Method for measuring film thickness deviation, film manufacturing method, device for measuring film thickness deviation, and film manufacturing device

Abstract

Provided are technology for measuring a film thickness deviation and technology for manufacturing a film, with which information on a film-thickness deviation can be acquired in a non-contact manner through a simple and safe device configuration. In a method for measuring a film-thickness deviation according to the present invention, a film to be measured is set as an object being measured, a preset background image can be imaged by an imaging means, a reference image in which the background image is captured by the imaging means in a state in which the object being measured is not interposed, or a reference image in which calculations are used to derive how the background image in the state in which the object being measured is not interposed is viewed by the imaging means is acquired, a measurement image, in which the background image is captured by the imaging means with the object being measured interposed between the background image and the imaging means, is acquired, and on the basis of the acquired reference image and measurement image, a displacement amount of the background image in the two images is calculated to obtain the film thickness deviation of the object being measured.

Description

Film thickness deviation measuring method, film manufacturing method, film thickness deviation measuring device, and film manufacturing device

[0001] The present invention relates to a technology for non-contact measurement of thickness deviation of a film that transmits visible light, and to the production of a film using the technology for measuring thickness deviation. The present invention particularly relates to a technology suitable for use when the visible light transmittance of the target film is equal to or greater than a specific value. The film that transmits visible light is made of, for example, resin or glass. The film of the present disclosure also includes a sheet.

[0002] The following method is widely used as an in-line, non-contact method for measuring the thickness of a resin film: This method involves transmitting light such as X-rays or infrared rays through the film and measuring the thickness in accordance with the Beer-Lambert law based on the attenuation rate of the transmitted light.

[0003] Furthermore, Patent Document 1 describes an optical interference film thickness measurement method that irradiates a film with white parallel light and calculates the film thickness from the spectral intensity of the reflected light. Patent Document 2 describes a capacitance film thickness measurement method. These film thickness measurement methods often employ a single sensor that moves back and forth in a direction perpendicular to the transport direction of the object being measured. However, with a reciprocating sensor, the faster the film transport speed, the longer the transport-direction interval between film thickness data at the same width position. This makes it difficult to detect quality or operational abnormalities, such as sudden changes in film thickness distribution. Furthermore, there is a problem in that it is difficult to separate film thickness variations in the transport direction from film thickness variations in the width direction using film thickness data.

[0004] A conventional technique for addressing this issue is described in Patent Document 3. Patent Document 3 uses a plurality of X-ray detector elements that are arranged in the width direction and detect the amount of transmitted X-rays. Patent Document 3 also discloses that the film thickness is measured on a straight line in a direction perpendicular to the conveyance direction by tilting the plurality of X-ray detector elements with respect to the width direction in accordance with the conveyance speed and measurement time.

[0005] JP 7-280523 A JP 2019-2918 A JP 2006-275750 A

[0006] "Improvement of the visualization method for density gradients based on the Background Oriented Schlieren method," Junichi Akatsuka et al., Transactions of the Japan Society of Mechanical Engineers (Part B), Vol. 77, No. 784, p. 2391, published December 25, 2011

[0007] According to the film thickness measurement device described in Patent Document 3, multiple X-ray detector elements are tilted with respect to the width direction in accordance with the transport speed and measurement time. This allows Patent Document 3 to measure the thickness on a straight line perpendicular to the film running direction. However, the film thickness measurement device described in Patent Document 3 requires multiple X-ray detector elements. Therefore, the film thickness measurement device described in Patent Document 3 has issues such as increased equipment costs and the need for calibration between X-ray detector elements, which leads to increased maintenance costs.

[0008] The present invention has been made in light of the above-mentioned points, and an object of the present invention is to provide a film thickness deviation measurement technique that can obtain information on film thickness deviation in a non-contact manner using a simple and safe device configuration, and a film manufacturing technique that uses the measurement technique.

[0009] The inventors conducted extensive research to solve the above-mentioned problems. As a result of their extensive research, they came up with the idea of ​​applying the background-directed Schlieren method, which is used to calculate the density gradient of a fluid, to measuring the film thickness deviation of a solid film. They then came up with the idea of ​​utilizing the amount of background displacement (amount of movement) between two images: a reference image captured without the object being measured, and a measurement image captured through the object being measured. This led to the discovery that it is possible to obtain the film thickness deviation of an object being measured using a simple and safe device configuration. The present invention was completed based on this discovery and further research.

[0010] To solve the problem, one aspect of the present invention is a method for measuring film thickness deviation, which comprises using the film to be measured as the object to be measured, making it possible to capture a preset background image with an imaging means, acquiring a reference image which is an image of the background image captured by the imaging means without the object to be measured in between, or a reference image derived by calculation of how the background image appears when viewed by the imaging means without the object to be measured in between, interposing the object to be measured between the background image and the imaging means, acquiring a measurement image which is an image of the background image captured by the imaging means in between, and calculating the amount of displacement of the background image in these two images based on the acquired reference image and measurement image, thereby determining the film thickness deviation of the object to be measured.

[0011] The film thickness deviation is calculated, for example, using the following formula (1).

[0012]

[0013] where t is the film thickness deviation obtained by subtracting the film thickness of the object to be measured at the origin from the film thickness of the object to be measured at the measurement point, and n is the target : refractive index of the object to be measured n ａｉｒ : Refractive index of the atmosphere d: Distance between the background image and the object to be measured Δh: Two-dimensional vector representing the amount of displacement of the background image of interest on the measurement image from the same background image in the reference image v: Unit tangent vector of path C ∫ ｃ Δh·vdl": an operation of performing a line integration of the inner product of the above Δh and v over an integral path C from the origin to the measurement point.

[0014] Another aspect of the present invention is a film manufacturing method in which molten resin is discharged from a die having a plurality of lip gap adjustment means to form a film that is transmissive to visible light, the method comprising the steps of: transporting the film discharged from the die along a predetermined pass line; arranging imaging means on either side of the pass line to capture a background image and a background image; and acquiring a measurement image that is an image of the background image captured by the imaging means through the film; acquiring a reference image that is an image of the background image captured by the imaging means without the film being interposed between the background image and the imaging means, or acquiring a reference image derived by calculation of how the background image would appear through the imaging means without the film being interposed; comparing the acquired reference image with the acquired measurement image and calculating the amount of displacement of the background image in the film width direction between the two images; and operating the lip gap adjustment means based on the calculated film thickness deviation profile so that the film thickness deviation at each point becomes a target value.

[0015] According to an aspect of the present invention, it is possible to obtain information on film thickness deviation in a non-contact manner using a simple and safe device configuration for a film made of resin, glass, or the like. Furthermore, according to an aspect of the present invention, it is possible to obtain information on film thickness deviation in a non-contact manner, for example, over the entire length and width of the film. As a result, according to an aspect of the present invention, when producing a film, it is possible to produce a film with higher accuracy by, for example, quickly reflecting the obtained information on film thickness deviation in the control of the lip gap adjustment means.

[0016] Fig. 1 is a conceptual diagram of an optical system for the BOS method used for quantitative visualization of the density gradient of a fluid. Fig. 2 is a schematic outline configuration diagram of a film thickness deviation measuring device according to an embodiment based on the present invention. Fig. 3 is a diagram showing an example configuration of a calculation unit. Fig. 4 is a outline configuration diagram of a film manufacturing device according to an embodiment based on the present invention. Fig. 5 is a diagram showing the amount of background displacement in the MD direction and the TD direction in an example. Fig. 6 is a diagram showing the film thickness deviation distribution within the film plane in an example.

[0017] Hereinafter, embodiments based on the present invention will be described with reference to the drawings. Note that the embodiments described below are merely examples of the present invention, and the invention is not limited to the following contents.

[0018] (Regarding the Principles of the Present Disclosure) The inventors conceived of applying the background-oriented schlieren method, which is used to calculate the density gradient of a fluid, to measuring the film thickness deviation of a solid film. The background-oriented schlieren method is also called the BOS method. After various studies, the inventors came up with the idea of ​​using the amount of movement (displacement) of the background between two images: a reference image captured without a measurement object and a measurement image captured with a measurement object in between. The inventors then discovered that by calculating the amount of background displacement between the two images, the film thickness deviation of the measurement object can be obtained with a simple and safe device configuration.

[0019] The BOS method is a quantitative visualization technique for the density gradient of a fluid. The BOS method measures the density gradient by utilizing the phenomenon in which, when a background is viewed through a fluid, the background appears shifted due to a difference in refractive index caused by the density gradient (see FIG. 1). In this disclosure, this measurement principle is applied to a solid film to measure the film thickness deviation. The BOS method has two advantages over the conventional Schlieren method. These advantages are that the optical system for measurement is simple and that the density gradient can be quantified.

[0020] A conceptual diagram of the optical system of the BOS method is shown in Figure 1. As shown in Figure 1, the optical system of the BOS method consists of a fluid object 11, a background image 12, and an imaging means 13. Figure 1 shows an apparent optical path 14 from the imaging means 13 and a true optical path 15 due to the object 11. In this BOS method, the background image 12 is captured through the object 11. At this time, the position of the background image on the image captured by the imaging means 13 is shifted in proportion to the density gradient of the object 11.

[0021] In this phenomenon, the total deflection angle ε can be approximated by the ratio of the amount of shift Δh of the background image in the captured image to the distance d between the object to be measured and the background image. In other words, it can be expressed as "ε = Δh / d". The total deflection angle ε is expressed by the following equation (2). In other words, the total deflection angle ε can be found by integrating the density gradient ∂n / ∂x in the direction perpendicular to the optical axis of the refractive index n of the object to be measured over the thickness 2Δz of the object to be measured.

[0022]

[0023] The refractive index n and density ρ have the relationship n=ρG+1, where G is the constant of the Gladstone-Dale law. Therefore, the density gradient is expressed by the following equation (3).

[0024] In this embodiment, the optical system of the BOS method is used to obtain the gradient of the film thickness t (film thickness deviation) of a solid film. In this embodiment, the Gladstone-Dale law is not used. Instead, in this embodiment, it is approximated that the refractive index has a gradient in the x direction only at the portion of the film thickness that is in contact with air. This makes it possible to obtain the following equation (4). Note that the x direction is an arbitrary direction selected from the surface direction of the film being measured. The surface direction of the film is a two-dimensional direction perpendicular to the z direction.

[0025]

[0026] Furthermore, ε=Δh / d. Therefore, the following equation (5) can be obtained from the equation "ε=Δh / d" and equation (4).

[0027]

[0028] Then, by integrating equation (5) in the x direction, the film thickness deviation can be measured. The film thickness deviation is the amount of change in film thickness in the x direction from the set reference position x0. By extending this concept, it is also possible to calculate the film thickness deviation in two dimensions, for example, the x and y directions along the film surface.

[0029] A specific method for determining the film thickness deviation is, for example, a method using the following formula (1). In this specification, the term "reference image" refers to an image of a background image captured by an imaging means without an object to be measured in between. The term "measurement image" refers to an image of a background image captured by an imaging means through an object to be measured.

[0030]

[0031] Here, t: film thickness deviation obtained by subtracting the film thickness of the object to be measured at the origin from the film thickness of the object to be measured at the measurement point, n target : refractive index of the object to be measured n ａｉｒ : Refractive index of the atmosphere d: Distance between the background image and the object to be measured Δh: Two-dimensional vector representing the amount of displacement of the background image of interest on the measurement image from the same background image in the reference image v: Unit tangent vector of path C ∫ ｃ Δh·vdl": an operation of linearly integrating the inner product of the above Δh and v over an integral path C from the origin to the measurement point.

[0032] (Configuration of Film Thickness Deviation Measuring Device) Next, an example of a film thickness deviation measuring device of this embodiment will be described. Fig. 2 shows a schematic configuration of a film thickness deviation measuring device 20 of this embodiment. As shown in the schematic diagram of Fig. 2, the film thickness deviation measuring device 20 of this embodiment includes a background image 22, an imaging means 23 that captures the background image 22, and a calculation unit 24. The object to be measured 21 is a film to be measured.

[0033] <Object to be Measured> When acquiring a measurement image, the object to be measured 21 is placed in front of the background image with the film thickness direction facing the imaging direction. The object to be measured in this embodiment is a film that transmits visible light. Therefore, the imaging means is able to image the background image 22 through the object to be measured 21. In principle, the background image 22 is imaged through the object to be measured 21, and although this depends on the film thickness of the object to be measured, the visible light transmittance of the object to be measured 21 is preferably 10% or more. Furthermore, the visible light transmittance is more preferably 30% or more, and even more preferably 70% or more.

[0034] Furthermore, regardless of the visible light transmittance value, the object 21 preferably has a haze value of 20% or less. Furthermore, the haze value is more preferably 15% or less, and even more preferably 5% or less. In this embodiment, the visible light transmittance is the total light transmittance in the wavelength range of 400 to 700 nm measured using a haze meter. Furthermore, the haze value is the ratio of the total light transmittance to the diffuse transmittance, also measured using a haze meter.

[0035] <Background Image> The background image 22 may be a pattern or design that allows for the detection of the amount of displacement when the position of the background image captured by the imaging means changes depending on whether the object under test 21 is present or not. For example, a spangle pattern may be present on the top surface 25A of the platform 25 at a position corresponding to the background image 22, and the pattern may be captured in detail by the imaging means 23. In this case, the spangle pattern serves as the background image 22, and a separate image is not required. If no suitable pattern is available for the background image 22, a random dot image or a stripe image with a periodic luminance distribution is preferred. The random dot image and stripe image may be black and white or grayscale, or may be a color image painted in three colors, for example, red, green, and blue. The background image 22 may be created by directly drawing random dots on the top surface 25A of the platform 25 or by attaching a paper with a background image printed on it to the top surface 25A.

[0036] Furthermore, a display device for displaying a background image may be provided on the top surface of the stand 25, and the background image may be switched depending on the passage of time or the conditions of the object to be measured. In such a configuration in which a background image is displayed on a display, the image may be switched at high speed to vibrate the background image, thereby improving the effective resolution. The setting of the background image is not limited to these, and various known methods may be used.

[0037] <Imaging Means> The imaging means is a device that captures a background image. A CCD camera, a CMOS camera, or other digital camera can be used as the imaging means 23. The imaging means 23 may also be an area camera or a line camera. However, if it is desired to measure the film thickness deviation in only one specific direction, a line camera may be used as the imaging means 23. In this case, improved processing speed and reduced equipment costs can be expected. The imaging means 23 may also be composed of a set of multiple cameras. In this case, it is possible to reduce the imaging area of ​​the object to be measured that each camera is responsible for. This allows for improved resolution and the use of a narrow-angle camera. In this case, the reduction in the area of ​​the background image 22 can be expected to reduce the scale of the device.

[0038] Furthermore, the camera of the imaging means 23 may be a monochrome camera when the background image 22 is black and white or grayscale. Also, when the background image 22 is a color image, a device configuration in which separate colors are detected by multiple cameras may be adopted, for example, by installing color filters between the imaging means 23 and the object to be measured 21. In this case, using a monochrome camera as the camera of the imaging means 23 is preferable from the viewpoints of cost-effectiveness and spatial resolution.

[0039] <Arrangement and Others> When capturing a measurement image, the object to be measured is interposed between the imaging means and the background image. It is preferable that the object to be measured 21 and the background image 22 be arranged in a parallel relationship from the viewpoint of analytical calculation. However, if parallel arrangement is not possible due to device layout constraints, appropriate corrections can be made during calculation. Furthermore, the greater the separation between the object to be measured 21 and the background image 22, the better the film thickness deviation resolution. For this reason, the opposing distance between the object to be measured 21 and the background image 22 is preferably 30 mm or more, more preferably 50 mm or more, and even more preferably 200 mm or more. On the other hand, the closer the object to be measured 21 and the background image 22 are, the easier it is to focus when capturing the image. For this reason, the opposing distance between the object to be measured 21 and the background image 22 is preferably 2000 mm or less, more preferably 1000 mm or less, and even more preferably 600 mm or less. The distance between the object 21 and the background image 22 is preferably between 30 and 2000 mm, more preferably between 50 and 1000 mm, and even more preferably between 200 and 600 mm. The distance between the object 21 and the background image 22 is preferably determined by the required film thickness deviation resolution and the lens configuration of the imaging means 23.

[0040] It is preferable that the optical axis of the imaging means 23 be perpendicular to the object to be measured 21 and the background image 22 from the viewpoints of analytical calculation and preventing out-of-focus. The distance between the imaging means 23 and the object to be measured 21 is preferably determined taking into account the angle of view and focal length of the imaging means 23. The distance is preferably set so that the object to be measured 21 and the background image 22 do not become out of focus and the range of the object to be measured 21 falls within the imaging range. The axis connecting the object to be measured 21, the background image 22, and the imaging means 23 may be set vertically (perpendicular to the background image) as shown in Figure 2. However, the axis may be set horizontally or at any other angle depending on the orientation of the object to be measured 21, the background image 22, etc., depending on the installation environment and other constraints.

[0041] Other structures (not shown), such as glass windows, may be present between the object to be measured 21 and the background image 22, or between the object to be measured 21 and the imaging means 23. However, it is preferable that the structure have a visible light transmittance of 10% or more so that the object to be measured 21 and the background image 22 can be recognized through the imaging means 23. Furthermore, the visible light transmittance is more preferably 30% or more, and even more preferably 70% or more. Furthermore, if the structure has a film thickness deviation, it is preferable to correct the displacement of the background image in the captured image or the calculated film thickness deviation of the object to be measured 21. However, if the distance between the background image 22 and the structure is short and has only an acceptable effect on the calculated film thickness deviation of the object to be measured 21, correction may not be necessary. It is more preferable that the structure has no film thickness deviation.

[0042] <Calculation Unit> The calculation unit 24 executes a process of calculating the film thickness deviation of the object to be measured based on the images captured by the imaging means. Specifically, the calculation unit 24 executes a process of calculating the amount of displacement of the background image between two images, a reference image and a measurement image, thereby calculating the film thickness deviation of the object to be measured. The amount of displacement of the background image is determined, for example, by the amount of movement of a region selected from within the background image. As shown in FIG. 3, the calculation unit of this embodiment includes a reference image acquisition unit 24A, a measurement image acquisition unit 24B, and a film thickness deviation calculation unit 24C.

[0043] [Reference Image Acquisition Unit] The reference image acquisition unit 24A performs a process of acquiring a reference image. The reference image is an image that serves as a reference for comparison with a measurement image. Examples of the process of acquiring the reference image include a first processing method and a second processing method.

[0044] [First Processing Method] In the first processing method, an image of the background image 22 captured by the imaging means 23 is acquired as a reference image. This reference image is acquired when it is determined that the object to be measured 21 is not interposed between the imaging means and the background image. The reference image is an image that serves as a reference for calibration. Therefore, when determining the film thickness deviation online, it is sufficient to perform this before acquiring information on the film thickness deviation for the object to be measured.

[0045] [Second Processing Method] In the second processing method, an image derived by calculation of how the background image 22 appears when the object under measurement 21 is not present as viewed by the imaging unit 23 is obtained as a reference image. One method for calculating the reference image is to perform projective transformation on the printing data of the background image 22 according to the positional relationship between the imaging unit 23 and the background image 22. Another method for calculating the reference image is to detect an image of a marker on the background image that is located outside the area of ​​the object under measurement in the measurement image. Then, a projective transformation matrix is ​​calculated from the marker position, and the reference image is calculated by projectively transforming the printing data that constitutes the background image 22. In this reference image calculation method, printing data of the background image 22 is used instead of image data of the background image 22. This reference image calculation method converts the printing data into how the object under measurement 21 appears when viewed by the imaging unit. Because the second processing method obtains the reference image through calculation, it is possible to obtain a reference image for calibration regardless of whether the object under measurement 21 is present.

[0046] <Measurement Image Acquisition Unit> The measurement image acquisition unit 24B acquires image data of a background image captured by the imaging means when the object to be measured is present between the background image and the imaging means. The measurement image acquisition unit 24B may acquire a measurement image each time the position of the object to be measured changes relatively. Based on the multiple measurement images acquired in this way, image processing may be performed to integrate the multiple measurement images and convert them into a single measurement image.

[0047] <Film Thickness Deviation Calculation Unit> The film thickness deviation calculation unit 24C calculates the amount of displacement between the background images based on the reference image acquired by the reference image acquisition unit 24A and the measurement image acquired by the measurement image acquisition unit 24B. Then, the film thickness deviation calculation unit 24C calculates the film thickness deviation of the object to be measured from the amount of displacement. The film thickness deviation calculation unit 24C may execute a process to calculate the film thickness deviation each time a measurement image is acquired by the measurement image acquisition unit 24B.

[0048] [Calculation of Background Displacement Amount] Various known methods can be used to calculate the amount of displacement between the background image in the reference image and the background image in the measurement image. Examples of such methods include template matching and wavelet analysis. Template matching is a technique used in particle image velocimetry (PIV). When using template matching, the reference image is divided into small inspection windows, for example, 64 pixels square. Then, the cross-correlation coefficient between the image within the inspection window and an area of ​​the same size as the measurement image is calculated repeatedly for the entire area or a portion of the measurement image. The point with the highest cross-correlation coefficient is calculated as the coordinate of the inspection window. Note that the above description of template matching is an example. An inspection window may be set on the measurement image side, and the scanned image may be used as the reference image. In addition, the inspection windows may overlap, and the size of the inspection window may be rectangular or other shapes other than square. In addition, to reduce the amount of calculation, it is preferable to limit the scanning range by setting a safety factor to the maximum value reasonably considered as the gradient of the film thickness deviation.

[0049] Several methods are known for calculating the cross-correlation coefficient. Among these, the application of the normalized cross-correlation coefficient, particularly the zero-mean normalized cross-correlation coefficient (ZNCC), is preferred from the viewpoint of robustness to brightness fluctuations. However, the normalized cross-correlation coefficient requires a large amount of calculation. Therefore, when calculation speed is particularly important, the application of the sum of absolute differences (SAD) or the sum of squared differences (SSD) is preferred. Furthermore, it is known that erroneous vectors occur in template matching. Various known methods can be used to deal with erroneous vectors. For example, a method of setting a cutoff value for the cross-correlation coefficient, a method of allowing erroneous vectors at a small number of calculation points by focusing on the fact that the film thickness deviation is an integral value, or a combination of these can be easily applied to deal with erroneous vectors.

[0050] Furthermore, wavelet analysis is preferably applied when measuring film thickness deviation in only one specific direction. When calculating the displacement amount using wavelet analysis, the background image 22 is preferably a stripe image composed of waves with a predetermined spatial frequency and having a periodic brightness distribution. For example, a stripe image composed of a brightness distribution expressed as a sine wave adopted over the entire region, with the displacement amount calculated using a value with a safety factor set to the maximum value reasonably considered as the gradient of the film thickness deviation, can be used as the background image. An example of a method using wavelet analysis in the BOS method for fluids is the method described in Non-Patent Document 1. Here, the continuous wavelet transform is expressed by the following equation (6):

[0051]

[0052] Here, the number of elements in one column is represented by n, the i-th luminance is represented by x(i), and the mother wavelet is represented by Ψ(i). The conjugate complex number of the wavelet function is represented by Ψ with an overline (superscript bar). The inverse of the scale parameter a corresponds to the frequency. The shift parameter b corresponds to the spatial coordinate. Various functions have been proposed for the mother wavelet Ψ(i), including, for example, the Morlet complex wavelet function represented by the following equation (7).

[0053]

[0054] In continuous wavelet transform, one-dimensional luminance distributions along the stripe direction are extracted from the reference image and the measurement image. Then, continuous wavelet transform can be performed to obtain intensity versus frequency and spatial coordinates. The amount of background displacement can be calculated by calculating the phase of the luminance distribution at each spatial coordinate from the spatial distribution of intensity obtained by the wavelet transform, and then calculating the amount of background displacement from the phase difference between the reference image and the measurement image. However, it is preferable to use a complex wavelet function, which allows direct acquisition of phase information in addition to intensity during continuous wavelet transform, to calculate the amount of background displacement from the phase difference between the reference image and the measurement image. Multiplying the phase difference by the period of the luminance distribution determines the amount of displacement Δh on the background image 22. Furthermore, if a specific function other than a sine wave is used to represent the luminance distribution of the background image 22, it is preferable to use a complex mother wavelet similar to that function. Furthermore, particularly when the frequency variation of the stripes due to film thickness deviation is not very large, a constant window size relative to the frequency does not have a significant effect. Therefore, using a short-time Fourier transform instead of a continuous wavelet transform simplifies implementation.

[0055] [Calculation of Film Thickness Deviation] The amount of background displacement Δh obtained as described above can be used to calculate the film thickness deviation, for example, using the following equation (8), which sets an integral path that first proceeds from the origin in the longitudinal direction of the film being measured and then in a direction perpendicular to the outward direction.

[0056]

[0057] Here, t(x, y): film thickness deviation obtained by subtracting the film thickness of the object to be measured at the origin (0, 0) from the film thickness of the object to be measured at the coordinates (x, y) n ｔａｒｇｅｔ : refractive index of the object to be measured n ａｉｒ : is the refractive index of the atmosphere, d: the distance between the background image and the object to be measured, Δh: a two-dimensional vector representing the amount of displacement of the background image of interest in the measurement image from the same background image in the reference image, Δhx: the x-direction component of the amount of displacement of the background, and Δx: the x-direction length per pixel.

[0058] In the above equation (8), the integral path is specifically written out. However, more generally, it can be expressed in the form of the following equation (1), for example.

[0059]

[0060] Here, t: film thickness deviation obtained by subtracting the film thickness of the object to be measured at the origin from the film thickness of the object to be measured at the measurement point, n target : refractive index of the object to be measured n ａｉｒ : Refractive index of the atmosphere d: Distance between the background image and the object to be measured Δh: Two-dimensional vector representing the amount of displacement of the background image of interest on the measurement image from the same background image in the reference image v: Unit tangent vector of path C ∫ ｃ Δh·vdl": an operation of performing a line integration of the inner product of the above Δh and v over an integral path C from the origin to the measurement point.

[0061] (Functions and Others) The film thickness deviation of the object to be measured obtained by the above method can be displayed by a known visualization method. When the film thickness deviation is obtained only in one specific direction, it is preferable to display it as a line graph. Furthermore, when the film thickness deviation is obtained in an in-plane direction, it is preferable to display it as an overlaid line graph or a heat map. In this case, the film thickness deviation is, for example, the film thickness of the object to be measured at the measurement point minus the film thickness of the object to be measured at the origin. For this reason, it may be more preferable to display the film thickness as an absolute value by adding the film thickness of the object to be measured at the origin obtained by an appropriate method. Examples of methods for obtaining the film thickness of the object to be measured at the origin include a method of measuring only one point using a known film thickness meter such as a contact type or transmission type, and a method of adopting a nominal film thickness regardless of the actual film thickness.

[0062] (Film Manufacturing Apparatus (Film Manufacturing Method)) Next, the film manufacturing apparatus (film manufacturing method) of this embodiment will be described. FIG. 4 shows a schematic configuration of a film manufacturing apparatus 30 of this embodiment. The film manufacturing apparatus 30 of this embodiment has an extruder 35, a T-die 36, a cooling roll 37, a stretching machine 38, and a winder 39. The T-die 36 has a plurality of lip gap adjustment means 361. The film manufacturing apparatus 30 of this embodiment also has a film thickness deviation measuring device. The film thickness deviation measuring device includes the background image 32, the imaging means 33, and the calculation unit 34, which are configured as described above. Reference numeral 40 denotes a lip gap adjustment unit.

[0063] <Extruder> The extruder 35 is equipped with one or more feeders or hoppers, and melts and kneads multiple types of resins and fillers as necessary. The extruder 35 can be either a single-screw extruder or a twin-screw extruder, and may be a multi-stage extruder. Furthermore, multiple extruders 35 may be provided and connected to a T-die 36. Note that it is preferable to provide a filter, a vent, or a gear pump to improve the quality of the extruded resin and stabilize the extrusion rate. The extruder 35 may have a known configuration.

[0064] <T-die> The T-die 36 discharges the molten resin extruded from the extruder 35 through a gap and continuously molds a resin film of the target thickness. When the film to be manufactured has a multilayer structure, either the feed block method or the multi-manifold method can be applied. The feed block method is a method in which the molten resins are joined together before the T-die. The multi-manifold method is a method in which the single layers are spread inside the T-die and then joined together near the lip.

[0065] [Lip Gap Adjusting Means 361] The T-die 36 has a plurality of lip gap adjusting means 361 arranged along the lip longitudinal direction of the T-die 36. The lip gap adjusting means 361 adjusts the width of the lip gap in accordance with a command from the lip gap adjusting unit 40 or manual operation to adjust the film thickness of the molten resin (film). The lip gap adjusting unit 40 generates and outputs a command for feedback control of the film thickness based on information on the film thickness deviation in the width direction from the calculation unit 34. A known method may be used as the lip gap adjusting mechanism of the lip gap adjusting means 361. Suitable examples of such mechanisms include a method in which the amount of press-in is changed by rotating a threaded bolt, a heat bolt method in which a cartridge heater is built in to thermally expand and contract, and a method in which adjustment is performed by applying hydraulic pressure to the bolt or the T-die.

[0066] <Chilling Roll 37> The chilling roll 37 cools and solidifies the molten resin extruded from the lip gap of the T-die 36, forming it into a film. The chilling roll 37 is preferably equipped with a pinning device such as an electrostatic application type, a suction chamber type, or an air knife type, or a touch roll, for the purposes of suppressing variations in film thickness and improving surface properties. Furthermore, in order to improve the cooling rate of the molten resin, a part of the chilling roll may be immersed in water or water may be sprayed onto it.

[0067] <Stretching machine 38> The stretching machine is a device that stretches the film formed by the cooling roll 37. Stretching may be performed as needed, and the film to be produced does not need to be stretched. The stretching machine 38 constitutes a stretching step. When stretching is performed using the stretching machine 38, stretching in the machine direction (MD direction: longitudinal direction) and stretching in the width direction (TD direction) can be performed alone or in combination. That is, a method of stretching only in the MD direction, a method of stretching only in the TD direction, a method of sequentially stretching in the MD direction and the TD direction, or a method of simultaneously stretching in the MD direction and the TD direction can be used. Furthermore, various stretching methods can be applied, such as a method of sequentially stretching in the MD direction and the TD direction and then re-stretching again in the MD direction or the TD direction. Regarding the stretching process, it is preferable to select and use a stretching method, including non-stretching, depending on the resin type and application of the film.

[0068] Furthermore, with regard to longitudinal stretching in the MD direction, examples include single-stage stretching using a pair of rolls with a speed difference, and multi-stage stretching using multiple pairs of rolls. Both of these can be suitably used. Furthermore, examples of methods for raising the temperature of the film for stretching include a method using a preheating roll and a method of heating without contact using an infrared heater or the like. Both of these can be suitably used. Furthermore, near the outlet of the stretching machine 38, surface treatments such as heat setting treatment for relaxation while heating or corona treatment may be performed. The measured value of the film thickness deviation obtained by the film thickness deviation measuring device can also be used to control the stretching machine 38.

[0069] <Winding machine 39> The winding machine 39 is a device that winds up the manufactured film into a roll. The film wound into a roll becomes the product to be shipped. It is preferable to install quality assurance equipment such as a surface inspection device, a pinhole inspection device, and a film thickness meter just before the winding machine 39. It is also preferable to adjust the product width to the desired width using a slitter or trimmer. Oscillation winding may be performed to prevent gauge bands during winding.

[0070] <Film Thickness Deviation Measuring Device> The film thickness deviation measuring device may employ a configuration similar to that of the film thickness deviation measuring device of the present embodiment described above. The background image and the imaging means (imaging means) are arranged opposite each other across the film pass line. In the example of FIG. 4 , the background image 32 and imaging means 33 are arranged at the pass line position between the chill roll 37 and the stretching machine 38. However, the background image 32 and imaging means 33 may also be arranged before or downstream of the stretching machine 38. That is, the film thickness deviation measuring unit consisting of the background image 32 and imaging means 33 can be installed anywhere from downstream of the T-die 36 to upstream of the winder 39. Considering the speed of feedback control and the possibility that the refractive index of the film 31 may change due to processing by the stretching machine 38, it is preferable to install it upstream of the stretching machine 38. Furthermore, if the background image 32 and imaging means 33 increase the distance between the lip of the T-die 36 and the chill roll 37, there is a risk of increased film thickness unevenness in the MD and necking in the TD. For this reason, it is preferable that the film thickness deviation measuring unit be installed downstream of the cooling roll 37 .

[0071] As described above, the background image 32 may be an image that allows detection of the amount of displacement when the position on the measurement image changes depending on the presence or absence of the film 31. In particular, a random dot image or a stripe image with a periodic brightness distribution is preferable as the background image 32 from the viewpoint of resolution. As described above, various digital cameras can be used for the imaging means 33. In this example, the use of a line camera in particular is expected to improve processing speed and reduce equipment costs. The imaging means 33 may also be a collection of multiple cameras. Incidentally, by having the imaging means 33 also function as a surface inspection device, foreign matter, defects, and the like can be detected early. In this case, a surface inspection calculation unit (not shown) is provided that performs a directional inspection based on the image captured by the imaging means 33.

[0072] It is preferable that the background image 32 and the imaging means 33 are installed on the same structure. In this case, misalignment between the background image 32 and the imaging means 33 due to vibrations during operation can be suppressed. Furthermore, in order to calibrate the misalignment, it is more preferable that the installed structure has a mechanism that allows it to be pulled out offline during operation to capture a reference image. Furthermore, from the perspective of analytical calculations and preventing out-of-focus images, it is preferable that the optical axis of the imaging means 33 be set so as to be perpendicular to the film 31 and background image 32. The axis connecting the film 31, background image 32, and imaging means 33 may be arranged vertically as shown in Figure 4. However, the axis may be installed horizontally or at any other angle depending on the installation environment and other constraints.

[0073] There may be other visible light-transmitting structures, such as glass windows, in the gap between the film 31 and the background image 32, or between the film 31 and the imaging means 33. However, the structures must be such that the film 31 and the background image 32 can be recognized through the imaging means 33. If the structures have a film thickness deviation, it is preferable to correct the displacement of the background image 32 or the calculated film thickness deviation of the film 31. However, if the distance between the background image 32 and the structures is short and has only an acceptable effect on the calculated film thickness deviation of the object 31, correction may not be necessary. It is more preferable that the structures have no film thickness deviation.

[0074] As described above, the calculation unit 34 calculates the amount of background displacement between the reference image and the measurement image to calculate the film thickness deviation. Information on the film thickness deviation in the width direction is then supplied to the lip gap adjustment unit 40. The reference image is an image that has been captured or derived by calculation, as described above. The measurement image is an image of the background image 32 captured by the imaging means 33 with the film 31 interposed, as described above. In this example, the object to be measured is transported along the pass line at a predetermined transport speed. The transport speed at the film thickness deviation measurement position may be relatively slowed using looper equipment, etc.

[0075] As described above, various known methods can be used to calculate the amount of displacement of the background image between the reference image and the measurement image. However, in this example, information on the film thickness deviation is fed back to the control of the lip gap adjustment means. From this perspective, it is preferable to calculate the film thickness deviation only in the TD direction. It is particularly preferable to use wavelet analysis or short-time Fourier analysis to calculate the amount of displacement of the background image. The film thickness deviation is calculated using the amount of displacement of the background image calculated by the above method. As described above, various integral paths can be used for this calculation method. However, in this example, it is preferable to calculate only in the TD direction from the perspective of feeding back the film thickness deviation to the control of the lip gap adjustment means. For example, it is more preferable to use the following equation (9).

[0076]

[0077] where t(x) is the film thickness deviation obtained by subtracting the film thickness of the film 31 at the origin x=0 from the film thickness of the film 31 at the TD coordinate x, and n ｔａｒｇｅｔ : refractive index of film 31 n ａｉｒ : is the refractive index of the atmosphere, d: the distance between the background image 32 and the film 31, Δhx: the distance the background image of interest in the measurement image is moved in the TD direction from the same background image in the reference image, and Δx: the length in the x direction per pixel.

[0078] The thickness deviation of the film 31 obtained by the above method can be displayed by a known visualization method. In this embodiment, it is preferable to display only the thickness deviation in the TD direction, from the viewpoint of feeding back information on the calculated thickness deviation to the control of the lip gap adjustment means. Furthermore, in order to know the fluctuation of the thickness deviation over time, it is more preferable to display, for example, a line graph with the TD coordinate on the horizontal axis and the thickness deviation on the vertical axis, with the color of the line changing over time.

[0079] The lip gap adjustment unit 40 compares the film thickness deviation profile of the actual film 31 obtained as described above with the ideal film thickness deviation profile of the film 31 input in advance to the calculation unit 34. Then, the lip gap adjustment means 361 is controlled to increase the lip gap corresponding to the TD coordinate in order to thicken the portion where the actual film thickness is thinner than the ideal film thickness. The lip gap adjustment means 361 is also controlled to decrease the lip gap corresponding to the TD coordinate in order to thin the portion where the actual film thickness is thicker than the ideal film thickness. That is, the measured film thickness deviation profile is compared with the ideal film thickness deviation profile to determine the film thickness difference at the position corresponding to each lip gap adjustment means 361 in the film width direction. Then, each lip gap adjustment means 361 is feedback-controlled so that the difference falls within a predetermined range.

[0080] Here, necking occurs when the film is discharged from the T-die. Therefore, the distance from the TD center of the film 31 to a specific position when imaged by the imaging device 33 generally differs from the distance from the TD center of the molten resin when discharged from the T-die lip. Therefore, the lip gap adjustment unit 40 preferably uses various known methods to determine which lip gap adjustment device 361 to issue a command to when the difference between the actual film thickness deviation and the ideal film thickness deviation is detected at a specific TD coordinate. It is also known that moving multiple lip gap adjustment devices 361 simultaneously, taking into account an influence coefficient, rather than moving only one, improves accuracy. Therefore, it is preferable to weight the results calculated using various known methods and operate multiple lip gap adjustment devices 361 simultaneously.

[0081] The above describes the embodiments of the present invention. However, the present invention is not limited to these embodiments, and various modifications are possible without departing from the spirit of the invention. Furthermore, the film in the above description includes resin compositions that are generally called sheets because they are relatively thick.

[0082] (Other) The present disclosure may also have the following configurations. (1) A method for measuring a film thickness deviation, comprising: using a film to be measured as an object to be measured; enabling an imaging means to capture a preset background image; acquiring a reference image that is an image of the background image captured by the imaging means without the object to be measured, or a reference image derived by calculation of how the background image appears to the imaging means without the object to be measured; acquiring a measurement image that is an image of the background image captured by the imaging means with the object to be measured interposed between the background image and the imaging means; and calculating the amount of displacement of the background image in these two images based on the acquired reference image and measurement image, thereby determining the film thickness deviation of the object to be measured. (2) The film thickness deviation is determined using the above-mentioned formula (1). (3) The integral path C is a path that proceeds from a preset origin in a first direction, which is either the longitudinal direction of the object or a direction perpendicular to the longitudinal direction of the object, and then in a second direction perpendicular to the first direction. (4) A method for measuring a film thickness deviation of an object being transported between the background image and the imaging means, wherein when the film thickness deviation of the object being measured is a film thickness deviation in the width direction of the object, the integral path C is a line segment that proceeds from the origin in a direction perpendicular to the transport direction. (5) The background image is a random dot image, and the displacement amount Δh of the background image is calculated by template matching using a normalized cross-correlation coefficient. (6) The background image is a striped image composed of waves with a predetermined spatial frequency and having a periodic brightness distribution, and the displacement amount Δh of the background image is calculated by wavelet analysis. (7) The visible light transmittance of the object being measured is 10% or more.(8) A film manufacturing method for molding a film that is capable of transmitting visible light by discharging molten resin from a die having a plurality of lip gap adjustment means, the film discharged from the die being transported along a predetermined pass line; imaging means for capturing a background image and the background image being disposed on either side of the pass line; and acquiring a measurement image that is an image of the background image captured by the imaging means through the film; acquiring a reference image that is an image of the background image captured by the imaging means in a state where the film is not interposed between the background image and the imaging means, or acquiring a reference image derived by calculation of how the background image appears through the imaging means in a state where the film is not interposed; comparing the acquired reference image with the acquired measurement image and calculating the amount of displacement of the background image in the film width direction between the two images, thereby calculating a film thickness deviation profile in the width direction of the film; and operating the lip gap adjustment means based on the calculated film thickness deviation profile so that the film thickness deviation at each point becomes a target value. (9) The visible light transmittance of the film extruded and molded from the die is 10% or more, and a stretching process for stretching the film is provided downstream of the position on the pass line where the measurement image is acquired. (10) A film thickness deviation measuring device for visible light transmission, comprising: a film to be measured as a measurement object; a background image; an imaging means for capturing the background image; a reference image acquisition unit for acquiring a reference image, which is an image of the background image captured by the imaging means without the measurement object interposed, or a reference image derived by calculation from how the background image appears when viewed by the imaging means without the measurement object interposed; a measurement image acquisition unit for acquiring a measurement image, which is an image of the background image captured by the imaging means with the measurement object interposed between the background image and the imaging means; and a film thickness deviation calculation unit for calculating the amount of displacement of the background image in the two images based on the acquired reference image and measurement image. (11) The background image is a random dot image. (12) The background image is a striped image that is composed of waves having a predetermined spatial frequency and has a periodic brightness distribution.(13) The object to be measured has a visible light transmittance of 10% or more. (14) A film manufacturing apparatus that extrudes molten resin from a die having a plurality of lip gap adjustment means to form a film that is transmissive to visible light, the film extruded from the die being transported along a predetermined pass line, imaging means for capturing background images and background images being disposed on either side of the pass line, the film manufacturing apparatus comprising: a measurement image acquisition unit that acquires a measurement image that is an image of the background image captured by the imaging means through the film; a reference image acquisition unit that acquires a reference image that is an image of the background image captured by the imaging means in a state where the film is not interposed between the background image and the imaging means, or a reference image derived by calculation of how the background image appears through the imaging means in a state where the film is not interposed; a film thickness deviation information acquisition unit that compares the acquired reference image with the acquired measurement image and calculates the amount of displacement of the background image in the film width direction between the two images, thereby calculating a film thickness deviation profile in the film width direction; and a lip gap adjustment unit that operates the lip gap adjustment means based on the calculated film thickness deviation profile so that the film thickness deviation at each point becomes a target value. (15) The visible light transmittance of the film extruded and molded from the die is 10% or more, and a stretching machine for stretching the film is provided downstream of the position on the pass line where the image is acquired by the imaging means.

[0083] Examples based on this embodiment will be described below. However, the following examples are merely intended to explain one example of the present invention in more detail and are not intended to limit the scope of the present invention.

[0084] (Measurement Conditions for this Example) In this example, a PET film was used as the film under test 21. A photograph of a random dot image was used as the background image 22. A single-lens reflex camera was used as the imaging means 23. As shown in FIG. 2 , the imaging means 23 (SLR camera), the object under test 21 (PET film), and the background image 22 (random dot image) were arranged vertically. The vertical direction is the direction perpendicular to the surface of the background image 22. Here, the PET film in this example was an unstretched film with an average thickness of 95 μm extruded using a T-die method. The PET film had a thickness standard deviation of 1.8 μm in the MD direction and 25 μm in the TD direction. The PET film and the random dot image were spaced 50 mm apart, and the PET film and the single-lens reflex camera were spaced 600 mm apart.

[0085] (Processing for Measuring Film Thickness Deviation) First, a background image was captured without a PET film to obtain a reference image. That is, a photograph of a random dot image was captured with a single-lens reflex camera without a PET film. Next, a background image was captured with a PET film interposed to obtain a measurement image. That is, a photograph of a random dot image was captured with a single-lens reflex camera through the PET film. The two acquired images were then imported into a personal computer, and calculation processing was performed by the calculation unit 24 to determine the film thickness deviation. In this example, first, the amount of displacement of the background image between the two images in the MD and TD directions was calculated by pattern matching using ZNCC. The background displacement amount distribution of the calculated displacement amount is shown in Figure 5. ZNCC is a normalized cross-correlation of the template matching method.

[0086] Next, the film thickness deviation at each point was calculated from the background displacement distribution using the above-mentioned equation (8), and the film thickness deviation distribution shown in Figure 6 was obtained. In Figures 5 and 6, the horizontal direction is the MD direction, and the vertical direction is the TD direction. In Figures 5 and 6, both ends of the MD direction are areas where there is no film and the background image is directly visible. The shading indicates the background displacement or film thickness deviation. As can be seen from Figure 6, the PET film used in this example has almost no film thickness variation in the MD direction. On the other hand, periodic film thickness variation is observed in the TD direction. Furthermore, the film thickness variation period in the TD direction coincides with the spacing of the heat bolts installed in the T-die, confirming that this reflects the width of the T-die lip gap caused by the heat bolts.

[0087] As described above, it has been found that the present invention makes it possible to measure film thickness deviation data (film thickness deviation information) in one-dimensional and two-dimensional directions along the film surface in a non-contact manner using a simple and safe device configuration.

[0088] The entire contents of Japanese Patent Application No. 2023-110735 (filed July 5, 2023), from which this application claims priority, are incorporated herein by reference. While the present invention has been described with reference to a limited number of embodiments, the scope of the invention is not limited thereto, and modifications of each embodiment based on the above disclosure would be obvious to those skilled in the art.

[0089] 20 Film thickness deviation measuring device 21 Object to be measured 22 Background image 23 Imaging means 24 Calculation unit 24A Reference image acquisition unit 24B Measurement image acquisition unit 24C Film thickness deviation calculation unit 25 Stand 25A Top plate surface 30 Film manufacturing device 31 Film (object to be measured) 32 Background image 33 Imaging means 34 Calculation unit 35 Extruder 36 T-die 37 Cooling roll 38 Stretching machine 39 Winder 40 Lip gap adjustment unit 361 Lip gap adjustment means

Claims

1. A method for measuring film thickness deviation, comprising the steps of: The film to be measured is the object to be measured. A preset background image can be captured by an imaging means; A reference image is obtained by capturing the background image by the imaging means without the object to be measured, or a reference image is obtained by calculating how the background image appears when captured by the imaging means without the object to be measured; The object to be measured is interposed between the background image and the imaging means, and a measurement image is obtained by imaging the background image with the imaging means; calculating a displacement amount of a background image in each of the acquired reference image and measurement image based on the acquired reference image and measurement image, thereby determining a film thickness deviation of the object to be measured; Method for measuring film thickness deviation.

2. 2. The method for measuring thickness deviation according to claim 1, wherein the thickness deviation is determined using the following formula (1): [0010] Where: t: film thickness deviation obtained by subtracting the film thickness of the measured object at the origin from the film thickness of the measured object at the measurement point n target : Refractive index of the object to be measured n ａｉｒ : Refractive index of the atmosphere d: Distance between the background image and the object being measured Δh: A two-dimensional vector representing the amount of displacement (amount of movement) of a background image of interest in a measurement image from the same background image in a reference image v: unit tangent vector of path C "∫ ｃ "Δh · vdl": An operation of integrating the inner product of the above Δh and the above v over an integral path C from the origin to the measurement point It is.

3. The integral path C is a path that proceeds from a preset origin in a first direction that is either the longitudinal direction of the object to be measured or a direction perpendicular to the longitudinal direction of the object to be measured, and then proceeds in a second direction that is perpendicular to the first direction. The film thickness deviation measuring method according to claim 2.

4. a film thickness deviation measuring method for the object to be measured, the object being transported so as to pass between the background image and the imaging means, When the thickness deviation of the object to be measured is a thickness deviation in the width direction of the object, The integral path C is a line segment proceeding from the origin in a direction perpendicular to the conveying direction. The film thickness deviation measuring method according to claim 2.

5. The background image is a random dot image, The displacement amount Δh of the background image is calculated by template matching using a normalized cross-correlation coefficient. The method for measuring thickness deviation according to any one of claims 1 to 4.

6. the background image is a striped image having a periodic luminance distribution composed of waves having a predetermined spatial frequency, The displacement amount Δh of the background image is calculated by wavelet analysis. The film thickness deviation measuring method according to claim 1, 2 or 4.

7. The visible light transmittance of the object to be measured is 10% or more. The method for measuring thickness deviation according to any one of claims 1 to 4.

8. The visible light transmittance of the object to be measured is 10% or more. The film thickness deviation measuring method according to claim 5.

9. The visible light transmittance of the object to be measured is 10% or more. The film thickness deviation measuring method according to claim 6.

10. A method for producing a film, comprising discharging molten resin from a die having a plurality of lip gap adjustment means to form a film capable of transmitting visible light, the method comprising the steps of: The film discharged from the die is transported along a preset path line, A background image and an imaging means for imaging the background image are disposed on either side of the pass line; acquiring a measurement image, which is an image obtained by capturing the background image through the film by the imaging means; acquiring a reference image, which is an image of the background image captured by the imaging means without the film being interposed between the background image and the imaging means, or a reference image derived by calculation of how the background image appears when viewed by the imaging means without the film being interposed; a step of calculating a film thickness deviation profile in the width direction of the film by comparing the acquired reference image with the acquired measurement image and calculating a displacement amount of the background image between the two images in the width direction of the film; a step of operating the lip gap adjusting means so that the film thickness deviation at each point becomes a target value based on the calculated film thickness deviation profile; A film manufacturing method comprising:

11. The visible light transmittance of the film extruded from the die is 10% or more; A stretching step of stretching a film is provided downstream of a position in the pass line where a measurement image is acquired. The method for producing a film according to claim 10.

12. An apparatus for measuring film thickness deviation of a film that transmits visible light, comprising: The film to be measured is the object to be measured. a background image; and an imaging means for imaging the background image; a reference image acquisition unit that acquires a reference image, which is an image of the background image captured by the imaging means without the object to be measured being present, or a reference image derived by calculation of how the background image appears when captured by the imaging means without the object to be measured being present; a measurement image acquisition unit that acquires a measurement image, which is an image obtained by capturing the background image by the imaging means, in a state in which the object to be measured is interposed between the background image and the imaging means; a thickness deviation calculation unit for calculating a displacement amount of a background image in the reference image and the measurement image based on the acquired reference image and the measurement image, thereby obtaining a thickness deviation of the object to be measured; Equipped with Film thickness deviation measuring device.

13. The background image is a random dot image. The film thickness deviation measuring device according to claim 12.

14. The background image is a striped image having a periodic luminance distribution composed of waves having a predetermined spatial frequency. The film thickness deviation measuring device according to claim 12.

15. The object to be measured has a visible light transmittance of 10% or more. The film thickness deviation measuring device according to any one of claims 12 to 14.

16. A film manufacturing apparatus that extrudes molten resin from a die having a plurality of lip gap adjustment means to form a film that is capable of transmitting visible light, The film discharged from the die is transported along a preset path line, Imaging means for capturing a background image and a background image are disposed on either side of the pass line; a measurement image acquisition unit that acquires a measurement image, which is an image obtained by capturing the background image through the film by the imaging means; a reference image acquisition unit that acquires a reference image obtained by capturing the background image by the imaging means without the film being interposed between the background image and the imaging means, or a reference image derived by calculation of how the background image appears when captured by the imaging means without the film being interposed; a film thickness deviation information acquisition unit that calculates a film thickness deviation profile in the film width direction by comparing the acquired reference image with the acquired measurement image and calculating a displacement amount of a background image between the two images in the film width direction; a lip gap adjusting unit that operates the lip gap adjusting means based on the calculated film thickness deviation profile so that the film thickness deviation at each point becomes a target value; Film manufacturing equipment.

17. The visible light transmittance of the film extruded from the die is 10% or more; a stretching machine for stretching a film is provided downstream of a position in the pass line where an image is captured by the imaging means; 17. The film manufacturing apparatus according to claim 16.

Images