Intelligent monitoring and analysis method of textile fabrics based on intelligent manufacturing of textiles and clothing
Through high-resolution image acquisition and analysis technology, the problems of low efficiency and low accuracy of traditional textile fabric monitoring methods are solved, and efficient, accurate monitoring and optimization of textile fabric quality are achieved.
Patent Information
- Application Number
- CN202411034424.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-07-31
AI Technical Summary
Traditional textile fabric monitoring methods rely on manual visual inspection and simple sensors, and there are problems of low monitoring efficiency and low accuracy.
High-resolution image acquisition equipment is used to scan the textile fabric image, and the quality level monitoring and analysis is carried out through grayscale conversion, structural texture feature analysis and chromatic defect quantization calculation, and structural defects and chromatic defect factors are combined.
The efficiency and accuracy of textile fabric quality monitoring have been improved, defects and color defects are identified in a timely manner, the production process is optimized, and the product pass rate and production efficiency have been improved.
Smart Images

Figure CN118570193B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of image processing, and in particular to an intelligent monitoring and analysis method for textile fabrics based on intelligent manufacturing of textile garments. Background Art
[0002] In the textile and garment manufacturing industry, the application of intelligent manufacturing technology is becoming more and more extensive, especially in the production of textile fabrics. The development of intelligent monitoring and analysis methods is of great significance for improving production efficiency, ensuring product quality and reducing costs. With the development of the Internet of Things, big data and image processing technology, the intelligent monitoring and analysis method of textile fabrics based on image processing technology has become a new trend in the development of the industry. By using high-resolution camera equipment to shoot the corresponding textile fabric surface on the textile and garment intelligent manufacturing production line, the physical parameters (such as thickness, density, strength, etc.) and quality indicators (such as color difference, textile structure, fiber material, etc.) of the textile fabric are collected in real time. At the same time, by combining image processing technology, real-time analysis and intelligent recognition of textile fabric images are realized, and defects, foreign matter, production anomalies and other problems in textile fabrics can be quickly and accurately identified. For example, damage, defects or abnormal textile structure in the fabric can be detected, and early warning notifications can be issued in time to notify production personnel to handle, thereby improving the qualified rate and production efficiency of clothing products. However, traditional textile fabric monitoring methods mainly rely on manual visual inspection and simple sensor data collection, and often have problems of low monitoring efficiency and low accuracy. Summary of the invention
[0003] Based on this, it is necessary for the present invention to provide an intelligent monitoring and analysis method for textile fabrics based on intelligent manufacturing of textile garments to solve at least one of the above technical problems.
[0004] To achieve the above purpose, a textile cloth intelligent monitoring and analysis method based on textile and clothing intelligent manufacturing includes the following steps:
[0005] Step S1: Scanning the textile cloth image in the textile and clothing intelligent manufacturing process by a high-resolution image acquisition device to obtain a textile cloth surface scan image; performing grayscale conversion processing on the textile cloth surface scan image to obtain a textile cloth surface structure grayscale image;
[0006] Step S2: performing structural texture feature analysis on the grayscale image of the textile surface structure to obtain the textile surface structure texture feature data; performing structural defect quality impact analysis on the grayscale image of the textile surface structure based on the textile surface structure texture feature data to obtain the textile surface structure defect quality impact factor;
[0007] Step S3: performing a quantitative calculation of the surface color difference defects on the scanned image of the textile surface to obtain the textile surface color difference defect degree coefficient; performing a color difference defect quality impact analysis on the scanned image of the textile surface based on the textile surface color difference defect degree coefficient to obtain the textile surface color difference defect quality impact factor;
[0008] Step S4: Based on the quality influencing factors of textile surface structural defects and the quality influencing factors of textile surface color difference defects, the quality grade of the corresponding textile fabrics in the textile clothing intelligent manufacturing process is monitored and analyzed to obtain the quality grade monitoring results of the textile fabrics.
[0009] The present invention firstly scans the surface structure of textile cloth in the intelligent manufacturing process of textile clothing by using high-resolution image acquisition equipment, which can capture the tiny details and texture of the cloth surface, and provide a sufficient data basis for subsequent image processing and analysis. The key to this step is to ensure the clarity and integrity of the image, and follow the acquisition conditions designed in advance to ensure that the acquired image can accurately reflect the real state and characteristics of the cloth. At the same time, by grayscale conversion processing of the acquired surface scanned image of the textile cloth, the color image is converted into a grayscale image, so as to better analyze and process the structure and characteristics of the cloth surface. The grayscale processing can simplify the information complexity of the image, retain the main characteristics of the cloth surface, and reduce the complexity and calculation cost of data processing. The key to this step is to provide a clearer and easier-to-process data form for subsequent image analysis, thereby providing important support for automated decision-making and control in the intelligent manufacturing process. Secondly, by performing surface structure texture feature analysis on the grayscale image of the surface structure of the textile cloth, the purpose is to extract and analyze the texture feature data of each structural area. By analyzing the texture features of the cloth surface in detail, we can identify the texture type, density and distribution pattern of different areas, and provide quantitative data basis for cloth quality assessment and defect detection. The key to this step is to accurately describe and quantify the subtle changes and characteristics of the cloth surface structure, so as to provide a scientific basis for subsequent defect assessment and quality control decisions. By analyzing the impact of structural defects on the grayscale image of the surface structure of textile cloth based on the surface structure texture feature data of textile cloth, we aim to analyze the factors affecting the surface structure quality of the cloth by the degree of defects in the surface structure image. Through detailed quality impact analysis, we can deeply understand the degree of impact of defects on product appearance and performance, and provide a basis for improvement in product design and production process. The key to this step is to optimize the product quality control strategy, minimize the appearance of defective products, and thus improve the production efficiency and product qualification rate of intelligent manufacturing of textiles and garments. Then, by performing quantitative calculation of surface color difference defects on the surface scanned images of textile fabrics, this step aims to evaluate the degree of color difference between various areas on the surface of the fabric, and by comparing the difference between the expected color and the actual color, the color difference defect degree coefficient is calculated, which can provide a quantitative evaluation of the color consistency and quality problems on the surface of the fabric, and help to accurately locate and analyze possible color difference defect areas, thereby guiding subsequent quality improvement measures and production optimization. By performing color difference defect quality impact analysis on the surface scanned images of textile fabrics based on the color difference defect degree coefficient of the surface of textile fabrics, this step aims to determine the actual impact of color difference defects on the surface quality of the fabric, and identify the key factors that cause color difference defects.By analyzing the influencing factors, the key parameters that cause color difference in the production process can be determined and optimized to improve the color consistency and quality stability of the product, which helps producers to deeply understand the root causes of the color difference problem, thereby providing a scientific basis for formulating effective quality control strategies and improving product competitiveness. Finally, by combining the quality influencing factors of textile surface structural defects and the quality influencing factors of textile surface color difference defects, the quality grade of the corresponding textile fabrics in the intelligent manufacturing process of textile clothing is monitored and analyzed. The key to this step is to convert the comprehensive quality score value obtained by statistical calculation of the quality influencing factors into a specific quality grade assessment according to the pre-set quality standards and rating system, which can timely discover and identify the quality problems and improvement space of textile fabrics, and help producers and quality managers formulate corresponding quality improvement strategies and control measures, thereby improving the monitoring efficiency and accuracy of textile fabric quality. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:
[0011] Figure 1 The present invention is a schematic diagram of the steps of the intelligent monitoring and analysis method for textile fabrics based on intelligent manufacturing of textile garments;
[0012] Figure 2 for Figure 1 Detailed step flow diagram of step S1;
[0013] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION
[0014] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.
[0015] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.
[0016] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.
[0017] To achieve this, please refer to Figures 1 to 3 The present invention provides a textile cloth intelligent monitoring and analysis method based on textile clothing intelligent manufacturing, the method comprising the following steps:
[0018] Step S1: Scanning the textile cloth image in the textile and clothing intelligent manufacturing process by a high-resolution image acquisition device to obtain a textile cloth surface scan image; performing grayscale conversion processing on the textile cloth surface scan image to obtain a textile cloth surface structure grayscale image;
[0019] Step S2: performing structural texture feature analysis on the grayscale image of the textile surface structure to obtain the textile surface structure texture feature data; performing structural defect quality impact analysis on the grayscale image of the textile surface structure based on the textile surface structure texture feature data to obtain the textile surface structure defect quality impact factor;
[0020] Step S3: performing a quantitative calculation of the surface color difference defects on the scanned image of the textile surface to obtain the textile surface color difference defect degree coefficient; performing a color difference defect quality impact analysis on the scanned image of the textile surface based on the textile surface color difference defect degree coefficient to obtain the textile surface color difference defect quality impact factor;
[0021] Step S4: Based on the quality influencing factors of textile surface structural defects and the quality influencing factors of textile surface color difference defects, the quality grade of the corresponding textile fabrics in the textile clothing intelligent manufacturing process is monitored and analyzed to obtain the quality grade monitoring results of the textile fabrics.
[0022] In the embodiment of the present invention, please refer to Figure 1 As shown, it is a schematic diagram of the steps of the intelligent monitoring and analysis method of textile cloth based on intelligent manufacturing of textile clothing of the present invention. In this example, the intelligent monitoring and analysis method of textile cloth based on intelligent manufacturing of textile clothing includes the following steps:
[0023] Step S1: Scanning the textile cloth image in the textile and clothing intelligent manufacturing process by a high-resolution image acquisition device to obtain a textile cloth surface scan image; performing grayscale conversion processing on the textile cloth surface scan image to obtain a textile cloth surface structure grayscale image;
[0024] In an embodiment of the present invention, a high-resolution image acquisition device (including a professional scanner or camera system with sufficient resolution and color depth to capture the tiny details and color changes on the surface of the cloth) is used to perform real-time image scanning of textile fabrics in the intelligent manufacturing process of textile clothing, so as to capture the tiny details and textures on the surface of the textile fabrics in the intelligent manufacturing process in real time, thereby scanning and obtaining a scanned image of the surface of the textile fabric. Then, the corresponding grayscale algorithm (such as the component method or the grayscale projection method) is used to perform grayscale conversion on the scanned image of the surface of the textile fabric after image preprocessing, so as to convert the color or enhanced image into a grayscale image, so as to better analyze and process the structure and characteristics of the surface of the fabric, and finally obtain a grayscale image of the surface structure of the textile fabric.
[0025] Step S2: performing structural texture feature analysis on the grayscale image of the textile surface structure to obtain the textile surface structure texture feature data; performing structural defect quality impact analysis on the grayscale image of the textile surface structure based on the textile surface structure texture feature data to obtain the textile surface structure defect quality impact factor;
[0026] In an embodiment of the present invention, a texture feature analysis method is used to perform texture feature recognition and analysis on a grayscale image of the surface structure of a textile cloth, so as to extract surface texture features from the image. These features can be used to describe and distinguish texture differences in different parts. For example, a local binary pattern (LBP) algorithm or a gray level co-occurrence matrix (GLCM) and other techniques can be used to calculate texture features of each region in the image. The LBP algorithm can describe the gray distribution pattern of pixels around each pixel, thereby forming a texture feature with strong descriptiveness, while the GLCM can calculate the gray level co-occurrence relationship between pixels, measure the statistical characteristics of the texture in the image, such as texture features such as contrast, energy and entropy, and thus obtain texture feature data of the surface structure of the textile cloth. Then, by combining the textile fabric surface structure texture feature data obtained in the previous analysis, the structural defect situation of the corresponding textile fabric surface structure grayscale image is evaluated and analyzed, so as to evaluate the degree of defect of the fabric surface structure according to the change and distribution of the texture features, and by using the quality impact assessment algorithm, the corresponding structural quality problems in the textile fabric surface structure grayscale image are evaluated and analyzed, so as to evaluate and quantify the potential influencing factors of different structural features on the fabric surface quality, and by combining the defect degree obtained in the previous analysis, a regression analysis method (including linear regression, logistic regression or other statistical methods) is used to perform defect impact regression analysis on each quality influencing factor, so as to quantitatively analyze the specific correlation degree of the defect degree to the quality influencing factor, and screen out the quality influencing factors corresponding to the preset threshold value to determine them as defect influencing factors, so as to identify the factors that have the most significant impact on the fabric surface quality, including specific texture features, particle distribution, pattern complexity, etc., and finally obtain the textile fabric surface structure defect quality influencing factors.
[0027] Step S3: performing a quantitative calculation of the surface color difference defects on the scanned image of the textile surface to obtain the textile surface color difference defect degree coefficient; performing a color difference defect quality impact analysis on the scanned image of the textile surface based on the textile surface color difference defect degree coefficient to obtain the textile surface color difference defect quality impact factor;
[0028] In an embodiment of the present invention, a color space conversion is performed on a scanned image of a textile cloth surface by using a LAB color space conversion method, so as to convert the scanned image of the textile cloth surface into a LAB color space, and the image LAB color space is subdivided into smaller areas according to the position of the pixel points or specific rules. At the same time, the chromaticity value of the pixel in each area is quantitatively calculated by calculating the specific color values of each color space sub-area on the L, a, and b channels, and the surface color difference defects of the corresponding scanned image of the textile cloth surface are quantitatively calculated, so as to use a distance measurement method of the color space to calculate the distance between each pixel sub-area and the ideal color or the degree of color difference defects between adjacent sub-areas, thereby obtaining a color difference defect degree coefficient of the textile cloth surface. Then, by combining the color difference defect degree coefficient of the textile cloth surface obtained by quantitative calculation, the quality impact analysis of the color difference defects is carried out on the scanned image of the entire textile cloth surface. This analysis aims to identify and understand the actual impact of color difference defects on the cloth surface quality, so as to determine which factors are particularly important when there are color difference defects, and determine and analyze the key parameters that cause color difference in the intelligent manufacturing production process of textile clothing, including color material formula ratio, dyeing temperature, dyeing time, pressure, etc., and finally obtain the quality influencing factors of color difference defects on the textile cloth surface.
[0029] Step S4: Based on the quality influencing factors of textile surface structural defects and the quality influencing factors of textile surface color difference defects, the quality grade of the corresponding textile fabrics in the textile clothing intelligent manufacturing process is monitored and analyzed to obtain the quality grade monitoring results of the textile fabrics.
[0030] In an embodiment of the present invention, a statistical method or a machine learning algorithm is used to quantitatively calculate the influence of different influencing factors on the corresponding textile fabrics in the intelligent manufacturing process of textiles and garments, and a quantitative evaluation is performed to quantitatively determine the contribution of each structural defect and color difference defect influencing factor to the overall textile fabric quality, and a variety of methods are used, such as the analytic hierarchy process (AHP), the principal component analysis method (PCA), or a data-driven machine learning model, to determine the specific influence weight of each structural defect influencing factor and the color difference defect influencing factor on the overall fabric quality, and a weighted average mathematical operation method is used for quantitative calculation, so as to apply the allocated weight data to the specific quality influencing factor to calculate the comprehensive quality score value of the textile fabric surface. For example, the specific implementation formula is: , where It is the comprehensive quality score of textile fabric surface. is the total number of factors affecting the quality of textile fabric surface structural defects, is the quantitative index parameter of the factors affecting the quality of textile surface structural defects, is the total number of factors affecting the quality of color difference defects on the surface of textile fabrics, is the quantitative index parameter of the factors affecting the quality of color difference defects on the surface of textile fabrics, For the The weight of cloth quality influence of structural defect influence factors, For the Factors affecting the quality of textile fabric surface structural defects: For the The weight of the cloth quality influencing factor of the color difference defect, For the The quality influencing factors of color difference defects on the surface of textile fabrics are determined, and then the comprehensive quality score values obtained by quantitative calculation are compared and matched by using the preset textile fabric quality grade judgment standards (for example, quality score values between 0 and 40 are determined as ordinary, 41 to 60 are determined as good, 61 to 100 are determined as excellent, etc., which are determined according to threshold setting, division of numerical range or classification method based on experience, etc.), so as to monitor and determine the quality grade of each textile fabric by the correspondence between the pre-set score value and the quality grade, and finally obtain the textile fabric quality grade monitoring result.
[0031] The present invention firstly scans the surface structure of textile cloth in the intelligent manufacturing process of textile clothing by using high-resolution image acquisition equipment, which can capture the tiny details and texture of the cloth surface, and provide a sufficient data basis for subsequent image processing and analysis. The key to this step is to ensure the clarity and integrity of the image, and follow the acquisition conditions designed in advance to ensure that the acquired image can accurately reflect the real state and characteristics of the cloth. At the same time, by grayscale conversion processing of the acquired surface scanned image of the textile cloth, the color image is converted into a grayscale image, so as to better analyze and process the structure and characteristics of the cloth surface. The grayscale processing can simplify the information complexity of the image, retain the main characteristics of the cloth surface, and reduce the complexity and calculation cost of data processing. The key to this step is to provide a clearer and easier-to-process data form for subsequent image analysis, thereby providing important support for automated decision-making and control in the intelligent manufacturing process. Secondly, by performing surface structure texture feature analysis on the grayscale image of the surface structure of the textile cloth, the purpose is to extract and analyze the texture feature data of each structural area. By analyzing the texture features of the cloth surface in detail, we can identify the texture type, density and distribution pattern of different areas, and provide quantitative data basis for cloth quality assessment and defect detection. The key to this step is to accurately describe and quantify the subtle changes and characteristics of the cloth surface structure, so as to provide a scientific basis for subsequent defect assessment and quality control decisions. By analyzing the impact of structural defects on the grayscale image of the surface structure of textile cloth based on the surface structure texture feature data of textile cloth, we aim to analyze the factors affecting the surface structure quality of the cloth by the degree of defects in the surface structure image. Through detailed quality impact analysis, we can deeply understand the degree of impact of defects on product appearance and performance, and provide a basis for improvement in product design and production process. The key to this step is to optimize the product quality control strategy, minimize the appearance of defective products, and thus improve the production efficiency and product qualification rate of intelligent manufacturing of textiles and garments. Then, by performing quantitative calculation of surface color difference defects on the surface scanned images of textile fabrics, this step aims to evaluate the degree of color difference between various areas on the surface of the fabric, and by comparing the difference between the expected color and the actual color, the color difference defect degree coefficient is calculated, which can provide a quantitative evaluation of the color consistency and quality problems on the surface of the fabric, and help to accurately locate and analyze possible color difference defect areas, thereby guiding subsequent quality improvement measures and production optimization. By performing color difference defect quality impact analysis on the surface scanned images of textile fabrics based on the color difference defect degree coefficient of the surface of textile fabrics, this step aims to determine the actual impact of color difference defects on the surface quality of the fabric, and identify the key factors that cause color difference defects.By analyzing the influencing factors, the key parameters that cause color difference in the production process can be determined and optimized to improve the color consistency and quality stability of the product, which helps producers to deeply understand the root causes of the color difference problem, thereby providing a scientific basis for formulating effective quality control strategies and improving product competitiveness. Finally, by combining the quality influencing factors of textile surface structural defects and the quality influencing factors of textile surface color difference defects, the quality grade of the corresponding textile fabrics in the intelligent manufacturing process of textile clothing is monitored and analyzed. The key to this step is to convert the comprehensive quality score value obtained by statistical calculation of the quality influencing factors into a specific quality grade assessment according to the pre-set quality standards and rating system, which can timely discover and identify the quality problems and improvement space of textile fabrics, and help producers and quality managers formulate corresponding quality improvement strategies and control measures, thereby improving the monitoring efficiency and accuracy of textile fabric quality.
[0032] Preferably, step S1 comprises the following steps:
[0033] Step S11: Designing the collection angle and illumination conditions for the textile and clothing intelligent manufacturing process to obtain the collection angle conditions and illumination conditions for the textile and clothing raw material cloth;
[0034] Step S12: Scanning the textile cloth image of the textile and clothing intelligent manufacturing process using a high-resolution image acquisition device based on the collection angle conditions and lighting conditions of the textile and clothing raw material cloth to obtain a textile cloth surface scan image;
[0035] Step S13: performing image denoising processing on the scanned image of the textile cloth surface to obtain a denoised image of the textile cloth surface;
[0036] Step S14: performing edge contrast enhancement processing on the denoised image of the textile cloth surface to obtain an edge enhanced optimized image of the textile cloth surface;
[0037] Step S15: grayscale conversion processing is performed on the textile cloth surface edge enhancement optimization image to obtain a textile cloth surface structure grayscale image.
[0038] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in FIG. 1 , in this embodiment, step S1 includes the following steps:
[0039] Step S11: Designing the collection angle and illumination conditions for the textile and clothing intelligent manufacturing process to obtain the collection angle conditions and illumination conditions for the textile and clothing raw material cloth;
[0040] In an embodiment of the present invention, the acquisition angle of the cloth image is designed by taking into account the surface characteristics and required resolution of the textile cloth in the intelligent manufacturing process of textiles and garments. Generally, the angle perpendicular to the cloth surface can maximize the display of the cloth's texture and details, so this angle is the first choice, and the specific angle will vary due to the special texture or design requirements of the cloth, so the acquisition angle needs to be adjusted according to the specific situation in the intelligent manufacturing process of textiles and garments. At the same time, the color temperature and brightness of the acquisition light source are adjusted by taking into account the color and reflection characteristics of the textile cloth, and finally the design obtains the acquisition angle conditions and lighting conditions of the textile and clothing raw material cloth.
[0041] Step S12: Scanning the textile cloth image of the textile and clothing intelligent manufacturing process using a high-resolution image acquisition device based on the collection angle conditions and lighting conditions of the textile and clothing raw material cloth to obtain a textile cloth surface scan image;
[0042] In an embodiment of the present invention, by combining the previously designed textile and clothing raw material cloth collection angle conditions and lighting conditions, a high-resolution image acquisition device (including a professional scanner or camera system with sufficient resolution and color depth to capture the tiny details and color changes on the cloth surface) is used to perform real-time image scanning of textile cloth in the intelligent manufacturing process of textile and clothing, so as to capture the tiny details and textures on the surface of the textile cloth in the intelligent manufacturing process in real time, and ensure that the device maintains a stable distance and angle with the cloth, and finally scan the surface of the textile cloth to obtain a scanned image.
[0043] Step S13: performing image denoising processing on the scanned image of the textile cloth surface to obtain a denoised image of the textile cloth surface;
[0044] In an embodiment of the present invention, since the scanned image of the surface of the textile cloth obtained by real-time scanning may be affected by the noise interference of the shooting equipment, light conditions or transmission process, these noises will affect the results of subsequent image processing. In order to effectively remove these noises, various image denoising algorithms can be used, such as median filtering, Gaussian filtering or wavelet transform, etc., to denoise the scanned image of the surface of the textile cloth obtained by real-time scanning to eliminate the noise in the image and retain useful image information. For example, for a scanned image of a textile cloth, if spot noise appears in the image due to uneven lighting or sensor failure during the shooting process, a median filtering algorithm can be applied. This algorithm replaces the value of each pixel with the median of its surrounding pixels, thereby eliminating redundant noise points while retaining the main structure of the image, and finally obtaining a denoised image of the surface of the textile cloth.
[0045] Step S14: performing edge contrast enhancement processing on the denoised image of the textile cloth surface to obtain an edge enhanced optimized image of the textile cloth surface;
[0046] In an embodiment of the present invention, the denoised image of the textile fabric surface after denoising is enhanced by using image enhancement techniques such as gradient calculation, histogram equalization or using a specific filtering operator, so as to highlight the edge features and detail information in the image by adjusting the contrast of the image pixels, so that the edges in the image are clearer and more prominent, and finally an edge-enhanced optimized image of the textile fabric surface is obtained.
[0047] Step S15: grayscale conversion processing is performed on the textile cloth surface edge enhancement optimization image to obtain a textile cloth surface structure grayscale image.
[0048] In an embodiment of the present invention, a corresponding grayscale conversion is performed on the edge enhanced and optimized image of the textile cloth surface that has undergone edge enhancement and optimization processing by using a corresponding grayscale algorithm (such as a component method or a grayscale projection method) to convert the color or enhanced image into a grayscale image, so as to better analyze and process the structure and features of the cloth surface, and finally obtain a grayscale image of the textile cloth surface structure.
[0049] The present invention firstly designs the acquisition angle and illumination conditions for the intelligent manufacturing process of textile and clothing, and accurately designs the acquisition angle and illumination conditions, so as to ensure that high-quality textile cloth surface images are obtained in the subsequent image scanning process. This step not only affects the clarity and accuracy of the image, but also directly affects the effectiveness of the subsequent processing steps and the reliability of the results. Through reasonable design, the influence of illumination conditions on image quality can be minimized, and the acquired images can truly reflect the surface characteristics and texture of the cloth, providing accurate data support and basis for intelligent manufacturing. Secondly, by applying high-resolution image acquisition equipment to image scan textile cloth based on the designed acquisition angle and illumination conditions, the tiny details and texture of the cloth surface can be captured, providing a sufficient data basis for subsequent image processing and analysis. The key to this step is to ensure the clarity and integrity of the image, and at the same time follow the acquisition conditions designed in advance to ensure that the acquired images can accurately reflect the real state and characteristics of the cloth. Then, by performing image denoising on the acquired textile cloth surface scanned image, the purpose is to eliminate noise and interference in the image, thereby improving the quality and clarity of the image. Denoising can effectively remove random noise introduced by factors such as lighting and equipment during the acquisition process, making subsequent image analysis and processing more accurate and reliable. The key to this step is to improve the signal-to-noise ratio of the image, thereby enhancing the details and recognition of the image, and providing more accurate data support for subsequent visual analysis and pattern recognition. Next, edge contrast enhancement processing is performed on the denoised image of the textile surface, aiming to highlight the edges and contours of the cloth surface in the image. By enhancing the edge contrast, the texture and structure of the cloth can be made clearer and more visible, further improving the visual effect and recognition ability of the image. The key to this step is to optimize the visual performance of the image, making the details and features of the cloth more prominent, which is helpful for subsequent intelligent manufacturing applications such as automatic detection, classification and quality control of the cloth. Finally, the color image is converted into a grayscale image by performing grayscale conversion on the edge enhancement optimized image, so as to better analyze and process the structure and features of the cloth surface. The grayscale processing can simplify the information complexity of the image and retain the main features of the cloth surface, while reducing the complexity and computational cost of data processing. The key to this step is to provide a clearer and easier-to-process data form for subsequent image analysis, thereby providing important support for automated decision-making and control in the intelligent manufacturing process.
[0050] Preferably, step S2 comprises the following steps:
[0051] Step S21: performing surface structure contour recognition analysis on the grayscale image of the textile surface structure to obtain the edge contour range of the textile surface structure;
[0052] Step S22: performing structural edge region segmentation on the grayscale image of the surface structure of the textile cloth based on the edge contour range of the surface structure of the textile cloth, so as to obtain a segmented image of the surface structure of the textile cloth;
[0053] Step S23: performing structural texture feature analysis on the textile surface structural region segmentation image to obtain textile surface structural texture feature data;
[0054] Step S24: performing surface structure defect evaluation calculation on the segmented image of the surface structure region of the textile cloth based on the surface structure texture feature data of the textile cloth to obtain the degree of surface structure defect of the textile cloth;
[0055] Step S25: Based on the degree of the surface structural defects of the textile fabric, the grayscale image of the surface structural defect of the textile fabric is analyzed for its quality impact, and a quality impact factor of the surface structural defect of the textile fabric is obtained.
[0056] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed step flow diagram of step S2 in FIG. 1 , in this embodiment, step S2 includes the following steps:
[0057] Step S21: performing surface structure contour recognition analysis on the grayscale image of the textile surface structure to obtain the edge contour range of the textile surface structure;
[0058] In an embodiment of the present invention, the grayscale image of the surface structure of a textile fabric that has been grayscaled is subjected to recognition and analysis of the surface structure contour by using an edge detection algorithm such as a Canny edge detection algorithm or a Sobel operator to find out the edges and contour lines in the image. These edges and contour lines reflect the structural features and shape of the surface of the textile fabric. For example, for a grayscale image of the surface of a textile fabric, the Sobel operator can be firstly applied to calculate the gradient of the image, and then the positions where the gradient changes significantly are found. These positions often correspond to the edges of the object. Then, the non-maximum suppression (NMS) technique can be applied to refine and extract these edge lines, and finally the edge contour range of the surface structure of the textile fabric is obtained.
[0059] Step S22: performing structural edge region segmentation on the grayscale image of the surface structure of the textile cloth based on the edge contour range of the surface structure of the textile cloth, so as to obtain a segmented image of the surface structure of the textile cloth;
[0060] In an embodiment of the present invention, the structural edge area in the corresponding grayscale image of the textile fabric surface structure is segmented by combining the textile fabric surface structure edge contour range obtained by previous identification and analysis, so as to segment the grayscale image of the fabric surface into different structural areas according to the structural edge contour. For example, based on the determined edge contour range, a watershed algorithm or a region growing-based segmentation method can be applied to perform region segmentation on the image, wherein the watershed algorithm can identify the dividing line between different regions in the image, while the region growing algorithm gradually grows regions with similar grayscale features through the similarity and connectivity of pixels, and finally obtains a segmented image of the textile fabric surface structure region.
[0061] Step S23: performing structural texture feature analysis on the textile surface structural region segmentation image to obtain textile surface structural texture feature data;
[0062] In an embodiment of the present invention, a texture feature analysis method is used to perform texture feature recognition and analysis on a segmented image of a textile surface structure region obtained through edge segmentation, so as to extract surface texture features from the image. These features can be used to describe and distinguish texture differences in different parts. For example, a local binary pattern (LBP) algorithm or a gray level co-occurrence matrix (GLCM) and other techniques can be used to calculate texture features of each region in the image. The LBP algorithm can describe the gray distribution pattern of pixels around each pixel point, thereby forming a texture feature with strong descriptiveness, while the GLCM can calculate the gray level co-occurrence relationship between pixels, measure the statistical characteristics of the texture in the image, such as texture features such as contrast, energy and entropy, and finally obtain the surface structure texture feature data of the textile.
[0063] Step S24: performing surface structure defect evaluation calculation on the segmented image of the surface structure region of the textile cloth based on the surface structure texture feature data of the textile cloth to obtain the degree of surface structure defect of the textile cloth;
[0064] In an embodiment of the present invention, the structural defect situation of the corresponding textile fabric surface structure area segmentation image is evaluated and analyzed by combining the textile fabric surface structure texture feature data obtained by previous analysis, so as to evaluate the degree of defect of the fabric surface structure according to the change and distribution of texture features, and accurately detect and analyze the defects or abnormal conditions existing on the fabric surface, and finally obtain the degree of defect of the textile fabric surface structure.
[0065] Step S25: Based on the degree of the surface structural defects of the textile fabric, the grayscale image of the surface structural defect of the textile fabric is analyzed for its quality impact, and a quality impact factor of the surface structural defect of the textile fabric is obtained.
[0066] In an embodiment of the present invention, a feature analysis is performed on a grayscale image of a textile surface structure obtained after grayscale processing by using a structural feature analysis method to extract key surface structure features from the image, including fiber density, texture uniformity, particle distribution, etc., and a quality impact assessment algorithm is used to evaluate and analyze the corresponding structural quality problems in the grayscale image of the textile surface structure, so as to evaluate, analyze and quantify the potential impact of different structural features on the surface quality of the fabric, thereby helping to understand and predict the overall quality influencing factors of the textile fabric, including structural influencing factors such as texture, particle size, and shape. Then, by combining the degree of surface structural defects of the textile fabric obtained by previous analysis, a regression analysis method (including linear regression, logistic regression or other statistical methods) is used to perform a defect impact regression analysis on each structural feature quality influencing factor, so as to quantitatively analyze the specific correlation degree of the defect degree with the fabric structural quality influencing factor, and screen out the structural quality influencing factors corresponding to a preset threshold value or greater than or equal to the preset threshold value to determine as defect influencing factors, thereby identifying the factors that have the most significant impact on the surface quality of the fabric, including specific texture features, particle distribution, pattern complexity, etc., and finally obtaining the textile surface structural defect quality influencing factors.
[0067] The present invention firstly analyzes the edge contour range in the image by performing surface structure contour recognition analysis on the grayscale image of the surface structure of the textile cloth, so as to accurately determine the main structural features and contour boundaries of the surface of the cloth. The key of this step is to extract and identify the structural edge information of each part of the surface of the cloth, and provide a clear data basis for subsequent image processing and analysis. Effective contour recognition analysis can ensure the accurate segmentation and feature analysis of the surface structure area of the cloth in the subsequent steps, thereby laying a solid foundation for defect detection and quality assessment. Secondly, by performing structural edge area segmentation on the grayscale image of the surface structure of the textile cloth based on the edge contour range of the surface structure of the textile cloth, the grayscale image of the surface of the cloth is segmented into different structural areas according to the structural edge contour. The key of this step is to accurately segment the surface of the cloth into independent structural areas, and provide accurate data support for subsequent texture feature analysis and defect assessment. By segmenting the image area, the complex image processing task can be effectively simplified, so that the analysis of each structural area is more refined and effective, which helps to improve the production accuracy and quality control level in intelligent manufacturing. Then, the structural texture feature analysis is performed on the segmented image of the surface structure area of the textile cloth, aiming to extract and analyze the texture feature data of each structural area. By analyzing the texture features of the cloth surface in detail, the texture type, density and distribution law of different areas can be identified, providing quantitative data basis for the quality assessment and defect detection of the cloth. The key of this step is to accurately describe and quantify the subtle changes and characteristics of the surface structure of the cloth, so as to provide a scientific basis for subsequent defect assessment and quality control decisions. Next, the surface structure defect evaluation calculation is performed on the segmented image of the surface structure area of the textile cloth based on the surface structure texture feature data of the textile cloth, aiming to evaluate the degree of defect of the surface structure of the cloth according to the change and distribution of the texture features. Through effective evaluation calculation, the defects or abnormalities on the surface of the cloth can be accurately detected and analyzed, and then corresponding quality control and repair measures can be taken. The key of this step is to improve the accuracy and reliability of defect detection, thereby ensuring the stability and controllability of product quality in the manufacturing process, and providing important support for enterprises to save costs and enhance competitiveness. Finally, the structural defect quality impact analysis is performed on the grayscale image of the textile surface structure based on the degree of the textile surface structure defect, aiming to analyze the influencing factors of the degree of defect on the surface structure quality of the fabric. Through detailed quality impact analysis, we can deeply understand the degree of influence of defects on the appearance and performance of the product, and provide a basis for improvement in product design and production process. The key of this step is to optimize the quality control strategy of the product, minimize the occurrence of defective products, and thus improve the production efficiency and product qualification rate of intelligent manufacturing of textiles and garments.
[0068] Preferably, step S24 comprises the following steps:
[0069] Step S241: performing surface structure defect detection on the segmented image of the surface structure region of the textile cloth based on the surface structure texture feature data of the textile cloth to obtain image defects of the surface structure region of the textile cloth;
[0070] In an embodiment of the present invention, advanced computer vision technology and image processing algorithms are used to detect and analyze the corresponding segmented image of the surface structure of the textile fabric by combining the texture feature data of the surface structure of the textile fabric obtained in the previous analysis, so as to detect the defects or defect points in the surface structure of the fabric according to the texture features in the image, such as breaks, stains or uneven fiber distribution points, and finally obtain the defects in the surface structure area of the textile fabric.
[0071] Step S242: locating the defect position coordinates of the regional image of the textile surface structure defect to obtain the regional image spatial position coordinates of each textile surface structure defect;
[0072] In an embodiment of the present invention, an image coordinate system is established to locate the defect position coordinates of the previously identified textile fabric surface structure regional image defects, so as to determine the specific position coordinates of each defect in the fabric surface structure regional image space, and accurately describe the position and distribution of the defects, and finally obtain the regional image space position coordinates of each textile fabric surface structure defect.
[0073] Step S243: based on the spatial position coordinates of the regional images of the surface structural defects of each textile cloth, the defect region connection marking is performed on the textile cloth surface structural regional segmentation image to obtain a textile cloth surface structural defect distribution region marking image;
[0074] In an embodiment of the present invention, by combining the spatial position coordinates of the regional image of each textile fabric surface structure defect previously located and using a pixel-level connection algorithm, such as a region growing algorithm or an edge detection algorithm, adjacent defects on the textile fabric surface structure regional segmentation image are connected together to form a continuous defect area, and finally a marked image of the textile fabric surface structure defect distribution area is obtained.
[0075] Step S244: performing defect distribution density analysis on the marked image of the textile fabric surface structural defect distribution area to obtain the textile fabric surface structural defect distribution density;
[0076] In an embodiment of the present invention, by using the pixel statistics and area counting functions in the image processing software and mathematical methods such as statistical distribution analysis or spatial statistical analysis, a statistical analysis of the distribution density of the corresponding defect distribution areas in the textile cloth surface structure defect distribution area marking image is performed to quantitatively analyze the distribution density of the defect area on the cloth surface, thereby helping to determine the severity and distribution of quality problems in different defect areas, and finally obtaining the distribution density of the textile cloth surface structure defect.
[0077] Step S245: Based on the surface structure defect distribution density of the textile fabric, a surface structure defect degree calculation formula is used to perform surface structure defect evaluation calculation on the marked image of the surface structure defect distribution area of the textile fabric to obtain the surface structure defect degree of the textile fabric.
[0078] In an embodiment of the present invention, a suitable surface structure defect degree calculation formula is formed by combining the total area size of the region, the distribution density of surface structure defects of textile fabrics, the regional coordinate parameters, the surface structure defect influence weight coefficient, the gray value, the average gray value of the textile fabric surface, the gray value influence degree coefficient, the defect distribution density energy potential function and related parameters to perform surface structure defect evaluation calculation on the marked image of the surface structure defect distribution area of the textile fabric, so as to quantitatively calculate and determine the overall defect degree of the textile fabric surface, and finally obtain the surface structure defect degree of the textile fabric.
[0079] The present invention firstly detects surface structure defects on the segmented image of the surface structure area of the textile cloth based on the surface structure texture feature data of the textile cloth, so as to analyze the texture features in the image, and can detect the defects or defect points in the surface structure of the cloth, such as breaks, stains or uneven fiber distribution points. The key to this step is to improve the accuracy and sensitivity of the detection, and ensure that all potential defect points can be effectively identified and marked. Accurate defect detection lays a solid foundation for subsequent position positioning and defect analysis, thereby helping to achieve immediate quality control and problem solving in the intelligent manufacturing production process. Secondly, positioning the position coordinates of the defects in the image of the surface structure area of the textile cloth is an important step for further processing the defect data. By determining the specific position coordinates of each defect in the image space, the position and distribution of the defect can be accurately described. The key to this step is to provide accurate positioning information for each defect, which provides necessary data support for the subsequent connection and analysis of the defect area. Accurate position coordinate positioning helps to accurately evaluate the degree of influence of the defect on the quality of the cloth, thereby providing a reliable basis for quality improvement in the intelligent manufacturing process of textile clothing. Then, the defect area connection marking is performed on the segmented image of the textile surface structure area based on the spatial position coordinates of the regional image of each textile surface structure defect, in order to combine adjacent or related defects into a larger defect area to form a complete defect distribution area marking image. The key to this step is to provide an in-depth understanding of the correlation between defects, and to identify a wider range of defect areas, so as to more comprehensively evaluate the quality problems of the surface of the cloth. Through effective connection marking, the error in the analysis process can be reduced, the comprehensiveness and accuracy of defect evaluation can be ensured, and a more reliable basis for quality management decisions can be provided. Next, the defect distribution density analysis of the marked image of the defect distribution area of the textile surface structure is to quantify the distribution density and frequency of defects on the surface of the cloth. Detailed analysis of the defect distribution density can help determine the severity of quality problems in different areas and provide important data support for subsequent quality evaluation and improvement. The key to this step is to provide an in-depth understanding of the overall distribution characteristics of the quality problems on the surface of the cloth, which can help formulate targeted quality control strategies and production optimization plans. Finally, the surface structural defect evaluation calculation is performed on the marked image of the surface structural defect distribution area of the textile cloth by using the surface structural defect degree calculation formula based on the surface structural defect distribution density of the textile cloth. The final defect evaluation calculation is based on the data analysis and quantification results in the previous steps, and uses a specially designed evaluation formula to determine the overall defect degree of the cloth surface, which can provide a comprehensive evaluation of the cloth quality issues, not only considering the number and distribution of defects, but also combining their impact on the overall quality of the product. Through accurate defect degree evaluation, necessary control measures can be taken in a timely manner, thereby improving product quality and customer satisfaction in the intelligent manufacturing process of textiles and garments.
[0080] Preferably, the surface structure defect degree calculation formula in step S245 is specifically:
[0081] ;
[0082] ;
[0083]
[0084] In the formula, It is the degree of surface structural defects of textile fabrics. The total size of the area of the image marked with the distribution area of the surface structural defects of the textile cloth, is the distribution density of structural defects on the surface of textile fabrics, is the horizontal coordinate parameter of the distribution area of the surface structural defects of the textile fabric, is the ordinate parameter of the distribution area of the surface structural defects of textile fabrics, is the weight coefficient of surface structural defects, Mark the distribution area of the textile fabric surface structure defect. The distribution area of the textile fabric surface structure defect in the image is at the position The gray value at is the average gray value of the textile surface, is the gray value influence coefficient, Mark the distribution area of the textile fabric surface structure defect. The distribution area of the textile fabric surface structure defect in the image is at the position The defect distribution density energy potential function at It is the correction factor for the degree of surface structural defects of textile fabrics.
[0085] The present invention obtains a surface structure defect degree calculation formula by using a specific mathematical model and after verification, which is used to evaluate and calculate the surface structure defect of the marked image of the surface structure defect distribution area of the textile cloth. The surface structure defect degree calculation formula can comprehensively evaluate the degree of the surface structure defect of the textile cloth by combining the distribution density of the defect, the change of the defect in space and the change of the gray value. By considering the influence and weight of different parameters, the formula can accurately quantify the influence of the surface structure defect, thereby providing an important reference for quality control and production process optimization. Among them, the defect distribution density energy potential function describes the distribution change of the defect in space, and can consider the local change of the defect distribution, so as to reveal the shape and boundary information of the defect. In addition, the formula also introduces a correction coefficient to allow the evaluation result to be adjusted according to the actual situation to adapt to different application requirements or standard settings, thereby enhancing the accuracy and adaptability of the evaluation calculation. In summary, the detailed analysis and application of the formula can effectively help detect and evaluate surface structure defects in the textile production process, thereby improving product quality and production efficiency. Therefore, the formula fully considers the degree of surface structure defects of textile cloth. , the total size of the area of the marked image of the surface structural defect distribution area of the textile cloth , distribution density of structural defects on the surface of textile fabrics , the horizontal coordinate parameter of the distribution area of the surface structural defects of textile fabrics , the ordinate parameter of the distribution area of the surface structural defects of textile fabrics , surface structural defects affect the weight coefficient , the distribution area of the textile fabric surface structure defect is marked in the image. The distribution area of the textile fabric surface structure defect is at the position Gray value at , the average gray value of the textile surface , gray value influence coefficient , the distribution area of the textile fabric surface structure defect is marked in the image. The distribution area of the textile fabric surface structure defect is at the position The defect distribution density energy potential function at , correction factor for the degree of surface structural defects of textile fabrics , where the total area of the image marked by the distribution area of the textile surface structural defects is , the horizontal coordinate parameter of the distribution area of the surface structural defects of textile fabrics , the ordinate parameter of the distribution area of the surface structural defects of textile fabrics , the distribution area of the textile fabric surface structure defect is marked in the image. The distribution area of the textile fabric surface structure defect is at the position Gray value at , the average gray value of the textile surface And the gray value influence coefficient It constitutes a distribution density of structural defects on the surface of textile fabrics. Functional relationship , and also through the distribution density of structural defects on the surface of textile fabrics , the horizontal coordinate parameter of the distribution area of the surface structural defects of textile fabrics , the ordinate parameter of the distribution area of the surface structural defects of textile fabrics The related operation method constitutes a method for marking the distribution area of the surface structural defects of textile fabrics. The defect distribution density energy potential function at Functional relationship , according to the degree of surface structural defects of textile fabrics The correlation between the above parameters forms a functional relationship:
[0086] ;
[0087] This formula can realize the surface structure defect evaluation calculation process of the marked image of the distribution area of the textile surface structure defect. At the same time, the correction coefficient of the degree of the textile surface structure defect is used. The introduction of can be adjusted according to the errors occurring in the calculation process, thereby improving the accuracy and applicability of the calculation formula for the degree of surface structural defects.
[0088] Preferably, step S25 comprises the following steps:
[0089] Step S251: performing structural feature analysis on the grayscale image of the textile surface structure to obtain textile surface structure feature data;
[0090] In an embodiment of the present invention, a feature analysis is performed on a grayscale image of the surface structure of a textile fabric obtained after grayscale processing by using a structural feature analysis method to extract key surface structural features from the image. These structural features may include fiber density, texture uniformity, particle distribution, etc., and enable them to effectively describe the physical structural features of the surface of the fabric, and finally obtain surface structural feature data of the textile fabric.
[0091] Step S252: performing a structural quality impact assessment analysis on the grayscale image of the textile surface structure based on the textile surface structure feature data to obtain a data set of factors affecting the textile surface structure quality;
[0092] In an embodiment of the present invention, a quality impact assessment algorithm is used to evaluate and analyze the corresponding structural quality problems in the grayscale image of the textile fabric surface structure by combining the textile fabric surface structure feature data obtained in the previous analysis, so as to evaluate, analyze and quantify the potential impact of different structural features on the surface quality of the fabric, thereby helping to understand and predict the overall quality influencing factors of the textile fabric, including structural influencing factors such as texture, particle size, and shape, so as to better understand how different structural features affect the appearance and performance of the textile fabric surface, and finally obtain a data set of textile fabric surface structure quality influencing factors.
[0093] Step S253: performing defect impact regression analysis on the data set of factors affecting the surface structure of textile fabrics based on the degree of surface structure defects of the textile fabrics, so as to obtain the degree of influence correlation between the surface structure defects and the factors affecting the quality;
[0094] In an embodiment of the present invention, a regression analysis method (including linear regression, logistic regression or other statistical methods) is used to perform a regression analysis of the impact of defects on each structural feature quality influencing factor in the textile fabric surface structure quality influencing factor data set by combining the surface structure defect degree of the textile fabric obtained by previous analysis, so as to quantitatively analyze the specific impact of the defect degree on the fabric structure quality influencing factors, and establish a quantitative relationship between the defect degree and the structural quality influencing factors to determine which structural features are significantly associated with the occurrence of defects, and finally obtain the impact correlation degree between the surface structure defects and the quality influencing factors.
[0095] Step S254: performing defect influencing factor identification analysis on the textile fabric surface structure quality influencing factor data set according to the degree of influence correlation between the surface structure defects and the quality influencing factors, and obtaining the textile fabric surface structure defect quality influencing factors.
[0096] In an embodiment of the present invention, the influence correlation degree between the surface structural defects and the quality influencing factors obtained by the previous quantitative calculation is compared and judged according to a preset influence correlation threshold, so as to screen out the influence correlation degree between the surface structural defects and the quality influencing factors that is greater than or equal to the preset threshold, and determine the corresponding structural quality influencing factors in the textile cloth surface structure quality influencing factor data set as defect influencing factors, thereby identifying the factors that have the most significant impact on the cloth surface quality, these factors include specific texture features, particle distribution, pattern complexity, etc., and finally obtain the textile cloth surface structure defect quality influencing factors.
[0097] The present invention firstly analyzes the structural features of the grayscale image of the surface structure of the textile cloth to extract key surface structural feature data from the image. These feature data may include fiber density, texture uniformity, particle distribution, etc. These features can effectively describe the physical structural features of the cloth surface. The key to this step is that it can provide a quantitative description of the overall structural characteristics of the cloth, which lays the foundation for subsequent quality impact assessment and defect analysis. Through accurate feature extraction and analysis, the structural differences between different cloth samples can be identified, thereby providing data support for the quality management and product optimization of the intelligent manufacturing process of textile clothing. Secondly, by performing structural quality impact assessment and analysis on the grayscale image of the surface structure of the textile cloth based on the surface structural feature data of the textile cloth, the purpose is to understand the actual impact of these features on the surface structural quality of the cloth through quantitative analysis. By establishing a correlation model between structural feature data and quality impact, it can be determined which structural features have a decisive impact on the quality of the final product in actual production, which helps producers and R&D personnel understand and optimize the manufacturing process of the cloth, and improve product quality and performance by adjusting specific structural parameters, thereby enhancing market competitiveness. Then, by performing defect impact regression analysis on the dataset of factors affecting the surface structure quality of textile fabrics based on the degree of surface structural defects of textile fabrics, the purpose is to explore the specific degree of influence of defects on the factors affecting the surface structure quality of fabrics. By studying the relationship between surface defects and structural feature data through regression analysis, the influence of different types of defects on fabric quality can be quantified, and a prediction model can be established to evaluate the potential impact of defects on product performance, which can provide an in-depth understanding of the root causes of fabric surface quality problems and provide a scientific basis for formulating improvement measures and optimizing production processes. Finally, by performing defect impact factor identification analysis on the dataset of factors affecting the surface structure quality of textile fabrics based on the degree of correlation between surface structural defects and quality influencing factors, the purpose is to determine which structural features have a decisive influence on the quality of the final product in actual production. By analyzing the degree of influence of different defect types on various structural features, key defect impact factors can be identified. These factors directly affect the quality and performance of the product, which helps producers and R&D personnel to optimize the production process in a targeted manner, reduce the possibility of defects, and improve product consistency and reliability, thereby enhancing customer satisfaction and market competitiveness of intelligent manufacturing of textiles and apparel.
[0098] Preferably, step S3 comprises the following steps:
[0099] Step S31: performing color space conversion on the scanned image of the textile cloth surface to obtain the LAB color space of the textile cloth surface image;
[0100] In an embodiment of the present invention, the color space of the scanned image of the textile cloth surface is converted by using the LAB color space conversion method to separate the color information from the brightness information in the scanned image of the textile cloth surface, so that the description of the color is more independent and accurate, thereby converting the scanned image of the textile cloth surface into the LAB color space, and ensuring that the RGB value of each pixel can be accurately mapped to the corresponding LAB color space coordinate, and finally obtaining the LAB color space of the textile cloth surface image.
[0101] Step S32: performing pixel point area division on the LAB color space of the textile cloth surface image to obtain pixel point color space sub-regions of each textile cloth image;
[0102] In an embodiment of the present invention, each pixel point in the LAB color space of the textile fabric surface image is divided into regions by using an image segmentation algorithm (for example, a method based on region growing or a method based on edge detection), so as to subdivide the LAB color space of the textile fabric surface image into smaller regions according to the position of the pixel points or specific rules, thereby more accurately analyzing the color characteristics and color distribution of each region, and ensuring that the pixel colors in the corresponding sub-regions are relatively consistent, and finally obtaining the color space sub-regions of each textile fabric image pixel point.
[0103] Step S33: Calculate the pixel chromaticity of each color space sub-region of the textile fabric image pixel point to obtain the pixel chromaticity value of each color sub-region of the textile fabric pixel point;
[0104] In an embodiment of the present invention, the chromaticity value of the pixel in each area is quantitatively calculated by calculating the specific color values of the color space sub-area of each textile fabric image pixel point on the L, a, and b channels, and it is made to reflect the specific color attributes of each pixel point in the LAB color space, and finally the pixel chromaticity value of the color sub-area of each textile fabric pixel point is obtained.
[0105] Step S34: performing a quantitative calculation of the surface color difference defects of the textile cloth surface scanned image based on the color space sub-regions of each textile cloth image pixel point and the pixel chromaticity value of each textile cloth pixel point color sub-region to obtain the textile cloth surface color difference defect degree coefficient;
[0106] In an embodiment of the present invention, by combining the color space sub-regions of the pixel points of each textile fabric image obtained by previous division and the pixel chromaticity values of the color sub-regions of each textile fabric pixel point obtained by quantitative calculation, the surface color difference defects of the corresponding textile fabric surface scan image are quantitatively calculated, and the distance measurement method of the color space is used to calculate the distance between each pixel sub-region and the ideal color or the color difference between adjacent sub-regions, and the degree of color difference defects in each region is evaluated and analyzed, and finally the color difference defect degree coefficient of the textile fabric surface is obtained.
[0107] Step S35: performing a color difference defect quality impact analysis on the scanned image of the textile cloth surface based on the color difference defect degree coefficient of the textile cloth surface to obtain a color difference defect quality impact factor of the textile cloth surface.
[0108] In an embodiment of the present invention, the quality impact analysis of color difference defects is performed on the scanned image of the entire textile cloth surface by combining the color difference defect degree coefficient of the textile cloth surface obtained by quantitative calculation. This analysis aims to identify and understand the actual impact of color difference defects on the cloth surface quality, so as to determine which factors are particularly important when color difference defects occur, and determine and analyze the key parameters that cause color difference in the intelligent manufacturing production process of textile clothing, including color material formula ratio, dyeing temperature, dyeing time, pressure, etc., and finally obtain the quality influencing factors of color difference defects on the textile cloth surface.
[0109] The present invention first converts the scanned image of the textile surface into color space in order to more accurately capture and analyze the color characteristics of the cloth surface. The LAB color space includes brightness (L) and two chromaticity channels (a and b), which can provide a wider range of color expression capabilities and help distinguish subtle color changes and color differences. The key to this step is to ensure that the color characteristics of the cloth surface can be accurately evaluated in subsequent analysis, thereby laying the foundation for detecting and identifying color defects. Secondly, the LAB color space of the textile surface image is divided into pixel areas in order to divide the image into multiple pixel color space sub-areas. This area division can be performed based on the position of the pixel or specific rules, so as to subdivide the image into smaller areas, so as to more accurately analyze the color characteristics and color distribution of each area, thereby providing a refined analysis capability of the local color characteristics of the cloth surface, which helps to accurately locate and identify the existing color difference areas. Then, by calculating the pixel chromaticity of each textile fabric image pixel color space sub-region, that is, calculating the chromaticity value of the pixel in each region, these chromaticity values reflect the specific color attributes of each pixel in the LAB color space, including the color brightness, saturation and hue information, so as to provide a quantitative description of the color characteristics of each pixel on the surface of the fabric, and provide the necessary data basis for the subsequent quantitative calculation of color difference defects. Next, by quantitatively calculating the surface color difference defects of the textile fabric surface scanned image based on the pixel chromaticity values of each textile fabric image pixel color space sub-region and each textile fabric pixel color sub-region, this step aims to evaluate the degree of color difference between various regions on the fabric surface, and calculate the color difference defect degree coefficient by comparing the difference between the expected color and the actual color, so as to provide a quantitative evaluation of the color consistency and quality problems of the fabric surface, and help to accurately locate and analyze the possible color difference defect areas, so as to guide the subsequent quality improvement measures and production optimization. Finally, the color difference defect quality impact analysis is performed on the surface scanned image of the textile fabric based on the color difference defect degree coefficient of the textile fabric surface. This step aims to determine the actual impact of the color difference defect on the surface quality of the fabric and identify the key factors that cause the color difference defect. By analyzing the influencing factors, the key parameters that cause color difference in the production process can be determined and optimized to improve the color consistency and quality stability of the product, which helps manufacturers to deeply understand the root causes of the color difference problem, thereby providing a scientific basis for formulating effective quality control strategies and improving product competitiveness.
[0110] Preferably, step S34 includes the following steps:
[0111] Step S341: performing regional position positioning on the color space sub-region of each textile fabric image pixel point to obtain the color space position coordinates of the color sub-region of each textile fabric pixel point;
[0112] In an embodiment of the present invention, by performing regional position positioning on the color space sub-region of the pixel point of the textile cloth image to obtain the color space position coordinates of each sub-region, it is first necessary to determine the position of each pixel point, which can be achieved through coordinate mapping and pixel position indexing in image processing. For the image in the LAB color space, each pixel point has three components: brightness (L), green-red hue (a), and blue-yellow hue (b). By traversing each pixel, its row and column position in the image can be obtained, and further mapped to specific coordinates in the LAB color space, and finally the color space position coordinates of the color sub-region of each textile cloth pixel point are obtained.
[0113] Step S342: performing pixel chromaticity vector conversion on the color space position coordinates of the color sub-region of each textile fabric pixel and the pixel chromaticity value to obtain the chromaticity space distribution vector of each textile fabric pixel;
[0114] In an embodiment of the present invention, the color space position coordinates and pixel chromaticity values of the color sub-region of each textile fabric pixel are converted into a pixel chromaticity space distribution vector. This conversion is to better describe the distribution characteristics of the color in each sub-region. In the specific implementation process, it is necessary to combine the color space position coordinates and the corresponding chromaticity values of each pixel to form a vector representation. This vector usually includes a combination of position information and chromaticity information, such as the position coordinates and the corresponding L, a, and b values in the LAB color space, to ensure that the vector can accurately reflect the color distribution in each sub-region, and finally obtain the chromaticity space distribution vector of each textile fabric pixel.
[0115] Step S343: performing vector space distance calculation on the chromaticity space distribution vector of each textile fabric pixel point to obtain the vector space distance between each textile fabric pixel point distribution vector; performing color deviation statistical calculation on the chromaticity space distribution vector of each textile fabric pixel point based on the vector space distance between each textile fabric pixel point distribution vector to obtain the pixel color difference value between the color sub-regions of each textile fabric pixel point;
[0116] In an embodiment of the present invention, the spatial distance of the chromaticity space distribution vector of each textile fabric pixel is calculated by using a distance measurement method such as Euclidean distance, Manhattan distance or cosine similarity to quantitatively calculate the vector space distance between each distribution vector, thereby obtaining the vector space distance between each textile fabric pixel distribution vector. At the same time, by combining the vector space distance between each textile fabric pixel distribution vector calculated previously and the preset vector space distance threshold between each textile fabric pixel chromaticity space distribution vector, the degree of deviation is statistically calculated to evaluate and analyze the degree of color difference between different areas of the fabric surface through a numerical method, and finally obtain the pixel color difference value between the color sub-areas of each textile fabric pixel.
[0117] Step S344: performing pixel color difference defect area detection on the surface scanned image of the textile cloth based on the pixel color difference values between the color sub-areas of each pixel point of the textile cloth, and obtaining the pixel color difference defect area of the surface image of the textile cloth;
[0118] In an embodiment of the present invention, the color difference defect area of the scanned image of the textile cloth surface is detected by combining the pixel color difference values between the color sub-areas of each textile cloth pixel point obtained by previous statistical calculation, so as to determine which areas or pixels have significant color differences based on the statistical results of color deviation, and accurately locate and identify the areas where the color difference defects exist in the cloth surface structure, and finally obtain the pixel color difference defect area of the textile cloth surface image.
[0119] Step S345: performing area statistics on the color difference defect area of the textile surface image pixels to obtain the area size of the color difference defect area on the textile surface;
[0120] In an embodiment of the present invention, an area measurement method is used to perform statistical calculations on the areas of previously identified pixel color difference defect areas on the surface of a textile fabric, so as to calculate the number of pixels or areas marked as color difference defect areas, quantify the size and distribution of these areas, and finally obtain the area size of the color difference defect area on the surface of the textile fabric.
[0121] Step S346: Based on the size of the color difference defect area on the surface of the textile cloth, a surface color difference defect degree calculation formula is used to perform a quantitative calculation of the surface color difference defect on the scanned image of the textile cloth surface to obtain a color difference defect degree coefficient on the surface of the textile cloth.
[0122] In an embodiment of the present invention, a suitable surface color difference defect degree calculation formula is formed by combining the number parameters of the color difference defect area on the surface of the textile cloth, the area size of the color difference defect area on the surface of the textile cloth, the spatial coordinate range parameter, the spatial coordinate variable parameter, the pixel color change value, the weight coefficient, the pixel color change defect influence index and related parameters to perform a quantitative calculation of the surface color difference defects on the scanned image of the textile cloth surface, so as to quantitatively calculate the color difference defect quality status of the entire textile cloth surface, and finally obtain the textile cloth surface color difference defect degree coefficient.
[0123] The present invention first locates the regional position of the color space sub-region of each textile cloth image pixel point, and can determine the accurate position and spatial distribution of each sub-region in the overall image. The key to this step is that it can provide basic data for subsequent color analysis and defect detection, thereby ensuring accurate analysis and quantitative evaluation of each local area on the cloth surface. At the same time, by performing pixel chromaticity vector conversion on the color space position coordinates and pixel chromaticity values of the color sub-region of each textile cloth pixel point, this vectorization processing helps to convert complex color information into a mathematically processable form, so that subsequent numerical calculations and statistical analysis can be more convenient, which can provide convenience and accuracy for color deviation statistics and color difference analysis, and enhance the in-depth understanding and quantitative description capabilities of the color characteristics of the cloth surface. Secondly, by calculating the vector space distance of the chromaticity space distribution vector of each textile fabric pixel point, the vector space distance between different pixel point distribution vectors is calculated. The color deviation statistics of the chromaticity space distribution vector of each textile fabric pixel point are calculated by combining the vector space distance between the previously calculated pixel point distribution vectors. This process can evaluate and analyze the color difference between different areas of the fabric surface through a numerical method, providing a quantitative basis for subsequent color deviation statistics and defect detection, so that the color distribution similarities and differences of each area can be accurately measured and compared, helping to identify potential color difference problems and quality defects. Then, the pixel color difference defect area of the textile fabric surface scanned image is detected based on the pixel color difference value between the color sub-areas of each textile fabric pixel point. This step aims to accurately locate and identify the area with color difference defects on the fabric surface structure according to the statistical results of color deviation, and provide an operational information basis for subsequent quality control and improvement measures. This can achieve accurate identification and localization of color difference problems on the fabric surface, thereby providing key data support for timely adjustment and improvement in the production process. Next, the area statistics of the pixel color difference defect areas on the surface of textile fabrics are performed. This step is to quantitatively evaluate the size and impact of different color difference defect areas on the surface of the fabric, and provide the necessary numerical basis for subsequent color difference defect degree calculation and quality analysis, thereby helping producers and quality managers understand and evaluate the severity of different defect areas, so as to guide the optimization of the production process of intelligent manufacturing of textiles and garments and improve product quality.Finally, the surface color difference defects of the textile fabric surface scanned image are quantitatively calculated by using the surface color difference defect degree calculation formula based on the area size of the textile fabric surface color difference defect area. This step aims to convert the area size and severity of different color difference defect areas into comparable quantitative indicators through mathematical models or algorithms, so as to evaluate and improve product quality. It can provide an objective and repeatable evaluation method to help manufacturers accurately evaluate the color difference problems of products and take effective control and improvement measures to improve product quality and market competitiveness.
[0124] Preferably, the surface color difference defect degree calculation formula in step S346 is specifically:
[0125] ;
[0126] In the formula, is the coefficient of color difference defect degree on the surface of textile fabrics, is the total number of color difference defect areas on the textile surface in the textile surface scan image. is the index parameter of the color difference defect area on the surface of textile cloth, For the The area of the color difference defect area on the surface of the textile cloth, is the spatial coordinate range parameter of the color difference defect area on the surface of textile cloth, is the spatial coordinate variable parameter of the color difference defect area on the surface of the textile cloth, For the The color difference defect area on the surface of the textile cloth is located in the spatial coordinates The pixel color change value at For the The weight coefficient of the color difference defect area on the surface of the textile cloth, For the The pixel color change defect impact index of the color difference defect area on the surface of the textile cloth is It is the correction value of the color difference defect degree coefficient on the surface of textile fabrics.
[0127] The present invention obtains a surface color difference defect degree calculation formula by using a specific mathematical model and after verification, which is used to quantitatively calculate the surface color difference defects of the scanned image of the textile cloth surface. The surface color difference defect degree calculation formula combines the area, color change and spatial distribution information of each color difference defect area, and can comprehensively and quantitatively evaluate the color difference defects on the surface of the textile cloth. By calculating the color difference characteristics of each area and considering its importance and weight, the degree of surface color difference can be accurately measured, thereby providing a basis for quality control and production process optimization. In addition, the correction coefficient in the formula allows adjustment according to actual conditions to adapt to different application scenarios or user needs, thereby improving the accuracy and adaptability of the evaluation. In summary, the detailed analysis and application of the formula can effectively help to detect and evaluate surface color difference problems in real time during the textile production process, thereby improving product quality and production efficiency. Therefore, the formula fully considers the textile cloth surface color difference defect degree coefficient. , the total number of color difference defect areas on the textile surface in the textile surface scan image , item index parameter of color difference defect area on textile cloth surface , No. The area of the color difference defect area on the surface of the textile cloth , the spatial coordinate range parameters of the color difference defect area on the surface of textile cloth , the spatial coordinate variable parameters of the color difference defect area on the surface of textile cloth , No. The color difference defect area on the surface of the textile cloth is located in the spatial coordinates The pixel color change value at , No. Weight coefficient of color difference defect area on the surface of textile cloth , No. Defect impact index of pixel color change in the color difference defect area of the textile surface , Correction value of the color difference defect degree coefficient of textile cloth surface , according to the color difference defect degree coefficient of textile cloth surface The correlation between the above parameters forms a functional relationship:
[0128] ;
[0129] This formula can realize the quantitative calculation process of the surface color difference defects of the scanned image of the textile cloth surface, and at the same time, the correction value of the degree coefficient of the color difference defect of the textile cloth surface is obtained. The introduction of can be adjusted according to the errors occurring in the calculation process, thereby improving the accuracy and applicability of the calculation formula for the degree of surface color difference defects.
[0130] Preferably, step S4 comprises the following steps:
[0131] Step S41: Based on the quality influencing factors of textile surface structural defects and the quality influencing factors of textile surface color difference defects, an important evaluation analysis of the quality of textile fabrics corresponding to the textile fabrics in the intelligent manufacturing process of textile garments is performed to obtain the importance of the fabric quality influence of each structural defect influence factor and color difference defect influence factor;
[0132] In an embodiment of the present invention, by collecting and sorting out the influencing factors of structural defects and color difference defects related to the quality of textile fabrics, wherein the influencing factors of structural defects may include influencing factors related to the structure of the fabric such as broken yarn, weaving defects, and abrasions, while the influencing factors of color difference defects include influencing factors related to the color quality of the fabric such as uneven color and excessive color difference, and at the same time, by using data analysis tools or mathematical models for processing, for example, using statistical methods or machine learning algorithms to quantitatively calculate the impact of different influencing factors on the corresponding textile fabrics in the intelligent manufacturing process of textiles and garments, a quantitative evaluation is performed to quantitatively determine the contribution of each structural defect and color difference defect influencing factor to the overall textile fabric quality, and finally obtain the importance of the fabric quality impact of each structural defect influencing factor and color difference defect influencing factor.
[0133] Step S42: assigning defect quality impact weights to the textile fabric surface structural defect quality impact factors based on the cloth quality impact importance of each structural defect impact factor, and obtaining the cloth quality impact weights of each structural defect impact factor;
[0134] In an embodiment of the present invention, influence weights are allocated to corresponding textile cloth surface structural defect quality influencing factors by combining the importance of the cloth quality influence of each structural defect influencing factor obtained by previous quantitative calculation, so as to adopt a variety of methods, such as the analytic hierarchy process (AHP), principal component analysis (PCA) or data-driven machine learning models, to determine the specific influence weight of each structural defect influencing factor on the overall cloth quality, and ensure that the final weight allocation result can reflect the actual influence of various structural defect influencing factors on cloth quality, and finally obtain the cloth quality influence weight of each structural defect influencing factor.
[0135] Step S43: allocating color difference quality impact weights to the color difference quality impact factors on the surface of textile cloth based on the cloth quality impact importance of each color difference defect impact factor, and obtaining the cloth quality impact weights of each color difference defect impact factor;
[0136] In an embodiment of the present invention, similarly, by combining the cloth quality influence importance of each color difference defect influencing factor obtained by previous quantitative calculation, the influence weights of the corresponding textile cloth surface color difference defect quality influencing factors are allocated, so as to adopt a variety of methods, such as the analytic hierarchy process (AHP), the principal component analysis method (PCA) or a data-driven machine learning model, etc., to determine the specific influence weight of each color difference defect influencing factor on the overall cloth quality, and ensure that the weight allocation can objectively and accurately reflect the influence of the color difference defect on the overall cloth quality, and finally obtain the cloth quality influence weight of each color difference defect influencing factor.
[0137] Step S44: performing cloth quality weighted average calculation on the textile cloth surface structure defect quality influence factor and the textile cloth surface color difference defect quality influence factor based on the cloth quality influence weights of each structure defect influence factor and the color difference defect influence factor to obtain a comprehensive textile cloth surface quality score;
[0138] In the embodiment of the present invention, the weighted average mathematical operation method is used to perform quantitative calculation by combining the product of the cloth quality influence weights of each structural defect influence factor previously allocated and the textile cloth surface structural defect quality influence factor and the product of the cloth quality influence weights of each color difference defect influence factor and the textile cloth surface color difference defect quality influence factor, so as to apply the allocated weight data to the specific quality influence factor to calculate the comprehensive quality score value of the textile cloth surface. For example, the specific implementation formula is: , where It is the comprehensive quality score of textile fabric surface. is the total number of factors affecting the quality of textile fabric surface structural defects, is the quantitative index parameter of the factors affecting the quality of textile surface structural defects, is the total number of factors affecting the quality of color difference defects on the surface of textile fabrics, is the quantitative index parameter of the factors affecting the quality of color difference defects on the surface of textile fabrics, For the The weight of cloth quality influence of structural defect influence factors, For the Factors affecting the quality of textile fabric surface structural defects: For the The weight of the cloth quality influencing factor of the color difference defect, For the The influencing factors of color difference defects on the surface of textile fabrics are analyzed, and finally the comprehensive quality score of the surface of textile fabrics is obtained.
[0139] Step S45: Performing quality grade monitoring and analysis on the comprehensive quality score value of the textile cloth surface according to the preset textile cloth quality grade judgment standard to obtain the textile cloth quality grade monitoring result.
[0140] In an embodiment of the present invention, by using a preset textile fabric quality grade judgment standard (for example, a quality score value between 0 and 40 is determined as ordinary, 41 to 60 is determined as good, 61 to 100 is determined as excellent, etc., which are determined according to threshold setting, numerical range division or experience-based classification methods, etc.), the textile fabric surface comprehensive quality score values previously obtained by quantitative calculation are compared and matched, so as to monitor and determine the quality grade of each textile fabric based on the correspondence between the pre-set score value and the quality grade, and finally obtain the textile fabric quality grade monitoring result.
[0141] The present invention firstly conducts an important evaluation and analysis of the quality of textile fabrics corresponding to the intelligent manufacturing process of textile clothing by combining the quality influencing factors of textile fabric surface structural defects and the quality influencing factors of textile fabric surface color difference defects. The key to this step is to determine the importance of different quality influencing factors to the overall quality of the fabric through systematic evaluation and analysis, which provides a theoretical basis and practical guidance for subsequent quality improvement and control. By quantifying and comparing the influence of different factors, the production process can be optimized and product quality can be improved in a targeted manner. Secondly, the weight of the influence of the quality influencing factors of the structural defects on the surface of the textile fabric is allocated based on the importance of the influence of the quality of the fabric quality of each structural defect influencing factor. The key to this step is to accurately calculate and allocate the actual influence weights of different defect factors on the quality of the fabric according to the results of the previous evaluation. Through weight allocation, the contribution of various structural defects to the overall quality of the fabric can be more accurately understood and quantified, which is helpful to optimize the production process and improve product design, thereby improving quality standards and reducing production costs. Then, the color difference quality impact weights are allocated to the color difference quality impact factors on the surface of textile fabrics based on the importance of the color difference defect impact factors on the fabric quality. The key to this step is to accurately allocate the weights of various color difference factors on the fabric quality according to the degree and importance of the color difference defect. Through the analysis of weight allocation, the negative impact of color difference defects on the overall quality of the fabric can be more accurately evaluated and controlled, providing manufacturers with specific directions for optimizing production strategies and improving product quality. Next, the weighted average calculation of the fabric quality of the structural defect quality impact factors and the color difference defect quality impact factors on the surface of textile fabrics is performed based on the cloth quality impact weights of the various structural defect impact factors and the color difference defect impact factors. The key to this step is to comprehensively consider and balance the influence of various factors, and obtain the comprehensive quality score of the textile fabric surface through the weighted average method. Through this method, the overall quality status of the fabric can be comprehensively evaluated, thereby providing a quantitative basis and reference indicators for formulating quality control strategies and improving product quality. Finally, the quality grade of the textile fabric surface comprehensive quality score is monitored and analyzed according to the preset textile fabric quality grade judgment standard. The key to this step is to convert the comprehensive quality score into a specific quality grade assessment according to the pre-set quality standards and rating system. Through quality grade monitoring and analysis, the quality problems and improvement space of the fabric can be discovered and identified in time, helping producers and quality managers to formulate corresponding quality improvement strategies and control measures, thereby improving overall production efficiency and product competitiveness.
[0142] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.
[0143] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may 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 the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.
Claims
1. A textile cloth intelligent monitoring and analysis method based on textile and clothing intelligent manufacturing, characterized in that: The following steps are involved: Step S1: Scanning the textile cloth image in the textile and clothing intelligent manufacturing process by using a high-resolution image acquisition device to obtain a surface scan image of the textile cloth; Performing grayscale conversion processing on the scanned image of the textile surface to obtain a grayscale image of the textile surface structure; Step S2: Performing structural texture feature analysis on the grayscale image of the textile surface structure to obtain the textile surface structure texture feature data; performing structural defect quality impact analysis on the grayscale image of the textile surface structure based on the textile surface structure texture feature data to obtain the textile surface structure defect quality impact factor; wherein step S2 includes the following steps: Step S21: performing surface structure contour recognition analysis on the grayscale image of the textile surface structure to obtain the edge contour range of the textile surface structure; Step S22: performing structural edge region segmentation on the grayscale image of the surface structure of the textile cloth based on the edge contour range of the surface structure of the textile cloth, so as to obtain a segmented image of the surface structure of the textile cloth; Step S23: performing structural texture feature analysis on the textile surface structural region segmentation image to obtain textile surface structural texture feature data; Step S24: performing surface structure defect evaluation calculation on the segmented image of the surface structure region of the textile cloth based on the surface structure texture feature data of the textile cloth to obtain the degree of surface structure defect of the textile cloth; wherein step S24 includes the following steps: Step S241: performing surface structure defect detection on the segmented image of the surface structure region of the textile cloth based on the surface structure texture feature data of the textile cloth to obtain image defects of the surface structure region of the textile cloth; Step S242: locating the defect position coordinates of the regional image of the textile surface structure defect to obtain the regional image spatial position coordinates of each textile surface structure defect; Step S243: based on the spatial position coordinates of the regional images of the surface structural defects of each textile cloth, the defect region connection marking is performed on the textile cloth surface structural regional segmentation image to obtain a textile cloth surface structural defect distribution region marking image; Step S244: performing defect distribution density analysis on the marked image of the textile fabric surface structural defect distribution area to obtain the textile fabric surface structural defect distribution density; Step S245: Based on the distribution density of the surface structural defects of the textile fabric, the surface structural defect degree calculation formula is used to perform surface structural defect evaluation calculation on the marked image of the surface structural defect distribution area of the textile fabric to obtain the surface structural defect degree of the textile fabric; wherein the surface structural defect degree calculation formula is specifically: ; ; In the formula, It is the degree of surface structural defects of textile fabrics. The total size of the area of the image marked with the distribution area of the surface structural defects of the textile cloth, is the distribution density of structural defects on the surface of textile fabrics, is the horizontal coordinate parameter of the distribution area of the surface structural defects of the textile fabric, is the ordinate parameter of the distribution area of the surface structural defects of textile fabrics, is the weight coefficient of surface structural defects, Mark the distribution area of the textile fabric surface structure defect. The distribution area of the textile fabric surface structure defect in the image is at the position The gray value at is the average gray value of the textile surface, is the gray value influence coefficient, Mark the distribution area of the textile fabric surface structure defect. The distribution area of the textile fabric surface structure defect in the image is at the position The defect distribution density energy potential function at It is the correction factor for the degree of surface structural defects of textile fabrics; Step S25: analyzing the influence of structural defects on the grayscale image of the surface structure of the textile cloth based on the degree of the surface structural defects of the textile cloth, and obtaining the quality influence factor of the surface structural defects of the textile cloth; wherein step S25 includes the following steps: Step S251: performing structural feature analysis on the grayscale image of the textile surface structure to obtain textile surface structure feature data; Step S252: performing a structural quality impact assessment analysis on the grayscale image of the textile surface structure based on the textile surface structure feature data to obtain a data set of factors affecting the textile surface structure quality; Step S253: performing defect impact regression analysis on the data set of factors affecting the surface structure of textile fabrics based on the degree of surface structure defects of the textile fabrics, so as to obtain the degree of influence correlation between the surface structure defects and the factors affecting the quality; Step S254: performing defect influencing factor identification analysis on the textile fabric surface structure quality influencing factor data set according to the influence correlation between the surface structure defects and the quality influencing factors, and obtaining the textile fabric surface structure defect quality influencing factors; Step S3: quantifying and calculating the surface color difference defects of the textile cloth surface scanned image to obtain the textile cloth surface color difference defect degree coefficient; performing color difference defect quality impact analysis on the textile cloth surface scanned image based on the textile cloth surface color difference defect degree coefficient to obtain the textile cloth surface color difference defect quality impact factor; wherein step S3 includes the following steps: Step S31: performing color space conversion on the scanned image of the textile cloth surface to obtain the LAB color space of the textile cloth surface image; Step S32: performing pixel point area division on the LAB color space of the textile cloth surface image to obtain pixel point color space sub-regions of each textile cloth image; Step S33: Calculate the pixel chromaticity of each color space sub-region of the textile fabric image pixel point to obtain the pixel chromaticity value of each color sub-region of the textile fabric pixel point; Step S34: performing a quantitative calculation of the surface color difference defects of the textile cloth surface scanned image based on the color space sub-regions of each textile cloth image pixel point and the pixel chromaticity value of each textile cloth pixel point color sub-region to obtain the textile cloth surface color difference defect degree coefficient; wherein step S34 includes the following steps: Step S341: performing regional position positioning on the color space sub-region of each textile fabric image pixel point to obtain the color space position coordinates of the color sub-region of each textile fabric pixel point; Step S342: performing pixel chromaticity vector conversion on the color space position coordinates of the color sub-region of each textile fabric pixel and the pixel chromaticity value to obtain the chromaticity space distribution vector of each textile fabric pixel; Step S343: performing vector space distance calculation on the chromaticity space distribution vector of each textile fabric pixel point to obtain the vector space distance between each textile fabric pixel point distribution vector; performing color deviation statistical calculation on the chromaticity space distribution vector of each textile fabric pixel point based on the vector space distance between each textile fabric pixel point distribution vector to obtain the pixel color difference value between the color sub-regions of each textile fabric pixel point; Step S344: performing pixel color difference defect area detection on the surface scanned image of the textile cloth based on the pixel color difference values between the color sub-areas of each pixel point of the textile cloth, and obtaining the pixel color difference defect area of the surface image of the textile cloth; Step S345: performing area statistics on the color difference defect area of the textile surface image pixels to obtain the area size of the color difference defect area on the textile surface; Step S346: quantifying the surface color difference defects of the textile surface scanned image using a surface color difference defect degree calculation formula based on the size of the surface color difference defect area of the textile cloth, so as to obtain the surface color difference defect degree coefficient of the textile cloth; wherein the surface color difference defect degree calculation formula is specifically: ; In the formula, is the color difference defect degree coefficient of the textile cloth surface, is the total number of color difference defect areas on the textile surface in the textile surface scan image. is the index parameter of the color difference defect area on the surface of textile cloth, For the The area of the color difference defect area on the surface of the textile cloth, is the spatial coordinate range parameter of the color difference defect area on the surface of textile cloth, is the spatial coordinate variable parameter of the color difference defect area on the surface of the textile cloth, For the The color difference defect area on the surface of the textile cloth is located in the spatial coordinates The pixel color change value at For the The weight coefficient of the color difference defect area on the surface of the textile cloth, For the The pixel color change defect impact index of the color difference defect area on the surface of the textile cloth is It is the correction value of the color difference defect degree coefficient of the textile cloth surface; Step S35: performing a color difference defect quality impact analysis on the scanned image of the textile surface based on the color difference defect degree coefficient of the textile surface, and obtaining a color difference defect quality impact factor of the textile surface; Step S4: Based on the quality influencing factors of textile surface structural defects and the quality influencing factors of textile surface color difference defects, the quality grade of the corresponding textile fabrics in the textile clothing intelligent manufacturing process is monitored and analyzed to obtain the quality grade monitoring results of the textile fabrics; wherein step S4 includes the following steps: Step S41: Based on the quality influencing factors of textile surface structural defects and the quality influencing factors of textile surface color difference defects, an important evaluation analysis of the quality of textile fabrics corresponding to the textile fabrics in the intelligent manufacturing process of textile garments is performed to obtain the importance of the fabric quality influence of each structural defect influence factor and color difference defect influence factor; Step S42: assigning defect quality impact weights to the textile fabric surface structural defect quality impact factors based on the cloth quality impact importance of each structural defect impact factor, and obtaining the cloth quality impact weights of each structural defect impact factor; Step S43: allocating color difference quality impact weights to the color difference quality impact factors on the surface of textile cloth based on the cloth quality impact importance of each color difference defect impact factor, and obtaining the cloth quality impact weights of each color difference defect impact factor; Step S44: performing cloth quality weighted average calculation on the textile cloth surface structure defect quality influence factor and the textile cloth surface color difference defect quality influence factor based on the cloth quality influence weights of each structure defect influence factor and the color difference defect influence factor to obtain a comprehensive textile cloth surface quality score; Step S45: Performing quality grade monitoring and analysis on the comprehensive quality score value of the textile cloth surface according to the preset textile cloth quality grade judgment standard to obtain the textile cloth quality grade monitoring result.
2. The intelligent monitoring and analysis method for textile fabrics based on intelligent manufacturing of textile garments according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Designing the collection angle and illumination conditions for the textile and clothing intelligent manufacturing process to obtain the collection angle conditions and illumination conditions for the textile and clothing raw material cloth; Step S12: Scanning the textile cloth image of the textile and clothing intelligent manufacturing process using a high-resolution image acquisition device based on the collection angle conditions and lighting conditions of the textile and clothing raw material cloth to obtain a textile cloth surface scan image; Step S13: performing image denoising processing on the scanned image of the textile cloth surface to obtain a denoised image of the textile cloth surface; Step S14: performing edge contrast enhancement processing on the denoised image of the textile cloth surface to obtain an edge enhanced optimized image of the textile cloth surface; Step S15: grayscale conversion processing is performed on the textile cloth surface edge enhancement optimization image to obtain a textile cloth surface structure grayscale image.
