A quality inspection system applied to fiber cloth

Abstract

The application provides a quality inspection system applied to fiber cloth, which is characterized by comprising a fan, a database, an image acquisition module, a flaw identification module, an alien object type analysis module and a flaw marking module.

Description

Technical Field

[0001] This invention relates to the technical field of quality inspection systems, and more specifically to a quality inspection system applied to fiber fabrics. Background Technology

[0002] After the fiber fabric is produced, the products need to undergo quality inspection, and qualified products are selected for packaging. Currently, this is mainly done manually using tools such as rulers and scales to measure the size, weight, and shape of the products, inspect them to see if they meet production standards, and issue inspection reports. This process is cumbersome, inefficient, and not suitable for large-scale product inspection. Furthermore, during the inspection process, there may be defects on the fabric surface. Manually inspecting for fabric defects is time-consuming and may result in missed defects, making it too inefficient. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a quality inspection system for fiber fabrics to overcome the above-mentioned defects in the existing technology.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A quality inspection system for fiber fabrics, including a fan, and A database containing pre-set fabric information, including fabric patterns; An image acquisition module is used to acquire an image of the fabric surface as an image to be analyzed. The defect identification module is used to acquire the image to be analyzed in the image acquisition module, determine whether there are defects on the fabric surface through a semantic segmentation model, and if so, generate a defect signal and perform morphological analysis on the defect to obtain the defect morphology, which includes stain morphology and foreign object morphology. The foreign object type analysis module, when the defect form is a foreign object form, acquires the fabric image to be analyzed under different fan speeds, analyzes the outline in the image to be analyzed to obtain the swing amplitude of the foreign object, and indexes the foreign object type in the database based on the swing amplitude. The defect marking module is used to acquire defect signals from the defect recognition module, mark the location of defects and the shape of foreign objects in the image to be analyzed, and generate a detection report based on the defect marking information.

[0005] Preferably, the defect identification module is equipped with a defect location marking unit. The defect location marking unit is used to acquire the image to be analyzed, and to cut the fabric area map by a preset semantic segmentation model. The fabric area map is then subjected to binarization image processing to obtain a binarized image. The binarized image is compared with the fabric pattern binarized image in the database to obtain the defect area. The contour line of the defect area is then extracted by a contour algorithm.

[0006] Preferably, the defect recognition module is further provided with the angle recognition unit, which is used to acquire the image to be analyzed at different angles, acquire the outer contour line of the defect area in the image to be analyzed, calculate the fit of the outer contour line with the fabric in the binary image, and determine whether the defect is a stain or a foreign object based on the fit.

[0007] Preferably, the defect recognition module is further provided with a comparison and judgment unit. The comparison and judgment unit is used to acquire the defect area, determine the edge point and LBP value corresponding to the defect area, and preset a standard defect area. By comparing the edge point and LBP value of the defect area and the standard defect area, the similarity value between the defect area and the standard defect area is determined, and the defect morphology of the defect area is judged based on the similarity value.

[0008] Preferably, the comparison and judgment unit further includes an edge judgment subunit and an LBP judgment subunit. The edge judgment subunit is used to obtain the slope of the line connecting adjacent corners of the edge corresponding to the standard defect area, and determine the first similarity between the defect area and the pre-standard defect area based on the slope of the line connecting adjacent corners of the edge corresponding to each defect area and the slope of the line connecting adjacent corners of the edge corresponding to the standard defect area. The LBP determination subunit is used to obtain the LBP values ​​of each corner point of the edge corresponding to the standard defect area, and determine the second similarity between the defect area and the standard defect area based on the LBP values ​​of each corner point of the edge corresponding to the defect area and the LBP values ​​of each corner point of the edge corresponding to the standard defect area. The comparison and judgment unit is used to obtain a first similarity and a second similarity, and to obtain a similarity value based on the first similarity and the second similarity.

[0009] Preferably, the system also includes a quality inspection unit, which is used to obtain fabric parameters such as fabric quality and thickness.

[0010] Preferably, the system also includes a preliminary inspection module. The preliminary inspection module is used to acquire the image to be analyzed, cut the fabric region map through a preset semantic segmentation model, perform outlining processing on the fabric region map, acquire the fabric outline, compare the fabric outline with the fabric information in the database, and generate an outline similarity. If the outline similarity is higher than the threshold, the preliminary inspection is qualified; if the outline similarity is lower than the threshold, the preliminary inspection is unqualified.

[0011] Preferably, the types of foreign matter include adhesives, burrs, etc.

[0012] The beneficial effects of this invention are: it realizes full automation from image acquisition to defect identification, marking and report generation, improving detection efficiency; it adopts advanced semantic segmentation models and image processing technology to ensure the accuracy and reliability of defect identification; it can automatically analyze the types of foreign objects, providing strong support for production improvement; the system modules are flexibly designed and easy to integrate with other production equipment or management systems. Attached Figure Description

[0013] Figure 1 This is an overall structural diagram of the present invention. Detailed Implementation

[0014] 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.

[0015] It should be noted that when a component is described as "fixed to" another component, it can be directly on the other component or may have a component in between. When a component is considered "connected to" another component, it can be directly connected to the other component or may have a component in between. When a component is considered "set on" another component, it can be directly set on the other component or may have a component in between. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0017] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings: A quality inspection system for fiber fabrics includes a fan. The fan provides different airflow speeds during the inspection process to observe and analyze the dynamic characteristics of foreign objects on the fabric surface, providing fundamental data for subsequent foreign object type analysis. The database contains pre-set fabric information, including various pre-set information required by the fabric pattern storage system, such as fabric information (e.g., fabric patterns, quality parameters, etc.), defect sample data, foreign object types and their characteristic data, etc. The image acquisition module is used to acquire images of the fabric surface as images to be analyzed, and to capture images of the fabric surface in real time as images to be analyzed. The defect identification module acquires the image to be analyzed from the image acquisition module. It uses a semantic segmentation model to determine if defects exist on the fabric surface. If so, a defect signal is generated, and the defect's morphology is analyzed to obtain its form, which includes stain and foreign object forms. The module receives the image to be analyzed from the image acquisition module, accurately segments the fabric area using a preset semantic segmentation model, and identifies defects on the fabric surface. Based on the characteristics of the defect area, it performs morphological analysis on the defects, distinguishing them into stain and foreign object forms. Foreign objects include removable foreign objects such as adhesives and frayed edges. Stains are color contamination present on the fabric surface during dyeing and weaving processes.

[0018] The defect identification module includes a defect location marking unit. This unit acquires the image to be analyzed and segments it into a fabric region map using a pre-defined semantic segmentation model. The fabric region map is then binarized to obtain a binarized image. This binarized image is compared with binarized fabric pattern images in the database to identify the defect areas. A contour algorithm is then used to extract the contour lines of these defect areas. After identifying the defect, this unit further refines the defect location by binarizing the fabric region map and comparing it with the binarized fabric pattern images in the database. Difference analysis is used to determine the defect areas. Finally, a contour algorithm is used to extract the contour lines of the defect areas, providing precise location information for subsequent processing. The foreign object type analysis module acquires images of the fabric at different fan speeds when the defect is identified as a foreign object. Based on the contour changes in these images, it calculates the swing amplitude of the foreign object and indexes the foreign object type in the database using this amplitude. This module activates when the defect is determined to be a foreign object. It first acquires images of the fabric at different fan speeds to observe the dynamic behavior of the foreign object; by analyzing the contour changes in the images, it calculates the swing amplitude. Then, using the swing amplitude as an index, it searches the database for the corresponding foreign object type information. During the detection process, when foreign objects such as frayed edges are present on the fabric surface, the frayed edges swing under the action of the fan, changing their shape. The swing amplitude of the foreign object determines its type and the connection between it and the fabric.

[0019] The defect marking module is used to acquire defect signals from the defect recognition module, mark the defect location and foreign object shape in the image to be analyzed, and generate an inspection report based on the defect marking information. It receives defect signals from the defect recognition module, accurately marks the defect location and foreign object shape in the image to be analyzed based on the location information and shape analysis results provided by the defect location marking unit, and generates a detailed inspection report based on the marking results, including information such as defect location, shape, type, and possible causes.

[0020] The defect recognition module also includes an angle recognition unit. This unit acquires images of the target area from different angles, obtains the outer contour line of the defect region within the image, calculates the fit between the outer contour line and the fabric in the binarized image, and determines whether the defect is a stain or a foreign object based on the fit. It receives fabric photographs from different angles, extracts the outer contour line of the defect region, compares the extracted contour line with the shape and texture features of the fabric itself, and calculates the fit between the contour line and the fabric. The fit describes the degree of matching between the contour line and the fabric edge or texture. Based on the calculated fit value, the angle recognition unit determines the morphology of the defect. If the fit is high, meaning the contour line closely matches the fabric edge or texture, the defect is more likely a stain. Conversely, if the fit is low, with a large deviation between the contour line and the fabric edge or texture, the defect is more likely a foreign object. The angle recognition unit can quickly determine the morphology of the defect.

[0021] The defect recognition module also includes a comparison and judgment unit. This unit is used to acquire defect areas, determine the corresponding edge points and LBP values ​​of the defect areas, and preset standard defect areas. By comparing the edge points and LBP values ​​of the defect areas with those of the standard defect areas, the similarity value between the defect areas and the standard defect areas is determined. Based on the similarity value, the defect morphology of the defect areas is judged. By calculating the similarity between the defect areas and the standard defect areas, this unit can automatically determine the morphology of the defects.

[0022] The comparison and judgment unit also includes an edge judgment subunit and an LBP judgment subunit. The edge detection subunit is used to obtain the slope of the line connecting adjacent corner points of the edge corresponding to the standard defect area. Based on the slope of the line connecting adjacent corner points of the edge corresponding to each defect area and the slope of the line connecting adjacent corner points of the edge corresponding to the standard defect area, the first similarity between the defect area and the pre-standard defect area is determined. The edge detection subunit is mainly responsible for acquiring and analyzing the edge features of the defect area and the standard defect area. By identifying adjacent corner points on the edge and calculating the slope of the line connecting these corner points, this subunit can quantify the similarity of the defect area and the standard defect area in edge shape. The edge point data of the defect area and the standard defect area are acquired, the edge points are sorted to extract adjacent corner points in order, the slope of the line connecting each pair of adjacent corner points is calculated to form a slope sequence, and the slope sequence of the defect area and the standard defect area are compared to calculate the first similarity between them. The LBP judgment subunit is used to obtain the LBP values ​​of each corner point of the edge corresponding to the standard defect area. Based on the LBP values ​​of each corner point of the edge corresponding to the defect area and the LBP values ​​of each corner point of the edge corresponding to the standard defect area, the second similarity between the defect area and the standard defect area is determined. The similarity between the two in texture features is evaluated by calculating the LBP values ​​of each corner point on the edge of the defect area and the standard defect area and comparing these values. A certain number of corner points on the edge of the defect area and the standard defect area are selected as sample points. The LBP algorithm is applied to each sample point to calculate its LBP value. The second similarity between the two is calculated by comparing the LBP values ​​of each sample point in the defect area and the standard defect area. The comparison and judgment unit is used to obtain a first similarity and a second similarity, and to obtain a similarity value based on the first similarity and the second similarity. The weighting strategy can be adjusted according to the actual application scenario. The first similarity and the second similarity are weighted and summed to obtain the final similarity value. Based on the comparison result of the similarity value and a preset threshold, it is determined whether the defect morphology of the defective area matches the standard defective area.

[0023] It also includes a quality inspection unit, which is used to obtain fabric parameters such as fabric quality and thickness; it uses sensors (such as thickness sensors, mass sensors, etc.) to perform non-contact or contact measurements on the fabric and collect parameters such as fabric thickness and quality.

[0024] It also includes a preliminary inspection module, which is used to acquire the image to be analyzed, cut out the fabric region map through a preset semantic segmentation model, perform outline processing on the fabric region map, obtain the fabric outline, compare the fabric outline with the fabric information in the database and generate an outline similarity. If the outline similarity is higher than the threshold, the preliminary inspection is qualified; if the outline similarity is lower than the threshold, the preliminary inspection is unqualified. The preliminary inspection module is the front-end of the quality inspection system, used to perform preliminary processing and screening of the input image to be analyzed. Using semantic segmentation models and image processing technology, the initial inspection module can automatically cut out fabric region images and extract fabric outlines, comparing them with standard fabric information in the database to quickly determine the initial inspection pass rate of the fabric. It receives the fabric image to be analyzed as input data for the initial inspection, processes the image using a preset semantic segmentation model, automatically identifies and cuts out fabric region images, and performs outline processing on the fabric region images to clearly define the fabric outlines. The extracted fabric outlines are compared with the standard fabric outlines stored in the database to calculate the outline similarity. Based on the comparison result of the outline similarity with a preset threshold, it determines whether the fabric passes the initial inspection. If the similarity is higher than the threshold, the initial inspection is passed; if the similarity is lower than the threshold, the initial inspection is failed.

[0025] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A quality inspection system for fiber fabrics, characterized in that, Including wind turbines, and A database containing pre-set fabric information, including fabric patterns; An image acquisition module is used to acquire an image of the fabric surface as an image to be analyzed. The defect identification module acquires the image to be analyzed from the image acquisition module, determines whether there are defects on the fabric surface through a semantic segmentation model, generates a defect signal if so, and performs morphological analysis on the defect to obtain the defect morphology, which includes stain morphology and foreign object morphology. The defect identification module is equipped with a defect location marking unit and an angle recognition unit. The defect location marking unit acquires the image to be analyzed and cuts it into a fabric area map through a preset semantic segmentation model. The fabric area map is then binarized to obtain a binarized image. The binarized image is compared with a fabric pattern binarized image in the database to obtain the defect area, and the contour line of the defect area is extracted through a contour algorithm. The angle recognition unit is used to acquire images to be analyzed from different angles, acquire the outer contour line of the defect area in the image to be analyzed, calculate the fit of the outer contour line with the fabric in the binarized image, and determine whether the defect is a stain or a foreign object based on the fit. The foreign object type analysis module, when the defect form is a foreign object form, acquires the fabric image to be analyzed under different fan speeds, dynamically analyzes the contour in the image to be analyzed to obtain the swing amplitude of the foreign object, and indexes the foreign object type in the database based on the swing amplitude. The defect marking module is used to acquire defect signals from the defect recognition module, mark the location of defects and the shape of foreign objects in the image to be analyzed, and generate a detection report based on the defect marking information.

2. The quality inspection system for fiber fabrics according to claim 1, characterized in that, The defect recognition module is further equipped with a comparison and judgment unit. The comparison and judgment unit is used to acquire the defect area, determine the edge point and LBP value corresponding to the defect area, and preset a standard defect area. By comparing the edge point and LBP value of the defect area and the standard defect area, the similarity value between the defect area and the standard defect area is determined, and the defect morphology of the defect area is judged based on the similarity value.

3. The quality inspection system for fiber fabrics according to claim 2, characterized in that, The comparison and judgment unit further includes an edge judgment subunit and an LBP judgment subunit. The edge judgment subunit is used to obtain the slope of the line connecting adjacent corner points of the edge corresponding to the standard defect area, and determine the first similarity between the defect area and the standard defect area based on the slope of the line connecting adjacent corner points of the edge corresponding to each defect area and the slope of the line connecting adjacent corner points of the edge corresponding to the standard defect area. The LBP determination subunit is used to obtain the LBP values ​​of each corner point of the edge corresponding to the standard defect area, and determine the second similarity between the defect area and the standard defect area based on the LBP values ​​of each corner point of the edge corresponding to the defect area and the LBP values ​​of each corner point of the edge corresponding to the standard defect area. The comparison and judgment unit is used to obtain a first similarity and a second similarity, and to obtain a similarity value based on the first similarity and the second similarity.

4. The quality inspection system for fiber fabrics according to claim 1, characterized in that, It also includes a quality inspection unit, which is used to obtain the quality and thickness of the fabric.

5. The quality inspection system for fiber fabrics according to claim 1, characterized in that, It also includes a preliminary inspection module, which is used to acquire the image to be analyzed, cut the fabric region map through a preset semantic segmentation model, perform outlining processing on the fabric region map, acquire the fabric outline, compare the fabric outline with the fabric information in the database and generate an outline similarity. If the outline similarity is higher than the threshold, the preliminary inspection is qualified; if the outline similarity is lower than the threshold, the preliminary inspection is unqualified.

6. The quality inspection system for fiber fabrics according to claim 1, characterized in that, The types of foreign objects include adhesives and rough edges.

Images