Information processing system, feature point extraction method, and control program

Abstract

The information processing system includes: a comparison unit (513) that compares first feature point information of the web in first inspection data with second feature point information of the web in second inspection data, and extracts feature points that are the same in the first inspection data and the second inspection data; an analysis unit (514) that generates relational information indicating a relationship between a difference value of coordinates of each of the feature points extracted by the comparison unit in the first inspection data and the second inspection data and a coordinate position in the entire web; and an output unit (516) that outputs the relational information.

Description

Technical Field

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

[0002] Liquid crystal displays (LCDs) have begun to be used in large-screen televisions and large monitors, which has led to a demand for wider films used in the display surfaces of LCDs. For example, there is a need for wide films with a width of 2000 mm or more. In addition, in order to anticipate substrate loss (film loss) or reduce transportation costs, it is also necessary to manufacture film rolls with a winding length of 1000 m or more, and further 3000 m or more.

[0003] In the post-processing stages of manufacturing products based on film and other roll materials, when quality defects occur, it is necessary to distinguish whether the defect originated in the previous process (i.e., during the manufacturing of the roll material and was already present on it) or a defect arose in the post-processing stage. When it is not clearly possible to distinguish whether the defect originated in the post-processing stage, improvements to the previous process are sometimes required. In attempting to respond to these improvement requests, the shipping specifications of the previous process become unnecessarily stringent, resulting in over-spec. Furthermore, sometimes, even though the impact on the subsequent processes is not clearly known, to prevent quality defects, estimates are made and the shipping specifications during roll material manufacturing are set too strictly. This also leads to over-spec. Over-spec leads to decreased yield and increased costs, which is undesirable from an environmental perspective and is also undesirable for both those working in the previous and post-processing stages.

[0004] Although it is necessary to distinguish whether the defect is pre-existing on the roll material or generated in the subsequent processing, it is still necessary to establish a correspondence between the coordinate system on the roll material in the previous process and the coordinate system in the subsequent process.

[0005] Patent Document 1 discloses a technique for synchronizing the inspection results of a first inspection device and a second inspection device when performing inspection using the first and second inspection devices in an optical thin film inspection system. In this inspection system, the synchronization of the two inspection results is achieved by using the distance between the first inspection device and the second inspection device located at the rear end of the transport direction, as well as an encoder signal indicating the amount of film movement.

[0006] Patent Document 1: Japanese Patent Application Publication No. 2016-161576

[0007] The technology in Patent Document 1 obtains synchronization of inspection results based on the mutual distance and movement distance between the inspection devices, but synchronization cannot be obtained when the film stretches or contracts. Furthermore, synchronization cannot be obtained when an offline process of winding and unwinding the film is inserted between the two inspection devices. Summary of the Invention

[0008] The present invention was made in view of the above circumstances, and its object is to efficiently collect information that helps improve processes and set shipment specifications in both the manufacturing of roll materials and the manufacturing process using them for subsequent processing.

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

[0010] (1) An information processing system, wherein:

[0011] The acquisition unit acquires first inspection data in a first manufacturing process of manufacturing rolled materials or performing post-processing on the manufactured rolled materials, and second inspection data in a second manufacturing process after the first manufacturing process in which post-processing treatment using the rolled materials is performed.

[0012] The comparison unit compares the first feature point information of the roll material in the first inspection data with the second feature point information of the roll material in the second inspection data, and extracts the same feature points in the first inspection data and the second inspection data.

[0013] The analysis unit generates relational information, which represents the relationship between the difference values ​​of the coordinates of the multiple feature points extracted by the comparison unit in the first inspection data and the second inspection data, and their coordinate positions in the entire roll material; and

[0014] The output section outputs the aforementioned relational information.

[0015] (2) According to the information processing system described in (1) above, the relational information is a chart showing the relationship between the difference value of the coordinates and the coordinate position in the whole.

[0016] (3) The information processing system described in (1) or (2) above, wherein the relational information is statistical information representing the relationship between the difference value of the coordinates and the coordinate position in the whole.

[0017] (4) According to the information processing system described in (1) above, wherein,

[0018] The aforementioned information processing system also includes a receiving unit, which receives the coordinate correction values ​​of the first inspection data and / or the second inspection data.

[0019] After the comparison unit corrects the coordinates of the feature points using the correction amount received by the receiving unit, it compares the first feature point information of the roll material in the first inspection data with the second feature point information of the roll material in the second inspection data, and extracts the same feature points in the first inspection data and the second inspection data.

[0020] (5) The information processing system described in (4) above, wherein,

[0021] After the output unit displays the relational information, the receiving unit accepts the correction amount.

[0022] In response to the receiving unit accepting the correction amount, the comparison unit extracts the same feature points again, and the parsing unit generates the relational information again.

[0023] The output section then displays the aforementioned relational information again.

[0024] (6) According to the information processing system described in (1) above, wherein,

[0025] The aforementioned relational information consists of a scatter plot and a regression equation showing the relationship between the difference values ​​of the coordinates and their coordinate positions within the whole.

[0026] The aforementioned information processing system also includes an extraction unit, which classifies the feature points based on the distance from the feature points in the scatter plot to the regression equation.

[0027] (7) According to the information processing system described in (1) above, wherein,

[0028] The aforementioned relational information includes the variance of the difference values ​​or the correlation coefficient representing the relationship between the difference values ​​of the coordinates and their coordinate positions in the whole.

[0029] If the absolute value of the correlation coefficient is below a first predetermined threshold or if the variance is above a second predetermined threshold, the analysis unit generates an alarm, and the output unit reports the alarm.

[0030] (8) A method for extracting feature points, which is a method for extracting feature points of roll materials, wherein it comprises:

[0031] Step (a): Obtain first inspection data in the first manufacturing process of manufacturing the roll material or performing post-processing on the manufactured roll material;

[0032] Step (b) is to obtain the second inspection data in the second manufacturing process that follows the first manufacturing process and involves post-processing of the roll material.

[0033] Step (c) compare the first feature point information of the roll material in the first inspection data with the second feature point information of the roll material in the second inspection data.

[0034] Step (d): Based on the comparison results of step (c) above, extract the same feature points in the first inspection data and the second inspection data above;

[0035] Step (e) generates relational information, which represents the relationship between the difference values ​​of the coordinates of the multiple feature points extracted in step (d) in the first inspection data and the second inspection data, and their coordinate positions in the entire roll material; and

[0036] Step (f) outputs the relational information generated in step (e) above.

[0037] (9) According to the feature point extraction method described in (7) above, the relational information is a graph showing the relationship between the difference value of the coordinates and the coordinate position in the whole.

[0038] (10) A control program for causing a computer to perform the extraction method described in (7) or (8) above.

[0039] The information processing system of the present invention includes: an acquisition unit that acquires first inspection data in a first manufacturing process of manufacturing a rolled material or performing post-processing on the manufactured rolled material, and second inspection data in a second manufacturing process after the first manufacturing process in which post-processing using the rolled material is performed. Furthermore, the information processing system includes: a comparison unit that compares first feature point information of the rolled material in the first inspection data with second feature point information of the rolled material in the second inspection data, and extracts identical feature points in both the first and second inspection data; a parsing unit that generates relational information, which represents the relationship between the difference values ​​of the coordinates of each of the multiple feature points extracted by the comparison unit in the first and second inspection data and their coordinate positions in the entire rolled material; and an output unit that outputs the relational information. Thus, information conducive to process improvement and shipment specification setting can be efficiently collected in both the manufacturing process of the rolled material and the manufacturing process in which it is used for post-processing. Attached Figure Description

[0040] The advantages and features provided by one or more embodiments of the present invention will be more fully understood through the following detailed description and accompanying drawings, but these are for illustrative purposes only and are not intended to limit the invention.

[0041] Figure 1 This is a schematic diagram showing a thin film roll manufacturing line.

[0042] Figure 2 yes Figure 1 A top-view diagram of the extended processes, drying processes, and finishing processes of the manufacturing line.

[0043] Figure 3A It is a schematic diagram showing the structure of the inspection device.

[0044] Figure 3B It is a schematic diagram showing the structure of the inspection device.

[0045] Figure 3C It is a schematic diagram showing the structure of the inspection device.

[0046] Figure 4 This is a schematic diagram illustrating an application example of the information processing system involved in this embodiment.

[0047] Figure 5 This is a table used to illustrate the first to third types of feature points extracted.

[0048] Figure 6 It is a block diagram that represents the general structure of an information processing system.

[0049] Figure 7 These are examples of various types of data stored in the storage department.

[0050] Figure 8 This is an example of a check data database stored in the storage department.

[0051] Figure 9 This is a flowchart representing the generation process of the first inspection data in the first manufacturing process.

[0052] Figure 10 This is a flowchart illustrating the generation process of second inspection data during the second manufacturing process.

[0053] Figure 11 This is a flowchart representing the feature point extraction process performed in an information processing system.

[0054] Figure 12 This is an example of a screen that accepts scaling.

[0055] Figure 13 This is a schematic diagram illustrating the feature point extraction process.

[0056] Figure 14 This is a flowchart of the subroutine for comparison processing 1 in step S35.

[0057] Figure 15 This is an example of a probability density function representing the location and intensity of a feature point, calculated through kernel density estimation.

[0058] Figure 16 This is an example of a list of corresponding points.

[0059] Figure 17A This is an example of a distribution diagram for relational information.

[0060] Figure 17B This is an example of a distribution diagram for relational information.

[0061] Figure 17C This is an example of a distribution diagram for relational information.

[0062] Figure 17D This is an example of a distribution diagram for relational information.

[0063] Figure 18 This is an example of a distribution diagram for relational information.

[0064] Figure 19 This is an example of an operation screen for receiving calibration values.

[0065] Figure 20 This is a flowchart of the subroutine for comparison processing 2 in step S50.

[0066] Figure 21 This is an example of a distribution map after applying corrections.

[0067] Figure 22 This is an example of feature point extraction conditions.

[0068] Figure 23 This is a diagram illustrating the classification based on the distance to the regression equation.

[0069] Figure 24 This is a flowchart illustrating the feature point extraction process in the second embodiment.

[0070] Figure 25 This is a diagram showing an example of a distribution map and alarm display. Detailed Implementation

[0071] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, the scope of the present invention is not limited to the disclosed embodiments. Furthermore, in the description of the drawings, the same reference numerals are used for the same elements, and repeated descriptions are omitted. Additionally, the dimensions of the drawings are exaggerated for ease of explanation and sometimes differ from the actual scale.

[0072] In this embodiment, the roll material is a sheet material comprising a resin film and a metal film. In the following description, the roll material will be described as a long strip of resin film, a film roll, and an article processed thereon. The processing also includes processes such as applying a coating liquid and overlapping another film-like material.

[0073] First, refer to Figure 1 , Figure 2 The manufacturing line for film roll 80 is described. Figure 1 This is a schematic diagram showing a thin film roll manufacturing line. Figure 1 , Figure 2 The manufacturing line shown illustrates film formation by solution casting, but is not limited to this; film formation can also be achieved by melt extrusion.

[0074] like Figure 1 As shown, the film roll manufacturing apparatus 1000 includes a casting section 01, a first drying section 02, an extension section 03, a trimming section 05, a second drying section 06, and a winding section 07 (also called a winding device).

[0075] In the film roll manufacturing apparatus 1000, an inspection device 90 is arranged between multiple processes. The inspection device 90 detects the conveying status of the film F8 by photographing the surface of the film F8 being conveyed in the manufacturing line as the monitoring area and analyzing the obtained frame images. The structure of the inspection device 90 will be described later. Figure 3A (etc.). In Figure 1 In the diagram, hollow arrows indicate the direction of transport for materials such as thin films.

[0076] The casting section 01 has a mirror-finished strip-shaped metal casting belt (hereinafter referred to as the belt) 01a with an endless support body that runs in an endless manner (in the direction of the arrow in the figure), and a die 01b for casting the raw liquid, which is made by dissolving resin in a solvent, onto the belt 01a. In addition, in order to stabilize the raw liquid film flowing out of the die 01b, a depressurization chamber (not shown) may be provided on the upstream side of the die 01b relative to the conveying direction of the belt, and a pressurization chamber (not shown) may be provided on the downstream side.

[0077] The peeling roller 01d peels off the cast film 01c formed on the tape 01a, generating an unstretched film F8.

[0078] The first drying section 02 (first drying process) has a drying box 02a with a drying air inlet 02b and an outlet 02c, and a set of upper and lower conveying rollers 02d for conveying the film F8.

[0079] In the first drying section 02, the amount of solvent contained in the unstretched film F8 before entering the stretching section 03 (stretching process) can be adjusted. This adjustment of the solvent amount is performed by changing the drying temperature and conveying speed in the first drying section 02 through the control device 08.

[0080] (Extension 03, Repair 05)

[0081] Figure 2 yes Figure 1A top-view diagram of the extension process (also known as the tenter frame process), drying process, and finishing process of the manufacturing line. Figure 2 , Figure 3A In this design, the vertical direction is designated as the Z-direction, the conveying direction of film F8 is designated as the Y-direction, and the direction orthogonal to the conveying direction and the width direction of film F8 is designated as the X-direction. The Y-direction and the conveying direction are also referred to as MD (Machine Direction). Furthermore, the X-direction and the width direction are also referred to as TD (Transverse Direction) or the left-right direction. In addition, the Y-direction is also referred to as the long side direction, and as a counterpart to the long side direction, the width direction is also marked as the short side direction.

[0082] The extension section 03 has an MD extension section 03a and a TD extension section 03b. The extension section 03 extends the unextended film F8 conveyed from the first drying section 02. Figure 2 As shown, the extension 03 clamps both ends of the heated film F8 by a plurality of left and right pairs of clamping parts 301. The extension 03 includes clamping parts 301, chain, drive part, shut-off device, opener device, etc. (except for clamping parts 301, which are not shown in the figure).

[0083] In the MD extension section 03a (MD extension process) located on the upstream side, it moves while accelerating as it is conveyed downstream, thereby gradually increasing its speed in the conveying direction (in Figure 2 (Represented by solid arrows in the middle). Thus, in the MD stretching process, the unstretched film F8 is stretched and conveyed while being stretched in the MD direction.

[0084] In the next section, TD extension 03b (TD extension process), film F8 and the like are conveyed by clamping parts 301. Clamping parts 301 are, for example, clamps, with each clamping part 301 connected to an endless chain. The chain, which is suspended from a sprocket, is rotated by a drive unit, thereby moving the clamping parts 301 in the conveying direction. A closing device is located at the upstream end of the TD extension process in extension 03, and an opening device is located at the downstream end. Clamping parts 301 are sequentially closed when they reach the closing device and sequentially released when they reach the opening device. The clamping parts 301 in the closed state clamp the side ends of film F8 and convey them. In this TD extension 03b (TD extension process), the pair of left and right clamping parts 301 clamping film F8 and the like gradually move outward in the width direction while moving in the conveying direction (solid arrow). Therefore, in the TD stretching process, the distance (gap) between a pair of clamping parts 301 in the width direction gradually increases, and the film F8 and the like held by them are stretched and transported in the TD direction.

[0085] Alternatively, an oblique stretching process other than the TD stretching process can be applied in this stretching process. The TD stretching process is a process in which the film is stretched in a direction parallel to the width direction. That is, in the TD stretching process, the film F8 is stretched at an angle of 0° with respect to the width direction. On the other hand, in the oblique stretching process, in order to adjust the optical properties of the film, the two ends of the heated film are held by multiple grippers and stretched in an oblique direction at an angle in the range of 40 to 50° with respect to the width direction of the film (called "oblique stretching"). Furthermore, in this case, the cooling process may include a heat-relieving process for stress relief heat treatment while the stretched film is being continuously transported.

[0086] The trimming section 05 (trimming process) includes a cutting section 05a and a collection section 05b. The cutting section 05a includes two edge trimmers 5110L and R on the left and right sides, and multiple rollers 5120, etc. The edge trimmers 5110L and R are, for example, rotatably pivoted circular or disc-shaped blades, which are rotated by a drive unit (not shown). Regarding the rotation of the edge trimmers 5110L and R, they can be driven to rotate at approximately the same speed in the same direction as the conveying direction of the film F8 at the cutting position, or they can be driven to rotate in the opposite direction. Through the two edge trimmers 5110L and R, the film F8 after the cutting position is separated into the central portion of the film F8 and trimmed films F801 and F802 (also called ears or ears) as the side ends on both sides.

[0087] The central portion of the film F8 serves as the product area, and its width is, for example, in the range of 1000mm to 2500mm. The widths of the trimming films F801 and F802 (also referred to as trimming widths) are both around tens to hundreds of millimeters (e.g., 170mm). The central portion of the film F8 is conveyed downstream in the conveying direction and supplied to the subsequent winding process. On the other hand, the trimming films F801 and F802 are conveyed by changing their orientation downwards by 90 degrees via roller 5120 and are recovered by the subsequent recovery unit 05b. The recovery unit 05b has a rotary cutter and a suction pipe (neither shown). By means of the rotary cutter 522, the conveyed trimming films F801 and F802 are cut into small fragments of a few millimeters in size and recovered into the dust collection box in the downstream sweeper. Furthermore, if the inspection device 90 is arranged before the trimming process, the trimming width can also be input during pre-treatment (described later). Figure 11 In step S32), a coordinate transformation corresponding to the trim width is performed from the captured data.

[0088] The central portion of the film F8 is conveyed to the downstream winding section 07. An embossing process can also be performed between the trimming section 05 and the winding section 07. In the embossing process, embossing is formed at both ends of the trimmed film F8.

[0089] The second drying section 06 (second drying process) has the same basic structure as the first drying section 02, so the description is omitted.

[0090] (Wrapping section 07)

[0091] The winding section 07 includes: a winding machine 07a that winds a film F8 with embossed ends in the trimming section 05, an accompanying air volume control device 07b, a contact or non-contact linear encoder 07c for detecting the travel speed of the extended film F8, a winding shaft speed measuring machine 07d, a tension control device 07e, and a thickness measuring device 6f.

[0092] like Figure 1 As shown, the casting section 01 discharges the raw liquid through the die head 01b onto the endlessly circulating belt 01a. This raw liquid is prepared by dissolving the resin of the raw material in a solvent and adding various additives such as plasticizers, ultraviolet absorbers, anti-deterioration agents, lubricants, and peeling accelerators as needed. Next, the cast film formed on the belt 01a is peeled off from the belt after the solvent is removed to a certain extent on the endless support. Then, it is passed through various conveying units to the drying section and the extension section 03, and the two ends are trimmed. After appropriate embossing, it is wound onto the winding shaft by the winding section 07, thereby producing an optical thin film.

[0093] Figure 1 , Figure 2 The material of the manufactured film F8 is not particularly limited, but generally, examples include polycarbonate resin, polysulfone resin, acrylic resin, polyolefin resin, cyclic olefin resin, polyether resin, polyester resin, polyamide resin, polysulfide resin, unsaturated polyester resin, epoxy resin, melamine resin, phenolic resin, diallyl phthalate resin, polyimide resin, polyurethane resin, polyvinyl acetate resin, polyvinyl alcohol resin, styrene resin, cellulose acetate resin, and polyvinyl chloride resin. Furthermore, considering factors such as productivity and quality, the width of the film F8 is preferably 1000 mm to 3200 mm. Considering factors such as quality and operability, the thickness is preferably 15 μm to 500 μm.

[0094] Considering productivity, winding quality, etc., it is wound onto winding shaft 82 (refer to...). Figure 1 The length of the optical film F8 of the film roll 80 (also called the winding shaft or core) is preferably 2000m to 8000m. The winding length is a value calculated from the speed and time. The winding speed is, for example, 100m / min.

[0095] The inspection device 90 takes pictures of the thin film, etc., and generates image data. Figure 1In the example shown, multiple inspection devices 90 are arranged before and after each process in the thin film manufacturing line, including the stretching process. Figure 1 In this process, the inspection device 90 is configured to photograph the surface of the film F8 immediately after the casting process (casting section 01), immediately after the drying process (first drying section 02), and immediately after the finishing process (finishing section 05). Hereinafter, refer to... Figures 3A-3C The structure of the inspection device 90 will be described.

[0096] (Inspection device 90)

[0097] Using transparent materials such as thin film F8 as the object to be inspected, and as a device to detect irregularities on its surface, bubbles, cracks, strain in the internal structure, etc., there are the following methods.

[0098] (1) A transmissive inspection device that irradiates the object being inspected with light and receives light that has passed through the object being inspected, thereby detecting defects in the object being inspected.

[0099] (2) A reflective inspection device that detects defects in an object by receiving light reflected from the object being inspected.

[0100] Furthermore, in both transmittance and reflectance types, depending on the positional relationship between the camera's optical axis, the light source, and the inspected object, there are bright-field inspection devices that receive non-scattered light from the surface and dark-field inspection devices that receive scattered light. In a bright-field inspection device, when there are no defects, light from the light source is incident on the light detection unit unobstructed because there is no light scattering. When a defect exists, the light is blocked by the defect and does not incident on the light detection unit. Therefore, the defect is observed as a dark spot or dark stripe against a bright background. In contrast, in a dark-field inspection device, when there are no defects, light is not scattered and does not incident on the light detection unit. However, if a defect exists, the light is scattered by the defect and thus incident on the light detection unit. Therefore, the defect is observed as a bright spot or stripe against a dark background. As the inspection device 90 in this embodiment, any type of inspection device can be used. The first inspection data and the second inspection data are preferably acquired by the same type of inspection device, but are not limited to this.

[0101] (Reflective inspection device)

[0102] Figure 3A This is a schematic diagram of the structure of the reflective inspection device 90 as viewed from the width direction (X direction). Figure 3BThis is a schematic diagram of the structure of the inspection device 90 as viewed from the transport direction (Y direction). The inspection device 90 includes a light source 91, a camera 92 as an optical sensor, an image analysis unit 93 as a data processing unit, and a storage unit 94. The inspection device 90 is an apparatus for optically inspecting feature points (hereinafter also referred to as defects) generated in the film F8 during transport. The inspection device 90 uses the camera 92 to optically inspect 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 dynamic image data composed of continuous still images in a time series. In addition, as inspection data, feature points can be extracted directly from signal data without image digitization. The number of cameras 92, the viewing angle, and the distance to the film surface are set such that the entire width direction of the film F8 becomes the inspection area (shooting range). Regarding the number of cameras, this is to arrange multiple cameras in the width direction when one camera cannot properly capture the full width of the film. Figure 3B This diagram illustrates a configuration where two cameras 92 are arranged as an example in the width direction (X direction). The image analysis unit 93 can combine multiple images obtained from continuous shooting by one camera 92 to generate a single image data containing the entire film surface of the film roll 80, or it can store multiple image data in the storage unit 94 in association with the shooting time. Furthermore, it can combine the same image data obtained from multiple cameras 92 arranged in the width direction. By referring to the stored transport speed (winding speed or delivery speed), the image analysis unit 93 can determine the position of the film F8 in the long side direction based on the shooting time corresponding to the image data. Hereinafter, we will describe the storage of multiple image data obtained from continuous shooting in association with the shooting time for one film roll 80. The image analysis unit 93 generates defect information by analyzing the image data. The inspection device 90 takes defects generated during manufacturing processes such as the winding of the long strip of film F8 as the inspection target.

[0103] Light source 91 illuminates the inspection area of ​​film F8. Light source 91 illuminates the rolled film F8 uniformly in the width direction (orthogonal to the long side of film F8 and parallel to the film surface). Here, uniform means that the illuminance on film F8 is approximately the same throughout the width direction of film F8 (the difference between the maximum and minimum values ​​is below a specified value, etc.).

[0104] Camera 92 is an optical sensor that optically reads the inspection area of ​​thin film F8. Camera 92 includes an imaging element such as a CCD (Charge-Coupled Device) or CMOS (Complementary Metal Oxide Semiconductor), a lens, etc. Camera 92 is an area array sensor that generates two-dimensional image data based on the output signals of each imaging element. Camera 92 detects diffuse light in the light reflected from the inspection area of ​​thin film F8 by light source 91. Here, either a color camera or a black-and-white camera (monochrome camera) can be used as camera 92.

[0105] One or more cameras 92 have a shooting range covering the entire width direction of the thin film F8, and simultaneously read the entire width direction of the thin film F8 in a single shot. The camera 92 can be a camera that detects light in the visible light region or a camera that detects light in the infrared region.

[0106] Furthermore, in the output signal of the camera 92, it is desirable that the contrast between the signal value corresponding to the illuminated portion of the thin film F8 that is illuminated by the light source 91 and the signal value corresponding to the non-illuminated portion that is not illuminated by the light source 91 is at least a predetermined value. That is, it is desirable that only the area (illuminated portion) on the thin film F8 illuminated by the light from the light source 91 appears bright.

[0107] Contrast ratio is represented by the difference or ratio of two values ​​of the processed object (in this case, the signal value corresponding to the irradiated part and the signal value corresponding to the non-irradiated part). The greater the difference between the two values, the greater the contrast ratio. In order to increase the contrast ratio between the irradiated part and the non-irradiated part, it is desirable to use a powerful light source 91 with high directivity.

[0108] Here, "powerful" means that when the illuminance at an illumination distance of 50mm is set to E50, the illuminance E50 is 50,000 lx or more. In addition, "high direct penetration" means that when the illuminance at an illumination distance of 50mm is set to E50 and the illuminance at an illumination distance of 100mm is set to E100, (E50-E100) / E50 < 0.5.

[0109] The image analysis unit 93 consists of a CPU, RAM, etc. It reads various processing programs stored in the storage unit 94 and expands them into RAM, and performs various processing in cooperation with the program.

[0110] The storage unit 94, composed of HDD, SSD (Solid State Drive), etc., stores various processing programs and data required to execute those programs. Furthermore, the storage unit 94 stores captured image data (inspection data) in association with the capture time. The storage unit 94 stores the winding speed (e.g., 100 m / min) in the film roll manufacturing apparatus 1000, or the unwinding conditions (e.g., 30 m / min) of the film F8 in the product manufacturing apparatus 2000. These winding speeds and unwinding conditions can be included in the inspection list of the inspection DB (see [reference]). Figure 8 Table T3).

[0111] The image analysis unit 93 detects feature points (position, intensity) such as defects in the thin film F8 by processing the output signal of the camera 92 (optical sensor). The data processing includes: image processing of image data obtained from the output signal of the camera 92; defect determination processing to determine defects based on the image-processed data; and quantitative evaluation processing to quantitatively evaluate defects based on the image-processed data.

[0112] (Relative position of camera 92 and light source 91)

[0113] The camera 92 can also be positioned to receive orthogonal reflected light from the light source 91 (in the case of a bright-field inspection mode that receives non-scattered light).

[0114] Furthermore, the camera 92 can also be positioned to avoid receiving direct reflections of light from the light source 91 (in the case of dark-field inspection methods that receive diffused light). That is, it is preferable to position it to receive diffused light reflected from the object being inspected.

[0115] (Through-type inspection device)

[0116] Figure 3C This is an example of a transmission-type inspection device 90. A transmission-type inspection device 90 can also be used, in which the light source 91 is positioned opposite the camera 92 through the thin film F8.

[0117] Figure 4 This is a schematic diagram illustrating an application example of the information processing system 50 according to this embodiment. For example... Figure 4 As shown, the information processing system 50 communicates with terminal devices 70 in factories A and B via a network. The network is a data communication network or other communication line. In some networks, wired LANs or wireless LANs (e.g., LANs conforming to the IEEE 802.11 standard) may also be used.

[0118] Terminal device 70 is, for example, 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.

[0119] Factory A is equipped with the aforementioned film roll manufacturing apparatus 1000. Factory A is operated or managed, for example, by a film manufacturer. The first manufacturing process of manufacturing film roll 80 is carried out in factory A. The film surface of film roll 80 is inspected by inspection device 90.

[0120] Factory B is equipped with a product manufacturing facility 2000. Factory B is operated or managed by, for example, a coating manufacturer (hereinafter also referred to as a user company or user). There are multiple user companies operating various factories B. In factory B, products are manufactured using film rolls 80 shipped and transported from factory A. In factory B, post-processing, namely the first or second manufacturing process, is carried out by unwinding film (film F8 described below) from film rolls 80 and coating it, or by overlapping or bonding it with other films.

[0121] In a typical example of this embodiment, the process of manufacturing the film (roll) is referred to as the first manufacturing process, and the post-processing performed on the manufactured film at factory B is referred to as the second manufacturing process. For example, in the post-processing process, a stretching process is performed, or a stretching process and a coating process for imparting a functional layer to the surface are performed. In the second manufacturing process, the film surface of the film roll 80 is also inspected by the inspection device 90.

[0122] As an atypical example of this embodiment, when the manufactured film undergoes multiple post-processing steps at factory B, the earlier post-processing step is sometimes referred to as the first manufacturing step, and the subsequent post-processing step is sometimes referred to as the second manufacturing step. Furthermore, sometimes the film roll 80 manufactured at factory A is unwound and post-processed in a subsequent step. In this case, the process of manufacturing the film roll 80 is sometimes referred to as the first manufacturing step, and the subsequent post-processing step performed at the same factory A is sometimes referred to as the second manufacturing step. The second manufacturing step performed at factory A includes stretching and drying steps.

[0123] (Overview of feature point extraction and processing)

[0124] Details regarding feature point extraction will be described later; please refer to the following: Figure 4 The outline of the feature point extraction process will be explained below. In the following, as a typical example, the process of manufacturing the film (roll) will be described as the first manufacturing process, and the post-processing performed on the manufactured film at factory B will be described as the second manufacturing process.

[0125] In the first manufacturing process of factory A, during product inspection, inspection device 90 performs optical inspection on the film and generates inspection data (also known as fault data or defect data). Specifically, by analyzing the image data obtained from photographing the film surface, inspection data containing feature point information (feature point, position, intensity) is generated (hereinafter referred to as first inspection data). Information processing system 50 obtains the first inspection data from terminal device 70 of factory A (step S1).

[0126] Here, feature points refer to defects on the thin film, generated by analyzing image data. Image analysis can be performed using known techniques, extracting pixels whose pixel values ​​deviate from the surrounding average by more than a specified margin (difference greater than a specified margin) from image data obtained from photographing the thin film surface, or it can be calculated using the "image processing for feature point generation" method described later. From one or more images of a thin film roll 80 (several km in length), tens to thousands of feature points are often generated. Defects include both malfunctions reaching the level of product defects and minor malfunctions not reaching the level of product defects. Feature points include defects related to poor adhesion or axial unevenness during the bonding of thin films together midway (e.g., ultrasonic welding). Feature point information includes size and position (xy coordinates). Furthermore, as feature point information, information that combines multiple adjacent feature points into one (clustered) can also be used. Image processing for feature point generation will be described later.

[0127] The film roll 80 manufactured in factory A is conveyed and transported to factory B. In the second manufacturing process of factory B, during product inspection, the film of the film roll 80 is optically inspected by inspection device 90, and inspection data is generated. By analyzing the image data obtained by photographing the film surface, inspection data containing feature points (hereinafter referred to as second inspection data) is generated. Information processing system 50 obtains the second inspection data from terminal device 70 in factory B (step S2). The inspection device 90 in factory A (first manufacturing process) and the inspection device 90 in factory B (second manufacturing process) are preferably the same, that is, the same measurement system and the same measurement conditions, but are not limited thereto. The performance, quality, and product specifications (hereinafter referred to as product specifications, etc.) required in factory A and factory B are sometimes different, and sometimes an inspection device 90 with appropriate measurement systems and measurement conditions is used according to the respective product specifications, etc.

[0128] The information processing system 50 performs feature point extraction processing (step S3) by comparing first inspection data and second inspection data related to the same thin film roll 80. Specifically, the information processing system 50 performs feature point extraction processing by comparing feature points at corresponding positions on the thin film surface in the first inspection data and second inspection data. In this specification, detecting feature points from image data is referred to as "generating feature points". Classifying feature points into any one of the following first to third types of feature points by comparing the first inspection data and second inspection data is referred to as "extracting feature points".

[0129] Figure 5 This is a table used to illustrate the first to third types of feature points extracted through feature point extraction processing. Multiple feature points generated from inspection data of a thin film roll 80 are sometimes classified into the first to third types of feature points. For example, some of several hundred feature points are extracted as first-type feature points, several as second-type feature points, and the rest as third-type feature points.

[0130] (First type of feature point)

[0131] The first type of feature point is a feature point that exists in the first inspection data but not in the second inspection data. This first type of feature point is a feature point that disappears in the second manufacturing process (e.g., the coating process). This first type of feature point is a feature point that does not need to be managed in the first manufacturing process. In this case, the manufacturing conditions that cause the generation of the first type of feature point can be subject to specification relaxation in the first manufacturing process.

[0132] (Second type of feature point)

[0133] 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 originates from the second manufacturing process, it can be used to improve the second manufacturing process.

[0134] (Third type of feature point)

[0135] The third type of feature point exists in both the first and second inspection data. This third type of feature point originates from the first manufacturing process and is a feature point that needs to be managed. Because this third type of feature point originates from the first manufacturing process, it can be used to improve the first manufacturing process.

[0136] The information processing system 50 can also feed back the feature point extraction results to users such as employees of factory A and factory B (step S4). For example, in response to access from the terminal device 70 of the user who is the manufacturing source of the film roll 80, and the terminal device 70 of the user to whom the film roll 80 is delivered, the extraction results are sent to the terminal device 70. The above is an overview of the feature point extraction process. More detailed information about the process will be described later.

[0137] (Information Processing System 50)

[0138] The following is for reference Figures 6-8 The information processing system 50 is described below. Figure 6 This is a block diagram representing the general structure of the information processing system 50. The information processing system 50 is, for example, a server. Figure 6 As shown, the information processing system 50 includes a control unit 51, a storage unit 52, and a communication unit 53.

[0139] (Control Department 51)

[0140] 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 or the like that executes the control or various arithmetic operations of the above-mentioned units according to the program. The various functions of the information processing system 50 are performed by the CPU executing its corresponding program.

[0141] The control unit 51 functions as an acquisition unit 511 and a receiving unit 512 in cooperation with the communication unit 53. Furthermore, the control unit 51 functions as a comparison unit 513, an analysis unit 514, an extraction unit 515, and an output unit 516. The acquisition unit 511 acquires first inspection data and second inspection data obtained through inspections in the first and second manufacturing processes. The receiving unit 512 receives correction input from the user. Additionally, the receiving unit 512 can also receive stretch rate input from the user. Here, the correction amount includes X-direction, Y-direction, θ-direction, r-direction, and surface / interior correction. Hereinafter, the correction amount will be described as X-direction, Y-direction, and θ-direction. Furthermore, hereafter, the received stretch rate and correction amount will be collectively referred to as correction amount, etc. The comparison unit 513 extracts feature points from the first and second inspection data respectively, and performs comparison processing using the received correction amount, etc. Through comparison processing, the comparison unit 513 explores corresponding feature points between the first and second inspection data. In the comparison processing, the comparison unit 513 generates descriptors of feature points (descriptor 2, described later) for two inspection data (image data), and uses these descriptors to perform feature point matching between the inspection data, outputting the comparison result (described later). Figure 16The analysis unit 514 uses the list of corresponding points to calculate indicators representing the relationship (hereinafter referred to as relationship information). The relationship information includes charts and statistical information such as average, regression formula, variance, standard deviation, and correlation coefficient. In addition, the charts include scatter plots, histograms, Pareto charts, bubble charts, etc. The regression formula includes regression lines and multinomial regressions of quadratic or higher. In addition, as a regression formula (regression model), in addition to single regression, multiple regression, logistic regression, and other regression methods can also be used. The main extraction unit 515 uses the comparison results to extract (classify) the first to third type of feature points. The output unit 516 responds to requests from the terminal device 70, etc., and sends the feature point extraction results to the terminal device 70 or displays them on the display unit (not shown).

[0142] (Storage Department 52)

[0143] Storage unit 52 is a large-capacity auxiliary storage device that stores various programs or data, including the operating system. As storage, it may employ a hard disk, solid-state drive, flash memory, ROM, etc. Storage unit 52 stores user lists, batch lists, inspection data databases, etc. The management and registration of user lists and batch lists are performed by an administrator accessing terminal device 70. For example, this administrator might be a staff member in the department responsible for operating factory A.

[0144] (User List)

[0145] Figure 7 These are examples of various data stored in storage unit 52. Figure 7 Table T1 shown is an example of a user list. The user list stores user IDs, usernames, contact addresses, etc. Furthermore, access permissions are set for each user to the retrieval data DB (inspection database), granting them access to various data (inspection data, extracted data, etc.) related to the film roll 80 (identified by batch ID) involved with that user.

[0146] (Batch list)

[0147] Figure 7 Table T2 is an example of a batch list. The batch list records the batch ID assigned to each film roll, the product name (also known as variety), the delivery location user ID (orderer), and multiple manufacturing conditions, dimensions (width, length, thickness), manufacturing date, etc.

[0148] (Check the database)

[0149] Figure 8 This is an example of a check database (DB) stored in storage unit 52. The check database stores data such as... Figure 8The data shown includes the first inspection data, the second inspection data, and the feature point extraction results, etc., related to the inspection of various film rolls 80. As described above, the first inspection data is data obtained during inspection in the first manufacturing process. The second inspection data is data obtained during inspection in the second manufacturing process. The feature point extraction results are data generated by the information processing system 50 using these first and second inspection data.

[0150] Figure 8 Table T3 shown is an example of an inspection list registered in the inspection data DB. The inspection list stores information such as inspection ID, batch ID, inspection device ID, inspection data, and inspection date and time.

[0151] Figure 8 Table T4 shows an example of the contents of the inspection data (inspection ID: i0101) in the inspection list. The inspection data describes the feature point IDs that are automatically assigned to each feature point with consecutive numbers, and the feature point descriptors 1, 2 (hereinafter referred to as descriptor 1, etc.) for each feature point ID.

[0152] Descriptor 1 contains individual information about the feature point, recording its XY coordinate position and intensity. Intensity refers to the feature point's level, which will be described later. Furthermore, as intensity information, it may also include information about the feature point's size (diameter, area) and brightness. The XY coordinate position is based on the starting point of the film surface (e.g., the left end of the front end). X is the coordinate in the width direction of the film, which, depending on the film size (refer to Table T2), can range from 0 to 3000 mm. Y is the coordinate in the long side direction of the film, which, depending on the film size, can range from 0 to 10000 μm. Descriptor 1 is generated by the image analysis unit 93 of the inspection device 90.

[0153] Descriptor 2 represents surrounding information, which is a vector or array representing information such as the relationship with other feature points. For example, SIFT feature values ​​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 descriptor 2 is mainly performed by the comparison unit 513.

[0154] Figure 8 Table T5 shows an example of the extracted result data (hereinafter referred to as the extracted data). The extracted data records the inspection IDs of the first and second inspection data (which serve as the basis), as well as the extraction result for each feature point. The extracted results (types one through three) are as described above. Figure 5The classification is shown. The composite feature point ID is an automatically assigned identifier with consecutive numbers, corresponding to the feature point present in either the first or second inspection data. The number of composite feature point IDs is greater than or equal to the number of feature point IDs in the first and second inspections. Furthermore, the extracted data is updated with the input of correction values, etc. This is because, as described later, descriptors 1 and 2 are updated by accepting correction values, etc. Figure 14 , Figure 15 Steps S402 and S502 are performed, and the updated descriptors 1 and 2 are used for comparison.

[0155] Furthermore, for a single batch ID, multiple first inspection data and second inspection data may sometimes be generated by multiple inspection devices. For example, sometimes in a second manufacturing process, the film roll 80 in its master roll state is inspected (photographed), and inspections are also performed in several downstream processes, thereby generating multiple second inspection data. In this case, the information processing system 50 can also generate multiple feature point extraction result data for a single first inspection data, in a one-to-many relationship with multiple second inspection data. Additionally, it can be configured to allow the user to select which second inspection data corresponds to (e.g., as described later). Figure 12 Buttons b1, b2, etc.

[0156] (Ministry of Communications 53)

[0157] The communications section 53 is also an interface for network connection with external devices such as PCs.

[0158] (Generation and processing of the first and second inspection data)

[0159] The following is for reference Figure 9 , Figure 10 The generation and processing of the first inspection data and the second inspection data performed in the first manufacturing process and the second manufacturing process are explained. Figure 9 This is a flowchart representing the generation process of the first inspection data in the first manufacturing process.

[0160] (Generation and processing of the first inspection data)

[0161] (Step S11)

[0162] In the above typical example, in the first manufacturing process, film roll 80 is manufactured by film roll manufacturing apparatus 1000.

[0163] (Step S12)

[0164] The inspection device 90 takes a picture of the film and saves the image data. The inspection device 90 is for... Figure 3A The apparatus described in the document.

[0165] (Step S13)

[0166] The image analysis unit 93 performs the image processing described below on the image data to generate multiple feature points.

[0167] (Image processing for feature point generation)

[0168] The image analysis unit 93 acquires the two-dimensional image data generated by the camera 92 and stored in the storage unit 94.

[0169] The image analysis unit 93 performs data processing on the image data (inspection data) acquired from the camera 92.

[0170] The image analysis unit 93 divides the image data into multiple regions. For example, the image analysis unit 93 divides the image data into n (e.g., several to dozens) regions (hereinafter referred to as region a1 to region an) in the width direction.

[0171] Next, the image analysis unit 93 acquires image data of region a1 and performs mathematical processing on the image data of region a1. Appropriate mathematical processing is prepared in advance according to the type of defect of the detected object (stripes, longitudinal wrinkles, diagonal wrinkles, uneven bright spots, light leakage, etc.).

[0172] Mathematical processing includes preprocessing, enhancement processing, signal processing, and image feature extraction.

[0173] As a preprocessing step, the following steps can be taken:

[0174] • Image cropping;

[0175] • Low-pass filter, high-pass filter, Gaussian filter, median filter, bilateral filter;

[0176] • Morphological transformations, color transformations (L*a*b*, sRGB, HSV, HSL), contrast adjustment, noise removal, restoration of blurred / shaky images, mask processing, Hough transform, projective transformation, etc.

[0177] Examples of enhancement methods include the Sobel filter, Scharr filter, Laplacian filter, Gabor filter, and the Canny method.

[0178] As a form of signal processing, the following processing methods can be cited:

[0179] • Solving for basic statistics (maximum, minimum, mean, median, standard deviation, variance, quartiles), root mean square, difference, sum, product, ratio, distance matrix processing, calculus, threshold processing (binarization, adaptive binarization, etc.).

[0180] • Fourier transform, wavelet transform, peak detection (peak value, number of peak values, full width at half maximum, etc.), etc.

[0181] Examples of image feature extraction methods include template matching and SIFT features.

[0182] Next, the image analysis unit 93 performs thresholding on the values ​​(feature quantities) obtained from the image data of region a1 through mathematical processing. Thresholding is a process that determines whether a defect is a target for detection based on a predetermined defect determination threshold and determines the level (intensity) of the defect.

[0183] 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 levels based on thresholds is equivalent to "quantitative evaluation processing."

[0184] For example, for parameters (features) with values ​​from 1 to 100, defects can be classified into multiple levels. For instance, levels can be classified based on defect size (diameter or area). Furthermore, the size-based classifications can be further subdivided based on the parameter values.

[0185] The image analysis unit 93 processes regions other than region a1 in the same way.

[0186] 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 data corresponding to the level of the detected defect and its location (xy coordinates) for each region (each position of the thin film F8 in the width direction).

[0187] After data processing, the image analysis unit 93 saves the processing results to the storage unit 94. The image analysis unit 93 performs this data processing on each of the multiple image data obtained from inspecting a thin film roll 80 to obtain a processing result. By summarizing these processing results, a result is generated as follows: Figure 8 The inspection data is shown in Table T4.

[0188] (Step S14)

[0189] The terminal device 70, located in the first manufacturing process, sends inspection data containing information on multiple feature points obtained through processing up to step S13 to the information processing system 50. The acquisition unit 511 of the information processing system 50 saves the acquired inspection data as first inspection data in the inspection data DB of the storage unit 52.

[0190] (Generation and processing of the second inspection data)

[0191] Figure 10This is a flowchart illustrating the generation process of second inspection data during the second manufacturing process.

[0192] (Step S21)

[0193] In the second manufacturing process, after the first manufacturing process, the product manufacturing apparatus 2000 performs a post-processing using the film roll 80 to manufacture a product using the film F8.

[0194] (Step S22)

[0195] The inspection device 90 takes pictures of the surface of the film F8 before post-processing, or during or after post-processing, and saves the image data.

[0196] (Step S23)

[0197] The image analysis unit 93 performs the same process as in step S13, storing the inspection data containing feature point information of the generated multiple feature points in the storage unit 94.

[0198] (Step S24)

[0199] The terminal device 70, located in the second manufacturing process, sends inspection data containing information on multiple feature points obtained through processing up to step S23 to the information processing system 50. The acquisition unit 511 of the information processing system 50 saves the acquired inspection data as second inspection data in the inspection data DB of the storage unit 52. The second inspection data, like the first inspection data shown in Table T4, is data described by feature point ID and feature point descriptors 1 and 2.

[0200] (Feature point extraction and processing)

[0201] The following is for reference Figures 11-18 The feature point extraction process performed in the information processing system 50 is explained. Figure 11 This is a flowchart representing the feature point extraction process. Figure 12 This is an example of an operation screen displayed on terminal device 70. Figure 13 This is a schematic diagram illustrating the feature point extraction process.

[0202] The information processing system 50 starts processing after step S31 in response to a start instruction from the user via the operation screen of the terminal device 70, or when the second check data is registered in the check data DB of the storage unit 52 and a pair of the first check data and the second check data are complete. Figure 12This is an example of an operation screen 701 displayed on the terminal device 70. After selecting a batch, the user selects first inspection data and second inspection data from multiple inspection data associated with that batch using buttons b1 and b2. The second inspection data is inspection data obtained in a process downstream of the first inspection data. Furthermore, in this embodiment, the case where there is an extended process between the location where the first inspection data is collected and the location where the second inspection data is collected will be described.

[0203] (Step S31)

[0204] The acquisition unit 511 acquires the same batch, i.e., a pair of first and second inspection data, from the inspection data DB. The same batch is associated based on the batch ID assigned to the inspection ID. Furthermore, regarding batch IDs, the branch number ID or new batch ID assigned in a certain process is sometimes not a one-to-one correspondence, but rather a one-to-n or n-to-one correspondence. For example, sometimes a batch of 3000m film in a previous process is divided into 3 parts of 1000m in a subsequent process. In this case, the branch number or new batch ID is assigned to the batch ID in the subsequent process. Conversely, sometimes two or three film rolls are joined together in a previous process and used as one film in a subsequent process. In this case, there are two or three batch IDs in the previous process, but only one batch ID in the subsequent process. In either case, the correspondence between the batch IDs of each process is recorded in the batch list (see reference). Figure 7 ).

[0205] (Steps S32 and S33)

[0206] The comparison unit 513 performs preprocessing on the first inspection data under the first condition and preprocessing on the second inspection data under the second condition.

[0207] like Figure 13 As shown, as a preprocessing step for the second condition, the comparison unit 513 performs a preprocessing step to reverse the Y-coordinate (vertical) for matching purposes when there is a difference between winding (first manufacturing process) and unwinding (second manufacturing process) in the second inspection data. Furthermore, in the second manufacturing process, based on information that the shooting area of ​​the camera 92 is set to the surface or back of the film F8, the comparison unit 513 performs a preprocessing step to reverse the X-coordinate (left-right) for the second inspection data (or the first inspection data). Moreover, if the film F8 stretches or contracts due to heating temperature or stretching settings during post-processing in the second manufacturing process, and if the stretching rate can be estimated in advance, the first or second inspection data is transformed using the stretching rate received by the receiving unit 512.

[0208] Furthermore, the comparison unit 513 performs at least one of the following processes as noise removal processing included in the first and second conditions for the first inspection data and the second inspection data.

[0209] (1) Remove low-level feature points.

[0210] (2) Remove extremely small feature points;

[0211] (3) Remove consecutive dots;

[0212] (4) Remove concentrated dots along the short side. This is especially common at the head or tail of film F8.

[0213] (Step S34)

[0214] The receiving unit 512 receives the scaling factor from the user. For example, the receiving unit 512 receives the scaling factor via, as shown in the example... Figure 12 The operation screen 701 displayed on the terminal device 70 accepts input of the scaling factor. The user can input any value within a specified range (e.g., 50% to 300%) via the numeric keypad using buttons b3 and b4. The receiving unit 512 accepts input of the scaling factor in the Y direction (long side direction) via the user's input to button b3. Furthermore, the receiving unit 512 accepts input of the scaling factor in the X direction (short side direction) via the user's input to button b4. Figure 12 The operation screen 701 shows the states where 140% has been entered. If the value is greater than 100%, it is assumed that the film is more elongated in the second inspection data compared to the first inspection data.

[0215] In addition, Figure 12 The diagram shows an example where the extension section 03 receives elongation rates in the X and Y directions respectively for TD extension in the X direction and MD extension in the Y direction. However, it is not limited to this; for example, in the case of an oblique extension process performed by the extension section 03, the receiving section 512 may also be configured to receive the oblique extension ratio. In this case, the receiving section 512 receives the input of the extension angle value within a specified range (40 to 50°).

[0216] (Step S35)

[0217] The comparison unit 513 retrieves feature points from the inspection data of one party that are the same as or correspond to feature points existing in the inspection data of the other party through the comparison processing 1, and establishes correspondence (matching) between the feature points. Figure 14 This is a flowchart of the subroutine that represents the processing of step S35.

[0218] (Step S401)

[0219] The comparison unit 513 moves the position of a feature point in one of the first and second inspection data according to the scaling factor received in step S34. That is, the comparison unit 513 transforms the coordinate system. Hereinafter, the explanation will assume that the comparison unit 513 moves the feature point of the first inspection data according to the scaling factor. For example, in... Figure 12 In the input example, the comparison unit 513 transforms the Y-coordinate of each feature point of the first inspection data into a value obtained by multiplying the stretch rate by the origin (0). Here, the origin is the head position (winding start position) of the film F8 during the production of the film roll 80 (the tail end during unwinding). Furthermore, the comparison unit 513 transforms the X-coordinate of each feature point of the first inspection data into a value obtained by multiplying the stretch rate by the center point. That is, the first inspection data is transformed to extend to both sides based on the center point. Here, the center point is the center position of the film in the width direction of the first inspection data. As another example, in the case of oblique extension, based on the input extension angle and elongation rate, the comparison unit 513 calculates the elongation rate in the X direction and the elongation rate in the Y direction corresponding to the position in the width direction of the film in the first inspection data, and applies this elongation rate. Additionally, as yet another example, in the case of transformation on the second inspection data side, the comparison unit 513 can perform coordinate transformation using the same processing as in the case of transformation on the first inspection data side, except for dividing by the stretch rate.

[0220] (S402)

[0221] The comparison unit 513 generates descriptors 2 for each feature point of the first inspection data and the second inspection data. The comparison unit 513 may use SIFT feature quantities as descriptors, or use the probability density function of the feature points calculated using kernel density estimation as descriptors.

[0222] (Steps S403, S404)

[0223] The comparison unit 513 compares the feature points of the first inspection data and the second inspection data with each other and retrieves the most corresponding point. The comparison unit 513 evaluates the similarity between the feature points based on descriptor 1 and descriptor 2, and regards the most similar point as the corresponding feature point.

[0224] For example, in comparing feature points of the first inspection data and the second inspection data, the comparison unit 513 retrieves feature points from the feature points of the first inspection data that correspond to the object feature points of the second inspection data. At this time, the comparison unit 513 determines the feature points of the first inspection data that are identical to or have the closest intensity to the X and Y coordinates of the feature points of the second inspection data as identical points. Alternatively, it determines the points whose distance (Euclidean distance) between the X and Y coordinates of the descriptors 1 is closest as identical points (corresponding points). The comparison unit 513 excludes feature points that are further apart than the set threshold from the determination of corresponding points. Furthermore, in this embodiment, regarding the threshold, the threshold in the Y direction is set to be 2 to 3 orders of magnitude larger than that in the X direction. For example, the threshold in the X direction is several millimeters, and the threshold in the Y direction is several meters. The threshold in the Y direction is 2 to 3 orders of magnitude larger than that in the X direction because of the greater variation.

[0225] Furthermore, the comparison unit 513 can also use descriptor 2 in conjunction with descriptor 1 to extract the most similar feature points based on the distances between vectors in the high-dimensional vector space. In this case, the comparison unit 513 can also use the probability density function calculated by kernel density estimation as described above as descriptor 2. Figure 15 This is an example of a probability density function representing the location and intensity (density) of a feature point, calculated using kernel density estimation. Figure 15 In the diagram, the horizontal and vertical axes represent the X and Y coordinates, and the higher the concentration, the higher the density.

[0226] The comparison unit 513 determines that a 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 registered as corresponding points in the corresponding point list.

[0227] Figure 16 This is an example of a list of corresponding points stored in storage unit 52. The list of corresponding points is formed by establishing correspondences between each feature point of the second inspection data and the most similar feature point in the first inspection data. Furthermore, the list of corresponding points describes the XY coordinates of the feature points of the second inspection data, the Euclidean distance between the corresponding feature points, the difference dx of the X coordinate, and the difference dy of the Y coordinate. A ratio can also be used as the difference.

[0228] (Step S405)

[0229] The analysis unit 514 generates correlation information between feature points that exist simultaneously in both the first and second inspection data. Specifically, for feature points in the corresponding point list that have corresponding feature points, the analysis unit 514 generates any one of four scatter plots as data representing the correlation information. Furthermore, principal component analysis can be performed on the following four relationships to generate scatter plots projected onto a coordinate system that maximizes the variance of the data.

[0230] (1) Horizontal axis X coordinate, vertical axis dx (X-dx scatter plot);

[0231] (2) Horizontal axis X coordinate, vertical axis dy (X-dy scatter plot);

[0232] (3) Horizontal axis Y coordinate, vertical axis dy (Y-dy scatter plot);

[0233] (4) Horizontal axis Y coordinate, vertical axis dx (Y-dx scatter plot).

[0234] Figure 17A This is an example of a scatter plot corresponding to (1) above. Figure 17B This is an example of a scatter plot corresponding to (2) above. Figure 17C This is an example of a scatter plot corresponding to (3) above. Figure 17D This is an example of a scatter plot corresponding to (4) above. Figure 18 This is another example of a scatter plot corresponding to (1) above, an example of bow-shaped stretching produced during the extension process. What is common in these plots is that the horizontal axis represents the X (or Y) coordinate of the feature points in the second inspection data. Furthermore, the analysis unit 514 may replace the scatter plot or be used in conjunction with the scatter plot. Figure 1 Statistical information (such as regression equations, correlation coefficients, etc.) is calculated to serve as correlation information. Figures 17A to 17D , Figure 18 The regression equation for the scatter plot is also shown in the diagram. Alternatively, a histogram, Pareto chart, or bubble chart can be used instead of a scatter plot.

[0235] (Step S36)

[0236] The output unit 516 displays or outputs alarm information. For example, it displays the information on the terminal device 70. The output unit 516 displays the relevant information on the operation screen of the terminal device 70. In addition, the output unit 516 can also notify the user of the alarm through sound information based on a warning tone, or through tactile information such as vibration of the mobile terminal. Figure 19An example of an operation screen 702 for receiving correction values, showing a scatter plot as correlation information, is provided. This operation screen 702 is displayed immediately following operation screen 701. For example, a scatter plot representing the correlation information generated in step S405 is displayed. Figure 19 The preview area b10 of the operation screen 702 shown. Additionally, correlation coefficients can also be overlaid on the preview area b10. Users can refer to... Figure 19 The correlation information displayed in preview area b10 can be used to determine whether the correspondence is appropriate. By referring to charts such as scatter plots, users can review the correction values ​​to determine whether the correspondence between the feature points of the two checked data is good.

[0237] As a variation, regarding the display range, it can also be configured to accept the display range (preview X-axis) set by the user for the long or short side, and perform a preview display within the accepted range. Alternatively, instead of applying the correction amount to the entire range, it can be divided into segments for the display range, or the setting can be used to accept interval settings, and the correction amount can be set separately for each interval.

[0238] In addition, as another variation, only the correlation coefficient can be displayed, or statistical information such as the correlation coefficient can be displayed in a window in the preview area b10.

[0239] Furthermore, as another variation, the following approach can be taken. For example, when a correction amount is entered and the feature points recalculated based on that correction amount are displayed in the preview area b1, the plot before applying the correction amount (result data) and the plot after applying the correction amount can be displayed side by side, or they can be overlapped using color differentiation. This allows the user to compare the result data before and after correction, thereby assessing the appropriateness of the entered correction amount.

[0240] (Step S37)

[0241] If a user observes a scatter plot and determines that the correlation is poor, they may input a correction value to try and improve the situation. In response to the receiving unit 512 receiving the correction value from the user (Yes), the control unit 51 proceeds the process to step S50.

[0242] Here, the types of correction amounts received by the receiving unit 512 are as follows. Additionally, a reversal correction can also be added as a correction amount. A reversal correction reverses the X or Y coordinate relative to a reference (the center of the short side or the tail of the long side).

[0243] (a) Oblique correction amount (rotation) (θ direction);

[0244] (b) X-coordinate correction (X direction);

[0245] (c) Y-coordinate correction (Y direction).

[0246] By accepting (a) a slant correction amount, the coordinates of one of the two inspection data points are transformed. This slant correction amount corresponds to the scaling factor. For example, in Figure 17, Figure 18 In this process, by applying a slant correction, the same effect as rotating the entire scatter plot is achieved. Furthermore, the input in (a) can also be accepted as an X-correction in the correction position of the X-coordinate (a11) or a Y-correction in the correction position of the Y-coordinate (a12). In (b) the X-coordinate correction, the comparison unit 513 shifts the overall X-coordinate of one of the inspection data based on the input correction. For example, the comparison unit 513 shifts the X-coordinate of the first inspection data based on the input correction. In this case, Figure 18 In this case, the overall Y-coordinate of the data is shifted in the vertical direction, similar to the effect of a scatter plot where the entire plot is offset. The same applies to (c) the Y-coordinate correction. In this case, one side checks if the overall Y-coordinate of the data is shifted according to the input correction. For example, in the example in Figure 17, the entire data is shifted in the vertical direction according to the input Y-coordinate correction.

[0247] Figure 19 The input example shown illustrates an example of (a12). Figure 19 In operation screen 702, the user selects a Y-dy scatter plot via button b11, and a Y-dy scatter plot as shown in Figure 17 is displayed in preview area b10 corresponding to this selection. The user inputs -6.0m in input area b12 and 0m in b13. By operating the recalculation button b14, a recalculation is performed in response to this input correction, and the display in preview area b10 is updated (redisplayed). In this case, in the regression equation, the center is fixed, and the leftmost position (b13: Y=0m) moves by -6.0m. That is, the entire coordinate system rotates counterclockwise so that the left end in the regression equation decreases by 6.0m.

[0248] (Step S50)

[0249] Figure 20 This is a flowchart representing the subroutine of step S50, comparison processing 2.

[0250] (Step S501)

[0251] The comparison unit 513 moves the position of one of the feature points in the first inspection data and the second inspection data according to the correction amount received in step S37 (hereinafter referred to as the input correction amount). That is, the comparison unit 513 performs a coordinate system transformation. Hereinafter, the comparison unit 513 will be described as moving the feature point of the first inspection data according to the input correction amount.

[0252] For example, in Figure 19 In the input example, the comparison unit 513 moves the Y-coordinate of a feature point located at the origin 0m by inputting a correction amount (-6.0m). For Y-coordinates outside the origin, the Y-coordinate of the feature point is moved by multiplying the input correction amount (-6.0m) by the ratio of the distance to the center position ((yc-y) / yc). Here, yc is the coordinate of the center position. yc is the value of half the distance from the origin to the end of the data. The end is the last data of the Y-coordinate of the first checked data, corresponding to the length of the film in the roll (refer to...). Figure 7 (The size of table T2). Figure 21 This is an example of applying correction values. Similarly, the X-coordinate is moved based on the user-input correction values. That is, the movement of the feature point's coordinates is based on the movement of the feature point's coordinates according to these processes, such as... Figure 21 As shown, the entire structure rotates counterclockwise (refer to the arrow) with the axis of rotation on the regression equation as the center.

[0253] (Steps S502 to S505)

[0254] The processing here and Figure 14 Steps S402 to S405 are the same. The comparison unit 513 uses the transformed coordinates from step S501 to regenerate feature point descriptor information, performs feature point matching, establishes correspondences between corresponding feature points, and saves them in a correspondence list. The parsing unit 514 refers to the updated correspondence list and generates correlation information between feature points that exist simultaneously in the first and second inspection data. The correlation information, as described above, includes at least one of a scatter plot and statistical information. Thus far, Figure 20 The processing is complete, return to Figure 11 The processing. For example, Figure 21 The scatter plot shown is displayed on Figure 19 The preview area b10 of the operation screen 702.

[0255] (Step S51)

[0256] The output unit 516 displays the correlation information again. The processing here is the same as in step S36. For example, in... Figure 19 In the preview area b10, the correlation information generated by the feature points after moving the coordinates according to the received correction amount is displayed in the preview area b10.

[0257] (Step S37: No)

[0258] If the user determines that the correlation is correct to some extent by referring to correlation information such as scatter plots, and determines that no correction is needed (re-correction) (No), the user presses the registration button b16 on the operation screen 701. The control unit 51 responds to the registration button b16 being activated, causing the process to proceed to step S38.

[0259] (Step S38)

[0260] Figure 22 An example of feature point extraction conditions that are determined and recorded in the check DB of the storage unit 52.

[0261] The control unit 51 associates and records the stretch rate and correction amount received in steps S34 and S37 with the batch and the combination of the first inspection data and the second inspection data. Subsequently, when using film rolls of the same type as that batch, by reading the feature point extraction conditions, adjustment can be avoided, or by comparing with previous batches under the same conditions, differences between batches can be identified. Alternatively, feature point extraction can be performed and recorded using a predetermined correction amount from previous processes. For example, feature point extraction (classification) can be performed each time based on an input correction amount.

[0262] Furthermore, based on the distance between the feature points in the first inspection data and the second inspection data, the extraction unit 515 classifies the feature points that correspond to each other in the first inspection data and the second inspection data into two categories: the third type a of level A and the third type b of level B. Figure 23 This is a schematic diagram illustrating classification based on the distance to the regression equation. The extraction unit 515 classifies feature points whose distance to the regression equation is below a predetermined threshold as type a, and feature points whose distance to the regression equation exceeds the predetermined threshold as type b. This classification is performed using the Y-dy scatter plot described above (3), but it can also be performed using other scatter plots (e.g., X-dx). Furthermore, classification can also be performed based on both the distance of the Y-dy scatter plot to the regression equation and the distance of the X-dx scatter plot to the regression equation, and the two can be compared and then a prompt is given. For example, if both are classified as level A, the classification is A1 (level S), and if only one is classified as level A, the classification is A2.

[0263] Furthermore, the distance between feature points and the regression equation can be determined using the distance along the vertical axis (dy), but it can also be determined using Euclidean distance. Additionally, a pre-set threshold is used, but it can also be set based on statistical information such as the standard deviation (σ). For example, the extraction unit 515 classifies feature points whose distance from the regression equation is in the range of +1σ to -1σ as a third type a, and classifies all other feature points as a third type b. The extraction unit 515 can also separately describe the third types a and b in a corresponding point list. Feature points classified as level A (third type a) have higher reliability as the same feature point.

[0264] (Step S39)

[0265] The extraction unit 515 classifies feature points according to the correspondence list, the first inspection data, and the second inspection data. Specifically, in the correspondence list, the extraction unit 515 classifies feature points in the second inspection data that do not correspond to feature points in the first inspection data as second-type feature points. Furthermore, the extraction unit 515 classifies feature points in the first inspection data that are not included in the correspondence list (feature points that do not correspond to feature points in the second inspection data) as first-type feature points (see reference). Figure 5 , Figure 8 Table T5).

[0266] (Step S40)

[0267] The output unit 516 registers the extraction result (extracted data) generated in step S38 in the inspection data DB, or sends the extraction result to the terminal device 70. Thus, Figure 11 The feature point extraction process shown is now complete (End).

[0268] In this embodiment, the information processing system includes a comparison unit that compares the first feature point information of the roll material in the first inspection data with the second feature point information of the roll material in the second inspection data, and extracts the same feature points in both the first and second inspection data. Furthermore, the information processing system includes a parsing unit that generates relational information representing the relationship between the difference values ​​of the coordinates of each of the multiple feature points extracted by the comparison unit in the first and second inspection data and their coordinate positions in the overall roll material; and an output unit that outputs the relational information. Thus, information that helps in process improvement and shipment specification setting can be efficiently collected during both the manufacturing of films or roll materials and in the manufacturing process of using them for post-processing. In particular, by having the user confirm the displayed correlation information, it is possible to determine whether the correspondence between the feature points in the first and second inspection data is correctly established. Furthermore, by accepting user-input correction values, the correspondence between feature points (third type of feature points) can be established with higher accuracy, thereby enabling more accurate classification of the first to third type of feature points.

[0269] (Second Implementation)

[0270] The second implementation method is to issue an alarm based on the correlation coefficient in the scatter plot. Figure 24 This is a flowchart illustrating the feature point extraction process in the second embodiment.

[0271] exist Figure 24 In the above, steps S71-S76 and S77-S91 respectively correspond to Figure 12 Steps 31 to S36 and S37 to S51 are shown in the first embodiment. In the second embodiment, step S765 is performed.

[0272] (Step S765)

[0273] The analysis unit 514 calculates the correlation coefficient and issues an alarm if the correlation coefficient is below a first predetermined threshold. For example, if the absolute value of the correlation coefficient is below 0.5, an alarm is displayed on the terminal device 70. Alternatively, if the regression equation is a straight line with a slope close to zero (i.e., flat), the variance of the difference values ​​(dy or dx) of the feature points is calculated, and an alarm is displayed if the variance is above a second threshold.

[0274] Figure 25 This is a diagram showing an example of a distribution map and alarm display. Figure 25 (A) represents a situation where the correlation coefficient is high or the variance of the difference value dy is small, indicating normality. On the other hand, Figure 25 (b) and (c) represent cases where the correlation coefficient is low or the variance of the difference value dy is large, indicating an anomaly. In this case, the output unit 516 displays an alarm on the display unit of the terminal device 70. By doing so, the user can understand that a certain fault (anomaly) has occurred.

[0275] Regarding the structure of the information processing system 50 described above, the main structure has been described to illustrate the features of the above embodiments, but it is not limited to the above structure, and various modifications can be made within the scope of the claims. Furthermore, structures typical of information processing apparatuses / systems are not excluded. For example, in this embodiment, an extension step is provided between the location where the first inspection data is collected and the location where the second inspection data is collected. Figure 11 , Figure 12 An example of accepting the stretch rate is shown, but the process of accepting the stretch rate can also be omitted (step S34). For example, if there is no extension process between the first and second inspection data, the acceptance of the stretch rate can be omitted. In this case, the comparison unit 513 performs a comparison process with a fixed stretch rate in step S35.

[0276] Furthermore, when batch selection is used, if there is a past track record of producing products under the same production conditions, and the correction amount set and determined by the user is used as a feature point extraction condition (see [reference]). Figure 22 If this information is recorded, it can also retrieve and display the correction values ​​applied in the same product in the past. This allows users to easily set appropriate correction values ​​by reducing the number of trial and error attempts.

[0277] Furthermore, for example, the information processing system 50 may 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 also perform the function of generating feature points for the image analysis unit 93 of the inspection device 90. In this case, the inspection device 90 sends image data obtained by photographing the film surface and its shooting conditions (transport speed, camera orientation, viewing angle, etc.) to the information processing system 50, and the feature point generation process is performed on the control unit 51 side.

[0278] Furthermore, the units and methods performing various processes in the information processing system 50 according to the above embodiments can be implemented by dedicated hardware circuits or a programmed computer. The aforementioned programs can be provided, for example, via computer-readable recording media such as USB memory or DVD (Digital Versatile Disc)-ROM, or online via a network such as the Internet. In this case, the programs recorded on the computer-readable recording media are typically transferred and stored in a storage unit such as a hard disk. Furthermore, the aforementioned programs can be provided as standalone application software or incorporated into the software of the device as a function of the device.

[0279] While embodiments of the invention have been described and illustrated in detail, the disclosed embodiments are for illustrative and exemplary purposes only and are not intended to be limiting. The scope of the invention should be interpreted according to the literal meaning of the appended claims.

[0280] This application is based on Japanese Patent Application No. 2024-056247, filed on March 29, 2024, the disclosure of which is incorporated herein by reference in its entirety.

[0281] Explanation of reference numerals in the attached drawings: 50…Information processing system; 51…Control unit; 511…Acquisition unit; 512…Receiving unit; 13…Comparison unit; 514…Analysis unit; 515…Extraction unit; 516…Output unit; 52…Storage unit; 90…Inspection device; 1000…Thin film roll manufacturing apparatus; 2000…Product manufacturing apparatus.

Claims

1. An information processing system, wherein, have: The acquisition unit acquires first inspection data in a first manufacturing process of manufacturing a roll or performing post-processing on the manufactured roll, and second inspection data in a second manufacturing process after the first manufacturing process in which the roll is used for post-processing. The comparison unit compares the first feature point information of the roll material in the first inspection data with the second feature point information of the roll material in the second inspection data, and extracts the same feature points in the first inspection data and the second inspection data. The analysis unit generates relational information, which represents the relationship between the difference values ​​of the coordinates of the multiple feature points extracted by the comparison unit in the first inspection data and the second inspection data and their coordinate positions in the whole roll material. as well as The output section outputs the relational information.

2. The information processing system according to claim 1, wherein, The relational information is a graph representing the relationship between the difference values ​​of the coordinates and their coordinate positions in the whole.

3. The information processing system according to claim 1 or 2, wherein, The relational information refers to statistical information representing the relationship between the difference values ​​of the coordinates and their coordinate positions within the whole.

4. The information processing system according to claim 1, wherein, The information processing system further includes a receiving unit that receives the correction values ​​for the coordinates of the first inspection data and / or the second inspection data. After the comparison unit corrects the coordinates of the feature points using the correction amount received by the receiving unit, it compares the first feature point information of the roll material in the first inspection data with the second feature point information of the roll material in the second inspection data, and extracts the same feature points in the first inspection data and the second inspection data.

5. The information processing system according to claim 4, wherein, After the output unit displays the relational information, the receiving unit accepts the correction amount. In response to the receiving unit accepting the correction amount, the comparison unit extracts the same feature points again, and the parsing unit generates the relational information again. The output unit then displays the relational information again.

6. The information processing system according to claim 1, wherein, The relational information is represented by a scatter plot and a regression equation showing the relationship between the difference values ​​of the coordinates and their coordinate positions in the whole. The information processing system also includes an extraction unit, which classifies the feature points based on the distance from the feature points in the scatter plot to the regression equation.

7. The information processing system according to claim 1, wherein, The relational information includes the variance of the difference values ​​or the correlation coefficient representing the relationship between the difference values ​​of the coordinates and their coordinate positions in the whole. If the absolute value of the correlation coefficient is below a first predetermined threshold or if the variance is above a second predetermined threshold, the analysis unit generates an alarm, and the output unit reports the alarm.

8. A method for extracting feature points, specifically a method for extracting feature points from roll materials, wherein, have: Step (a): Obtain first inspection data in the first manufacturing process of manufacturing the roll material or performing post-processing on the manufactured roll material; Step (b): Obtain second inspection data in a second manufacturing process that follows the first manufacturing process and involves post-processing using the coil material; Step (c): Compare the first feature point information of the roll material in the first inspection data with the second feature point information of the roll material in the second inspection data; Step (d): Based on the comparison results of step (c), extract the same feature points in the first inspection data and the second inspection data; Step (e) generates relational information, which represents the relationship between the difference values ​​of the coordinates of the multiple feature points extracted in step (d) in the first inspection data and the second inspection data and their coordinate positions in the whole roll material. as well as Step (f) outputs the relational information generated in step (e).

9. The feature point extraction method according to claim 8, wherein, The relational information is a graph representing the relationship between the difference values ​​of the coordinates and their coordinate positions in the whole.

10. A control program for causing a computer to perform the extraction method of claim 8 or 9.

Images