Film inspection device, film inspection method and program

Abstract

To easily detect defects generated in a film without requiring manual monitoring or operation.SOLUTION: A film inspection device for optically inspecting defects generated in overlapping films, irradiates the film with light by a light source (step S1); detects at least diffusion light reflected on the film by a camera (optical sensor) (step S2); extracts an inspection area at inspection timing from first image data obtained from an output signal of the camera (step S5); and detects defects by performing data processing on second image data corresponding to the inspection area (step S6).SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to a film inspection device, a film inspection method, and a program. [Background technology]

[0002] When a long resin film is wound into a roll, various defects occur depending on the film condition (film thickness deviation, residual stress, temperature, humidity) and the winding conditions (tension, speed, entrained air). Examples of defects in the film include gauge bands, vertical wrinkles, and diagonal wrinkles.

[0003] A roll in which a defect occurs must be detected as a defective product. Instead of manual monitoring, a film inspection device has been proposed that uses a camera or the like to photograph the film being wound and optically inspects the film for defects (see Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2023-017169 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, to observe a long film roll from the beginning to the end, it is necessary to photograph the roll at each timing. In this case, as the diameter of the roll changes, the distance between the roll surface and the fixed camera changes. As a result, the size of the film roll in the image captured by the camera changes.

[0006] Therefore, it is necessary to extract an inspection area from the film roll of different sizes in the image for each inspection (photography) timing. The above-mentioned conventional technology does not explain a method for extracting an inspection area from an image obtained by photography. Generally, methods for extracting an area of ​​interest from an image include a camera's ROI function and cropping the image. However, with general methods, although it is possible to extract a certain range, it is difficult to extract the correct inspection area. Alternatively, the user has to manually specify the extraction range for each inspection timing.

[0007] The present invention has been made in consideration of the problems in the conventional technology described above, and has an object to easily detect defects occurring in a film without requiring manual monitoring or operation. [Means for solving the problem]

[0008] In order to solve the above problem, the invention described in claim 1 is a film inspection device that optically inspects defects that occur in overlapping films, and includes a light source that irradiates light onto the film, an optical sensor that detects at least diffuse light of the light reflected by the film, an inspection area extraction unit that extracts an inspection area at the inspection timing from first image data obtained from the output signal of the optical sensor, and a defect detection unit that detects the defects by performing data processing on second image data corresponding to the inspection area.

[0009] The invention described in claim 2 relates to the film inspection device described in claim 1, which is provided with a plurality of the optical sensors, and the inspection area extraction unit further removes overlapping areas between the multiple first image data obtained from each of the multiple optical sensors.

[0010] The invention described in claim 3 is a film inspection device described in claim 1 or 2, in which the inspection area extraction unit performs processing in the inspection area extraction unit by optical calculation using dimensional information of the film at the inspection timing and setting information of the optical sensor.

[0011] The invention described in claim 4 is a film inspection device described in claim 1 or 2, in which the film is wound up in a roll, and the film inspection device inspects for defects that occur during the winding of the film.

[0012] The invention described in claim 5 is the film inspection device described in claim 1 or 2, wherein the optical sensor is an area sensor.

[0013] The invention described in claim 6 relates to a film inspection device described in claim 1 or 2, in which an anti-glare plate is provided to prevent specular light from entering the optical sensor among the light reflected in the inspection area of ​​the film, and the anti-glare plate is composed of a diffusion plate.

[0014] The invention described in claim 7 relates to the film inspection device described in claim 1 or 2, and further includes a display control unit that displays, on a display unit, information on the inspection area extracted by the inspection area extraction unit and / or information on defects detected by the defect detection unit.

[0015] The invention described in claim 8 is a film inspection method for optically inspecting defects that occur in overlapping films, and includes the steps of: irradiating light onto the film using a light source; detecting at least diffuse light of the light reflected by the film using an optical sensor; extracting an inspection area at the inspection timing from first image data obtained from the output signal of the optical sensor using an inspection area extraction unit; and detecting the defects by performing data processing on second image data corresponding to the inspection area using a defect detection unit.

[0016] The invention described in claim 9 is a program for causing a computer of a film inspection device that optically inspects for defects in the film, comprising a light source that irradiates light onto overlapping films and an optical sensor that detects at least diffuse light of the light reflected by the film, to function as an inspection area extraction unit that extracts an inspection area at the inspection timing from first image data obtained from the output signal of the optical sensor, and a defect detection unit that detects the defects by performing data processing on second image data corresponding to the inspection area. Effect of the Invention

[0017] According to the present invention, defects occurring in a film can be easily detected without requiring manual monitoring or operation. [Brief description of the drawings]

[0018] [Figure 1] FIG. 1 is a diagram showing a functional configuration of a film inspection device in a first embodiment of the present invention. [Diagram 2] FIG. 2 is a diagram for explaining the arrangement of a light source and a camera. [Diagram 3] FIG. 2 is a perspective view showing the positional relationship of a light source and a camera with respect to a roll of film. [Figure 4] FIG. 13 is a diagram for explaining the arrangement of an anti-glare plate. [Diagram 5] 4 is a flowchart showing a film defect detection process executed by the film inspection device. [Figure 6] 13 is a flowchart showing a first inspection area extraction process. [Figure 7] 1 is an example of a captured image. [Figure 8] 11A and 11B are diagrams for explaining a change in the size of an inspection area accompanying a change in the winding diameter of a film roll. [Figure 9] 11A and 11B are diagrams for explaining a method of calculating a boundary position when extracting an inspection area from a captured image. [Figure 10A] 1 is a graph showing the relationship between the winding diameter of a film roll and the cut-out upper position. [Figure 10B] 1 is a graph showing the relationship between the winding diameter of a film roll and the cut-out lower position. [Figure 11] FIG. 2 is a diagram for explaining the photographing range in the width direction of the film, etc. [Figure 12] 13 is a flowchart showing data processing. [Figure 13] 13 is an example of a test result screen. [Figure 14] 13 is a diagram for explaining the amount of overlap of the shooting ranges of the three cameras provided in the film inspection device in the second embodiment. FIG. [Figure 15] 13 is a flowchart showing a second inspection area extraction process. [Figure 16] FIG. 11 is a diagram for explaining a method for determining an overlap amount of a shooting range. [Figure 17A] FIG. 2 is a diagram for explaining an overlapping range for first image data obtained from each of three cameras. [Figure 17B] FIG. 13 is a diagram showing the first image data after the overlapping area has been removed. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of a film inspection device, a film inspection method, and a program according to the present invention will be described with reference to the drawings. However, the scope of the present invention is not limited to the illustrated examples.

[0020] [First embodiment] 1 shows the functional configuration of a film inspection device 10 in the first embodiment. Film inspection device 10 includes a light source 11, a camera 12 as an optical sensor, and an information processing device 20. Film inspection device 10 optically inspects film F that has been wound into a roll and is in a stacked state for defects. Film inspection device 10 inspects long film F for defects that occur during winding.

[0021] The material of film F is not particularly limited, but generally includes 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, phenol resin, diallyl phthalate resin, polyimide resin, urethane resin, polyvinyl acetate resin, polyvinyl alcohol resin, styrene resin, cellulose acetate resin, polyvinyl chloride resin, etc. Moreover, for example, the film F has a thickness of 1 to 1000 μm and a width of 0.1 to 5 m.

[0022] Defects in film F include gauge bands, vertical wrinkles, diagonal wrinkles, and the like. Gauge bands are defects that look darker than the surroundings, and are dark streaks along the circumference of the roll. Gauge bands occur when films F stick to each other. Gauge bands are also called black bands or lumps. Vertical wrinkles are stripe-like deformations in the circumferential direction of the roll. Vertical wrinkles are also called lantern buckling. The oblique wrinkles are deformations in a direction oblique to the circumferential and width directions of the roll, and have pyramidal or chain-like irregularities. The oblique wrinkles are also called diamond buckling.

[0023] Light source 11 irradiates light onto an area including the inspection area of ​​film F. Light source 11 irradiates light uniformly in the width direction of rolled film F. Here, the width direction of film F is a direction perpendicular to the longitudinal direction of film F and parallel to the film surface. Furthermore, "uniform in the width direction" means that the illuminance of film F is approximately uniform across the width direction of film F. "Approximately uniform illuminance" refers to, for example, a case where the difference between the maximum and minimum illuminance values ​​is equal to or less than a predetermined value.

[0024] The inspection area is an area to be inspected for defects at each inspection timing. The inspection area is a portion that is observed by diffused light among the light irradiated by the light source 11 and reflected by the film F, and is a range without reflection of the surroundings.

[0025] Camera 12 is an optical sensor that optically reads an area including the inspection area of ​​film F. Camera 12 includes an imaging element, a lens, and the like. As the imaging element, a CCD (Charge Coupled Device), a CMOS (Complementary Metal Oxide Semiconductor), or the like is used. Camera 12 is an area sensor that generates two-dimensional image data from the output signals of each imaging element. Camera 12 detects at least diffuse light of the light irradiated by light source 11 and reflected by film F.

[0026] In the first embodiment, camera 12 has a shooting range that spans the entire width of film F. That is, camera 12 simultaneously reads the entire range in the width direction of film F in one shooting session. Camera 12 may be one that detects light in the visible light region or one that detects light in the infrared region.

[0027] The information processing device 20 includes a control unit 21, I / F (Interface) 22, 23, a communication unit 24, a display unit 25, an operation unit 26, a memory unit 27, a timing unit 28, a film dimension measurement unit 29, etc. Each unit constituting the information processing device 20 is connected via a bus. The information processing device 20 is composed of a PC (Personal Computer) or the like.

[0028] The control unit 21 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), etc. The control unit 21 comprehensively controls the processing operations of each unit of the information processing device 20 (film inspection device 10). Specifically, the CPU reads out various processing programs stored in the storage unit 27, expands them in the RAM, and performs various processes in cooperation with the programs.

[0029] The I / F 22 is an interface for connecting to the camera 12. The I / F 22 transmits control signals to the camera 12, and receives image data (first image data) obtained by photographing the film F from the camera 12. The I / F 23 is an interface for connecting to the light source 11. The I / F 23 transmits, to the light source 11, a control signal for controlling the emission of light by the light source 11.

[0030] The communication unit 24 is configured by a network interface, etc. The communication unit 24 transmits and receives data to and from external devices connected via a communication network such as a LAN (Local Area Network).

[0031] The display unit 25 includes a monitor such as an LCD (Liquid Crystal Display), etc. The display unit 25 displays various screens according to instructions of a display signal input from the control unit 21.

[0032] The operation unit 26 includes a keyboard having cursor keys, letter and number input keys, various function keys, etc., and a pointing device such as a mouse. The operation unit 26 outputs operation signals input by key operations on the keyboard or mouse operations to the control unit 21.

[0033] The storage unit 27 is configured with a hard disk drive (HDD), a solid state drive (SSD), etc. The storage unit 27 stores various processing programs, data necessary for executing the programs, etc. The storage unit 27 stores setting information (camera setting information) of the camera 12. The camera setting information includes the focal length of the camera lens, the sensor size, the number of pixels of the sensor, the distance between the camera lens and the center of the film roll, and the like. The sensor size is the size of the imaging element of the camera 12. Here, the length of the imaging element in the width direction of the film F is used as the sensor size. The sensor pixel number is the number of pixels of the imaging element of the camera 12. Here, the number of pixels of the imaging element in the width direction of the film F is used as the sensor pixel number. The film roll is a roll-shaped portion around which the film F is wound.

[0034] The clock unit 28 has a clock circuit (RTC: Real Time Clock), and uses this clock circuit to clock the current date and time and output it to the control unit 21.

[0035] Film dimension measuring unit 29 measures the dimensions of film F while film F is being wound. That is, film dimension measuring unit 29 acquires dimensional information (film dimension information) of film F at the inspection timing. The film dimension information includes the film width, winding diameter, etc. The film width is the length in the width direction of the film F. The film dimension measuring unit 29 measures the film width using a laser distance sensor, a photoelectric sensor, or the like. The winding diameter is the diameter of the roll of film F. Film dimension measuring unit 29 measures the winding length of film F using a rotary encoder or the like. The winding length is the length of the wound film F in the longitudinal direction. Film dimension measuring unit 29 measures the film thickness of film F using a laser, ultrasonic waves, or the like. Film dimension measuring unit 29 calculates the winding diameter from the winding length and film thickness. Note that control unit 21 may obtain the winding length and film thickness from film dimension measuring unit 29 and calculate the winding diameter.

[0036] It should be noted that film dimension measuring unit 29 may use data measured by an external measuring device during the winding process of film F. In addition, since the film width does not change while the film F of the same width is being wound, a value that is preset for the film F being wound may be used.

[0037] The control unit 21 extracts an inspection area at the inspection timing from the first image data (entire photographed area) obtained from an output signal of the camera 12 (optical sensor). That is, the control unit 21 functions as an inspection area extraction unit.

[0038] The inspection timing may be set so that all areas along the entire length of the film F can be inspected without omission (full range inspection mode). Alternatively, the inspection timing may be set so that only a portion of the area along the length of the film F is inspected (sampling inspection mode). In the full range inspection mode, all areas on the film F are included in the inspection area at one of the inspection timings. The user may be able to specify whether the inspection will be performed in the full range inspection mode or the sampling inspection mode.

[0039] The control unit 21 extracts the inspection area by optical calculation using the film dimension information and camera setting information at the inspection timing.

[0040] The control unit 21 performs data processing on the second image data corresponding to the inspection area extracted from the first image data, thereby detecting defects in the film F. That is, the control unit 21 functions as a defect detection unit. For example, the control unit 21 detects the type, position, and intensity of defects in the film F.

[0041] The data processing includes image processing, defect determination processing, and quantitative evaluation processing. Here, the image processing is processing of image data obtained from an output signal of the camera 12. Specifically, the second image data corresponding to the inspection area extracted from the first image data is the processing target. The defect determination process is a process for determining defects based on the data after image processing. The quantitative evaluation process is a process for quantitatively evaluating defects based on the data after image processing.

[0042] The control unit 21 causes the display unit 25 to display the inspection area extracted from the first image data and / or information on the detected defects. That is, the control unit 21 functions as a display control unit.

[0043] Next, the arrangement of the light source 11 and the camera 12 will be described with reference to FIG. When observing the specular light irradiated by the light source 11 and reflected by the object under inspection (film F), the luminance component of the light source 11 is observed. In both visual observation by a human and machine vision such as camera 12, it is difficult to detect defects in film F using specular light. If camera 12 is placed at position P1 where it receives specularly reflected light (brightness component) from light source 11, camera 12 will basically capture an image of the inspection object with light source 11 reflected in it (an inspection object overlapping with the image of light source 11), which is not suitable for detecting defects.

[0044] The position of the camera 12 is to avoid position P1 where the camera 12 receives specularly reflected light from the light source 11. In addition, position P2, which is in the vicinity of position P1, is also not very suitable for detecting defects. Here, the vicinity of position P1 refers to an area that may be affected by specularly reflected light from the light source 11. The camera 12 may be disposed at any position (positions P3 to P6, etc.) other than position P1 and its vicinity (position P2, etc.), that is, at any position where it receives diffuse light out of the light reflected by the inspection object.

[0045] Furthermore, since photographing the surface of the object to be inspected from a low angle (positions P5, P6, etc.) is likely to result in aberration problems, it is desirable to install camera 12 at a position where the angle between the photographing direction of camera 12 and the surface to be inspected is equal to or greater than a predetermined value.

[0046] 3 is a perspective view showing the positional relationship of light source 11 and camera 12 with respect to roll-shaped film F. Light source 11 and camera 12 are disposed close to each other, and the roll surface of film F (overlapping films F) being wound up is the subject of inspection.

[0047] Furthermore, if the background (specular reflection) is reflected on the film surface in the observation range (corresponding to the inspection area) of the camera 12, this can cause a decrease in the accuracy of defect determination. Therefore, it is desirable to provide an anti-reflection plate in the film inspection device 10 so that specularly reflected light from the inspection area of ​​the film F does not enter the camera 12.

[0048] From immediately after the start of winding of film F until the end of winding, the winding diameter of the roll gradually increases. 4, roll 41 immediately after the start of winding has the smallest winding diameter during the time it is wound up to roll 42 at the end of winding. Therefore, immediately after the start of winding, roll 41 has the largest area subject to glare due to specular reflection (area where light specularly reflected from the film surface enters camera 12). Straight lines 43, 44 are drawn from camera 12 toward both ends 41A, 41B in the circumferential direction of the observation range of roll 41 immediately after winding begins, and straight lines 45, 46 (called reflection lines) are drawn in the directions in which these two straight lines 43, 44 are specularly reflected by the film surface. A range 47 between these two reflection lines 45, 46 is reflected in the observation range of camera 12. For this reason, it is preferable to have a reflection prevention plate 48 spanning this range 47.

[0049] When using images for analysis, it is necessary to ensure that there is a sufficient amount of light when taking the image. If the amount of light is insufficient, the difference between normal and abnormal areas on the film F will be small, and there is a concern that the accuracy of defect determination will decrease. To address such problems, the anti-glare plate 48 is made of a diffusion plate with uniform surface roughness and color. The surface of the anti-glare plate 48 facing the film F is preferably a highly reflective color such as white, blue, or gray.

[0050] Next, the operation of the film inspection device 10 will be described. 5 is a flowchart showing the film defect detection process executed by film inspection device 10. The film defect detection process is a process for detecting defects in overlapping films F during winding. This process is realized by software processing through cooperation between control unit 21 and a program stored in storage unit 27. Film F is wound so that the transport speed (linear speed) in the longitudinal direction of film F is constant.

[0051] First, the control unit 21 controls the light source 11 via the I / F 23 to irradiate light onto the film F (step S1). The control unit 21 controls the camera 12 via the I / F 22 to capture an image of an area of ​​the film F including the inspection area (step S2).

[0052] The control unit 21 acquires the two-dimensional first image data (the entire photographed area) generated by the camera 12 via the I / F 22 (step S3). Control unit 21 stores the acquired first image data in storage unit 27 (step S4). Specifically, control unit 21 acquires the current date and time from clock unit 28, sets it as the examination date and time, and stores the first image data in storage unit 27 in association with this examination date and time. Image data obtained in one shooting (first image data) targets a part of the area in the circumferential direction of film F being wound up. Therefore, the examination date and time associated with the first image data corresponds to a position in the longitudinal direction of film F.

[0053] Next, the control unit 21 performs a first inspection area extraction process on the first image data acquired from the camera 12 (step S5).

[0054] Now, the first inspection area extraction process will be described with reference to FIG. Control unit 21 acquires film dimension information and camera setting information at the inspection timing (step S11). Specifically, control unit 21 acquires film dimension information from film dimension measurement unit 29. The film dimension information includes film width, winding diameter, etc. Control unit 21 also acquires camera setting information stored in memory unit 27. The camera setting information includes the focal length of the camera lens, the sensor size, the number of pixels of the sensor, the distance between the camera lens and the center of winding of the film roll, etc.

[0055] Next, the control unit 21 extracts an inspection area from the first image data by optical calculation using the film dimension information and camera setting information (step S12).

[0056] 7 shows an example of a captured image 51. The captured image 51 corresponds to the entire area of ​​the first image data obtained by the camera 12. The photographed image 51 shows a film roll 52 (roll-shaped film F) and a winding shaft 53. The photographed image 51 includes an inspection area 54 at the inspection timing (time of photographing). The photographed image 51 may also include an area 55 where a light source is reflected.

[0057] As the winding diameter of the film roll 52 increases, the distance between the camera 12 and the surface of the film F decreases, and the size and position of the inspection area 54 in the captured image 51 changes.

[0058] 8 shows examples of captured images 51A, 51B, and 51C with different inspection timings. The symbols used in each of the captured images 51A, 51B, and 51C are the same as those in FIG. 7. Of the captured images 51A, 51B, and 51C, the wound diameter of the film roll 52 is the smallest in the captured image 51A, and the wound diameter of the film roll 52 is the largest in the captured image 51C. The larger the wound diameter of the film roll 52, the larger the size of the inspection area 54.

[0059] 9, a method for calculating the boundary position when extracting the inspection area 54 from the captured image 51 will be described. Since the position of the inspection area 54 changes depending on the winding diameter of the film roll 52, the range to be cut out is changed for each winding diameter.

[0060] As a preliminary step, the upper cut-out position 56 and the lower cut-out position 57 for each winding diameter are recorded from images previously acquired (from the start to the end of the roll). 10A shows the relationship between the winding diameter [mm] of the film roll 52 and the cut-out position [pixel position]. A linear approximation formula is created from this relationship. Cut-out top position [pixel position] = 0.24 x winding diameter [mm] + 338

[0061] 10B shows the relationship between the winding diameter [mm] of the film roll 52 and the cut-out lower position [pixel position]. A linear approximation formula is created from this relationship. Cut-out bottom position [pixel position] = 0.34 × winding diameter [mm] + 325

[0062] The control unit 21 inputs the winding diameter into the created linear approximation equation for the inspection area 54 on the captured image 51, and obtains the cut-out upper position 56 and the cut-out lower position 57. In other words, the control unit 21 uses the winding diameter to calculate the pixel positions of the ends of the inspection area 54 in the circumferential direction of the film roll 52.

[0063] Next, the control unit 21 obtains cut-out width positions 58, 59 in the width direction of the film F for the inspection area 54 on the photographed image 51 (see FIG. 9).

[0064] As shown in Fig. 11, the photographing range [mm] in the width direction of film F on the surface of film F at the inspection timing is h range The distance [mm] between the camera lens and the center of the film roll 52 is L r The radius [mm] of the film roll 52 at the time of inspection is r. The focal length [mm] of the camera lens is f, and the sensor size [mm] in the width direction of the film F is s. sThen, h range is calculated by the formula (1). Note that the winding radius r is half the winding diameter (roll diameter) included in the film dimension information.

number

[0065] The resolution [pix / mm] at the inspection timing is resol, and the number of sensor pixels [pix] in the width direction of film F is s p Then, resol can be calculated by equation (2).

number

[0066] The amount of cutting in the width direction [pix] is t pix If the film width [mm] at the inspection timing is w (see Figure 11), then t pix is calculated by equation (3). pix is the number of pixels corresponding to the film width w.

number

[0067] The control unit 21 determines whether the first image data is within ±t from the image center position in the width direction of the film F. pix The position of / 2 is set as the cut-out width position 58, 59. That is, the control unit 21 controls the winding radius r, the film width w, the focal length f, the distance L between the camera lens and the winding center r , sensor size s s , sensor pixel count s p is used to calculate the pixel position of the end of the inspection area 54 in the width direction of the film roll 52.

[0068] The control unit 21 extracts the inspection area 54 from the first image data (captured image 51) based on the upper cut-out position 56, the lower cut-out position 57, and the cut-out width positions 58, 59 obtained in this manner.

[0069] Next, the control unit 21 stores the second image data corresponding to the examination region in the storage unit 27 (step S13). Specifically, the control unit 21 stores the second image data in the storage unit 27 in association with the examination date and time that was associated with the first image data in step S4.

[0070] After the first inspection area extraction process, returning to FIG. 5, the control unit 21 performs data processing on the second image data corresponding to the extracted inspection area (step S6).

[0071] Now, with reference to FIG. 12, the data processing will be described. The control unit 21 divides the second image data into a plurality of regions in the width direction of the film F (step S21). Specifically, the control unit 21 divides the second image data into n regions a1 to an.

[0072] Next, the control unit 21 acquires image data of the region a1 (step S22) and performs mathematical processing on the image data of the region a1 (step S23). Appropriate mathematical processing is prepared according to the type of defect to be detected (gauge band, vertical wrinkle, diagonal wrinkle, etc.).

[0073] The mathematical processing includes preprocessing, enhancement processing, signal processing, image feature extraction, and the like. As pre-processing, image cropping, low-pass filter, high-pass filter, Gaussian filter, median filter, bilateral filter, morphological transformation, color transformation (L * a * b * , sRGB, HSV, HSL), contrast adjustment, noise removal, restoration of blurred and blurred images, mask processing, Hough transform, projection transformation, etc. Examples of enhancement processing include a Sobel filter, a Scharr filter, a Laplacian filter, a Gabor filter, and a Canny method. Signal processing includes basic statistics (maximum, minimum, average, median, standard deviation, variance, quartile), square root of the sum of squares, difference, sum, product, ratio, distance matrix calculation, differential and integral calculus, threshold processing (binarization, adaptive binarization, etc.), Fourier transform, wavelet transform, peak detection (peak value, number of peaks, half-width, etc.). Examples of image feature extraction include template matching and SIFT features. The mathematical processing in step S23 corresponds to the "image processing" performed on the second image data.

[0074] Specifically, in order to detect the gauge band, values ​​such as a low-pass filter, a Gaussian filter, a median filter, a morphological transformation, and mask processing are used as preprocessing, and basic statistics and threshold processing are used for signal processing. For vertical and diagonal wrinkles, a high-pass filter is used for pre-processing, a Sobel filter or a Gabor filter is used for enhancement processing, and the square root of the sum of squares, Fourier transform, basic statistics, etc. are used for signal processing.

[0075] Next, the control unit 21 performs threshold processing on the value (feature amount) calculated by the mathematical processing of the image data of the area a1 (step S24). The threshold processing is a process for determining whether or not the defect is a target defect based on one or more threshold values ​​that are predefined for each type of the defect, and for determining the rank (intensity) of the defect. In the threshold process of step S24, determining the presence and type of defects corresponds to a “defect determination process.” Also, in the threshold process of step S24, classifying defects into a plurality of ranks according to the thresholds corresponds to a “quantitative evaluation process.”

[0076] For example, the control unit 21 classifies defects into a plurality of ranks according to a parameter (feature) having a value of 1 to 100. Specifically, the control unit 21 classifies defects as follows: when the parameter value is 1 to 10, it is rank 1; when the parameter value is 11 to 30, it is rank 2; when the parameter value is 31 to 60, it is rank 3; and when the parameter value is 61 to 100, it is rank 4.

[0077] The same process is performed for the areas other than the area a1. For example, the process of steps S25 to S27 performed for the image data of the area an is the same as the process of steps S22 to S24.

[0078] After processing each of the regions a1 to an, the control unit 21 integrates the results for each of the regions a1 to an (step S28), and the data processing ends. Specifically, the control unit 21 generates data that associates the type and rank of the detected defects with each region (each position in the width direction of the film F).

[0079] Returning to FIG. 5, after the data processing, the control unit 21 stores the processing results of the first inspection area extraction processing and / or data processing in the storage unit 27 and causes the display unit 25 to display the processing results (step S7). For example, the control unit 21 causes the display unit 25 to display the inspection area extracted in the first inspection area extraction processing. The control unit 21 also causes the storage unit 27 to store the processing results of the data processing in association with the inspection date and time corresponding to the second image data. The processing results of the data processing include the type and rank of defects detected in each area (position) divided in the width direction. The control unit 21 causes the display unit 25 to display information such as the type, position, rank (strength) of the detected defects. This completes the film defect detection process.

[0080] 5 is a process showing an inspection of an area corresponding to one shooting. As the film F moves while being wound, the inspection area in the longitudinal direction of the film F is changed and the film defect detection process is repeated to obtain inspection results at each position in the longitudinal direction of the film F. 5, the case where the first inspection area extraction process (see FIG. 6) and the data processing (see FIG. 12) are performed each time an image is captured has been described. Alternatively, the control unit 21 may store image data obtained by multiple captures, and then perform the first inspection area extraction process and data processing collectively.

[0081] 13 shows an example of inspection result screen 251 displayed on display unit 25. Inspection result screen 251 displays the inspection results of film F being wound in real time. Inspection result screen 251 includes an inspection area image field 251A and a defect rank field 251B.

[0082] In the inspection area image field 251A, images of the inspection areas extracted at each inspection timing are displayed. Specifically, the control unit 21 arranges each inspection area that is long in the width direction of the film F along the winding length (corresponding to the inspection time). Here, the control unit 21 adjusts the magnification of the second image data corresponding to each inspection area so that the width direction length of each inspection area is aligned at each inspection timing. In the inspection area image field 251A, the horizontal direction indicates the position in the width direction of the film F. The vertical direction indicates the inspection timing based on the longitudinal position (winding length) of the film F. Note that the inspection targets the entire range of the film F, but the display of the inspection results in FIG. 13 is performed every 100 m of the winding length.

[0083] The defect rank column 251B displays the maximum rank of defects that occurred at each inspection timing. The control unit 21 graphs the maximum rank of defects detected from the inspection area obtained at each inspection timing, with the horizontal axis representing the winding length (corresponding to the inspection time). Here, the defect rank is shown for each type of defect (gauge band, diagonal wrinkle). The defect rank column 251B also displays defect ranks for every 100 m of winding length. The displayed defect rank value may be the maximum value of the rank of defects detected from the inspection area obtained at the inspection timing corresponding to the winding length. Alternatively, the maximum value of the rank of defects detected by inspection of 100 m of winding length, including inspection areas not displayed in the inspection area image column 251A, may be displayed in the defect rank column 251B.

[0084] In addition to the above examples, the control unit 21 may cause the display unit 25 to display alive monitoring information of the film inspection apparatus 10. For example, the control unit 21 causes the display unit 25 to display status information (operating / stopped / error) of each inspection process in the film inspection apparatus 10. Furthermore, when an error occurs, the control unit 21 may notify the error by an alert sound or the like. Furthermore, the control unit 21 may cause the display unit 25 to display a search screen for past test results, search results, and the like.

[0085] As described above, according to the first embodiment, the control unit 21 extracts an inspection area at the inspection timing from the first image data obtained from the output signal of the camera 12 (optical sensor). This allows the control unit 21 to extract an inspection area whose size and position change within the first image data. Furthermore, since the control unit 21 detects defects in the film F by targeting the second image data corresponding to the inspection area, defects can be detected efficiently. Therefore, the film inspection device 10 can easily detect defects that occur on the film F without requiring manual monitoring or operation.

[0086] Furthermore, the control unit 21 extracts the inspection area by optical calculation using the film dimension information and camera setting information at the inspection timing, which allows the control unit 21 to extract the inspection area with simple calculation.

[0087] Furthermore, by providing the anti-glare plate 48 in the film inspection device 10, it is possible to prevent a decrease in the accuracy of defect detection due to reflection in the inspection area. The anti-glare plate 48 prevents specular reflection of light reflected in the inspection area of ​​the film F from entering the camera 12, thereby preventing reflection of the background. Furthermore, light reflected by the diffusion plate (anti-glare plate 48) reaches the film F and is diffused by the film F, and the diffused light reaches the camera 12. This ensures a greater amount of light compared to the case where the anti-glare plate 48 is not provided.

[0088] Furthermore, the control unit 21 causes the display unit 25 to display the extracted inspection area, information on detected defects, and the like. This allows the user to check the quality of the film F in real time. This makes it possible to provide feedback to the production process in real time when quality deteriorates, and to improve the process through data analysis, leading to improved quality and stable production. Furthermore, the control unit 21 stores the inspection area images, defect detection results, and the like in the storage unit 27, making it possible to search and use past data.

[0089] [Second embodiment] Next, a second embodiment to which the present invention is applied will be described. Depending on the constraints on the installation position of the camera or the angle of view of the camera lens, it may be necessary to use multiple cameras in order to ensure a sufficient shooting range in the width direction of the film. The second embodiment differs from the film inspection device 10 shown in the first embodiment in that it is equipped with multiple cameras. The configuration and processing characteristic of the second embodiment will be described below. Here, FIG. 1 will be used, and a description of the same configuration as in the first embodiment will be omitted. It is assumed that the multiple cameras, including the lenses, are identical and are installed at the same distance from the film roll.

[0090] As shown in FIG. 14, the film inspection apparatus 10 in the second embodiment includes three cameras 12. The shooting range of each of the three cameras 12 covers the entire width of the film F. As the winding diameter of the film roll 52 changes, the distance between the cameras 12 and the film surface changes. As a result, the amount of overlap between the first image data (shooting ranges) obtained by adjacent cameras 12 changes according to the winding diameter of the film roll 52. Specifically, the overlap amount d1 of the shooting ranges when the winding radius r1 of the film roll 52 is relatively small is larger than the overlap amount d2 when the winding radius r2 is relatively large.

[0091] The camera setting information stored in the storage unit 27 further includes the inter-camera distance between the multiple installed cameras 12. The inter-camera distance is the distance between adjacent cameras 12 in the width direction of the film F.

[0092] The control unit 21 (inspection area extraction unit) removes overlapping areas between a plurality of first image data obtained from each of the plurality of cameras 12 (optical sensors). Note that the control unit 21 removes the overlapping areas so that each area included in the overlapping area remains in one of the first image data. The control unit 21 combines, in the width direction of the film F, the multiple first image data from which the overlapping areas have been removed. The control unit 21 extracts an inspection area at the inspection timing from the combined first image data. It should be noted that either the process of removing the overlapping area or the process of extracting the inspection area may be performed first.

[0093] The control unit 21 extracts the inspection area and removes overlapping areas by optical calculation using film dimension information and camera setting information at the inspection timing.

[0094] Next, the operation of film inspection apparatus 10 according to the second embodiment will be described. Here, differences from the film defect detection process (see FIG. 5) according to the first embodiment will be described. In the second embodiment, the control unit 21 acquires first image data from each of the multiple cameras 12 in step S3 of the film defect detection process. Moreover, the control unit 21 performs a second inspection area extraction process shown in FIG. 15 instead of step S5 of the film defect detection process.

[0095] In the second inspection area extraction process, first, control unit 21 acquires film dimension information and camera setting information at the inspection timing (step S31). Specifically, control unit 21 acquires film dimension information from film dimension measurement unit 29. The film dimension information includes film width, winding diameter, etc. Control unit 21 also acquires camera setting information stored in memory unit 27. The camera setting information includes the focal length of the camera lens, sensor size, sensor pixel count, distance between the camera lens and the winding center of the film roll, distance between the cameras, etc.

[0096] Next, the control unit 21 removes the overlapping area from the first image data by optical calculation using the film dimension information and the camera setting information (step S32).

[0097] A method for determining the amount of overlap of the shooting ranges will be described with reference to FIG. The overlap amount expressed in length (distance) [mm] is mm , the photographing range [mm] in the width direction of film F on the surface of film F at the inspection timing is h range , the distance between the cameras [mm] is L c Then, d mm is calculated by equation (4).

number

[0098] The overlap amount expressed in pixels [pix] is d pix , where resol is the resolution [pix / mm] at the inspection timing, d pix is calculated by equation (5).

number

[0099] That is, the control unit 21 controls the winding radius r, the focal length f, and the distance L between the camera lens and the winding center.r , sensor size s s , sensor pixel count s p , distance between cameras L c Using the overlap amount d pix Calculate.

[0100] As shown in FIG. 17A, the control unit 21 removes overlapping areas from each of the first image data 61, 62, and 63 obtained from each of the three cameras 12. Specifically, the control unit 21 removes overlapping area 61A from the first image data 61. The control unit 21 removes overlapping areas 62A and 62B from the first image data 62. The control unit 21 removes overlapping area 63A from the first image data 63. Here, the overlapping areas 61A, 62A, 62B, and 63A are represented by d pix For each of the first image data obtained by each camera 12, d pix By removing the overlapping range by 1 / 2, data leakage in the width direction of film F is prevented. Fig. 17B shows the first image data 61, 62, and 63 after the overlapping areas have been removed. Note that in Fig. 17A and Fig. 17B, flower-shaped marks and star-shaped marks are added to make the removed areas easier to understand, but they are not actually shown in the images.

[0101] Next, the control unit 21 combines the first image data from which the overlapping areas have been removed in the width direction of the film F (step S33). The image data generated by the combination can be treated in the same way as a continuous piece of image data (the first image data in the first embodiment).

[0102] Next, the control unit 21 extracts an inspection area from the combined image data by optical calculation using the film dimension information and the camera setting information (step S34). The process of step S34 is similar to the process of step S12 of the first inspection area extraction process (see FIG. 6) except that the combined image data is used.

[0103] Next, the control unit 21 stores the second image data corresponding to the inspection area in the storage unit 27 (step S35). The process of step S35 is similar to the process of step S13 of the first inspection area extraction process (see FIG. 6). This is the end of the second inspection area extraction process.

[0104] As described above, according to the second embodiment, in the film inspection device 10 equipped with the multiple cameras 12, the control unit 21 removes overlapping areas between the multiple first image data obtained from each of the multiple cameras 12. This allows the control unit 21 to obtain an inspection area without excess or deficiency from the multiple first image data.

[0105] The above-described embodiments are merely examples of the film inspection device, film inspection method, and program according to the present invention, and the present invention is not limited to these. The detailed configurations and operations of the components constituting the device may be modified as appropriate without departing from the spirit of the present invention. For example, the configurations or processes in the above embodiments may be combined.

[0106] In each of the above embodiments, a description has been given of the case where defects that occur when the film F is wound into a roll are detected. Alternatively, the film inspection device 10 may detect defects that occur when the film F cut to a predetermined size is stacked.

[0107] Furthermore, the control unit 21 may use a learning result obtained by machine learning when detecting defects from the second image data corresponding to the inspection area. Specifically, in the defect determination process, the control unit 21 uses a machine learning model constructed with the post-image processing data as an explanatory variable and the result of a defect determination made by a human as a target variable. In addition, in the quantitative evaluation process, the control unit 21 uses a machine learning model constructed with the post-image processing data as explanatory variables and the result of the quantitative evaluation of defects by a human as a target variable. The control unit 21 quantitatively evaluates the defects based on the post-image processing data using the machine learning results, and classifies the defects into a plurality of ranks. It is not necessary to execute the defect determination process and the quantitative evaluation process separately. The control unit 21 may execute the defect determination process and the quantitative evaluation process together by using the results of machine learning, with the post-image processing data as input data and the combinations of the types and ranks of defects evaluated by humans as output data.

[0108] The computer-readable medium for storing the program for executing each process is not limited to the above examples, and a portable recording medium may be used. A carrier wave may be used as a medium for providing the program data via a communication line. [Explanation of symbols]

[0109] 10 Film inspection equipment 11 Light source 12 Camera 20 Information processing device 21 Control section 25 Display section 26 Control section 27 Memory section 28 Timing section 29 Film Dimension Measurement Section 48 Anti-reflection plate 51 Images 52 Film Roll 53 Winding shaft 54 Inspection Area 56 Cut-out top position 57 Cut-out bottom position 58,59 Cutting width position 251 Test result screen F Film

Claims

1. A film inspection device that optically inspects for defects occurring on overlapping films, comprising: A light source that irradiates the film with light; an optical sensor that detects at least diffused light among the light reflected by the film; an inspection area extraction unit that extracts an inspection area at an inspection timing from first image data obtained from an output signal of the optical sensor; a defect detection unit that detects the defect by performing data processing on second image data corresponding to the inspection area; A film inspection device comprising:

2. A plurality of the optical sensors are provided, The film inspection device according to claim 1 , wherein the inspection area extraction section further removes overlapping areas between the plurality of first image data obtained from the plurality of optical sensors.

3. 3. The film inspection device according to claim 1, wherein the inspection area extraction section performs processing by optical calculation using dimensional information of the film at the inspection timing and setting information of the optical sensor.

4. The film is wound into a roll, 3. The film inspection apparatus according to claim 1, wherein the film inspection apparatus inspects for defects that occur during winding of the film.

5. 3. The film inspection device according to claim 1, wherein the optical sensor is an area sensor.

6. a reflection prevention plate is provided so that specular reflection light of the light reflected in the inspection area of ​​the film does not enter the optical sensor; 3. The film inspection device according to claim 1, wherein the anti-glare plate is a diffusion plate.

7. 3. The film inspection device according to claim 1, further comprising a display control unit that causes a display unit to display information about the inspection area extracted by the inspection area extraction unit and / or information about defects detected by the defect detection unit.

8. A film inspection method for optically inspecting for defects occurring on overlapping films, comprising the steps of: irradiating the film with light by a light source; detecting at least diffuse light of the light reflected by the film with an optical sensor; extracting an inspection area at an inspection timing from first image data obtained from an output signal of the optical sensor by an inspection area extraction unit; detecting the defect by performing data processing on second image data corresponding to the inspection area by a defect detection unit; A film inspection method comprising:

9. A film inspection device that includes a light source that irradiates light onto the overlapping films and an optical sensor that detects at least diffused light among the light reflected by the films, and optically inspects the films for defects, an inspection area extraction unit that extracts an inspection area at an inspection timing from first image data obtained from an output signal of the optical sensor; a defect detection unit that detects the defect by performing data processing on second image data corresponding to the inspection area; A program to function as a

Images