Jacquard fabric manufacturing method based on machine vision

Abstract

The invention discloses a jacquard fabric manufacturing method based on machine vision, which relates to the field of image data processing, and comprises the following steps: analyzing four continuous fabric reference images to obtain tissue information and material parameters of a first repeating unit; obtaining customization information, generating a second repetitive unit according to the customization information and the organization information and material parameters of each area, and generating fabric processing design information according to the second repetitive unit; weaving according to the fabric processing design information; and obtaining a fabric product image, analyzing the fabric product image to obtain a third repeating unit, and performing weaving quality analysis according to the second repeating unit and the third repeating unit. The method comprises the following steps: generating a second repetitive unit meeting a customization requirement according to customization information and organization information and material parameters of each region in a first repetitive unit; and weaving quality analysis is carried out according to the second repeating unit and the third repeating unit, so that a high-quality closed-loop system from design to production is realized.

Description

Technical Field

[0001] The present invention relates to the field of image data processing, and particularly to a method for manufacturing jacquard fabrics based on machine vision. Background Art

[0002] In the process of manufacturing jacquard fabrics, producing fabrics with complex patterns usually relies on accurate design drawings and a number of complex design parameters. The traditional fabric design process often requires manual drawing and adjustment of patterns, resulting in a long design cycle and high costs. In a rapidly changing market environment, consumers' demand for personalization and uniqueness is increasing day by day. However, the existing design and production methods have limited customization capabilities and are difficult to flexibly and quickly respond to customization requirements while retaining the design style of existing fabric products.

[0003] In response to the above problems, the technological development in recent years has begun to introduce automated and digital solutions, especially the application of machine vision and image processing technologies. Some enterprises have started to try using computer-aided design software combined with digital image processing technology to optimize the design process and shorten the production cycle. These technologies can improve the automation level of fabric design to a certain extent, but they are still insufficient in the overall quality control and design flexibility of the weaving process. Summary of the Invention

[0004] The present invention provides a method for manufacturing jacquard fabrics based on machine vision to solve at least one of the problems mentioned in the above background art.

[0005] The specific technical solutions provided by this application are as follows: A method for manufacturing jacquard fabrics based on machine vision, comprising the steps of: Analyzing a four-way continuous fabric reference image to obtain the tissue information and material parameters of the first repeating unit; Obtaining customization information, generating a second repeating unit according to the customization information, the tissue information and material parameters of each region, and generating fabric processing design information according to the second repeating unit; Weaving according to the fabric processing design information; obtaining a fabric product image, analyzing the fabric product image to obtain a third repeating unit, and performing weaving quality analysis according to the second repeating unit and the third repeating unit; The step of analyzing a four-way continuous fabric reference image to obtain the tissue information and material parameters of the first repeating unit includes the steps of: Performing grayscale and filtering processing on the fabric reference image; Extracting key edge information in the fabric reference image through an edge detection algorithm; Using an image feature extraction method to identify key feature regions in the fabric reference image; Based on the key edge information and key feature regions, the warp period and weft period of the fabric reference image are determined by the sliding window method; The first repeating unit is determined according to the warp period and weft period of the fabric reference image, and the tissue information and material parameters of each region within the first repeating unit are obtained.

[0006] As a preferred solution, the obtaining of the tissue information and material parameters of each region within the first repeating unit includes the steps of: Performing binarization and projection processing on the first repeating unit to obtain a binarized warp image and a binarized weft image, and obtaining the material parameters of each region; Merging the binarized warp image and the binarized weft image into the first repeating unit to generate a first warp and weft segmentation grid image, and identifying the positions of the tissue points; Judging the type of tissue points by calculating the maximum warp gray gradient and the maximum weft gray gradient of each tissue point region; Determining the tissue information of each region within the first repeating unit according to the type of tissue points.

[0007] As a preferred solution, the determining of the warp period and weft period of the fabric reference image by the sliding window method includes the steps of: Defining an initial sliding window; Matching the key edge information and key feature regions within the coverage of the sliding window with the remaining part of the fabric reference image; Recording the positions where the highest similarity appears during the window sliding process, and calculating the distances between the positions.

[0008] As a preferred solution, the filtering process is based on the Wiener filtering algorithm, expressed as: ; Wherein, represents the pixel value at the position (x, y) obtained by filtering, represents the local image variance, which represents the variance of the pixel values within the set local region; represents the noise variance; represents the pixel value at the position (x, y) of the original image; represents the average pixel value within the set local region.

[0009] As a preferred solution, the generating of the second repeating unit according to the customization information and the tissue information and material parameters of each region includes the steps of: Stitching the digital printing pattern with the pattern of the first repeating unit to generate a composite pattern; Obtaining a number of pixel points corresponding to the digital printing pattern in the composite pattern and recording them as the customized printing region, and designing the tissue information and material parameters of the customized printing region; Replace the tissue information and material parameters of the pixel points corresponding to the customized printing area in the first repeating unit to generate a second repeating unit.

[0010] As a preferred solution, replacing the tissue information and material parameters of the pixel points corresponding to the customized printing area in the first repeating unit specifically includes: traversing each pixel point of the first repeating unit; comparing the positions of the pixel points in the customized printing area with those of the pixel points in the first repeating unit, and replacing the tissue information and material parameters of the customized printing area with the corresponding pixels in the first repeating unit.

[0011] As a preferred solution, parsing the fabric product image to obtain a third repeating unit includes the steps of: Perform grayscale and filtering processing on the fabric product image; Extract the key edge information in the fabric product image through an edge detection algorithm; Use an image feature extraction method to identify the key feature areas in the fabric product image; Based on the key edge information and key feature areas, determine the warp period and weft period of the fabric product image through a sliding window method; Determine the third repeating unit according to the warp period and weft period of the fabric product image, and obtain the tissue information and material parameters of each area within the third repeating unit.

[0012] As a preferred solution, obtaining the tissue information and material parameters of each area within the third repeating unit includes the steps of: Perform binarization and projection processing on the third repeating unit to obtain a binarized warp image and a binarized weft image, and obtain the material parameters of each area; Merge the binarized warp image and the binarized weft image into the third repeating unit to generate a third warp and weft segmentation grid image, and identify the tissue point positions; Judge the tissue point type by calculating the maximum warp gray gradient and the maximum weft gray gradient of each tissue point area; Determine the tissue information of each area within the third repeating unit according to the tissue point type.

[0013] As a preferred solution, the customization information includes a digital pattern image and cross-sectional structure information; generating the second repeating unit according to the customization information and the tissue information and material parameters of each area further includes the steps of: Set the height value corresponding to each pixel in the second repeating unit according to the cross-sectional structure information; Optimize the tissue information of the second repeating unit according to the height value corresponding to each pixel.

[0014] As a preferred solution, optimizing the tissue information of the second repeating unit according to the height value corresponding to each pixel includes the steps of: Divide the second repeating unit into several search regions, and calculate the meridional gradient and zonal gradient of each region; If the meridional gradient or the zonal gradient is greater than the set gradient threshold, adjust the tissue information within the search region according to the gradient threshold.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: The jacquard fabric manufacturing method based on machine vision in this application accurately obtains the tissue information and material parameters of each region within each first repeating unit by analyzing the fabric reference image of the four-way continuous pattern; by generating a second repeating unit that meets specific requirements efficiently according to the customization information and the tissue information and material parameters of each region within the first repeating unit, to achieve personalized fabric product design; by analyzing the fabric product image to obtain a third repeating unit, and performing weaving quality analysis based on the second repeating unit and the third repeating unit, so as to timely discover the deviation during the weaving process, thereby realizing a high-quality closed-loop system from design to production to continuously optimize the design and process; By optimizing the tissue information of the second repeating unit according to the height value corresponding to each pixel in the embodiments of this application, it can ensure that there is no obvious pattern deformation or discontinuity when processing the fabric through the second repeating unit, and finally achieve the balance between the functionality and aesthetics of the fabric. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings here are incorporated into the specification and form a part of this specification, indicating the embodiments that conform to the present invention, and are used together with the specification to explain the principles of the present invention.

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a schematic flow chart of the jacquard fabric manufacturing method based on machine vision provided by the embodiments of the present invention; Figure 2 It is a schematic flow chart of analyzing the fabric reference image of the four-way continuous pattern to obtain the tissue information and material parameters of the first repeating unit provided by the embodiments of the present invention; Figure 3 It is a schematic flow chart of obtaining the tissue information and material parameters of each region within the first repeating unit provided by the embodiments of the present invention; Figure 4 It is a schematic flow chart of generating the second repeating unit according to the customization information and the tissue information and material parameters of each region provided by the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0020] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative positional relationship and movement conditions between components in a certain specific posture (as shown in the accompanying drawings). If this specific posture changes, the directional indications will also change accordingly.

[0021] In addition, the descriptions involving "first", "second", etc. in the present invention are only for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the ability of those of ordinary skill in the art to implement. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0022] Please refer to Figure 1 and Figure 2 , the present invention provides a jacquard fabric manufacturing method based on machine vision, including the steps of: Analyze the four-way continuous fabric reference image to obtain the tissue information and material parameters of the first repeat unit; Obtain customization information, generate a second repeat unit according to the customization information and the tissue information and material parameters of each region, and generate fabric processing design information according to the second repeat unit; Weave according to the fabric processing design information; obtain the fabric product image, analyze the fabric product image to obtain the third repeat unit, and perform weaving quality analysis according to the second repeat unit and the third repeat unit.

[0023] The step of analyzing the four-way continuous fabric reference image to obtain the tissue information and material parameters of the first repeat unit includes the steps of: Perform grayscale and filtering processing on the fabric reference image; Extract the key edge information in the fabric reference image through an edge detection algorithm; Use an image feature extraction method to identify the key feature regions in the fabric reference image; Based on the key edge information and key feature regions, the warp period and weft period of the fabric reference image are determined by the sliding window method; The first repeating unit is determined according to the warp period and weft period of the fabric reference image, and the tissue information and material parameters of each region within the first repeating unit are obtained.

[0024] The jacquard fabric manufacturing method based on machine vision in this application accurately obtains the tissue information and material parameters of each region within each first repeating unit by analyzing the fabric reference image with four-way continuity; by generating a second repeating unit that meets specific requirements efficiently according to the customization information and the tissue information and material parameters of each region within the first repeating unit, to achieve personalized fabric product design; by analyzing the fabric product image to obtain a third repeating unit, and performing weaving quality analysis based on the second repeating unit and the third repeating unit, so as to timely detect deviations in the weaving process and correct and optimize them.

[0025] Specifically, each step of the present invention is described in detail through the following content: A jacquard fabric manufacturing method based on machine vision includes the steps of: S1. Analyze the fabric reference image with four-way continuity to obtain the tissue information and material parameters of the first repeating unit; The fabric reference image with four-way continuity refers to an image having a periodic structure both horizontally and vertically. In this application, this periodic structure is denoted as the first repeating unit, and the first repeating unit can form a seamless and infinitely extensible pattern through replication and splicing.

[0026] Specifically, the fabric reference image can be a complete planar image of the fabric, or a partial planar image of the fabric including multiple first repeating units. As long as the image covers multiple first repeating units and can prove that these units are seamlessly connected in four directions, it can be used as a fabric reference image with four-way continuity. For example, a partial image including two repeating units horizontally and vertically can prove its continuity.

[0027] By processing the pattern through subsequent steps, the fabric product corresponding to the fabric reference image can be restored, or a fabric product with a different arrangement of the first repeating unit of the fabric corresponding to the fabric reference image can be woven, or a second repeating unit can be generated by splicing the first repeating unit with a new pattern and the fabric product can be woven according to the second repeating unit. Accordingly, designers can diversify the first repeating unit and flexibly adjust the design. In the application scenario of this embodiment, a four-way continuous fabric reference image and a pattern are obtained, and the designer can obtain the detailed tissue information and material parameters of the first repeating unit by analyzing the fabric reference image. In this way, the designer can splice the pattern without changing the design style of the fabric reference image, so as to meet the personalized needs of customers.

[0028] Further, the analyzing the four-way continuous fabric reference image to obtain the tissue information and material parameters of the first repeating unit includes the steps of: S11. Perform grayscale and filtering processing on the fabric reference image; Perform denoising processing on the image, and use a filtering algorithm to eliminate the noise points in the image, thereby improving the image quality. Then, perform grayscale processing on the image to convert the color image into a grayscale image, simplifying the data processing requirements of subsequent processing steps.

[0029] Further, the filtering processing is based on the Wiener filtering algorithm; Wiener filtering is a filtering method based on the minimum mean square error criterion, expressed as: ; where represents the pixel value at position (x, y) obtained by filtering, represents the local image variance, which represents the variance of the pixel values within a set local area and reflects the degree of change of the image in this area. If the variance is large, it means that the pixel values in this area change greatly and may contain more details or edges; represents the noise variance, which is used to reflect the noise intensity in the image; represents the pixel value at position (x, y) of the original image; represents the average pixel value within a set local area. Wiener filtering makes full use of the local mean and local variance, and aims to minimize the mean square error by balancing the contributions of noise and image content to generate the pixel value of the denoised image.

[0030] S12. Extract the key edge information in the fabric reference image through an edge detection algorithm; In one embodiment, the edge detection algorithm uses the Canny algorithm to identify the edge position by calculating the gradient of the image, ensuring that the boundaries or feature lines on the fabric pattern are clearly identified.

[0031] S13. Use an image feature extraction method to identify key feature regions in the image; In one embodiment, the image feature extraction method uses the SIFT algorithm. The obtained key feature regions are helpful for comparing and analyzing the pattern features of different parts, as well as subsequent image stitching and matching.

[0032] S14. Based on the key edge information and key feature regions, determine the warp period and weft period of the fabric reference image by the sliding window method; Specifically, step S14 of determining the warp period and weft period of the fabric reference image by the sliding window method includes: Sliding window preparation: Define an initial sliding window with a size of NxN pixels. The size of the window is usually slightly larger than the expected repeating unit to ensure that enough information can be included for periodic detection.

[0033] Sliding process: Start the sliding window from the upper left corner of the image and gradually move it to the right and down, moving 1 pixel each time (or set an appropriate step size). At each position, perform matching within the area covered by the sliding window through the key edge information and key feature regions extracted in the previous steps S12 and S13.

[0034] Match degree calculation: Match the key edge information and key feature regions within the area covered by the sliding window with the rest of the fabric reference image. In this embodiment, the normalized cross-correlation (NCC), structural similarity (SSIM), or feature point matching algorithm (such as the FLANN matcher) can be specifically used to calculate the similarity.

[0035] Determine the period: Record the positions where the highest similarity occurs during the window sliding process and calculate the distances between these positions. The warp distance and weft distance between the positions respectively represent the warp period and weft period of the fabric pattern.

[0036] For example, if high-similarity edge information and feature regions are detected at multiple positions during the horizontal sliding process, the horizontal spacing between these high-similarity positions can be calculated to obtain the warp period. Similarly, the vertical spacing corresponds to the weft period.

[0037] Verify the period: Use the determined warp period and weft period for verification. Verify whether the entire image can be seamlessly connected by copying and stitching the first repeating unit, so as to ensure the accuracy of the period. Use the determined warp period and weft period to crop the repeating unit of the image and copy and stitch it in the image to observe whether a seamless and periodically repeating pattern can be formed.

[0038] S15. Determine the first repeating unit according to the warp period and weft period of the fabric reference image, and obtain the tissue information and material parameters of each region within the first repeating unit.

[0039] Through the above steps, designers can accurately determine the warp period and weft period of the reference image of the seamless repeating fabric based on the key edge information and key feature regions, thereby determining the first repeating unit, improving the recognition accuracy of pattern features and the flexibility of design.

[0040] Further, please refer to Figure 3 , the steps for obtaining the tissue information and material parameters of each region within the first repeating unit include: S151: Perform binarization and projection processing on the first repeating unit to obtain a binarized warp image and a binarized weft image, and obtain the material parameters of each region; The material parameters at least include weft density, warp density, weft diameter, and warp diameter. Specifically, for binarization processing, first set a threshold to process the pixel values of the grayscale image: if the grayscale value of a certain pixel is higher than the threshold, set it to white (representing the yarn); if the grayscale value of a certain pixel is lower than the threshold, set it to black (representing the gap between the yarns). Project the entire binarized image in the arrangement direction of the warp yarns to obtain the warp image. The warp image should retain the warp stripes and display the gaps between the warp stripes. Similarly, project the entire binarized image in the arrangement direction of the weft yarns to obtain the weft image. The weft image should retain the weft stripes and display the gaps between the weft stripes. Based on the pixel amount of the first repeating unit and the pixel amount of the yarns, the weft density and warp density can be calculated.

[0041] S152: Merge the binarized warp image and the binarized weft image into the first repeating unit to generate a first warp and weft segmentation grid image, and identify the positions of the tissue points; the tissue point region is the intersection position of the warp yarns and weft yarns in the first warp and weft segmentation grid image; S153: Judge the tissue point type by calculating the maximum warp gray gradient and the maximum weft gray gradient of each tissue point region; the tissue point types include warp tissue points and weft tissue points. Specifically, if the maximum warp gray gradient is greater than the maximum weft gray gradient, then this tissue point is a warp tissue point; otherwise, it is a weft tissue point. Among them, the maximum warp gray gradient and the maximum weft gray gradient can be calculated by the Sobel operator or the Scharr operator.

[0042] S154: Determine the tissue information of each region within the first repeating unit according to the tissue point type.

[0043] For the fabric described in this embodiment, various weaves can be used within the first repeating unit to achieve specific functions, aesthetic effects, or structural characteristics. In one application scenario, the first repeating unit includes a bottom plate area and a surface jacquard area. The bottom plate area is woven with one weave, and the surface jacquard area is woven with several weaves. Specifically, the bottom plate area uses fixed weave structures, such as plain weave, twill weave, satin weave, etc., to ensure the stability and durability of the fabric. The surface jacquard area achieves unique decorative effects and rich patterns through a variety of weaves and diverse yarn combinations. Based on the weave information and material parameters of each area obtained through the above steps, ensure that the quality of the fabric produced subsequently meets the design requirements.

[0044] In this embodiment, the weave information is stored in the form of a weave diagram. A weave diagram is a matrix formed by the arrangement of weave points, used to represent the interlacing structure of the fabric. Each area within the first repeating unit is continuously repeated through its corresponding weave for weaving within the area.

[0045] S2. Obtain customization information, generate a second repeating unit according to the customization information, the weave information, and the material parameters of each area, and generate fabric processing design information according to the second repeating unit; The customization information is the detailed requirements and parameters provided to meet specific personalized needs. In one embodiment, the customization information includes a digital printing pattern. The designer combines the customized digital printing pattern with the first repeating unit to generate a composite pattern. This combination ensures that the newly generated pattern incorporates personalized customization elements while maintaining the design style of the first repeating unit.

[0046] Further, please refer to Figure 4 , the generating of the second repeating unit according to the customization information, the weave information, and the material parameters of each area includes the steps of: S21. Stitch the digital printing pattern with the pattern of the first repeating unit to generate a composite pattern; S22. Obtain several pixel points corresponding to the digital printing pattern in the composite pattern, denoted as the customized printing area, and design the weave information and material parameters of the customized printing area; S23. Replace the weave information and material parameters of the pixel points corresponding to the customized printing area in the first repeating unit to generate the second repeating unit.

[0047] Further, the replacing of the weave information and material parameters of the pixel points corresponding to the customized printing area in the first repeating unit specifically includes: Traverse each pixel point of the first repeating unit; Compare the positions of the pixels in the customized printed area with those in the first repeating unit, and replace the tissue information and material parameters of the customized printed area with the corresponding pixels in the first repeating unit. During the replacement process, if the tissue information of the two is significantly different, weighted averaging or edge interpolation methods are used for smooth transition to retain the design style.

[0048] Further, the fabric processing design information is an overall data set including detailed information required for production. This information not only includes the design of the second repeating unit, but also covers various parameters and guidelines required for further production, specifically including: Pattern design information: including the tissue information, position, warp and weft yarn arrangement of the second repeating unit, and the arrangement of the second repeating unit in the complete fabric pattern, etc.; Material parameters: involving all material details of the fabric, such as yarn type, color, density, etc.; Processing parameters: specific process parameters such as production equipment settings, temperature, pressure, etc.

[0049] S3. Weave according to the fabric processing design information; obtain the fabric product image, analyze the fabric product image to obtain the third repeating unit, and perform weaving quality analysis based on the second repeating unit and the third repeating unit.

[0050] Convert the fabric processing design information into control instructions and transmit them to the jacquard loom for weaving. For the fabric product output by the jacquard loom, a representative periodic structural unit needs to be extracted from the fabric product image for quality analysis. Therefore, in this embodiment, a high-definition image acquisition device is used to obtain the fabric product image, the fabric product image is analyzed to obtain the third repeating unit, and the second repeating unit provided in the design stage, i.e., step S2, and the third repeating unit are compared to analyze the weaving quality and find any potential defects or differences.

[0051] Further, the analysis of the fabric product image to obtain the third repeating unit includes the steps of: S31. Grayscale and filter the fabric product image; S32. Extract the key edge information in the fabric product image through an edge detection algorithm; S33. Use an image feature extraction method to identify the key feature areas in the fabric product image; S34. Based on the key edge information and the key feature areas, determine the warp period and weft period of the fabric product image through the sliding window method; S35. Determine the third repeating unit according to the warp period and weft period of the fabric product image, and obtain the tissue information and material parameters of each area within the third repeating unit.

[0052] Further, obtaining the tissue information and material parameters of each region within the third repeating unit includes the steps of: S351: Binarize and project the third repeating unit to obtain a binarized meridional map and a binarized zonal map, and obtain the material parameters of each region; S352: Merge the binarized meridional map and the binarized zonal map into the third repeating unit to generate a third meridional and zonal segmentation grid map, and identify the positions of tissue points; S353: Determine the types of tissue points by calculating the maximum meridional gray gradient and the maximum zonal gray gradient of each tissue point region; S354: Determine the tissue information of each region within the third repeating unit according to the types of tissue points.

[0053] Similar to the specific method of parsing the tissue information and material parameters of the first repeating unit from the four-way continuous fabric reference image in the aforementioned step S1, parsing the fabric product image to obtain the third repeating unit, that is, by obtaining the tissue information and material parameters of the third repeating unit, verifying the consistency between the finished fabric and the design drawing, and ensuring product quality; by continuously comparing the second repeating unit designed based on the first repeating unit and the third repeating unit of the fabric product, a high-quality closed-loop system from design to production can be realized to continuously optimize the design and process.

[0054] As a preferred embodiment, the customization information includes a digital pattern diagram and cross-sectional structure information; generating the second repeating unit according to the customization information and the tissue information and material parameters of each region further includes the steps of: S24: Set the height values corresponding to each pixel in the second repeating unit according to the cross-sectional structure information; Among them, the cross-sectional structure information can be represented in the form of a matrix or a three-dimensional model. By parsing the cross-sectional structure information, rasterize it into multiple height value data and map it to some pixel points of the second repeating unit; then, according to the meridional and zonal periods of the image, use the interpolation method to complete the height values of all pixels in the second repeating unit to ensure smooth transition of the height values.

[0055] S25: Optimize the tissue information of the second repeating unit according to the height values corresponding to each pixel.

[0056] In the actual implementation process, the height difference between adjacent pixel points can be calculated as the judgment condition for whether the tissue information needs to be optimized. Adjust the tissue information using the adjacent pixel height difference to compensate for the deformation of the pattern in the actual fabric product caused by a large height gradient.

[0057] Further, optimizing the tissue information of the second repeating unit according to the height values corresponding to each pixel includes the steps of: S251. Divide the second repeating unit into several search regions, and calculate the meridional gradient and zonal gradient of each region. The search region can be set as a 3×3 grid, and each 1×1 grid in the 3×3 grid represents a tissue. Then the meridional gradient and zonal gradient are respectively expressed as: ; where x and y represent the zonal and meridional positions, represents the tissue height at the position (x, y), represents the meridional gradient, represents the zonal gradient.

[0058] S252. If the meridional gradient or zonal gradient is greater than the set gradient threshold, adjust the tissue information within the search region according to the gradient threshold; wherein, the gradient threshold is set according to factors such as material properties and design requirements. Setting the tissue information within the search region can specifically be to adjust the density and increase the tissue points in the tissue diagram to maintain the continuity and smoothness of the pattern within the search region.

[0059] In an application scenario, for the jacquard fabric manufacturing method provided by the present application, the weaving raw materials include a first heating material, a second heating material, warp yarns, and weft yarns. The first heating material is arranged longitudinally, the second heating material is arranged latitudinally, and the first heating material and the second heating material are cross-arranged to form a grid structure. The first heating material and the second heating material are arranged in a warp and weft cross manner to form a grid-like fabric structure, which can better control the distribution and density of the heating material. Among them, the first heating material is copper wire and / or copper wire carbon fiber composite material, and the second heating material is carbon fiber. The copper wire can be purple copper wire, brass wire, phosphor copper wire, or red copper wire. Through external power connection, the current is conducted through the copper wire (longitudinal); the current contacts the carbon fiber at the intersection of the warp and weft, and then flows into the carbon fiber (latitudinal); due to its resistivity, the carbon fiber generates heat when passing through the current. Among them, the carbon fibers are evenly arranged longitudinally, so that each carbon fiber can generate heat, realizing uniform distribution of heat in the entire grid structure to ensure that the heat can be evenly diffused on the fabric surface and avoid local overheating or overcooling phenomena.

[0060] According to the set positions of the first heating material and the second heating material, the complete fabric pattern in the fabric processing design information includes a base fabric area, a first protection area, a second protection area, and several jacquard areas. Among them, the first protection area covers the first heating material, the second protection area covers the second heating material, and the first protection area and the second protection area are arranged crosswise to form a grid structure to protect the first heating material and the second heating material. Based on the foregoing scenario, since there are obvious protrusions in the first protection area and the second protection area in the fabric product, the part where the second repeating unit coincides with the first protection area or the second protection area is prone to pattern deformation.

[0061] By optimizing the tissue information of the second repeating unit according to the height value corresponding to each pixel, it can be ensured that there is no obvious pattern deformation or discontinuity when processing the fabric through the second repeating unit, and finally the balance between the functionality and aesthetics of the fabric is achieved.

[0062] In the embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or component libraries can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the modules can be in electrical, mechanical or other forms.

[0063] The modules described as separate components may or may not be physically separated. The components displayed as modules may or may not be physical modules, that is, they can be located in one place or distributed to multiple grid modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0064] In addition, in each embodiment of the present application, the functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0065] When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a grid device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, dynamic hard disks, read-only memories (ROM, read-only memory), random access memories (RAM, random access memory), magnetic disks, or optical discs.

[0066] The foregoing are only specific implementation manners of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather will conform to the broadest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for manufacturing jacquard fabric based on machine vision, characterized in that: Includes steps: Parsing the four-square continuous fabric reference image to obtain the organization information and material parameters of the first repeating unit; Obtaining customized information, generating a second repeating unit according to the customized information and the organization information and material parameters of each region, and generating fabric processing design information according to the second repeating unit; Weaving according to fabric processing design information; Acquire a fabric product image, analyze the fabric product image to obtain a third repeating unit, and perform weaving quality analysis based on the second repeating unit and the third repeating unit; The method of analyzing the four-dimensional continuous fabric reference image to obtain the tissue information and material parameters of the first repeating unit includes the following steps: Grayscale and filter the fabric reference image; Extract key edge information from fabric reference image by edge detection algorithm; Using image feature extraction methods, key feature areas in fabric reference images are identified; Based on the key edge information and key feature areas, the warp period and weft period of the fabric reference image are determined by the sliding window method. The first repeating unit is determined according to the warp period and the weft period of the fabric reference image, and the tissue information and material parameters of each area in the first repeating unit are obtained.

2. The method for manufacturing jacquard fabric based on machine vision according to claim 1, characterized in that: The step of obtaining tissue information and material parameters of each region in the first repeating unit comprises the following steps: Binarization and projection processing are performed on the first repeating unit to obtain a binary longitude map and a binary latitude map, and material parameters of each region are obtained; The binary longitude map and the binary latitude map are combined into a first repeating unit to generate a first longitude and latitude segmentation grid map, and the organization point positions are identified; The type of tissue point is determined by calculating the maximum longitudinal grayscale gradient and the maximum latitudinal grayscale gradient of each tissue point area; The tissue information of each region within the first repeating unit is determined according to the tissue point type.

3. The method for manufacturing jacquard fabric based on machine vision according to claim 1, characterized in that: The method of determining the warp period and weft period of the fabric reference image by the sliding window method comprises the following steps: Define an initial sliding window; Matching the key edge information and key feature areas within the coverage of the sliding window with the rest of the fabric reference image; The positions with the highest similarity during the sliding window are recorded, and the distance between the positions is calculated.

4. The method for manufacturing jacquard fabric based on machine vision according to claim 1, characterized in that: The filtering process is based on the Wiener filtering algorithm, which is expressed as: ； in, Represents the pixel value at the filtered position (x, y), Represents the local image variance, which means the variance of the pixel values ​​in the set local area; represents the noise variance; Represents the pixel value at the original image position (x, y); Indicates setting the average pixel value in the local area.

5. The method for manufacturing jacquard fabric based on machine vision according to claim 1, characterized in that: The customized information includes a digital printing pattern; the second repeating unit is generated according to the customized information and the organization information and material parameters of each area, including the steps of: splicing the digital printing pattern with the pattern of the first repeating unit to generate a composite pattern; Obtaining a number of pixel points corresponding to the digital printing image in the composite pattern as a customized printing area, and designing the organization information and material parameters of the customized printing area; The organizational information and material parameters of the pixel points corresponding to the customized printing area in the first repeating unit are replaced to generate a second repeating unit.

6. The method for manufacturing jacquard fabric based on machine vision according to claim 5, characterized in that: The organizational information and material parameters of the pixel points corresponding to the customized printing area in the first repeating unit are replaced, specifically including: traversing each pixel point of the first repeating unit; comparing the position of the pixel points of the customized printing area with the pixel points of the first repeating unit, and replacing the organizational information and material parameters of the customized printing area with the corresponding pixels of the first repeating unit.

7. The method for manufacturing jacquard fabric based on machine vision according to claim 1, characterized in that: The method of analyzing the fabric product image to obtain the third repeating unit comprises the steps of: Grayscale and filter the fabric product images; Extract key edge information from fabric product images through edge detection algorithm; Using image feature extraction methods, identify key feature areas in fabric product images; Based on the key edge information and key feature areas, the warp period and weft period of the fabric product image are determined by the sliding window method. The third repeating unit is determined according to the warp period and the weft period of the fabric product image, and the tissue information and material parameters of each area in the third repeating unit are obtained.

8. The method for manufacturing jacquard fabric based on machine vision according to claim 7, characterized in that: The step of obtaining tissue information and material parameters of each region in the third repeating unit comprises the following steps: Binarization and projection processing are performed on the third repeating unit to obtain a binary longitude map and a binary latitude map, and material parameters of each region are obtained; The binary longitude map and the binary latitude map are combined into a third repeating unit to generate a third longitude and latitude segmentation grid map, and the organization point position is identified; The type of tissue point is determined by calculating the maximum longitudinal grayscale gradient and the maximum latitudinal grayscale gradient of each tissue point area; The tissue information of each region within the third repeating unit is determined according to the tissue point type.

9. The method for manufacturing jacquard fabric based on machine vision according to claim 1, characterized in that: The customized information includes a digital pattern diagram and cross-sectional structure information; the second repeating unit is generated according to the customized information and the tissue information and material parameters of each region, and further includes the steps of: Setting the height value corresponding to each pixel in the second repeating unit according to the cross-sectional structure information; The organizational information of the second repeating unit is optimized according to the height value corresponding to each pixel.

10. The method for manufacturing jacquard fabric based on machine vision according to claim 9, characterized in that: The step of optimizing the tissue information of the second repeating unit according to the height value corresponding to each pixel comprises the following steps: Divide the second repeating unit into a plurality of search areas, and calculate the longitudinal gradient and the latitudinal gradient of each area; If the longitudinal gradient or the latitudinal gradient is greater than the set gradient threshold, the tissue information in the search area is adjusted according to the gradient threshold.

Images