Non-woven fabric thickness uniformity detection method and device based on image processing

Abstract

The invention belongs to the technical field of non-woven fabric detection, and particularly discloses a non-woven fabric thickness uniformity detection method and device based on image processing, and the detection method comprises the steps: collecting a non-woven fabric image, preprocessing the collected image, and calculating a color histogram and integral optical density of the image. Dividing image units, comparing the integral optical density to identify areas with uneven thickness, selecting a plurality of sample areas on the sample to evaluate the thickness uniformity, and calculating a thickness difference rate and an angle quality difference rate according to the integral optical density of the intermediate sample; the detection device comprises a roller set, a camera and computer equipment, the computer equipment processes image data and recognizes the area with the uneven thickness, a closed box and ground glass are arranged to isolate external interference and uniform illumination conditions, and optimization of the non-woven fabric production process and improvement of the product quality are promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of non-woven fabric detection, and particularly to a method and device for detecting the thickness uniformity of non-woven fabrics based on image processing. Background Art

[0002] Optical density, as a characterization of the light-shielding ability of a material, can indirectly reflect the relative content of a certain substance at a point in a sample. In the production of non-woven fabrics, especially in the production process of multi-layer non-woven fabrics, each layer of non-woven fabric is usually formed by coating glue or other adhesives on a single-layer non-woven fabric. However, during the gluing process, due to various factors, such as the accuracy of the gluing equipment, the gluing speed, the uniformity of the non-woven fabric surface, etc., the glue distribution is often uneven.

[0003] This uneven glue distribution will directly affect the thickness uniformity of the multi-layer non-woven fabric. Specifically, in the area with more glue, the adhesion between the non-woven fabric layers will be tighter, increasing the thickness of this area; while in the area with less glue, there may be poor adhesion between the layers, or even voids, resulting in a decrease in the thickness of this area. Therefore, the phenomenon of uneven thickness is likely to occur during the production of multi-layer non-woven fabrics.

[0004] To solve the above problems, the present invention proposes a method for detecting the thickness uniformity of non-woven fabrics based on image processing. By collecting the image of the non-woven fabric and performing a series of processes, it can accurately identify the areas with uneven thickness in the non-woven fabric, providing strong support for quality control during the production process. Summary of the Invention

[0005] In view of the above-mentioned disadvantages of the prior art, the purpose of the present invention is to provide a method and device for detecting the thickness uniformity of non-woven fabrics based on image processing, which are used to solve the problems in the prior art such as low detection accuracy of the thickness uniformity of non-woven fabrics, cumbersome detection process, easy influence of detection results by human factors, and inability to comprehensively evaluate the overall quality of non-woven fabrics.

[0006] To achieve the above purpose and other related purposes, the present invention provides a method for detecting the thickness uniformity of non-woven fabrics based on image processing, including the following steps: S1. Collect the image of the non-woven fabric; S2. Preprocess the collected image; S3. Calculate the color histogram and integral optical density of the image: Calculate the color histogram H(k) describing the global color distribution in the non-woven fabric image, where H(k) represents the proportion of pixels with gray level k in the total number of pixels, and the calculation formula is as follows: Where, Indicates the number of pixels with a gray level greater than or equal to , Indicates the number of pixels with a gray level greater than , Indicates the total number of pixels in the image; Calculate the integrated optical density IOD of the image through the histogram, and the calculation formula is as follows: where IOD represents the weighted sum of the gray values of all pixels in the image; S4. Divide the image into several units, calculate the integrated optical density for each unit, and identify the regions with uneven thickness in the image by comparing the differences between the integrated optical densities of each unit and the overall average integrated optical density; S5. Select multiple test regions on a single non-woven fabric sample, including multiple middle test samples located in the central region of the sample and multiple corner test samples located at the corners of the sample, and calculate the integrated optical density for each test region respectively; S6. Calculate the thickness difference rate according to the integrated optical density of the middle test samples. The thickness difference rate is used to reflect the thickness change degree of the central region of the non-woven fabric at different positions. Calculate the corner quality difference rate according to the integrated optical density of the corner test samples. The corner quality difference rate reflects the quality change degree of the non-woven fabric at the edge or corner positions.

[0007] Optionally, the image preprocessing steps include: Gray conversion: Convert the collected RGB image into a grayscale image, and the conversion calculation formula is as follows: In the formula, represents the gray value, represents the value of the red channel, represents the value of the blue channel, represents the value of the blue channel; Denoising processing: Remove the noise interference in the image; Enhance the contrast: Improve the contrast between the non-woven fabric region and the background in the image; Edge detection: Use an edge detection algorithm to extract the edge information of the non-woven fabric; Image segmentation: Segment the non-woven fabric region from the background according to the edge information.

[0008] Optionally, the identification of the regions with uneven thickness in the image includes: Calculate the average value of the integrated optical densities of all cells of the non-woven fabric with uniform thickness, and use it as the integrated optical density threshold; The integrated optical density calculated for each unit is compared with the integrated optical density threshold. If the difference between the integrated optical density of a unit and the threshold is greater than the set threshold, it is determined that the thickness of the unit is different as a whole, that is, it is an area of ​​uneven thickness in the image.

[0009] Optionally, the thickness difference rate is calculated as follows: in, is the thickness difference rate, in %, IOD_i is the integrated optical density of the i-th middle sample, is the average integrated optical density of m middle samples.

[0010] Optionally, the calculation formula of the angular mass deviation rate is as follows: in, is the angular quality difference rate, in %, and (IOD)_J represents the lowest value of the integrated optical density of multiple angular samples. is the average integrated optical density of m middle samples.

[0011] On the other hand, the present invention provides a non-woven fabric thickness uniformity detection device based on image processing, comprising: A roller assembly, between which the nonwoven fabric is wound; A camera, wherein the camera is used to capture images of the nonwoven fabric; A computer device, wherein the host computer is electrically connected to the camera, and the computer device includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the non-woven fabric thickness uniformity detection method based on image processing are implemented.

[0012] Optionally, the non-woven fabric is disposed in a closed box, and a light source is disposed in the closed box directly below the non-woven fabric.

[0013] Optionally, frosted glass is arranged between the light source and the non-woven fabric.

[0014] As described above, the non-woven fabric thickness uniformity detection method based on image processing proposed by the present invention has the following beneficial effects: The present invention can accurately capture the slight thickness difference on the surface of the non-woven fabric through image processing and calculation of the integrated optical density, thereby realizing high-precision detection of the thickness uniformity of the non-woven fabric, avoiding the errors and subjectivity that may exist in traditional detection methods, and improving the accuracy and reliability of detection; The present invention not only considers the thickness difference in the central region of the non-woven fabric, but also calculates the angular mass difference rate by selecting angular specimens, so as to comprehensively evaluate the thickness uniformity of the non-woven fabric over the entire sample. The comprehensive detection method helps to discover the quality changes in the edge or corner positions of the non-woven fabric. The present invention improves the consistency of product quality and is applicable to different types of non-woven fabrics, including single-layer non-woven fabrics and multi-layer non-woven fabrics. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It shows a schematic flow chart of a method for detecting the thickness uniformity of a non-woven fabric based on image processing according to an embodiment of the present invention; Figure 2 It shows a schematic diagram of sampling the non-woven fabric sample area according to an embodiment of the present invention; Figure 3 It shows a schematic structural diagram of a device for detecting the thickness uniformity of a non-woven fabric based on image processing according to an embodiment of the present invention.

[0016] Description of the reference numerals in the drawings: Non-woven fabric 1, closed box body 2, roller group 3, camera 4, computer device 5, ground glass 6, light source 7. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The following specific examples illustrate the embodiments of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

[0018] It should be noted that the diagrams provided in this embodiment only illustrate the basic concept of the present invention in a schematic manner. Therefore, only the components related to the present invention are shown in the diagrams, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and ratio of each component in actual implementation can be arbitrarily changed, and the component layout type may also be more complex. The structures, ratios, sizes, etc. shown in the diagrams of this specification are only used to cooperate with the content disclosed in the specification for those skilled in this technology to understand and read, and are not used to limit the limiting conditions for the implementation of the present invention. Therefore, they do not have technical substance. Any modification of the structure, change of the proportional relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope that can be covered by the technical content disclosed in the present invention. At the same time, the terms such as "upper", "lower", "left", "right", "middle", and "one" cited in this specification are only for the convenience of clear narration and are not used to limit the scope for the implementation of the present invention. The change or adjustment of their relative relationships, without substantial change in the technical content, should also be regarded as the scope within which the present invention can be implemented.

[0019] Referring to Figure 1 , the present invention provides a method for detecting the thickness uniformity of non-woven fabrics based on image processing, comprising the following steps: S1. Collect an image of the non-woven fabric; When collecting the image of the non-woven fabric, it is necessary to ensure that the light can evenly irradiate the surface of the non-woven fabric to obtain clear visual information. For this purpose, appropriate light sources and lighting conditions can be set during image collection to ensure that the light intensity on the surface of the non-woven fabric is consistent, so as to ensure that the collected image of the non-woven fabric can truly reflect the thickness of its surface during subsequent processing.

[0020] When collecting the image, a high-resolution camera needs to be used to ensure the clarity and richness of details of the image. The camera should be fixedly installed on the detection device to ensure the consistency of the position and angle of the images collected each time. In addition, parameters such as the exposure time and focal length of the camera can be adjusted according to actual needs to obtain the best image effect.

[0021] S2. Preprocess the collected image; Specifically, the image preprocessing steps include gray-scale conversion, denoising, contrast enhancement, and edge detection, etc., where: Gray-scale conversion: Convert the collected RGB image into a gray-scale image. The gray-scale image only contains brightness information and does not contain color information, so it can simplify subsequent processing steps. The calculation formula for gray-scale conversion is as follows: In the formula, represents the gray value, represents the value of the red channel, represents the value of the blue channel, represents the value of the blue channel; Denoising: Remove the noise interference in the image. Noise may be generated for various reasons, such as camera sensor noise, environmental light interference, etc. The denoising process uses a filtering algorithm. In this embodiment, at least one filtering algorithm is used for denoising, specifically including mean filtering, median filtering, and Gaussian filtering. Through the above filtering algorithms, the noise in the image can be smoothed while retaining the detail information of the image.

[0022] Contrast enhancement: Improve the contrast between the non-woven fabric area and the background in the image. In this embodiment, at least one contrast enhancement algorithm is used, specifically including histogram equalization and adaptive contrast enhancement algorithm. Through the above contrast enhancement algorithms, the gray-scale distribution of the image can be adjusted to make the contrast between the non-woven fabric area and the background more obvious.

[0023] Edge detection: Use edge detection algorithms to extract the edge information of the non-woven fabric. Edge detection is used to identify edge features in an image. In this embodiment, at least one edge detection algorithm is adopted, specifically including Canny edge detection and Sobel edge detection.

[0024] Image segmentation: Segment the non-woven fabric area from the background according to the edge information. Image segmentation is used to separate the target area from the background area in an image. In this embodiment, at least one segmentation algorithm is adopted, specifically including a threshold-based segmentation algorithm and a region-based segmentation algorithm, to accurately segment the non-woven fabric area from the background.

[0025] S3. Calculate the color histogram and integral optical density of the image; The color histogram is used to describe the global color distribution in the image. In this embodiment, by calculating the color histogram of the non-woven fabric image , the pixel distribution of different gray levels in the image can be obtained. The calculation formula of the color histogram is as follows: where represents the number of pixels with a gray level greater than or equal to , represents the number of pixels with a gray level greater than , represents the total number of pixels in the image; The integral optical density is the weighted sum of the gray values of all pixels in the image, reflecting the overall brightness of the image. By calculating the integral optical density of the non-woven fabric image, the gray distribution of the image can be further understood. The calculation formula of the integral optical density is as follows: where represents the gray level, represents the proportion of pixels with gray level k in the total pixels; S4. Divide the image into several units, calculate the integral optical density for each unit, and identify the uneven-thickness areas in the image by comparing the difference between the integral optical density of each unit and the overall average integral optical density. The specific steps are as follows: S41. Calculate the average value of the integral optical density of all cells of the evenly thick non-woven fabric and use it as the integral optical density threshold. This threshold is used as a benchmark for judging the thickness uniformity of the non-woven fabric.

[0026] S42. Compare the integral optical density calculated for each unit with the integral optical density threshold. If the difference value between the integral optical density of a certain unit and the threshold is greater than the set threshold, it is determined that the thickness of the unit is different as a whole, that is, the uneven-thickness area in the image.

[0027] S5. To more comprehensively evaluate the thickness uniformity of the non-woven fabric, multiple specimen areas are selected on a single non-woven fabric sample, including multiple intermediate specimens located in the central area of the sample and multiple corner specimens located at the corners of the sample. The integrated optical density is calculated for each specimen area respectively. The specific steps are as follows: S51. As Figure 2 shown, determine the positions of multiple specimen areas on the non-woven fabric sample. These specimen areas are evenly distributed in the central area and the corner areas of the sample to ensure the comprehensiveness and accuracy of the evaluation. Further, in this embodiment, 8 samples are taken in the central area of the sample area, and 1 sample is taken at each of the four corners of the sample, for a total of 4 corner areas; S52. Perform image acquisition and preprocessing on each specimen area to obtain a clear grayscale image; S53. Calculate the integrated optical density of each specimen area. By comparing the integrated optical densities of different specimen areas, understand the thickness variation of the non-woven fabric at different positions.

[0028] S6. Calculate the thickness difference rate according to the integrated optical density of the intermediate specimens. The thickness difference rate is used to reflect the thickness variation degree of the central area of the non-woven fabric at different positions. Calculate the corner quality difference rate according to the integrated optical density of the corner specimens. The corner quality difference rate reflects the quality variation degree of the non-woven fabric at the edge or corner positions Among them, the calculation formula of the thickness difference rate is as follows: Among them, is the thickness difference rate, with the unit of %, IOD_i is the integrated optical density of the i-th intermediate specimen, is the average value of the integrated optical densities of m intermediate specimens. The larger the thickness difference rate, the more obvious the thickness variation of the central area of the non-woven fabric.

[0029] The calculation formula of the corner quality deviation rate is as follows: Among them, is the corner quality difference rate, with the unit of %, (IOD)_J represents the lowest measured value of the integrated optical densities of multiple corner specimens, is the average value of the integrated optical densities of m intermediate specimens. The corner quality difference rate reflects the quality variation degree of the non-woven fabric at the edge or corner positions. If the corner quality difference rate is large, it indicates that the thickness variation of the non-woven fabric in the corner area is large, and there are quality problems, such as misalignment between layers of the non-woven fabric and uneven glue.

[0030] As Figure 3 shown, the present invention also provides a non-woven fabric thickness uniformity detection device based on image processing, including components such as a roller group 3, a camera 4, and a computer device 5; The roller set 3 is used to support and convey the non-woven fabric 1 sample. During the detection process, the non-woven fabric 1 sample is wound between the roller sets 3 and conveyed by the rotation of the rollers.

[0031] The camera 4 is used to collect images of the non-woven fabric 1. The camera 4 is fixedly installed on the detection device to ensure the consistency of the position and angle of the images collected each time. The selection of the camera 4 should be adjusted according to actual needs to ensure the clarity and richness of details of the images.

[0032] The computer device 5 is electrically connected to the camera 4 and is used to receive and process the image data collected by the camera 4. The computer device 5 includes a memory, a processor, etc. Computer-readable instructions are stored in the memory. When the processor executes the instructions, the steps of the above-mentioned non-woven fabric 1 thickness uniformity detection method based on image processing are realized. The computer device 5 has the capabilities of data processing and image analysis to quickly and accurately complete tasks such as preprocessing of images, calculation of color histograms, calculation of integrated optical density, and identification of non-uniform thickness regions.

[0033] In addition, the computer device 5 should also have a friendly user interface and interaction functions to facilitate users to operate and view the detection results. Users can input detection parameters, view detection results, and generate detection reports through the computer device 5.

[0034] Furthermore, the non-woven fabric 1 is arranged in the closed box body 2 to isolate external interference. The closed box body 2 can ensure the consistency of the lighting conditions during the detection process, avoid the influence of external light on the images collected by the camera 4. At the same time, the closed box body 2 can also reduce the pollution and interference of dust and sundries to the non-woven fabric 1. A light source 7 is arranged directly below the non-woven fabric 1 in the closed box body 2. The light source 7 can provide uniform and stable lighting conditions to ensure the consistency of the light intensity on the surface of the non-woven fabric 1.

[0035] Even further, a ground glass 6 is arranged between the light source 7 and the non-woven fabric 1. The ground glass 6 further homogenizes the lighting conditions, reduces the glare and reflection generated by the direct irradiation of the light source 7 on the camera 4 lens. At the same time, the ground glass 6 can also improve the softness and uniformity of the light source 7, making the collected images clearer and more accurate.

[0036] In summary, the non-woven fabric thickness uniformity detection method and device based on image processing of the present invention have many beneficial effects such as high precision, comprehensive detection, automation and intelligence, strong adaptability, easy implementation and maintenance, and improvement of product quality. These effects jointly promote the optimization of the non-woven fabric production process and the improvement of product quality.

[0037] The above embodiments are only illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by those with ordinary knowledge in the technical field without departing from the spirit and technical idea disclosed by the present invention should still be covered by the claims of the present invention.

Claims

1. A method for detecting thickness uniformity of nonwoven fabrics based on image processing, characterized in that: The following steps are involved: S1, collecting images of non-woven fabrics; S2, preprocessing the collected images; S3, calculating the color histogram and integrated optical density of the image: Calculate the color histogram H(k) describing the global distribution of colors in the non-woven fabric image, where H(k) represents the proportion of pixels with gray level k to the total pixels. The calculation formula is as follows: in, Indicates that the gray level is greater than or equal to The number of pixels, Indicates that the gray level is greater than The number of pixels, Indicates the total number of pixels in the image; The integrated optical density IOD of the image is calculated by the histogram, and the calculation formula is as follows: Among them, IOD represents the weighted sum of the grayscale values ​​of all pixels in the image; S4, dividing the image into a number of units, and calculating the integrated optical density of each unit, and identifying the uneven thickness area in the image by comparing the difference between the integrated optical density of each unit and the overall average integrated optical density; S5. Select multiple sample areas on a single non-woven fabric sample, including multiple middle samples located in the center area of ​​the sample and multiple corner samples located in the corners of the sample, and calculate the integrated optical density of each sample area; S6. Calculate the thickness difference rate based on the integrated optical density of the middle sample, and the thickness difference rate is used to reflect the thickness change degree of the central area of ​​the non-woven fabric at different positions. Calculate the angular quality difference rate based on the integrated optical density of the corner sample, and the angular quality difference rate reflects the quality change degree of the non-woven fabric at the edge or corner position.

2. The method for detecting thickness uniformity of nonwoven fabrics based on image processing according to claim 1, characterized in that: The image preprocessing step comprises: Grayscale conversion: Convert the collected RGB image to a grayscale image. The conversion calculation formula is as follows: In the formula, Represents the gray value, Represents the value of the red channel, represents the value of the blue channel, Represents the value of the blue channel; Denoising: remove noise interference in the image; Enhance contrast: Improve the contrast between the non-woven fabric area and the background in the image; Edge detection: Use edge detection algorithm to extract edge information of non-woven fabrics; Image segmentation: Segment the non-woven fabric area from the background based on edge information.

3. The method for detecting thickness uniformity of nonwoven fabrics based on image processing according to claim 2, characterized in that: The identifying of the uneven thickness area in the image includes: Calculate the average integrated optical density of all cells of the thick and uniform non-woven fabric and use it as the integrated optical density threshold; The integrated optical density calculated for each unit is compared with the integrated optical density threshold. If the difference between the integrated optical density of a unit and the threshold is greater than the set threshold, it is determined that the thickness of the unit is different as a whole, that is, it is an area of ​​uneven thickness in the image.

4. The method for detecting thickness uniformity of nonwoven fabrics based on image processing according to claim 3, characterized in that: The calculation formula of the thickness difference rate is as follows: in, is the thickness difference rate, in %, IOD_i is the integrated optical density of the i-th middle sample, is the average integrated optical density of m middle samples.

5. The method for detecting thickness uniformity of nonwoven fabrics based on image processing according to claim 4, characterized in that: The calculation formula of the angular mass deviation rate is as follows: in, is the angular quality difference rate, in %, and (IOD)_J represents the lowest value of the integrated optical density of multiple angular samples. is the average integrated optical density of m middle samples.

6. A non-woven fabric thickness uniformity detection device based on image processing, characterized in that: include: A roller assembly, between which the nonwoven fabric is wound; A camera, wherein the camera is used to capture images of the nonwoven fabric; A computer device, wherein the host computer is electrically connected to the camera, and the computer device includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the non-woven fabric thickness uniformity detection method based on image processing as described in any one of claims 1 to 5 are implemented.

7. The nonwoven fabric thickness uniformity detection device based on image processing according to claim 6, characterized in that: The non-woven fabric is arranged in a closed box, and a light source is arranged in the closed box and is located directly below the non-woven fabric.

8. The nonwoven fabric thickness uniformity detection device based on image processing according to claim 7 is characterized in that: Frosted glass is arranged between the light source and the non-woven fabric.

Images