of feature point is a feature point that does not require management in the first manufacturing process. In this case, the manufacturing conditions that contribute to the occurrence of the first type of feature point can be subject to specification deregulation in the first manufacturing process.

[0125] (Second type of feature point) The second type of feature point is a feature point that does not exist in the first inspection data but exists in the second inspection data. This second type of feature point is a newly generated feature point in the second manufacturing process. Since this second type of feature point is a feature point that originates from the second manufacturing process, it is effectively applied to the improvement of the second manufacturing process.

[0126] (Third type of feature point) The third type of feature point exists in both the first and second inspection data. This third type of feature point is the cause of the first manufacturing process and is also a feature point that needs to be managed. Because this third type of feature point is the cause of the first manufacturing process, it is effectively applied to the improvement of the first manufacturing process.

[0127] (Step S36) The control unit 51 outputs feature point extraction information extracted during the processing up to step S35. The output of the feature point extraction information is either registered in the inspection data DB by the control unit 51 or sent to the terminal device 70 by the display output unit 516 for display on its display.

[0128] Figure 21A shows an example of the extracted data (hereinafter referred to simply as extracted data). In the extracted data, the inspection IDs of the first and second inspection data as sources and the extraction results based on feature points are recorded. The extraction results (types 1-3) are classified as shown in Figure 20. The composite feature point ID is automatically assigned a consecutive number, corresponding to the feature points of one or both of the first and second inspection data, generating composite feature points. The number of composite feature point IDs is ≥ the number of feature point IDs of the first and second inspection data.

[0129] Figure 21B shows an example of DB inspection corresponding to Figure 13. In Figure 21B, although some of the descriptions of feature points 1 and 2 are omitted, they are the same as in Figure 13. In the second inspection data shown in Figure 21B, according to the processing in step S35, the extraction results of the second or third type of feature point are recorded in the rightmost column ("Extraction Result"). In addition, regarding the first inspection data, when multiple single data are obtained using multiple imaging units, the extraction results of the first or third type of feature point are also recorded in the same "Extraction Result" column.

[0130] Additionally, sometimes, for a single batch ID, multiple first and second inspection data are generated using multiple inspection devices. For example, in the second manufacturing process, the original roll of film 80 is inspected (photographed), and simultaneously, several downstream processes are inspected, thereby generating multiple second inspection data. In this case, the information processing system 50 can also be configured to generate multiple feature point extraction result data (one-to-many) between multiple second inspection data for a single first inspection data. Furthermore, establishing a correspondence with any second inspection data can also be made selectable by the user (e.g., button b1 in Figure 15 above).

[0131] (Steps S37 and S38) Control unit 51 determines whether at least one of the first and second inspection data is integrated inspection data. For example, the second inspection data is generated from the individual inspection data of the camera unit when using inspection device c (refer to FIG. 6) with two inspection methods, and these data are integrated (step S26 in FIG. 12). In this case, control unit 51 causes the process to proceed to step S38, displaying the feature point information. FIG. 22 is a subroutine flowchart illustrating the process for step S38.

[0132] (Steps S501, S502) The receiving unit 512 displays the user's selection of single-data or integrated data. Figure 23 shows an operation screen 702 for selection on the terminal device 70. The operation screen 702 shown in Figure 23 is displayed following the operation screen 701. On the operation screen 702, the user can select the data to be checked displayed on the preview screen using button b2. In the example of operation screen 702, the status of selecting the second set of data to be checked is displayed. The type of feature point and its display content can be selected.

[0133] The user can use button b3 to select which of the two camera sections to use. Selecting both camera sections results in data that is essentially the same as the combined data. Furthermore, button b4 can be used to select the type of feature points to be displayed and the type of preview graphic. For example, in the example shown on operation screen 702, camera section 4 and the third type of feature point were selected in the second data check. The feature point IDs extracted from the second data check shown in Figure 21B, with camera section "camera section 4" and extraction result "third type", are then displayed in the preview display area b10 according to the extracted data and the type of graphic selected by button b2.

[0134] Figure 24A shows an example of a scatter plot previewed according to the settings of operation screen 702. Furthermore, when both camera units 1 and 4 are selected using button b2, the data from both units are plotted on the scatter plot using patterns distinguishable by color separation, etc. In Figure 24A, the horizontal axis represents the X-coordinate of each feature point on the strip, and the vertical axis represents the Y-coordinate. The area in the figure (0.1 to 0.5) indicates the size (area) of the feature points. For example, a circle with an area of ​​0.1, corresponding to the size of the small circle shown in the figure, is plotted in the graphic area. Figure 24A (and Figure 24B are the same) shows the feature points obtained using the data from two camera units 1 and 4 in one pass, constituting the second inspection data selected by button b3 in Figure 23. Figure 24B shows an example of a histogram previewed according to the settings of operation screen 702, corresponding to the process in step S502. In Figure 24B, the vertical axis shows the frequency of occurrence of feature points according to the camera module, and the horizontal axis represents the area of ​​the feature points. Figure 24B shows the frequency of occurrence of two types of data in a single dataset. Optical A and B in the figure correspond to the camera module, respectively. For example, optical A is the single dataset obtained using camera module 1, and optical B is the single dataset obtained using camera module 4.

[0135] In this embodiment, the first feature point information of the strip in the first inspection data and the second feature point information of the strip in the second inspection data are compared. Based on the comparison result, a third type of feature point existing in both the first and second inspection data is extracted, and / or a first type of feature point existing in the first inspection data but not in the second inspection data, or a second type of feature point existing in the second inspection data but not in the first inspection data, is extracted. Furthermore, the inspection data of one of the first and second inspection data is obtained using one or more inspection methods, and the inspection data of the other is obtained using one or more inspection methods that include inspection methods different from those of one of the inspection data. This allows for the efficient collection of information that contributes to process improvement and shipment specification setting during both the manufacturing of the strip and the subsequent processing using this information.

[0136] Furthermore, the inspection data of at least one of the first inspection data or the second inspection data is inspection data that integrates feature point information extracted from multiple primary data obtained using multiple types of inspection methods in the same inspection area, and aligns their positions on the strip surface. Accordingly, appropriate inspection data can be obtained by utilizing one or more appropriate inspection methods corresponding to the state of the strip in each process.

[0137] Furthermore, in this embodiment, after extracting the third type of feature point as shown in Figure 22, the feature points of the aggregated inspection data are classified into feature points for each of the multiple primary data sets. This allows for the identification, for example, of feature points that are either highly reliable or low-reliability. For example, in the case of the third type of feature point, there is a high probability that it is an actual generated feature point. Conversely, in the case of the second type of feature point, which is a feature point generated (detected) by only one of the two inspection methods, there is a low probability of low reliability and high noise. For example, in the second inspection data for inspecting opaque strips, only reflection 1 or reflection 2 imaging methods can be used. In such cases, although there is often a lot of noise and a low signal-to-noise ratio, the user can more accurately identify the third type of feature point as a feature point even if it is detected as a feature point in the first inspection data.

[0138] The configuration of the information processing system 50 described above has been explained in terms of its main components when describing the features of the above-described embodiments. However, the system is not limited to the above configuration, and various modifications can be made within the scope of the patent application. Furthermore, this does not exclude components found in general information processing devices / systems.

[0139] Furthermore, the information processing system 50 may also include an inspection device 90 configured in the first manufacturing process and / or the second manufacturing process. Additionally, the control unit 51 of the information processing system 50 may perform the function of generating feature points for the image analysis unit 93 of the inspection device 90. In this case, image data of the film surface photographed by the inspection device 90, along with its photographic conditions (transport speed, camera orientation, angle of view, etc.), are sent to the information processing system 50, and the feature point generation process is performed on the control unit 51 side.

[0140] Furthermore, the means and methods for performing various processes in the information processing system 50 associated with the above-described embodiments can also be implemented using either dedicated hardware circuitry or a programmed computer. The aforementioned program can be provided, for example, using computer-readable recording media such as USB memory or DVD (Digital Versatile Disc)-ROM, or it can be provided online via a network such as the Internet. In this case, the program recorded on the computer-readable recording media is typically transferred and stored in a memory unit such as a hard drive. Furthermore, the aforementioned program can be provided as standalone application software or incorporated into the software of the device as a function of the device.

[0141] Although embodiments of the present invention have been described and illustrated in detail, the disclosed embodiments are for illustrative purposes only and are not intended to be limiting. The scope of the present invention should be interpreted in accordance with the wording of the claims.

[0142] This application is incorporated herein by reference, in its entirety, to the disclosure of Japanese Special Application filed on April 10, 2024 (Special Application No. 2024-063185).

[0143] 50: Information Processing System 51: Control Department 511: Acquisition Department 512: Receiving Unit 513: Comparison Section 514: Analysis Section 515: Extraction Department 516: Display Output Section 52: Memory Department 90, 90a1~90a2, 90b1~b4: Inspection device 91:Lighting 92: Camera 95: Photography Department 1000: Membrane roll manufacturing equipment 2000: Product manufacturing equipment

Claims

1. A method for extracting feature points of a strip material, comprising: step (a), which is to obtain first inspection data in a first manufacturing process of manufacturing the strip material or performing post-processing on the manufactured strip material; step (b), which is to obtain second inspection data in a second manufacturing process of performing post-processing treatment when using the strip material after the first manufacturing process; step (c), which is to compare the first feature point information of the strip material in the first inspection data with the second feature point information of the strip material in the second inspection data; and step (d), which is to extract, based on the comparison result of step (c), a third type of feature point existing in both the first and second inspection data, and / or a first type of feature point existing in the first inspection data but not in the second inspection data, or a second type of feature point existing in the second inspection data but not in the first inspection data; The inspection data of one of the aforementioned first inspection data and second inspection data is inspection data obtained using one or more inspection methods, and the inspection data of the other party is inspection data obtained using one or more inspection methods that include inspection methods different from those of the aforementioned inspection data; the aforementioned first inspection data and the aforementioned second inspection data are inspection data obtained by photographing the strip with an inspection device; the photographing method of the aforementioned other party's inspection data in the inspection device is inspection data obtained using one or more photographing methods that include photographing methods different from those of the aforementioned inspection data in the inspection device; the positional relationship between the lighting and photographing unit and the strip in the photographing method of the inspection device in the aforementioned first inspection data is different from the positional relationship between the lighting and photographing unit and the strip in the photographing method of the inspection device in the aforementioned other inspection data.

2. As in request item 1, the method for extracting feature points, wherein, The photographic method used in obtaining the inspection data of the aforementioned party includes at least one of the following, depending on the positional relationship between the lighting and the photographic unit: a first photographic method that photographs transmitted light traveling straight through the strip; a second photographic method that photographs transmitted light reflected within the strip; a third photographic method that photographs reflected light orthogonally reflected from the surface of the strip; and a fourth photographic method that photographs reflected light scattered from the surface of the strip; The photographic method for obtaining the inspection data of the aforementioned other party includes a photographic method that is different from the photographic method for obtaining the inspection data of the aforementioned party among the first to fourth photographic methods.

3. The feature point extraction method as described in request item 1, wherein, The aforementioned photographic methods include photographic methods using polarized light to detect the polarization state of light.

4. The feature point extraction method as described in request item 1, wherein, The inspection data of at least one of the aforementioned first inspection data or second inspection data is inspection data that integrates feature point information extracted from multiple primary data obtained using multiple types of inspection methods in the same inspection area by aligning their positions on the strip surface.

5. The feature point extraction method as described in request item 4, wherein, It further includes: step (e), which is to extract the third type of feature points using the aforementioned aggregated inspection data in the aforementioned step (d), and then classify the feature points of the aforementioned aggregated inspection data into feature points of each of the plurality of primary data.

6. The feature point extraction method as described in request item 5, wherein, It includes: step (f), which is the selector of the aforementioned inspection method for the inspection data of the aforementioned party; and step (g), which is the displayer of the feature point information of the primary data classified in the aforementioned step (e) corresponding to the selected inspection method, based on the selection in step (f).

7. A method for extracting feature points of a strip, comprising: step (a) obtaining first inspection data in a first manufacturing process of manufacturing the strip or performing post-processing on the manufactured strip; step (b) obtaining second inspection data in a second manufacturing process of post-processing performed after the first manufacturing process, where the strip is used; step (c) comparing the first feature point information of the strip in the first inspection data with the second feature point information of the strip in the second inspection data; and step (d) extracting a third type of feature point existing in both the first and second inspection data based on the comparison result of step (c); wherein at least one of the first and second inspection data is inspection data that integrates feature point information extracted from multiple primary data obtained in the same inspection area using multiple types of inspection methods, by aligning their positions on the surface of the strip. It further includes: step (e), which involves using the aforementioned consolidated inspection data, after extracting the aforementioned third type of feature points in step (d), classifying the feature points of the aforementioned consolidated inspection data into feature points of each of the plurality of primary data.

8. The feature point extraction method as described in request item 7, wherein, The aforementioned first inspection data and the aforementioned second inspection data are inspection data obtained by photographing the strip material with an inspection device. The inspection device for the aforementioned inspection data is inspection data obtained by using multiple photographic methods.

9. The feature point extraction method as described in request item 7, wherein, It includes: step (f), which is the selector of the aforementioned inspection method for the inspection data of the aforementioned party; and step (g), which is the displayer of the feature point information of the primary data classified in the aforementioned step (e) corresponding to the selected inspection method, based on the selection in step (f).

10. A control program for causing a computer to perform an extraction method such as any one of requests 1 to 9.

11. An information processing system comprising: an acquisition unit that acquires first inspection data in a first manufacturing process of manufacturing a strip or performing post-processing on the manufactured strip, and acquires second inspection data in a second manufacturing process of performing post-processing when using the strip after the first manufacturing process; a comparison unit that compares first feature point information of the strip in the first inspection data with second feature point information of the strip in the second inspection data; and an extraction unit that, based on the comparison result of the comparison unit, extracts a third type of feature point existing in both the first and second inspection data, and / or a first type of feature point existing in the first inspection data but not in the second inspection data, or a second type of feature point existing in the second inspection data but not in the first inspection data; The inspection data of one of the aforementioned first inspection data and second inspection data is inspection data obtained using one or more inspection methods, and the inspection data of the other party is inspection data obtained using one or more inspection methods that include inspection methods different from those of the aforementioned inspection data; the aforementioned first inspection data and the aforementioned second inspection data are inspection data obtained by photographing the strip with an inspection device; the photographing method of the aforementioned other party's inspection data in the inspection device is inspection data obtained using one or more photographing methods that include photographing methods different from those of the aforementioned inspection data in the inspection device; the positional relationship between the lighting and photographing unit and the strip in the photographing method of the inspection device in the aforementioned first inspection data is different from the positional relationship between the lighting and photographing unit and the strip in the photographing method of the inspection device in the aforementioned other inspection data.

12. An information processing system comprising: an acquisition unit that acquires first inspection data in a first manufacturing process of manufacturing a strip or performing post-processing on the manufactured strip, and acquires second inspection data in a second manufacturing process of performing post-processing on the strip after the first manufacturing process; a comparison unit that compares first feature point information of the strip in the first inspection data with second feature point information of the strip in the second inspection data; and an extraction unit that extracts a third type of feature point existing in both the first and second inspection data based on the comparison result of the comparison unit; wherein at least one of the first and second inspection data is inspection data that integrates feature point information extracted from each of a plurality of primary data obtained in the same inspection area using a plurality of different inspection methods, by aligning their positions on the surface of the strip. The aforementioned extraction unit, after extracting the aforementioned third type of feature points from the aggregated inspection data, classifies the feature points of the aggregated inspection data into feature points for each of the plurality of primary data.

Images