Weft yarn gap detection method and device based on image recognition

Abstract

The invention discloses a weft yarn gap detection method and device based on image recognition, which are used for detecting a glass fiber cloth cover before splitting operation, and the method comprises the following steps: turning on a background light source of a detection station; collecting a cloth cover image of the glass fiber cloth cover; identifying a plurality of shadow bright and dark areas of the cloth cover image by using the trained identification model; and analyzing according to each shadow light and shade area of the cloth cover image to obtain the weft yarn gap condition of the area. According to the method, the image of the glass fiber cloth cover is collected, the weft twisting form is judged according to the light intensity, the size and density of the weft gaps on the cloth roll are reflected according to the number of the weft twisting states, operation is easy, and the production process does not need to be interrupted.

Description

Technical Field

[0001] This invention specifically relates to a method and device for detecting weft gaps based on image recognition. Background Technology

[0002] In electronic-grade ultra-thin fiberglass cloth, the fiberglass yarn used is relatively fine, which will cause gaps between the weft yarns. When these gaps reach a certain level, they will cause pinhole defects in the downstream PP production. If not detected in time, they will cause short circuits in the PCB.

[0003] Currently, the main methods for determining the gap size are: measuring the air permeability of the sample fabric and measuring the width of the yarn bundles on the fabric surface. However, existing air permeability and yarn width measurements are offline and destructive tests that can only reflect the condition of the test sample and are used as a reference for the entire roll of fabric. When production fluctuates significantly, the detection error is relatively large. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a weft gap detection method and device based on image recognition.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] On one hand, this invention discloses an image recognition-based weft gap detection method for detecting the surface of fiberglass fabric before the fiber opening operation, including:

[0007] Step S1: Turn on the background light source at the inspection station;

[0008] Step S2: Acquire an image of the fiberglass fabric surface;

[0009] Step S3: Use the trained recognition model to identify several light and shadow areas in the fabric image;

[0010] Step S4: Analyze each light and shadow area of ​​the fabric image to determine the weft yarn gap in that area.

[0011] Based on the above technical solution, the following improvements can be made:

[0012] As a preferred option, step S1 specifically includes:

[0013] Step S1.1: Turn on the background light source at the inspection station;

[0014] Step S1.2: Acquire the light source detection image;

[0015] Step S1.3: Analyze the light source intensity of the acquired light source detection image to determine whether the intensity of the background light source meets the requirements.

[0016] If satisfied, proceed to step S2;

[0017] Otherwise, adjust the intensity of the background light source until it meets the requirements.

[0018] As a preferred option, step S2 specifically includes:

[0019] Step S2.1: Acquire an image of the fiberglass fabric surface;

[0020] Step S2.2: Preprocess the acquired fabric image;

[0021] Step S2.3: Perform preliminary screening on the preprocessed fabric images to determine whether they meet the image requirements.

[0022] If satisfied, proceed to step S3;

[0023] Otherwise, re-acquire images of the fabric until the image requirements are met.

[0024] As a preferred approach, before step S3 is executed, the recognition model is trained using several fabric images.

[0025] As a preferred option, step S4 specifically includes the following:

[0026] The grayscale value of each light and dark area in the fabric image is used to determine whether the difference between its grayscale value and that of other areas exceeds a threshold.

[0027] If this is exceeded, the warp and weft yarns in that area will interweave or the weft yarns will twist, resulting in gaps in subsequent weft yarns;

[0028] Otherwise, the area is normal.

[0029] On the other hand, the present invention also discloses a weft gap detection device based on image recognition, used to detect the surface of fiberglass fabric before the fiber opening operation, including:

[0030] The background light source control module is used to turn on the background light source of the inspection station.

[0031] Image acquisition module, used to acquire images of the fiberglass fabric surface;

[0032] The recognition module is used to identify several light and shadow areas in the fabric image using a trained recognition model.

[0033] The analysis module is used to analyze each light and shadow area of ​​the fabric image to determine the weft yarn gap in that area.

[0034] As a preferred embodiment, the background light source control module includes:

[0035] Background light source control unit, used to turn on the background light source of the inspection station;

[0036] A light source detection image acquisition unit is used to acquire light source detection images;

[0037] The light source intensity judgment unit is used to judge the light source intensity of the acquired light source detection image and determine whether the intensity of the background light source meets the requirements.

[0038] If the conditions are met, the image acquisition module will proceed with subsequent operations;

[0039] Otherwise, adjust the intensity of the background light source until it meets the requirements.

[0040] As a preferred embodiment, the image acquisition module includes:

[0041] Image acquisition unit, used to acquire images of the fiberglass fabric surface;

[0042] The preprocessing unit is used to preprocess the acquired fabric images;

[0043] The preliminary screening unit is used to perform a preliminary screening of the preprocessed fabric image to determine whether it meets the image requirements.

[0044] If the conditions are met, the recognition module will perform subsequent operations on the fabric image;

[0045] Otherwise, re-acquire images of the fabric until the image requirements are met.

[0046] As a preferred approach, the recognition model is trained using several fabric images before the recognition module is executed.

[0047] As a preferred embodiment, the analysis module specifically includes the following:

[0048] The grayscale value of each light and dark area in the fabric image is used to determine whether the difference between its grayscale value and that of other areas exceeds a threshold.

[0049] If this is exceeded, the warp and weft yarns in that area will interweave or the weft yarns will twist, resulting in gaps in subsequent weft yarns;

[0050] Otherwise, the area is normal.

[0051] This invention discloses a weft yarn gap detection method and device based on image recognition. It acquires images of the fiberglass fabric surface, determines the weft yarn twisting shape based on the brightness of the light, and reflects the size and density of the weft yarn gap on the fabric roll by the number of weft yarn twisting states. The operation is simple and does not require interruption of the production process. Attached Figure Description

[0052] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a flowchart of a weft yarn gap detection method provided in an embodiment of the present invention. Detailed Implementation

[0054] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] The expression “includes” is an “open-ended” expression, which means that there is a corresponding component or step, and should not be interpreted as excluding additional components or steps.

[0057] Through creative thinking, the inventor discovered that at the locations where pinhole defects occur, the yarn bundle narrows and the gap widens due to weft yarn twisting. Therefore, it was determined that during the fiber opening process of the fabric roll, the weft yarn width increases from 0.1-0.2 mm to 0.3-0.4 mm. If there is twisting in the weft yarn, the width of the yarn bundle cannot be expanded, resulting in a larger gap at this location, which easily leads to pinhole defects.

[0058] To achieve the objectives of this invention, in some embodiments of the weft gap detection method and device based on image recognition, the weft gap detection method is used to detect the surface of the fiberglass fabric before the fiber opening operation, such as... Figure 1 As shown, it includes:

[0059] Step S1: Turn on the background light source at the inspection station;

[0060] Step S2: Acquire an image of the fiberglass fabric surface;

[0061] Step S3: Use the trained recognition model to identify several light and shadow areas in the fabric image;

[0062] Step S4: Analyze each light and shadow area of ​​the fabric image to determine the weft yarn gap in that area.

[0063] The following is a detailed explanation of each step.

[0064] Step S1 specifically includes:

[0065] Step S1.1: Turn on the background light source at the inspection station;

[0066] Step S1.2: Acquire the light source detection image;

[0067] Step S1.3: Analyze the light source intensity of the acquired light source detection image to determine whether the intensity of the background light source meets the requirements.

[0068] If satisfied, proceed to step S2;

[0069] Otherwise, adjust the intensity of the background light source until it meets the requirements.

[0070] The above-mentioned light source detection images can be acquired using a board with a black and white checkered pattern, or directly using a fiberglass cloth surface.

[0071] Since the brightness of a fabric image can sometimes be affected by natural light or the external environment, it is necessary to adjust the intensity of the background light source.

[0072] Step S2 specifically includes:

[0073] Step S2.1: Acquire an image of the fiberglass fabric surface;

[0074] Step S2.2: Preprocess the acquired fabric image;

[0075] Step S2.3: Perform preliminary screening on the preprocessed fabric images to determine whether they meet the image requirements.

[0076] If satisfied, proceed to step S3;

[0077] Otherwise, re-acquire images of the fabric until the image requirements are met.

[0078] Image requirements can include requirements for image resolution, image pixels, image size, image color, and so on.

[0079] Before step S3 is executed, the recognition model is trained using several fabric images.

[0080] Step S4 specifically includes the following:

[0081] The grayscale value of each light and dark area in the fabric image is used to determine whether the difference between its grayscale value and that of other areas exceeds a threshold.

[0082] If this is exceeded, the warp and weft yarns in that area will interweave or the weft yarns will twist, resulting in gaps in subsequent weft yarns;

[0083] Otherwise, the area is normal.

[0084] Furthermore, embodiments of the present invention also disclose a weft gap detection device based on image recognition, used to detect the surface of fiberglass fabric before the fiber opening operation, including:

[0085] The background light source control module is used to turn on the background light source of the inspection station.

[0086] Image acquisition module, used to acquire images of the fiberglass fabric surface;

[0087] The recognition module is used to identify several light and shadow areas in the fabric image using a trained recognition model.

[0088] The analysis module is used to analyze each light and shadow area of ​​the fabric image to determine the weft yarn gap in that area.

[0089] Furthermore, the background light source control module includes:

[0090] Background light source control unit, used to turn on the background light source of the inspection station;

[0091] A light source detection image acquisition unit is used to acquire light source detection images;

[0092] The light source intensity judgment unit is used to judge the light source intensity of the acquired light source detection image and determine whether the intensity of the background light source meets the requirements.

[0093] If the conditions are met, the image acquisition module will proceed with subsequent operations;

[0094] Otherwise, adjust the intensity of the background light source until it meets the requirements.

[0095] Furthermore, the image acquisition module includes:

[0096] Image acquisition unit, used to acquire images of the fiberglass fabric surface;

[0097] The preprocessing unit is used to preprocess the acquired fabric images;

[0098] The preliminary screening unit is used to perform a preliminary screening of the preprocessed fabric image to determine whether it meets the image requirements.

[0099] If the conditions are met, the recognition module will perform subsequent operations on the fabric image;

[0100] Otherwise, re-acquire images of the fabric until the image requirements are met.

[0101] Furthermore, before the recognition module is executed, the recognition model is trained using several fabric images.

[0102] Furthermore, the analysis module specifically includes the following:

[0103] The grayscale value of each light and dark area in the fabric image is used to determine whether the difference between its grayscale value and that of other areas exceeds a threshold.

[0104] If this is exceeded, the warp and weft yarns in that area will interweave or the weft yarns will twist, resulting in gaps in subsequent weft yarns;

[0105] Otherwise, the area is normal.

[0106] This invention discloses a weft yarn gap detection method and device based on image recognition. It acquires images of the fiberglass fabric surface, determines the weft yarn twisting shape based on the brightness of the light, and reflects the size and density of the weft yarn gap on the fabric roll by the number of weft yarn twisting states. The operation is simple and does not require interruption of the production process.

[0107] This invention can reduce the inflow of potentially defective products to downstream customers due to production fluctuations, reduce abnormalities in downstream customers, and increase the market competitiveness of products.

[0108] It should be understood that the various techniques described herein can be implemented in combination with hardware or software, or a combination thereof. Thus, the methods and apparatus of the present invention, or certain aspects or portions thereof, can take the form of program code (i.e., instructions) embedded in a tangible medium, such as a floppy disk, CD-ROM, hard disk, or any other machine-readable storage medium, wherein when the program is loaded into and executed by a machine such as a computer, that machine becomes an apparatus for practicing the present invention.

[0109] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the present invention. Various changes and modifications can be made to the present invention without departing from its spirit and scope. All such changes and modifications fall within the scope of the present invention as claimed, which is defined by the appended claims and their equivalents.

Claims

1. A weft yarn gap detection method based on image recognition, characterized in that: Used to test the glass fiber cloth surface before fiber opening operation, including: Step S1: Turn on the background light source of the detection station; Step S2: collecting a surface image of the glass fiber cloth; Step S3: using the trained recognition model to identify several light and dark areas of the cloth image; Step S4: Analyze each light and dark area of ​​the cloth surface image to obtain the weft gap situation in the area.

2. The weft yarn gap detection method according to claim 1, characterized in that: The step S1 specifically includes: Step S1.1: Turn on the background light source of the detection station; Step S1.2: collecting light source detection images; Step S1.3: Perform light source intensity judgment on the collected light source detection image to determine whether the intensity of the background light source meets the requirements. If satisfied, proceed to step S2; Otherwise, adjust the intensity of the background light source until the intensity of the background light source meets the requirement.

3. The weft yarn gap detection method according to claim 1, characterized in that: The step S2 specifically includes: Step S2.1: collecting a surface image of the fiberglass cloth; Step S2.2: preprocessing the collected cloth image; Step S2.3: Perform a preliminary screening on the preprocessed cloth image to determine whether it meets the image requirements. If satisfied, proceed to step S3; Otherwise, re-capture the cloth image until the image requirements are met.

4. The weft yarn gap detection method according to claim 1, characterized in that: Before executing step S3, a recognition model is trained using a number of cloth images.

5. The weft yarn gap detection method according to claim 1, characterized in that: The step S4 specifically includes the following contents: Judging according to the gray value of each light and dark area of ​​the cloth surface image, whether the difference between the gray value of each area and the gray value of other areas exceeds a threshold; If it exceeds, the warp and weft yarns in this area will overlap or the weft yarns will be twisted, and gaps will be created in the subsequent weft yarns; Otherwise, the region is normal.

6. A weft yarn gap detection device based on image recognition, characterized in that: Used to test the glass fiber cloth surface before fiber opening operation, including: A background light source control module, used to turn on the background light source of the detection station; An image acquisition module, used for acquiring a surface image of the glass fiber cloth; A recognition module, used for recognizing a plurality of light and dark areas of a cloth surface image by using a trained recognition model; The analysis module is used to analyze each light and dark area of ​​the cloth surface image to obtain the weft yarn gap situation in the area.

7. The weft gap detection device according to claim 6, characterized in that: The background light source control module comprises: A background light source control unit, used to turn on the background light source of the detection station; A light source detection image acquisition unit, used for acquiring a light source detection image; The light source intensity judgment unit is used to judge the light source intensity of the collected light source detection image to determine whether the intensity of the background light source meets the requirements. If satisfied, the image acquisition module performs subsequent operations; Otherwise, adjust the intensity of the background light source until the intensity of the background light source meets the requirement.

8. The weft gap detection device according to claim 6, characterized in that: The image acquisition module comprises: An image acquisition unit, used for acquiring a surface image of the glass fiber cloth; A preprocessing unit, used for preprocessing the collected cloth surface image; The preliminary screening unit is used to perform preliminary screening on the pre-processed cloth image to determine whether it meets the image requirements. If the conditions are met, the recognition module performs subsequent operations on the cloth surface image; Otherwise, re-capture the cloth image until the image requirements are met.

9. The weft yarn gap detection device according to claim 6, characterized in that: Before the recognition module is executed, the recognition model is trained using a number of cloth images.

10. The weft gap detection device according to claim 6, characterized in that: The analysis module specifically includes the following contents: Judging according to the gray value of each light and dark area of ​​the cloth surface image, whether the difference between the gray value of each area and the gray value of other areas exceeds a threshold; If it exceeds, the warp and weft yarns in this area will overlap or the weft yarns will be twisted, and gaps will be created in the subsequent weft yarns; Otherwise, the region is normal.

Images