Quality detection system for lining cloth production

Abstract

The invention discloses a quality detection system for lining cloth production, and relates to the technical field of lining cloth production, and the quality detection system for lining cloth production comprises a data acquisition module, a wrinkle identification and division module, a wrinkle positioning and equipment labeling module, an operation parameter adjustment and optimization module and a quality evaluation and report output module. According to the method, the positions of the wrinkles are matched with the positions of the production equipment, so that the responsibility equipment can be quickly positioned, the problem can be quickly responded and handled, the interference of wrong operation on irrelevant equipment is effectively avoided, the smooth operation of a production line is guaranteed, the response speed and the problem handling efficiency are improved, the production process is optimized, and the production cost is reduced. And unnecessary downtime and maintenance cost are reduced.

Description

Technical Field

[0001] The invention relates to the technical field of lining cloth production, and in particular to a quality inspection system for lining cloth production. Background Art

[0002] Interlining is a special auxiliary material that uses woven fabrics, knitted fabrics and non-woven fabrics as base fabrics and is processed by special machinery. It is used for the inner layer of clothing, shoes and hats, home textiles and car interiors to play the role of reinforcement, stiffness, shaping and shape retention. In clothing production, interlining is often used in the inner layer of various parts of clothing, such as the front piece, inner edge, collar, lapel, back piece, pocket opening, pocket cover, belt, etc., to enhance the stiffness, three-dimensional sense, drape and comfort of clothing. It also has the function of increasing the thickness, fullness and thermal insulation of clothing. Therefore, interlining is often compared to the "skeleton" and "essence" of clothing, and its importance is self-evident.

[0003] As an internal auxiliary material for clothing, shoes, hats, home textiles and other products, the quality of interlining is undoubtedly crucial. Quality testing uses scientific methods to comprehensively evaluate key performance indicators such as dimensional stability, strength, wear resistance, and wrinkle resistance of interlining to ensure that each product can meet or even exceed the established quality standards. Through quality testing, potential problems in the interlining production process can be discovered in a timely manner, so that corrective measures can be taken in a timely manner to avoid problems accumulating and affecting the quality of the final product. This not only improves the reliability and durability of the product, but also greatly enhances consumers' confidence and satisfaction with the product.

[0004] Quality inspection is not only an inspection of finished products, but also a supervision of the production process. However, the traditional quality inspection system for interlining production is often limited to a simple assessment of the degree of wrinkles in the finished product, and is unable to deeply analyze and accurately identify the specific equipment or process links that cause wrinkles, thus hindering the optimal allocation of production resources and the refined management of the production process. It not only limits the further improvement of production efficiency, but may also cause unnecessary waste of resources and additional cost increases, ultimately affecting the overall quality stability and market competitiveness of interlining products.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In view of the problems in the related art, the present invention proposes a quality inspection system for lining cloth production to overcome the above technical problems existing in the existing related art.

[0007] To this end, the specific technical solution adopted by the present invention is as follows: A quality inspection system for lining cloth production, the quality inspection system for lining cloth production comprising: a data acquisition module, a wrinkle identification and classification module, a wrinkle positioning and equipment marking module, an operation parameter adjustment and optimization module and a quality assessment and report output module; The data acquisition module, the wrinkle identification and division module, the wrinkle positioning and equipment marking module, the operation parameter adjustment and optimization module and the quality assessment and report output module are connected in sequence; Data acquisition module, used to collect image data, operating parameters and position data of production equipment on the lining production line in real time; The wrinkle recognition and classification module is used to identify wrinkle features in image data using edge detection algorithms, and classify the identified wrinkle features into different levels in combination with preset wrinkle level evaluation criteria; The wrinkle positioning and equipment marking module is used to analyze the wrinkle characteristics after grade classification, extract the location data of the wrinkles, match the location data of the wrinkles with the location data of the production equipment, and mark the target production equipment; The operating parameter adjustment and optimization module is used to adjust and optimize the operating parameters of the target production equipment according to the wrinkle characteristics after grade classification; The quality assessment and report output module is used to assess the quality of the lining cloth after the operating parameters are adjusted according to the lining cloth quality standard, and output the lining cloth quality assessment report according to the assessment results; Among them, the wrinkle positioning and equipment marking module includes: wrinkle position extraction module, spatial feature extraction module, distance data acquisition module and target equipment selection and marking module; A fold position extraction module is used to extract fold position data from the fold features after hierarchical division using deep learning technology; A spatial feature extraction module is used to process the collected production equipment location data and extract the spatial features of the production equipment; A distance data acquisition module, used to calculate the shortest path between the wrinkle position data and the production equipment position data using an improved shortest path algorithm; The target equipment selection and labeling module is used to integrate the shortest path and the operating parameters of the production equipment, and use the target optimization algorithm to select and label the target production equipment.

[0008] Furthermore, the wrinkle recognition and division module includes: a preprocessing module, a wrinkle recognition module, a wrinkle matching module and a wrinkle division module; A preprocessing module, used for preprocessing the collected image data; A wrinkle recognition module is used to analyze the preprocessed image data using an edge detection algorithm, extract edge features in the image data, and recognize wrinkle features in the image data based on the edge features; The fold matching module is used to analyze the fold characteristics, obtain the fold parameters, and calculate the matching degree between the fold parameters and the fold grade evaluation standard in combination with the preset fold grade evaluation standard; The fold division module is used to divide the fold features into different levels based on the matching degree calculation results.

[0009] Furthermore, the distance data acquisition module includes: a production equipment map construction module, a wrinkle position definition module and a path calculation module; A production equipment graph construction module is used to construct a production equipment graph model by taking the spatial features of production equipment as nodes and the associations between production equipment as edges between nodes; A wrinkle position definition module is used to select nodes representing wrinkle positions in the production equipment graph model according to the extracted wrinkle position data; The path calculation module is used to calculate the shortest path from the fold position node to the production equipment node using an improved shortest path algorithm.

[0010] Furthermore, the spatial features of the production equipment are used as nodes, and the associations between the production equipment are used as edges between the nodes. The production equipment graph model is constructed including: Taking the spatial characteristics of production equipment as nodes, a node set is established; Analyze the correlation between production equipment, create edges between corresponding production equipment, and establish edge sets; According to the correlation strength of production equipment in the production process, determine the direction of each edge and calculate the weight of each edge; A production equipment graph model is constructed based on node sets, edge sets, edge directions and edge weights.

[0011] Furthermore, the edge weight calculation formula is: In the formula, w represents the weight of the edge; d(x i ，x j ) Representation Node i With Node j The distance between K Represents a constant.

[0012] Furthermore, using the improved shortest path algorithm, the shortest path from the wrinkle location node to the production equipment node is calculated including: The wrinkle position node is used as the source node, and the source node is added to the preset distance queue; Select node N from the distance queue and check whether node N is a production equipment node; If node N is a production equipment node, the path from the source node to node N is taken as the shortest path, the remaining nodes in the distance queue are released, and the shortest path from the fold position node to the production equipment node is output; If node N is not a production equipment node, the adjacent nodes of node N are processed until the distance queue is empty, and the shortest path from the wrinkle position node to the production equipment node is output.

[0013] Further, if node N is not a production equipment node, the adjacent nodes of node N are processed until the distance queue is empty, and the shortest path from the wrinkle position node to the production equipment node is output, including the following steps: S11. If node N is not a production equipment node, traverse the adjacent nodes of node N; S12, based on each adjacent node, if the adjacent node has been out of the queue, then the adjacent node is not processed; if the adjacent node is already in the queue, then the position of the adjacent node in the queue is adjusted according to the shortest path from the source node to the node N; if the adjacent node is not in the queue, then the adjacent node is added to the queue; S13, based on the processed node N, if the number of paths of the node N is zero, remove the node N from the distance queue; Repeat steps S11 to S13 until the distance queue is empty, and output the shortest path from the wrinkle location node to the production equipment node.

[0014] Further, the target equipment selection and marking module includes: a data integration module, a target production equipment selection module and a target production equipment marking module; Data integration module, used to integrate distance data with the operating parameters of production equipment to obtain production equipment configuration data; The target production equipment selection module is used to select the target production equipment according to the production equipment configuration data using the target optimization algorithm; The target production equipment marking module is used to mark the selected target production equipment using marking tools.

[0015] Furthermore, selecting target production equipment using a target optimization algorithm according to the production equipment configuration data includes the following steps: S21, setting the type and quantity of production equipment configuration data, and randomly generating an initial production equipment configuration according to the type and data; S22. According to the initial production equipment configuration, calculate the performance index of each initial production equipment configuration, and use the performance index as the fitness value of the initial production equipment configuration; S23, selecting the initial production equipment configuration with the highest fitness, and using the overall hybridization method to randomly screen several initial production equipment configurations for crossover operation to generate a new production equipment configuration; S24. Based on the new production equipment configuration, a simulated annealing algorithm is used to perform a local search to obtain a local optimal production equipment configuration; S25, iteratively generating the next generation of production equipment configuration according to the initial production equipment configuration with the highest fitness and the local optimal production equipment configuration; Repeat steps S23 to S25 until a preset number of iterations is reached, and output the production equipment configuration with the highest fitness in the final generation as the target production equipment.

[0016] Furthermore, the performance index calculation formula of the initial production equipment configuration is: In the formula, F´(n) Represents the initial production equipment configuration n performance index value; λ represents weight; ▽ f(n) Represents the initial production equipment configuration function f(n) Initial production equipment configuration n The gradient of ▽ f min A gradient representing the minimum initial production equipment configuration; ▽ f max represents the gradient of the highest initial production equipment configuration; F(n) Represents the initial production equipment configuration n The original performance index value of F min Represents the performance indicator value of the minimum initial production equipment configuration; f max Represents the performance indicator value of the highest initial production equipment configuration; ||·|| represents the norm.

[0017] The beneficial effects of the present invention are: 1. The present invention can quickly locate the responsible equipment by matching the wrinkle position with the production equipment position, facilitate rapid response and problem handling, effectively avoid the interference of erroneous operation on irrelevant equipment, ensure the smooth operation of the production line, not only improve the response speed and efficiency of problem handling, but also optimize the production process and reduce unnecessary downtime and maintenance costs.

[0018] 2. The present invention can adjust and optimize the operating parameters of related equipment for different levels of wrinkles by dividing the wrinkles, ensuring a smooth production process and improved product quality, thereby effectively reducing the scrap rate and reducing resource waste, while improving the market competitiveness of the product, thereby improving product quality, production efficiency and economic benefits.

[0019] 3. The present invention can reduce the occurrence of defects such as wrinkles by optimizing operating parameters, which not only improves the appearance quality and overall quality of the product, but also ensures the uniformity and stability of the material during the processing, thereby effectively avoiding the occurrence of undesirable phenomena such as wrinkles and distortions. It not only improves the market competitiveness of the product, but also reduces the scrap rate and rework costs, bringing significant economic benefits to the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 The invention is a block diagram of a quality inspection system for lining cloth production according to an embodiment of the invention.

[0022] In the figure: 1. Data acquisition module; 2. Wrinkle identification and division module; 3. Wrinkle positioning and equipment marking module; 4. Operation parameter adjustment and optimization module; 5. Quality assessment and report output module. DETAILED DESCRIPTION

[0023] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments. They can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. By referring to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention.

[0024] According to an embodiment of the present invention, a quality inspection system for lining cloth production is provided.

[0025] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the quality inspection system for lining cloth production of the embodiment of the present invention, the quality inspection system for lining cloth production comprises: a data acquisition module 1, a wrinkle identification and classification module 2, a wrinkle positioning and equipment marking module 3, an operation parameter adjustment and optimization module 4 and a quality assessment and report output module 5; The data acquisition module 1, the wrinkle identification and division module 2, the wrinkle positioning and equipment marking module 3, the operation parameter adjustment and optimization module 4 and the quality assessment and report output module 5 are connected in sequence; The data acquisition module 1 is used to collect image data, operating parameters and position data of the lining cloth production line in real time.

[0026] It should be supplemented that the type of data collected should be clearly defined, including image data, operating parameters of production equipment and location data; according to data collection requirements, select appropriate image sensors, temperature sensors, pressure sensors and other equipment and encoders, GPS and other equipment for obtaining location information; install sensors and equipment on the lining production line and related production equipment, and debug them to ensure that the equipment can work normally; use sensors and equipment to collect image data on the lining production line, operating parameters of production equipment and location data, and provide strong support for production process quality inspection.

[0027] In addition, the operating parameters of the production equipment include temperature parameters, humidity parameters, speed parameters, pressure parameters and processing parameters; among them, temperature parameters include baking temperature, heat setting temperature, etc. These parameters affect the shrinkage and expansion of the lining cloth, and thus affect the formation of wrinkles; humidity parameters include dryness, steam pressure, etc. These parameters affect the moisture content and softness of the lining cloth, and also affect wrinkles; speed parameters include the running speed and air supply speed of the lining cloth, etc. These parameters affect the moving speed and air flow of the lining cloth on the production line, and may indirectly affect the formation of wrinkles; pressure parameters include rolling pressure, tension, etc. These parameters directly affect the deformation and folding of the lining cloth, and are important factors in the formation of wrinkles; processing parameters include folding method, shear angle, etc. These parameters affect the final shape and structure of the lining cloth, and also affect the wrinkle characteristics.

[0028] The wrinkle recognition and classification module 2 is used to recognize wrinkle features in image data using an edge detection algorithm, and classify the recognized wrinkle features into different levels in combination with a preset wrinkle level evaluation standard.

[0029] Specifically, the wrinkle recognition and division module 2 includes: a preprocessing module, a wrinkle recognition module, a wrinkle matching module and a wrinkle division module; The preprocessing module is used to preprocess the collected image data.

[0030] It should be noted that the preprocessing of the collected image data includes: converting the color image in the image data into a grayscale image to reduce the complexity brought by the color information; using filters (such as Gaussian filters, median filters, etc.) to remove noise in the grayscale image; adjusting the contrast of the denoised image to make the histogram distribution of the denoised image more uniform; using sharpening filters (such as high-pass filters, Laplace operators, etc.) to process the contrast-processed image to increase the edge information of the image; adjusting the edge-enhanced image to a size suitable for processing, and normalizing the pixel values ​​of the image to a certain range (such as 0 to 1 or -1 to 1) for subsequent processing.

[0031] The wrinkle recognition module is used to analyze the preprocessed image data using an edge detection algorithm, extract edge features in the image data, and recognize wrinkle features in the image data based on the edge features.

[0032] It should be noted that a suitable edge detection algorithm, such as Canny edge detection, is selected according to the requirements, and the preprocessed image data is input into the Canny edge detection algorithm; first, the image is Gaussian filtered to smooth the image and reduce noise; then, the gradient strength and direction of each pixel in the image are calculated; next, the non-maximum suppression technique is applied to refine the edge, wherein the non-maximum suppression compares the gradient strength of each pixel with the adjacent pixels in its gradient direction, and only retains the pixels with the largest local gradient strength as edge points; finally, the double threshold method is used to determine the final edge, including setting two thresholds (high threshold and low threshold), determining the pixels with gradient strength higher than the high threshold as edge points, and excluding the pixels with gradient strength lower than the low threshold from the edge, and for the pixels with gradient strength between the two thresholds, if they are connected to the pixels that have been determined as edges, they are regarded as edge points; according to the results of the Canny edge detection, the edge features in the image are extracted, and in wrinkle recognition, the edge features usually correspond to the contours or boundaries of the wrinkles.

[0033] The fold matching module is used to analyze the fold characteristics, obtain the fold parameters, and calculate the matching degree between the fold parameters and the fold grade evaluation standard in combination with the preset fold grade evaluation standard.

[0034] It should be noted that machine learning algorithms (such as convolutional neural networks (CNN), random forests (RF), etc.) are used to analyze and learn the wrinkle features, and wrinkle parameters such as the length, width, shape index and distribution density of the wrinkles are automatically extracted; based on the wrinkle parameters, a wrinkle parameter list is established, and each parameter corresponds to a specific value or range; according to actual needs and application scenarios, a wrinkle grade evaluation standard is preset, and the preset wrinkle grade evaluation standard is usually a table or function set containing multiple grades and corresponding parameter thresholds, which is used to describe the characteristic range of wrinkles of different grades; the extracted wrinkle parameters are compared with the preset wrinkle grade evaluation standard, the wrinkle parameters are mapped to the corresponding parameters in the evaluation standard, and it is calculated whether they meet the threshold conditions of a certain grade; for each grade, the matching degree between the wrinkle parameters and the threshold of the grade is calculated, which is achieved by calculating the distance or angle difference between the parameter value and the threshold, such as the Euclidean distance, cosine similarity, etc., which calculates the distance or angle difference between the parameter value and the threshold.

[0035] In addition, the preset wrinkle grade evaluation criteria include: (1) Grade definition: Wrinkles are classified into different grades, such as mild, moderate, severe, etc., based on the severity, size, and distribution of the wrinkles.

[0036] (2) Parameter thresholds: Each level has one or more sets of parameter thresholds used to quantify wrinkle characteristics. These parameters may include the area, length, width, depth, edge roughness, etc. of the wrinkles.

[0037] For example, a simple wrinkle rating scale might look like Table 1: Table 1 Wrinkle grade evaluation standard table In the table, X, Y, and Z are predefined numerical thresholds. When the detected wrinkle feature meets the parameter threshold of a certain level, the wrinkle is classified into the corresponding level.

[0038] The fold division module is used to divide the fold features into different levels based on the matching degree calculation results.

[0039] The wrinkle positioning and equipment marking module 3 is used to analyze the wrinkle characteristics after grade classification, extract the location data of the wrinkles, match the location data of the wrinkles with the location data of the production equipment, and mark the target production equipment.

[0040] Among them, the wrinkle positioning and equipment marking module 3 includes: a wrinkle position extraction module, a spatial feature extraction module, a distance data acquisition module and a target equipment selection and marking module; The fold position extraction module is used to extract the fold position data from the fold features after hierarchical division using deep learning technology.

[0041] It should be noted that, based on the wrinkle features after hierarchical division, a data set is constructed and divided into a training set and a test set; according to the task requirements, a suitable deep learning model is selected, such as Faster R-CNN, YOLO, Mask R-CNN and other models for target detection, or models such as OpenPose, Hourglass and other models for key point detection are selected, and deep learning technology such as Faster R-CNN model is used to extract the position data of wrinkles from the wrinkle features after hierarchical division; the Faster R-CNN model is trained with the training set to learn and identify the wrinkle features in the image and predict the position data of wrinkles; the test set is used to evaluate the trained Faster R-CNN model to verify its performance and obtain the final Faster R-CNN model; the wrinkle features after hierarchical division are input into the final Faster R-CNN model, and the position data of wrinkles, such as the bounding box coordinates, key point positions, etc., are output.

[0042] The spatial feature extraction module is used to process the collected production equipment location data and extract the spatial features of the production equipment.

[0043] It should be noted that the collected production equipment location data is processed, including removing duplicate data, outliers (such as location coordinates beyond a reasonable range) and missing values, converting the location data into a unified format, and performing coordinate transformation on the production equipment location data.

[0044] The spatial characteristics of production equipment include location characteristics, distance characteristics, direction characteristics and distribution characteristics; based on the processed production equipment location data, read the plane coordinates of the production equipment; calculate the straight-line distance between production equipment or the distance based on the actual path; calculate the relative direction or angle based on the production equipment location data; use cluster analysis, kernel density estimation and other methods to analyze the spatial distribution pattern of production equipment.

[0045] The distance data acquisition module is used to calculate the shortest path between the wrinkle position data and the production equipment position data by using an improved shortest path algorithm.

[0046] Specifically, the distance data acquisition module includes: a production equipment map construction module, a wrinkle position definition module and a path calculation module; The production equipment graph construction module is used to construct a production equipment graph model by taking the spatial features of the production equipment as nodes and the associations between the production equipment as edges between the nodes.

[0047] Specifically, the spatial features of the production equipment are used as nodes, and the associations between the production equipment are used as edges between the nodes. The construction of the production equipment graph model includes: The spatial characteristics of production equipment are used as nodes to establish a node set.

[0048] Analyze the association between production equipment, create edges between corresponding production equipment, and establish edge sets.

[0049] According to the association strength of the production equipment in the production process, the direction of each edge is determined and the weight of each edge is calculated.

[0050] Specifically, the edge weight calculation formula is: In the formula, w represents the weight of the edge; d(x i ，x j ) Representation Node i With Node j The distance between K Represents a constant.

[0051] A production equipment graph model is constructed based on node sets, edge sets, edge directions and edge weights.

[0052] The wrinkle position definition module is used to select nodes representing wrinkle positions in the production equipment diagram model according to the extracted wrinkle position data.

[0053] It should be noted that for the position data of each fold (such as the center point coordinates), a new node is created in the production equipment graph model; the position attribute of the new node is set to the corresponding fold position data (such as the center point coordinates); if necessary, additional attributes can be set for the node based on other characteristics of the fold (such as size, shape, and grade).

[0054] The path calculation module is used to calculate the shortest path from the fold position node to the production equipment node using an improved shortest path algorithm.

[0055] Specifically, using the improved shortest path algorithm, calculating the shortest path from the fold location node to the production equipment node includes: The wrinkle position node is used as the source node, and the source node is added to the preset distance queue; Select node N from the distance queue and check whether node N is a production equipment node; If node N is a production equipment node, the path from the source node to node N is taken as the shortest path, the remaining nodes in the distance queue are released, and the shortest path from the fold position node to the production equipment node is output; If node N is not a production equipment node, the adjacent nodes of node N are processed until the distance queue is empty, and the shortest path from the wrinkle position node to the production equipment node is output.

[0056] Specifically, if node N is not a production equipment node, the adjacent nodes of node N are processed until the distance queue is empty, and the shortest path from the wrinkle position node to the production equipment node is output, including the following steps: S11. If node N is not a production equipment node, traverse the adjacent nodes of node N; S12, based on each adjacent node, if the adjacent node has been out of the queue, then the adjacent node is not processed; if the adjacent node is already in the queue, then the position of the adjacent node in the queue is adjusted according to the shortest path from the source node to the node N; if the adjacent node is not in the queue, then the adjacent node is added to the queue; S13, based on the processed node N, if the number of paths of the node N is zero, remove the node N from the distance queue; Repeat steps S11 to S13 until the distance queue is empty, and output the shortest path from the wrinkle location node to the production equipment node.

[0057] It should be noted that the distance queue is a priority queue, so that in the process of calculating the shortest path, the distance queue is evenly spread from the source node to the surrounding nodes; therefore, the closer the target node is to the source node, the shorter the time to calculate the shortest path.

[0058] The target equipment selection and labeling module is used to integrate the shortest path and the operating parameters of the production equipment, and use the target optimization algorithm to select and label the target production equipment.

[0059] Specifically, the target equipment selection and marking module includes: a data integration module, a target production equipment selection module and a target production equipment marking module; The data integration module is used to integrate the distance data with the operating parameters of the production equipment to obtain the production equipment configuration data.

[0060] The target production equipment selection module is used to select the target production equipment according to the production equipment configuration data using the target optimization algorithm.

[0061] Specifically, according to the production equipment configuration data, selecting the target production equipment using the target optimization algorithm includes the following steps: S21. Set the type and quantity of production equipment configuration data, and randomly generate an initial production equipment configuration according to the type and data.

[0062] S22. According to the initial production equipment configuration, a performance index of each initial production equipment configuration is calculated, and the performance index is used as the fitness value of the initial production equipment configuration.

[0063] Specifically, the performance index calculation formula for the initial production equipment configuration is: In the formula, F´(n) Represents the initial production equipment configuration n performance index value; λ represents weight; ▽ f(n) Represents the initial production equipment configuration function f(n) Initial production equipment configuration n The gradient of ▽ f min A gradient representing the minimum initial production equipment configuration; ▽ f max represents the gradient of the highest initial production equipment configuration; F(n) Represents the initial production equipment configuration n The original performance index value of F min Represents the performance indicator value of the minimum initial production equipment configuration; f max Represents the performance indicator value of the highest initial production equipment configuration; ||·|| represents the norm.

[0064] S23. Select the initial production equipment configuration with the highest fitness, and use the overall hybridization method to randomly screen several initial production equipment configurations for crossover operation to generate a new production equipment configuration.

[0065] S24. Based on the new production equipment configuration, a simulated annealing algorithm is used to perform a local search to obtain the local optimal production equipment configuration.

[0066] S25. Iteratively generate the next generation of production equipment configuration based on the initial production equipment configuration with the highest fitness and the local optimal production equipment configuration.

[0067] Repeat steps S23 to S25 until a preset number of iterations is reached, and output the production equipment configuration with the highest fitness in the final generation as the target production equipment.

[0068] The target production equipment marking module is used to mark the selected target production equipment using marking tools.

[0069] It should be noted that, based on the selected target equipment, the operating parameters and location data of the production equipment are analyzed, the operating parameters and location data of the target equipment are obtained, and a target folder is established; a labeling tool such as LabelImg is used to label the selected target production equipment, including: First, in the annotation interface of LabelImg, click the "Open Dir" button and select the target folder containing the target production equipment images. Then, LabelImg will automatically load all the images in the folder for the user to annotate. Next, on the loaded image, use the mouse to drag to create an accurate annotation box for the target production equipment. You can draw a rectangular box or other annotation box that fits the target shape to ensure that the annotation box tightly and accurately surrounds the target production equipment. After the annotation box is created, enter the category name of the target production equipment in the corresponding input box for identification in subsequent processing. If necessary, you can adjust the size, position and angle of the annotation box according to actual conditions to further improve the accuracy of the annotation. After completing all the annotation work, click the "Save" button to manually save the annotation results, or use LabelImg's auto-save function to automatically save. The annotation results will be saved as a file corresponding to the selected annotation format (for example, an XML file in Pascal VOC format), which contains information such as the precise position and size of the annotation box and the category name of the target production equipment.

[0070] The operating parameter adjustment and optimization module 4 is used to adjust and optimize the operating parameters of the target production equipment according to the wrinkle characteristics after grade classification.

[0071] It should be noted that a machine learning model, such as a regression analysis model, is used to analyze the relationship between the operating parameters of the target production equipment and the wrinkle feature level; historical data sets are collected and organized, and the regression analysis model is trained and verified using the historical data sets to obtain the final regression analysis model; the currently detected wrinkle feature level is input into the final regression analysis model, and the predicted values ​​of the operating parameters of the target production equipment that cause the wrinkles are output; the predicted values ​​of the operating parameters of the target production equipment, such as speed, temperature and pressure distribution, are analyzed, and a parameter adjustment strategy is formulated, including adjusting speed, changing temperature settings, optimizing pressure distribution, etc.; the operating parameters of the target production equipment are adjusted according to the formulated strategy; during the adjustment process, the operating status and product quality of the equipment are monitored in real time to ensure that the adjustment effect meets expectations; if the initial adjustment fails to achieve the expected effect, iterative optimization is performed based on the new monitoring data using the predictive ability of the machine learning model until the best combination of operating parameters is found.

[0072] Based on historical data and machine learning algorithms, a relationship model between operating parameters and wrinkle levels is established; based on the results of the model output, the operating parameters of the target production equipment are adjusted; The quality evaluation and report output module 5 is used to evaluate the quality of the lining cloth after the operation parameters are adjusted according to the lining cloth quality standard, and output the lining cloth quality evaluation report according to the evaluation result.

[0073] It should be noted that the interlining image data after the operating parameters are adjusted, the final operating parameters of the production equipment and the relevant production records should be collected from the production line in real time or regularly; the collected data should be integrated into a unified database to ensure that all data used for quality assessment are complete and consistent; according to the quality standards of the interlining, the specific indicators and acceptable range of the assessment should be determined; the integrated data should be compared and analyzed with the quality standards to assess whether the adjusted interlining meets the quality requirements; based on the assessment results, an interlining quality assessment report should be prepared, which should include the assessment purpose, assessment method, assessment results, problem analysis, improvement suggestions, etc.

[0074] In addition, interlining quality standards include: (1) Appearance quality: including color uniformity, surface flatness (i.e., degree of wrinkles), stains, damage, color difference, clarity and alignment of stripes or patterns, etc. In this system, the identification, grading, and location of wrinkles are directly related to the assessment of appearance quality.

[0075] (2) Physical properties: such as tensile strength, tear strength, abrasion resistance, dimensional stability (such as shrinkage), air permeability, softness, etc. These properties often need to be measured by professional testing equipment and methods.

[0076] (3) Chemical properties: including pH value, formaldehyde content, limits on harmful substances such as azo dyes, and whether the product has passed specific environmental or safety certifications (such as OEKO-TEX).

[0077] (4) Functional requirements: For interlinings used for specific purposes, there may be additional functional requirements, such as fire resistance, water resistance, antistatic properties, antibacterial properties, etc.

[0078] (5) Dimensions: including length, width, thickness, etc., which must meet the requirements of the design or order.

[0079] In summary, with the help of the above technical solution of the present invention, by matching the wrinkle position with the production equipment position, the responsible equipment can be quickly located, which is convenient for rapid response and problem handling, effectively avoiding the interference of erroneous operation on irrelevant equipment, and ensuring the smooth operation of the production line. It not only improves the response speed and the efficiency of problem handling, but also optimizes the production process and reduces unnecessary downtime and maintenance costs. By dividing the wrinkle levels, the operating parameters of the relevant equipment can be adjusted and optimized for wrinkles of different levels to ensure the smoothness of the production process and the improvement of product quality, thereby effectively reducing the scrap rate and reducing resource waste. At the same time, the market competitiveness of the product is improved, thereby improving product quality, production efficiency and economic benefits. By optimizing the operating parameters, the generation of defects such as wrinkles can be reduced, which not only improves the appearance quality and overall quality of the product, but also ensures the uniformity and stability of the material during the processing process, thereby effectively avoiding the occurrence of undesirable phenomena such as wrinkles and twisting, which not only improves the market competitiveness of the product, but also brings significant economic benefits to the enterprise by reducing the scrap rate and rework costs.

[0080] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A quality inspection system for lining cloth production, characterized in that: The quality inspection system for lining cloth production includes: data acquisition module, wrinkle identification and division module, wrinkle positioning and equipment marking module, operation parameter adjustment and optimization module and quality evaluation and report output module; The data acquisition module, the wrinkle identification and division module, the wrinkle positioning and equipment marking module, the operation parameter adjustment and optimization module and the quality assessment and report output module are connected in sequence; Data acquisition module, used to collect image data, operating parameters and location data of the lining production line in real time; The wrinkle recognition and classification module is used to identify wrinkle features in image data using edge detection algorithms, and classify the identified wrinkle features into different levels in combination with preset wrinkle level evaluation criteria; The wrinkle positioning and equipment marking module is used to analyze the wrinkle characteristics after grade classification, extract the location data of the wrinkles, match the location data of the wrinkles with the location data of the production equipment, and mark the target production equipment; The operating parameter adjustment and optimization module is used to adjust and optimize the operating parameters of the target production equipment according to the wrinkle characteristics after grade classification; The quality assessment and report output module is used to assess the quality of the lining cloth after the operating parameters are adjusted according to the lining cloth quality standard, and output the lining cloth quality assessment report according to the assessment results; Wherein, the wrinkle positioning and equipment marking module includes: a wrinkle position extraction module, a spatial feature extraction module, a distance data acquisition module and a target equipment selection and marking module; A fold position extraction module is used to extract fold position data from the fold features after hierarchical division using deep learning technology; A spatial feature extraction module is used to process the collected production equipment location data and extract the spatial features of the production equipment; A distance data acquisition module, used to calculate the shortest path between the wrinkle position data and the production equipment position data using an improved shortest path algorithm; The target equipment selection and labeling module is used to integrate the shortest path and the operating parameters of the production equipment, and use the target optimization algorithm to select and label the target production equipment.

2. A quality inspection system for lining cloth production according to claim 1, characterized in that: The fold recognition and division module includes: a preprocessing module, a fold recognition module, a fold matching module and a fold division module; A preprocessing module, used for preprocessing the collected image data; A wrinkle recognition module is used to analyze the preprocessed image data using an edge detection algorithm, extract edge features in the image data, and recognize wrinkle features in the image data based on the edge features; The fold matching module is used to analyze the fold characteristics, obtain the fold parameters, and calculate the matching degree between the fold parameters and the fold grade evaluation standard in combination with the preset fold grade evaluation standard; The fold division module is used to divide the fold features into different levels based on the matching degree calculation results.

3. A quality inspection system for lining cloth production according to claim 1, characterized in that: The distance data acquisition module includes: a production equipment map construction module, a wrinkle position definition module and a path calculation module; A production equipment graph construction module is used to construct a production equipment graph model by taking the spatial features of production equipment as nodes and the associations between production equipment as edges between nodes; A wrinkle position definition module is used to select nodes representing wrinkle positions in the production equipment diagram model according to the extracted wrinkle position data; The path calculation module is used to calculate the shortest path from the fold position node to the production equipment node using an improved shortest path algorithm.

4. A quality inspection system for lining cloth production according to claim 3, characterized in that: The spatial features of the production equipment are used as nodes, and the associations between the production equipment are used as edges between the nodes to construct the production equipment graph model, including: Taking the spatial characteristics of production equipment as nodes, a node set is established; Analyze the correlation between production equipment, create edges between corresponding production equipment, and establish edge sets; According to the correlation strength of production equipment in the production process, determine the direction of each edge and calculate the weight of each edge; A production equipment graph model is constructed based on node sets, edge sets, edge directions and edge weights.

5. A quality inspection system for lining cloth production according to claim 4, characterized in that: The edge weight calculation formula is: In the formula, w represents the weight of the edge; d(x i ，x j ) Representation Node i With Node j The distance between K Represents a constant.

6. A quality inspection system for lining cloth production according to claim 3, characterized in that: The method of calculating the shortest path from the wrinkle location node to the production equipment node by using the improved shortest path algorithm includes: The wrinkle position node is used as the source node, and the source node is added to the preset distance queue; Select node N from the distance queue and check whether node N is a production equipment node; If node N is a production equipment node, the path from the source node to node N is taken as the shortest path, the remaining nodes in the distance queue are released, and the shortest path from the fold position node to the production equipment node is output; If node N is not a production equipment node, the adjacent nodes of node N are processed until the distance queue is empty, and the shortest path from the wrinkle position node to the production equipment node is output.

7. A quality inspection system for lining cloth production according to claim 6, characterized in that: If the node N is not a production equipment node, the adjacent nodes of the node N are processed until the distance queue is empty, and the shortest path from the wrinkle position node to the production equipment node is output, including the following steps: S11. If node N is not a production equipment node, traverse the adjacent nodes of node N; S12, based on each adjacent node, if the adjacent node has been out of the queue, then the adjacent node is not processed; if the adjacent node is already in the queue, then the position of the adjacent node in the queue is adjusted according to the shortest path from the source node to the node N; if the adjacent node is not in the queue, then the adjacent node is added to the queue; S13, based on the processed node N, if the number of paths of the node N is zero, remove the node N from the distance queue; Repeat steps S11 to S13 until the distance queue is empty, and output the shortest path from the wrinkle location node to the production equipment node.

8. A quality inspection system for lining cloth production according to claim 1, characterized in that: The target equipment selection and marking module includes: a data integration module, a target production equipment selection module and a target production equipment marking module; Data integration module, used to integrate distance data with the operating parameters of production equipment to obtain production equipment configuration data; The target production equipment selection module is used to select the target production equipment according to the production equipment configuration data using the target optimization algorithm; The target production equipment marking module is used to mark the selected target production equipment using marking tools.

9. A quality inspection system for lining cloth production according to claim 8, characterized in that: The method of selecting target production equipment using a target optimization algorithm according to the production equipment configuration data comprises the following steps: S21, setting the type and quantity of production equipment configuration data, and randomly generating an initial production equipment configuration according to the type and data; S22. According to the initial production equipment configuration, calculate the performance index of each initial production equipment configuration, and use the performance index as the fitness value of the initial production equipment configuration; S23, selecting the initial production equipment configuration with the highest fitness, and using the overall hybridization method to randomly screen several initial production equipment configurations for crossover operation to generate a new production equipment configuration; S24. Based on the new production equipment configuration, a simulated annealing algorithm is used to perform a local search to obtain a local optimal production equipment configuration; S25, iteratively generating the next generation of production equipment configuration according to the initial production equipment configuration with the highest fitness and the local optimal production equipment configuration; Repeat steps S23 to S25 until a preset number of iterations is reached, and output the production equipment configuration with the highest fitness in the final generation as the target production equipment.

10. A quality inspection system for lining cloth production according to claim 9, characterized in that: The performance index calculation formula of the initial production equipment configuration is: In the formula, F´(n) Represents the initial production equipment configuration n performance index value; λ represents weight; ▽ f(n) Represents the initial production equipment configuration function f(n) Initial production equipment configuration n The gradient of ▽ f min A gradient representing the minimum initial production equipment configuration; ▽ f max represents the gradient of the highest initial production equipment configuration; F(n) Represents the initial production equipment configuration n The original performance index value of F min Represents the performance indicator value of the minimum initial production equipment configuration; f max Represents the performance indicator value of the highest initial production equipment configuration; ||·|| represents the norm.

Images