Release film production detection method and system

Abstract

The invention discloses a release film production detection method and system, and relates to the field of release film performance detection. Comprising a hardware entity terminal and a software control platform. The microscopic thickness change of silicone oil coating can be clearly captured through high-resolution imaging, the detection sensitivity is ensured, and in addition, the optical contrast ratio of uneven coating is enhanced by adopting a double-light-source scheme (white light and ultraviolet light); 5-degree inclined installation avoids mirror reflection interference and ensures stable image quality; and objective and quantitative defect judgment is realized through FFT (Fast Fourier Transform) frequency domain analysis and a high-frequency energy ratio threshold value in an algorithm level. And in the judgment process, noise suppression (Gaussian filtering) retains key textures, so that misjudgment is avoided. Edge interference is eliminated through ROI extraction, and the analysis accuracy is improved. The influence of uneven illumination is eliminated through gray normalization, and the robustness of the algorithm is enhanced. The problems that in the prior art, a direct detection method or an indirect detection method is not timely in feedback and cannot directly correspond to the coating amount due to interference of various factors are solved.

Description

Technical Field

[0001] This invention relates to the field of release film performance testing, and in particular to a release film production testing method and system. Background Technology

[0002] Release film refers to a film whose surface can be distinguished. When a release film comes into contact with a specific material under limited conditions, it does not have adhesiveness or has only slight adhesiveness.

[0003] During the production of release films, it is necessary to ensure that their performance meets the standards from multiple dimensions (such as peel force, residual adhesion rate, surface energy, etc.). Among them, the uniformity of silicone oil coating is one of the core indicators of release film performance, which directly affects the stability of peel force and residual adhesion rate.

[0004] Existing professional testing methods and judgment standards include direct testing methods (coating amount measurement) and indirect testing methods (performance correlation analysis). Direct testing methods are destructive testing methods and have the risk of missed detections in sampling, as well as the inability to provide real-time feedback and guidance for production. Indirect testing methods are affected by multiple factors such as environmental temperature and humidity and substrate surface condition, and cannot directly correspond to the coating amount.

[0005] Therefore, this invention proposes a method and system for testing the production of release films. Summary of the Invention

[0006] This invention provides a method and system for testing release film production, which can solve the problems of untimely feedback and inability to directly correspond to the coating amount due to interference from various factors in the existing direct or indirect testing methods.

[0007] A method for testing the production of release film, comprising: The first parameter image of the film surface after being coated with silicone oil during the production process is obtained using image detection equipment; The image preprocessing module preprocesses the first parameter image to obtain a processed image; The image transformation module performs a fast Fourier transform on the processed image based on the two-dimensional discrete Fourier transform formula to obtain a frequency domain image; The defect identification module performs defect identification on the frequency domain image processed by the image transformation module to achieve production inspection.

[0008] Preferably, the image detection device is installed on the release film production line, and the installation position of the image detection device is 300±5mm away from the film surface and at a 5° angle to the normal of the film surface.

[0009] Preferably, the image detection device is equipped with a coaxial white LED and ultraviolet light to excite the silicone oil fluorescent label.

[0010] Preferably, the preprocessing process includes noise reduction, analysis region extraction, and grayscale normalization.

[0011] Preferably, the noise reduction uses a Gaussian filter with σ=1.5; The analysis region is defined using ROI extraction technology. Gray-scale normalization maps pixel values ​​to the range of 0-255.

[0012] Preferably, the formula for the two-dimensional discrete Fourier transform is:

[0013] in, These are the original image pixel values. It is a complex matrix in the frequency domain, containing amplitude and phase information.

[0014] Preferably, the defect identification module performs defect identification on the frequency domain image processed by the image transformation module, including: By calculating the energy integral of the outer ring band of the spectrum, when the proportion of high-frequency energy is <B%, it indicates that the uniformity of silicone oil coating is normal. When the bright ring energy percentage is greater than 5%, it is considered that the uniformity of silicone oil coating is abnormal.

[0015] Preferably, the high-frequency region is defined as a ring-shaped region with a radius greater than 200 pixels; Set two levels of judgment thresholds: High-frequency energy percentage <3%: Release film is of superior quality; Release film with 3% ≤ high-frequency energy percentage ≤ 5% is considered qualified. If there is an abnormality in the uniformity of the silicone oil coating, it will manifest as a bright ring. The relationship between the bright ring radius r and the fringe spacing d is: d = N / r × pixel size; Where N is the image side length in pixels; When the bright ring energy ratio is greater than 5%, it is determined to be an abnormality in the uniformity of silicone oil coating.

[0016] A release film production inspection system, comprising: Hardware physical terminals and software control platforms; The hardware terminal includes an image detection device and a motion control module; the image detection device includes an imaging device installed on the release film production line for capturing images of the first parameter of the release film; the motion control module is used to control the movement and imaging of the image detection device, wherein the motion control module is equipped with a line scanning mode, which is coordinated with the high-speed production line. The software control platform is mainly used for preprocessing and defect analysis and identification of the first parameter image; including: Image preprocessing module, image transformation module, and defect recognition module; The image preprocessing module is used to perform noise reduction, analysis region extraction, and grayscale normalization on the first parameter image to obtain a processed image. The image transformation module is used to perform a fast Fourier transform on a single-processed image based on the two-dimensional discrete Fourier transform formula; The defect identification module is used to identify defects in the frequency domain image processed by the image transformation module, thereby enabling production inspection.

[0017] Preferably, the shooting device is a high-resolution industrial camera, specifically a 50-megapixel CMOS camera, paired with a telecentric lens.

[0018] Compared with the prior art, this application has the following beneficial effects: The release film production inspection method and system provided in this application, in terms of hardware architecture, can clearly capture the microscopic thickness changes of silicone oil coating through high-resolution imaging, ensuring detection sensitivity. Furthermore, a dual-light source scheme (white light + ultraviolet) is employed to enhance the optical contrast of coating unevenness; a 5° tilt installation avoids specular reflection interference, ensuring stable image quality; at the algorithm level, FFT frequency domain analysis + high-frequency energy ratio thresholding achieves objective and quantitative defect judgment. During the judgment process, noise suppression (Gaussian filtering) preserves key textures and avoids misjudgments. ROI extraction eliminates edge interference, improving analysis accuracy. Gray-level normalization eliminates the influence of uneven illumination, enhancing algorithm robustness. This solves the problems of untimely feedback and inability to directly correspond to coating amount due to interference from various factors in existing direct or indirect detection methods. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of a release film production and testing system provided by the present invention; Figure 2 This is a schematic diagram of a release film production and testing method provided by the present invention. Detailed Implementation

[0020] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0021] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0022] The release film production testing method provided in this application embodiment can be applied to a release film production testing system, such as... Figure 1 As shown, the detection system includes a hardware terminal and a software control platform; The hardware terminal is used to acquire the first parameter image of the film surface after it has been coated with silicone oil during the release film production process; the software control platform is connected to the serial communication module of the hardware terminal via RS485 bus and establishes a data transmission channel to achieve data transmission with the hardware terminal. As an example, the hardware terminal includes an image detection device and a motion control module; The image detection equipment includes a camera installed on the release film production line for capturing images of the first parameter of the release film. As an example, the shooting device is a high-resolution industrial camera, such as a 50-megapixel CMOS camera (e.g., Basler ace 2) with a telecentric lens (0.1% distortion rate), achieving a resolution of 5μm / pixel; or a 50-megapixel linear CCD camera (e.g., Teledyne DALSA Piranha4).

[0023] The high-resolution industrial camera is installed 300±5mm away from the film surface and at a 5° angle to the normal of the film surface to avoid specular reflection interference. It is obvious that a light source solution is required when using a high-resolution industrial camera. In order to make the captured first parameter image clear and to detect the uniformity of silicone oil coating on the release film, the high-resolution industrial camera is equipped with a coaxial white LED and ultraviolet light that can excite the silicone oil fluorescent label. The coaxial white LED has a wavelength of 450-650nm, and the illumination uniformity of the coaxial white LED light source is >90%, with a color temperature of 5500K±200K. The ultraviolet light uses a 365nm wavelength LED array, arranged at a 45° angle with the film surface, with a power density of 50mW / cm². The equipped coaxial white LED can not only ensure the capture of the first parameter image, but also reflect to a certain extent whether there are scratches on the surface of the release film, and can be used as a light source for detecting surface scratches. During the production of release film, the production line is a high-speed moving line. The fixed high-resolution industrial camera and light source scheme obviously cannot meet the requirement of no motion blur in the first parameter image. Therefore, this application also provides a motion control module. The motion control module is used to control the movement and shooting of the image detection equipment. The motion control module is equipped with a line scanning mode. By combining the line scanning mode with the high speed of the high-speed production line, the motion control module can reach a maximum speed of 120m / min. The motion control algorithm adopts PID + feedforward composite control, and the position tracking error is <0.1mm. The matching relationship between camera frame rate and production line speed: when the production line speed V≤60m / min, full frame mode (30fps) is used; when V>60m / min, switch to line scan mode. In addition, the motion control module is equipped with an encoder, and the image detection device works with the encoder to trigger the image taking, ensuring that the first parameter image is free of motion blur; the encoder is a 17-bit absolute encoder (131072ppr), which is installed on the drive roller shaft; The formula for calculating the line frequency in line scan mode is: f = V / (60×pixel_size), where V is the line velocity (m / min) and pixel_size is 5μm; and a trigger delay compensation mechanism is set: a hardware trigger delay compensation of 200μs is set to ensure that the image acquisition and the membrane position are strictly synchronized. Obviously, it needs to be pointed out in this application document that the first parameter image is the film surface after the release film is coated with silicone oil. Generally, the acquisition area of ​​the first parameter image is the film surface with a width of 1m and a resolution of 5μm / pixel. The release film production inspection system proposed in this application is mainly used to detect whether there are defects in the release film during the production process, that is, whether the silicone oil coating is uniform. The test results mainly include normal homogeneity and abnormal homogeneity. It is evident that the normal uniformity of silicone oil coating is characterized by a gray background in the first parameter image. Abnormalities in silicone oil coating uniformity manifest as periodic bright and dark stripes or random spots appearing in the first parameter image; The reason for the periodic bright and dark stripes is the light interference effect caused by the uneven thickness of the silicone oil coating. Based on the light interference effect, alternating bright and dark stripes will be formed (similar to Newton's rings). Obviously, the uniformity of silicone oil coating is indistinguishable to the naked eye, so the release film production and testing system is also equipped with a software control platform. The software control platform is mainly used for preprocessing and defect analysis and identification of the first parameter image; Specifically, preprocessing relies on the image preprocessing module; defect analysis and identification rely on the image transformation module and the defect identification module. Therefore, it can be known that the software control platform includes an image preprocessing module, an image transformation module, and a defect recognition module; The image preprocessing module proposed in this application is used to perform noise reduction, analysis region extraction, and grayscale normalization on the first parameter image to obtain a first-processed image. Noise reduction processing primarily aims to eliminate noise from the shooting equipment. It typically utilizes Gaussian filtering with σ=1.5, where Gaussian filtering, through linear smoothing, significantly suppresses noise that follows a Gaussian distribution. The larger the σ value, the more pronounced the filtering effect; however, σ=1.5 effectively reduces noise interference while preserving image details.

[0024] The purpose of region extraction is to define the analysis area, mainly to avoid edge interference. In a specific embodiment, the analysis area can be a 512×512 pixel block. In the specific release film silicone oil coating process, the effective uniform area of ​​silicone oil coating is usually located in the middle of the film material (80% of the width direction). ROI extraction can eliminate invalid data at the edges and prevent mechanical interference signals (such as high-frequency noise) from being mixed into the FFT spectrum of the edge area, which would lead to misjudgment of uneven coating.

[0025] Gray-scale normalization maps pixel values ​​to the range of 0-255, eliminating the effects of uneven lighting.

[0026] Gray-scale normalization is mainly considered to be the fact that the aging of light sources, angle shift or reflection of film surface in industrial sites will cause local overexposure (bright spots) or underexposure (dark areas) in the image, which directly interferes with the texture recognition of silicone oil coating. The original pixel value (such as 120-180) is stretched to the full dynamic range of 0-255 through linear transformation, forcing the gray-scale distribution of all areas to be uniform. The image transformation module is used to perform a fast Fourier transform on a single-processed image based on the two-dimensional discrete Fourier transform formula; The formula for the two-dimensional discrete Fourier transform is:

[0027] in, These are the original image pixel values. It is a complex matrix in the frequency domain, containing amplitude and phase information; and Spatial domain coordinates, x : The horizontal position of pixels in the image (column index, from 0 to M) 1).

[0028] y The vertical position of a pixel in the image (row index, from 0 to N). 1).

[0029] In a single image processing step, (x, y) represents the physical position of a point on the membrane surface (e.g., x corresponds to the width direction of the membrane material, and y corresponds to the direction of motion). For frequency domain coordinates, u: The frequency component in the horizontal direction (representing the periodic changes of the image in the x-direction).

[0030] u=0: DC component (average image brightness).

[0031] Increased u corresponds to higher frequency horizontal stripes (such as longitudinal stripes caused by uneven coating).

[0032] v: The frequency component in the vertical direction (representing the periodic changes of the image in the y-direction).

[0033] v=0: No vertical change.

[0034] Increased v corresponds to higher frequency transverse stripes (such as transverse ripples caused by uneven winding tension).

[0035] If the release film coating has periodic defects: Bright spots in the u direction → longitudinal stripe defects (related to the direction of coating head movement).

[0036] Bright spots in the v direction → transverse stripe defects (related to the direction of membrane material movement).

[0037] To facilitate intuitive interpretation in the frequency domain, the image transformation module shifts the low-frequency components to the center of the spectrum to generate a frequency domain image and calculates the logarithmic amplitude spectrum. In this application, the logarithmic amplitude spectrum = ; The defect identification module is used to identify defects in the frequency domain image processed by the image transformation module. It is evident that, for the frequency domain characteristics of the normal coating region, the energy distribution is concentrated in the central low-frequency region, exhibiting a uniform diffusion pattern (indicating the absence of a periodic structure). Specifically, the quantitative indicator is calculated by integrating the energy of the outer ring band of the spectrum. When the proportion of high-frequency energy is less than B%, it indicates that the uniformity of the silicone oil coating is normal. Generally, B=5. Specifically, the high-frequency region is defined as a ring-shaped region with a radius greater than 200 pixels. Set two levels of judgment thresholds: a) High-frequency energy percentage < 3%: Superior grade b) 3% ≤ High-frequency energy percentage ≤ 5%: Qualified product If there is an abnormality in the uniformity of the silicone oil coating, it will manifest as a bright ring. The relationship between the bright ring radius r and the fringe spacing d is: d = N / r × pixel size; Where N is the image side length in pixels; Furthermore, by calculating the proportion of bright ring energy, if the proportion of bright ring energy is greater than 5%, it is determined that the uniformity of silicone oil coating is abnormal.

[0038] Figure 2 The release film production testing method provided in the embodiments of this application, such as Figure 2 As shown, the method includes: S101. Obtain the first parameter image of the film surface after it has been coated with silicone oil during the production process using an image inspection device.

[0039] In one possible implementation, the user sets up a high-resolution industrial camera on the release film production line, with the camera positioned 300±5mm from the film surface and at a 5° angle to the normal of the film surface. As an example, high-resolution industrial cameras require a light source solution when in use, which consists of a coaxial white LED and ultraviolet light that can excite silicone oil fluorescent markers; S102, The image preprocessing module preprocesses the first parameter image to obtain a processed image.

[0040] In one possible implementation, the preprocessing process includes denoising, analysis region extraction, and grayscale normalization. As an example, noise reduction is primarily aimed at eliminating noise from the shooting equipment. It typically utilizes Gaussian filtering with σ=1.5, where Gaussian filtering, through linear smoothing, significantly suppresses noise that follows a Gaussian distribution. The larger the σ value, the more pronounced the filtering effect; however, σ=1.5 effectively reduces noise interference while preserving image details.

[0041] The purpose of region extraction is to define the analysis area, mainly to avoid edge interference. In a specific embodiment, the analysis area can be a 512×512 pixel block. In the specific release film silicone oil coating process, the effective uniform area of ​​silicone oil coating is usually located in the middle of the film material (80% of the width direction). ROI extraction can eliminate invalid data at the edges and prevent mechanical interference signals (such as high-frequency noise) from being mixed into the FFT spectrum of the edge area, which would lead to misjudgment of uneven coating.

[0042] Gray-scale normalization maps pixel values ​​to the range of 0-255, eliminating the effects of uneven lighting.

[0043] S103, the image transformation module performs a fast Fourier transform on the processed image based on the two-dimensional discrete Fourier transform formula to obtain a frequency domain image.

[0044] In one possible implementation, the formula for the two-dimensional discrete Fourier transform is:

[0045] in, These are the original image pixel values. It is a complex matrix in the frequency domain, containing amplitude and phase information; S104. The defect identification module performs defect identification on the frequency domain image processed by the image transformation module.

[0046] In one possible implementation, for the frequency domain characteristics of the normal coating region, the energy distribution is concentrated in the central low-frequency region, exhibiting a uniform diffusion pattern (indicating the absence of a periodic structure). Specifically, the quantitative indicator is calculated by integrating the energy of the outer ring band of the spectrum. When the proportion of high-frequency energy is less than B%, it indicates that the uniformity of the silicone oil coating is normal. Generally, B=5. Specifically, the high-frequency region is defined as a ring-shaped region with a radius greater than 200 pixels. Set two levels of judgment thresholds: a) High-frequency energy percentage < 3%: Superior grade b) 3% ≤ High-frequency energy percentage ≤ 5%: Qualified product If there is an abnormality in the uniformity of the silicone oil coating, it will manifest as a bright ring. The relationship between the bright ring radius r and the fringe spacing d is: d = N / r × pixel size; Where N is the image side length in pixels; Furthermore, by calculating the proportion of bright ring energy, if the proportion of bright ring energy is greater than 5%, it is determined that the uniformity of silicone oil coating is abnormal.

[0047] The release film production inspection method and system provided in this application, in terms of hardware architecture, can clearly capture the microscopic thickness changes of silicone oil coating through high-resolution imaging, ensuring detection sensitivity. Furthermore, a dual-light source scheme (white light + ultraviolet) is employed to enhance the optical contrast of coating unevenness; a 5° tilt installation avoids specular reflection interference, ensuring stable image quality; at the algorithm level, FFT frequency domain analysis + high-frequency energy ratio thresholding achieves objective and quantitative defect judgment. During the judgment process, noise suppression (Gaussian filtering) preserves key textures and avoids misjudgments. ROI extraction eliminates edge interference, improving analysis accuracy. Gray-level normalization eliminates the influence of uneven illumination, enhancing algorithm robustness. This solves the problems of untimely feedback and inability to directly correspond to coating amount due to interference from various factors in existing direct or indirect detection methods.

[0048] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0049] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of this application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

Claims

1. A method for testing the production of release film, characterized in that, include: The first parameter image of the film surface after being coated with silicone oil during the production process is obtained using image detection equipment; The image preprocessing module preprocesses the first parameter image to obtain a processed image; The image transformation module performs a fast Fourier transform on the processed image based on the two-dimensional discrete Fourier transform formula to obtain a frequency domain image; The defect identification module performs defect identification on the frequency domain image processed by the image transformation module to achieve production inspection.

2. The method for testing the production of release film as described in claim 1, characterized in that, The image inspection equipment is installed on the release film production line, and the installation position of the image inspection equipment is 300±5mm away from the film surface, with an inclination angle of 5° to the normal of the film surface.

3. The method for testing the production of release film as described in claim 2, characterized in that, The image inspection equipment is equipped with a coaxial white LED and ultraviolet light to excite the silicone oil fluorescent label.

4. The method for testing the production of release film as described in claim 1, characterized in that, The preprocessing process includes noise reduction, analysis region extraction, and grayscale normalization.

5. The method for testing the production of release film as described in claim 4, characterized in that, Noise reduction was achieved using a Gaussian filter with σ=1.5; The analysis region is defined using ROI extraction technology. Gray-scale normalization maps pixel values ​​to the range of 0-255.

6. The method for testing the production of release film as described in claim 1, characterized in that, The formula for the two-dimensional discrete Fourier transform is: in, These are the original image pixel values. It is a complex matrix in the frequency domain, containing amplitude and phase information.

7. The method for testing the production of release film as described in claim 1, characterized in that, The defect identification module performs defect identification on the frequency domain image processed by the image transformation module, including: By calculating the energy integral of the outer ring band of the spectrum, when the proportion of high-frequency energy is <B%, it indicates that the uniformity of silicone oil coating is normal. When the bright ring energy percentage is greater than 5%, it is considered that the uniformity of silicone oil coating is abnormal.

8. The method for testing the production of release film as described in claim 7, characterized in that, The high-frequency region is defined as a ring-shaped region with a radius greater than 200 pixels; Set two levels of judgment thresholds: High-frequency energy percentage <3%: Release film is of superior quality; Release film with 3% ≤ high-frequency energy percentage ≤ 5% is considered qualified. If there is an abnormality in the uniformity of the silicone oil coating, it will manifest as a bright ring. The relationship between the bright ring radius r and the fringe spacing d is: d = N / r × pixel size; Where N is the image side length in pixels; When the bright ring energy ratio is greater than 5%, it is determined to be an abnormality in the uniformity of silicone oil coating.

9. A release film production testing system, characterized in that, include: Hardware physical terminals and software control platforms; The hardware terminal includes an image detection device and a motion control module; the image detection device includes an imaging device installed on the release film production line for capturing images of the first parameter of the release film; the motion control module is used to control the movement and imaging of the image detection device, wherein the motion control module is equipped with a line scanning mode, which is coordinated with the high-speed production line. The software control platform is mainly used for preprocessing and defect analysis and identification of the first parameter image; including: Image preprocessing module, image transformation module, and defect recognition module; The image preprocessing module is used to perform noise reduction, analysis region extraction, and grayscale normalization on the first parameter image to obtain a processed image. The image transformation module is used to perform a fast Fourier transform on a single-processed image based on the two-dimensional discrete Fourier transform formula; The defect identification module is used to identify defects in the frequency domain image processed by the image transformation module, thereby enabling production inspection.

10. The release film production inspection system as described in claim 9, characterized in that, The shooting equipment is a high-resolution industrial camera, specifically a 50-megapixel CMOS camera paired with a telecentric lens.

Images