A clothing and apparel production optimization system, method and storage medium

Abstract

The invention discloses a clothing and apparel production optimization system, method and storage medium, and relates to the technical field of clothing production. The production optimization system comprises a number of functional modules that run in sequence, namely a cutting recognition module, a state monitoring module, a splicing supervision module, a test optimization module and a feedback adjustment module. The technical key points are as follows: the system aims to comprehensively monitor and optimize the splicing links in clothing production through intelligent and automated technical means, and the system can accurately identify and process residual abnormal points to ensure continuous improvement of product quality; combining a linear adjustment model with an exponential adjustment model, the system intelligently estimates and adjusts the suture tension according to the number of abnormal points and the tensile test results, which not only solves the quality problems caused by improper tension in the sewing process, but also avoids the efficiency loss caused by excessive adjustment; in particular, for different numbers of abnormal points, the system flexibly selects the adjustment model to achieve fine regulation of the suture tension.

Description

Technical Field

[0001] The present invention relates to the technical field of clothing production, and in particular to a clothing and apparel production optimization system, method and storage medium. Background Art

[0002] Clothing production refers to the process of using fabric as the processing object in the manufacturing industry to produce various types of clothing that meet people's daily needs through a series of process flows such as design, cutting, sewing, sizing and packaging. It is a labor-intensive industry that requires production personnel to have professional technical skills and operational experience. Clothing production not only focuses on the basic functions of clothing, such as covering the body and keeping warm, but also pays more attention to the fashion and aesthetics of clothing as consumer demand increases. Therefore, there are higher requirements and standards for the production process of clothing and accessories.

[0003] The technical solution pointed out in the document with the existing application number of 202311162623.4 and the name of a clothing production monitoring method, system and medium includes: obtaining clothing manufacturing process, obtaining clothing cutting information according to the clothing manufacturing process, and extracting clothing cutting features; preprocessing clothing cutting features; obtaining order information, and generating production equipment clothing cutting status information according to the order information and the optimized clothing cutting information; comparing the clothing cutting status information with the preset status information to obtain the state deviation rate; judging whether the state deviation rate is greater than or equal to the preset state deviation rate threshold; if greater than or equal to, correcting the clothing cutting information according to the correction information; if less than, transmitting the clothing cutting data to the terminal; adjusting the clothing cutting status information in real time and dynamically through the clothing manufacturing process so that the clothing cutting meets the requirements, thereby dynamically monitoring the clothing production status in real time, ensuring that the clothing cutting meets the requirements, and improving the clothing production efficiency; but it only monitors or optimizes the cutting production process, and does not optimize and adjust the two core steps of cutting and sewing in clothing production, and thus cannot provide a more refined solution;

[0004] In combination with the above documents and existing technologies, in the traditional clothing production process, more automated or semi-automated equipment is used to replace manual labor, such as intelligent cutting machines or sewing equipment. However, the running equipment cannot fully guarantee the quality of clothing after production. If the equipment parameters are not optimized and adjusted, a batch of clothing products will be unqualified. Among them, when sewing and splicing T-shirts, the sewing process may cause broken threads or skipped stitches. At this time, secondary seam reinforcement is required. If the cause of broken threads or skipped stitches is improper thread tension, the above problems will still exist even if multiple reinforcements are performed, which not only affects the quality of the final sewn product, but also affects the production efficiency to a certain extent. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the shortcomings of the prior art, the present invention provides a clothing and apparel production optimization system, method and storage medium, which comprehensively monitors and optimizes the splicing links in clothing production through intelligent and automated technical means. The system can accurately identify and effectively process residual abnormal points, solving the problems raised in the background technology.

[0007] (II) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] A clothing and apparel production optimization system, the system comprising:

[0010] The cutting and identification module uses RFID technology to assign a unique identification to each piece of fabric, record the fabric information in the identification, and cut the fabric to obtain the semi-finished product to be sewn;

[0011] The status monitoring module performs defect detection on the semi-finished products to be sewn;

[0012] If a defective point is detected, the material replacement instruction will be triggered until no defective point is detected;

[0013] If no defective points are detected, the edge detection mechanism is executed, and the edge features of the semi-finished product are extracted through the image processing algorithm, and compared with the preset cutting template to determine whether the cutting is complete; if it is incomplete, the secondary fabric cutting action is performed until the cutting is complete, and the sewing start signal is output;

[0014] The stitching supervision module, under the condition of receiving the stitching start signal, the corresponding stitching equipment performs stitching processing according to the preset stitching area position on the semi-finished product to obtain a preliminary finished product, triggers the stitching supervision mechanism, and executes the analysis and control strategy according to the result of the stitching supervision mechanism;

[0015] The test optimization module performs a tensile test on the preliminary finished product and triggers the suture supervision mechanism for the second time. If an abnormal point is identified and marked, the evaluation data set is collected according to the preset constraints, and the constructed linear adjustment model or exponential adjustment model is triggered to generate a suture tension estimate to guide the suture equipment to adjust the suture tension.

[0016] The feedback adjustment module continuously monitors the effect of adjusting the suture equipment based on the estimated suture tension value, obtains the total number of abnormal points obtained by the initial and secondary triggering of the suture supervision mechanism, and establishes a curve chart according to the changes in each preliminary finished product processing and the total number of abnormal points, analyzes the changing trend of the curve chart, and chooses whether to implement the dynamic threshold adjustment strategy based on the analysis results.

[0017] Furthermore, the fabric information recorded in the label at least includes: fabric type, batch, production date and size.

[0018] Furthermore, the process of defect detection is as follows: the semi-finished product to be sewn is laid on a light-transmitting plate, an image of light passing through the semi-finished product is detected from above by using optical principles, and an image processing software is used for analysis to identify whether a defect point is detected;

[0019] Among them, the image processing software analyzes the following contents:

[0020] Whether the image of the light passing through the semi-finished product is different from the pre-set normal area image, if there is a different area, it means that a defect point is detected; otherwise, it means that no defect point is detected.

[0021] Furthermore, the process of triggering the stitching supervision mechanism is: evaluating each stitch on each stitching area point by point; identifying whether each stitch has an abnormality;

[0022] If it exists, it is marked as an outlier;

[0023] If it does not exist, it will not be marked.

[0024] Further, the process of executing the analysis control strategy is as follows:

[0025] When any stitch is identified to be abnormal, the number of abnormal points in the seam area on the same side is obtained by counting;

[0026] When the number of abnormal points is 0, no response action is taken;

[0027] When the number of abnormal points is 1, a first-level reinforcement instruction is issued;

[0028] When the number of abnormal points is 2 and the distance between the two abnormal points is less than the preset L0, a secondary reinforcement instruction is issued; if the distance between the two abnormal points is not less than L0, a primary reinforcement instruction is issued to each of the two abnormal points;

[0029] When the number of abnormal points exceeds 2, a third-level reinforcement instruction is issued;

[0030] Among them, a first-level reinforcement instruction is issued: the suture equipment performs Xcm reinforcement suture operation with the current abnormal point as the center; a second-level reinforcement instruction is issued: the suture equipment performs 2Xcm reinforcement suture operation with the midpoint of the line segment formed by the two abnormal points as the center; a third-level reinforcement instruction is issued: the suture equipment performs a whole-segment reinforcement suture operation on the cutting area where more than two abnormal points are located; the value of X is 1~5.

[0031] Furthermore, the preset constraint condition is: comparing the number of abnormal points Nt obtained by the secondary triggering seam supervision mechanism with the preset delimitation threshold Mo;

[0032] When Nt≤Mo, the linear adjustment model is triggered;

[0033] When Nt>Mo, the index adjustment model is triggered;

[0034] Among them, the evaluation data set includes at least: the number of abnormal points N0 obtained by the initial triggering of the seam supervision mechanism, the number of abnormal points Nt obtained by the secondary triggering of the seam supervision mechanism, the tension value set when performing the tension test and its standard value, and the initially set suture tension.

[0035] Furthermore, the formula for constructing the linear adjustment model is as follows:

[0036] ;

[0037] In the formula, Indicates the estimated suture tension. Both represent weight coefficients, and their value ranges are [0, 1]. Fs represents the tension value set when performing the tension test. Indicates the standard value corresponding to the tension, Indicates the initial set suture tension;

[0038] The formula used to construct the index adjustment model is as follows:

[0039] ;

[0040] In the formula, Both represent adjustment coefficients, and their value ranges are [0, 1].

[0041] Further, the process of selecting whether to execute the dynamic threshold adjustment strategy based on the analysis results is as follows:

[0042] Two types of adjustment models are selected at the same time to make adjustments based on the two suture tension estimates calculated, and the effects of the two adjustments are compared, that is, the total number of two abnormal points is compared;

[0043] If the total number of outliers corresponding to the linear adjustment model is higher than the total number of outliers corresponding to the exponential adjustment model, the first-level threshold correction action is performed;

[0044] If the total number of outliers corresponding to the exponential adjustment model is higher than the total number of outliers corresponding to the linear adjustment model, the secondary threshold correction action is performed;

[0045] Among them, executing the first-level threshold correction action means:

[0046] The threshold Mo in the preset constraint condition is adjusted, and the clothing optimization process is performed according to the modified threshold Mox, and the modified threshold Mox is limited by the following method:

[0047]

[0048] In the formula, q represents the adjusted fixed value, and q is greater than 0;

[0049] Executing the secondary threshold correction action means:

[0050] The threshold Mo in the preset constraint condition is adjusted, and clothing optimization processing is performed according to the modified threshold Mox, wherein the modified threshold Mox is limited by the following method:

[0051]

[0052] In the formula, the value of q is not only greater than 0, but also less than Mo.

[0053] A method for optimizing the production of clothing and accessories comprises the following steps:

[0054] S1. Using RFID technology, a unique identification is assigned to each piece of fabric, the fabric information is recorded in the identification, and the fabric is cut to obtain a semi-finished product to be sewn;

[0055] S2. Perform defect detection on the semi-finished products to be sewn;

[0056] If a defective point is detected, the material replacement instruction will be triggered until no defective point is detected;

[0057] If no defective points are detected, the edge detection mechanism is executed, and the edge features of the semi-finished product are extracted through the image processing algorithm, and compared with the preset cutting template to determine whether the cutting is complete; if it is incomplete, the secondary fabric cutting action is performed until the cutting is complete, and the sewing start signal is output;

[0058] S3. Under the condition of receiving the sewing start signal, the corresponding sewing equipment performs sewing processing according to the preset sewing area position on the semi-finished product to obtain a preliminary finished product, triggering the sewing supervision mechanism, and executing the analysis control strategy according to the result of the sewing supervision mechanism;

[0059] S4. Perform a tensile test on the preliminary finished product and trigger the suture supervision mechanism for the second time. If an abnormal point is identified and marked, collect the evaluation data set according to the preset constraint conditions, and trigger the constructed linear adjustment model or exponential adjustment model to generate a suture tension estimate to guide the suture equipment to adjust the suture tension.

[0060] S5. Continuously monitor the effect of adjusting the suture equipment based on the estimated suture tension value, obtain the total number of abnormal points obtained by the initial and secondary triggering of the suture supervision mechanism, and establish a curve chart according to the changes in each preliminary finished product processing and the total number of abnormal points, analyze the changing trend of the curve chart, and choose whether to implement the dynamic threshold adjustment strategy based on the analysis results.

[0061] (III) Beneficial effects

[0062] The present invention provides a clothing and apparel production optimization system, method and storage medium, which have the following beneficial effects:

[0063] (1) This solution realizes the comprehensive automatic defect detection and edge integrity verification of semi-finished products to be sewn. By combining machine vision technology with optical principles, the system can efficiently identify various defects on the fabric, such as stains, holes, etc., and automatically trigger fabric replacement instructions to ensure that the quality of the fabric entering the sewing process is flawless;

[0064] (2) This solution achieves refined supervision and intelligent reinforcement of the sewing process. By evaluating the stitch quality of the sewing area point by point, the system can accurately identify and mark abnormal points, ensuring the fine quality of the sewn products. According to the number and distribution of abnormal points, the system intelligently executes analysis and control strategies, issues reinforcement instructions of different levels, and accurately reinforces the problem areas, which not only avoids excessive sewing that affects the wearing effect, but also ensures the strength and quality of the seams. By combining the automated intelligent sewing machine with a sophisticated reinforcement strategy, the sewing efficiency and product qualification rate are significantly improved.

[0065] (3) Through secondary seam supervision after the tensile test, this solution can accurately identify and process the remaining abnormal points to ensure continuous improvement of product quality. By combining the linear adjustment model and the exponential adjustment model, the system intelligently estimates and adjusts the suture tension according to the number of abnormal points and the tensile test results, which not only solves the quality problems caused by improper tension during the seam process, but also avoids the efficiency loss caused by excessive adjustment. In particular, according to the number of different abnormal points, the system flexibly selects the adjustment model to achieve fine control of the suture tension.

[0066] (4) This solution continuously monitors the adjustment effect of the sewing equipment and establishes a curve chart of the total number of abnormal points. The system can automatically identify and respond to the upward trend and execute a dynamic threshold adjustment strategy. This strategy compares the effects of the linear and exponential adjustment models, intelligently selects a better model, and corrects the defined threshold accordingly to achieve dynamic adjustment of the constraint conditions. It solves the quality problems caused by improper suture tension during the sewing process and improves the flexibility and accuracy of the adjustment strategy. Through the dynamic feedback mechanism, the system can continuously optimize the suture tension according to actual conditions, reduce the number of abnormal points, and improve the overall quality of sewn products. Therefore, the entire system provides an intelligent and refined suture tension adjustment solution for the garment manufacturing industry, effectively improving the quality and production efficiency of sewn products. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a modular schematic diagram of the optimization system in the present invention;

[0068] Figure 2 A schematic diagram showing a sample to be optimized for production in the present invention;

[0069] Figure 3 It is a schematic diagram of the overall steps of the optimization method in the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] Embodiment 1:

[0072] See also Figure 1 to Figure 2 This embodiment provides a clothing and apparel production optimization system, which aims to comprehensively monitor and optimize the splicing process in the production of clothing (for example, underwear, tights and other easy-to-process clothing) through intelligent and automated technical means to ensure the quality of each subsequent product;

[0073] The production optimization system includes several functional modules that run in sequence, namely, cutting recognition module, status monitoring module, splicing supervision module, test optimization module and feedback adjustment module, and targets clothing production links including cutting and sewing;

[0074] Functional module description

[0075] Crop recognition module:

[0076] Using RFID technology, each piece of fabric is assigned a unique identification, the fabric information is recorded in the identification, and the fabric is cut to obtain the semi-finished product to be sewn;

[0077] Among them, using RFID technology to assign a unique identification to each piece of fabric can greatly improve the efficiency and accuracy of fabric management and cutting processes. The identification is an RFID tag, which contains a chip and an antenna for storing and transmitting fabric information. Through this identification, digital management of fabrics is achieved; the fabric information recorded in the identification includes at least the type, batch, production date and size of the fabric;

[0078] The fabric cutting process is as follows:

[0079] Cutting preparation: before cutting, the fabric is placed within the effective range of the reader; the reader sends radio frequency signals through the antenna to activate the RFID tag on the fabric and read the information in the tag; cutting operation: according to the read fabric information, the cutting machine (i.e., clothing intelligent laser cutting machine) can cut according to the predetermined cutting plan; during the cutting process, RFID technology can monitor the position and status of the fabric in real time to ensure the accuracy and efficiency of cutting; cutting record: after cutting is completed, the RFID system can automatically record the cutting results, including the cutting size, quantity, remaining fabric and other information; this information will be used for subsequent sewing plans and inventory management;

[0080] When cutting fabrics, the intelligent clothing laser cutting machine is used. Through the AI ​​vision function, it can automatically identify the cutting track of the fabric and quickly cut a piece of clothing. After placing the fabric, the cutting action can be carried out.

[0081] The semi-finished products to be sewn can be referred to Figure 2 The cropped state shown;

[0082] The advantages of RFID technology in fabric cutting are as follows:

[0083] Improve efficiency: RFID technology can realize the rapid identification and positioning of fabrics, reducing the time and cost of manual operations; Enhance accuracy: The uniqueness of RFID tags and the real-time update of information can ensure the accuracy of the cutting and sewing process; Facilitate management: RFID technology can realize the full tracking and management of fabrics, from raw material procurement to finished product delivery, the information of each link can be recorded and traced; Support intelligent production: RFID technology can be combined with other intelligent manufacturing technologies, such as automated cutting machines, intelligent sewing equipment, etc., to jointly promote the intelligent upgrade of clothing production.

[0084] Condition Monitoring Module:

[0085] Use machine vision technology to detect defects in semi-finished products to be sewn;

[0086] If a defective point is detected, the material replacement instruction will be triggered until no defective point is detected;

[0087] If no defect points are detected, the edge detection mechanism is executed to extract the edge features of the semi-finished product through the image processing algorithm and compare them with the preset cutting template to determine whether the cutting is complete;

[0088] If it is incomplete, perform a second cloth cutting action until the cutting is complete;

[0089] If it is complete, then the seam start signal is output;

[0090] The process of defect detection is as follows:

[0091] The semi-finished product to be sewn is laid on a light-transmitting board, and the image of light passing through the semi-finished product is detected from above using optical principles, and processed and analyzed using image processing software to identify whether a defect point is detected;

[0092] The above method utilizes optical principles to determine whether there are defects on the fabric by the scattering, reflection or transmission of light when it passes through the fabric. When light passes through the fabric, defective areas (such as stains, holes, jumpers, stains, etc.) will have different effects on the light, causing the scattering, reflection or transmission of light in these areas to be different from that in normal areas. By detecting this difference, the defects can be identified and located.

[0093] The specific plan is as follows:

[0094] Equipment composition: Translucent board: Select a material with high light transmittance, flat surface and not easy to deform to make the translucent board. The translucent board should be able to stably support the fabric and allow light to pass through; Light source: Set a light source under the translucent board. The light source can be an LED light bar, a cathode cold light tube or an energy-saving lamp, etc. The light source should be able to provide uniform and sufficiently bright light to clearly display the defects on the fabric; Image acquisition element: Set a high-resolution camera or image sensor above the translucent board to capture the image of light passing through the fabric. The image acquisition element should maintain an appropriate distance and angle with the translucent board to ensure that a clear image is captured; Control system: Includes image processing software and an operating interface for processing, analyzing and displaying the captured images; The image processing software should be able to automatically identify defects, measure defect sizes, count the number of defects and classify them;

[0095] Testing process:

[0096] Laying the cloth: Lay the cloth to be tested flatly on the light-transmitting plate, ensuring that the cloth is in close contact with the light-transmitting plate and has no wrinkles; Turning on the light source: Turning on the light source under the light-transmitting plate to allow light to pass through the cloth to the image acquisition element; Capturing the image: Capturing the image of light passing through the cloth through the image acquisition element, and transmitting the image to the control system for processing; Image processing and analysis: The control system uses image processing software to process and analyze the captured image, automatically identifying defects on the cloth and measuring their size, counting the number and classifying them; Displaying the results: Displaying the test results on the operation interface for the operator to view, and generating a test report or issuing an alarm prompt as needed.

[0097] Triggering fabric replacement instruction: indicates replacing another semi-finished product to be sewn;

[0098] The specific process of implementing the edge detection mechanism is as follows:

[0099] Edge extraction: Use image processing algorithms, namely edge detection algorithms (such as Canny edge detection algorithms) to extract edge features of products. The algorithm first performs Gaussian filtering to eliminate noise, then calculates the size and direction of image gradients, and finally extracts edge points through non-maximum suppression and double threshold detection; Edge matching: Compare the extracted edge features with the preset cutting template. The preset cutting template should contain the complete edge shape and size information of the product. By comparing the shape, length, angle and other features of the edge, it is determined whether the cutting is complete; Decision and feedback: If the edge matching result does not conform to the preset template, an alarm signal is issued, and a prompt is given to re-cut or perform other processing (i.e., perform a secondary fabric cutting action); If the edge matching result conforms to the preset template, subsequent splicing operations are performed.

[0100] By adopting the above technical solutions, the fully automated defect detection and edge integrity verification of semi-finished products to be sewn are realized. By combining machine vision technology with optical principles, the system can efficiently identify various defects on the fabric, such as stains, holes, etc., and automatically trigger fabric replacement instructions to ensure that the quality of the fabric entering the sewing process is flawless.

[0101] At the same time, the introduction of edge detection mechanism uses advanced image processing algorithms to accurately extract fabric edge features and compare them with preset cutting templates to effectively determine the integrity of cutting. This process not only greatly improves detection efficiency and accuracy, but also significantly reduces the cost of manual intervention, and solves the problems of easy omissions and inconsistent detection standards in traditional detection methods.

[0102] In addition, the system can provide instant feedback based on the test results to guide secondary cutting or start sewing operations, realizing intelligent control of the production process and ensuring the overall quality and production efficiency of sewn products;

[0103] In summary, this technical solution effectively solves the technical problems of fabric defect detection and cutting integrity verification, and provides strong support for improving the level of automation and intelligence in the textile manufacturing industry.

[0104] Splicing supervision module:

[0105] Under the condition of receiving the sewing start signal, the corresponding sewing equipment performs sewing processing according to the preset sewing area position on the semi-finished product to obtain a preliminary finished product, triggering the sewing supervision mechanism to evaluate each stitch on each sewing area point by point to identify whether each stitch has an abnormality. If so, it is marked as an abnormal point; if not, it is not marked;

[0106] Execute analytical control strategies based on the results of the seam supervision mechanism;

[0107] The process of analyzing the control strategy is as follows:

[0108] When any stitch is identified to be abnormal, the number of abnormal points in the seam area on the same side is obtained by counting;

[0109] When the number of abnormal points is 0, no response action is taken;

[0110] When the number of abnormal points is 1, a first-level reinforcement instruction is issued;

[0111] When the number of abnormal points is 2 and the distance between the two abnormal points is less than the preset L0, a secondary reinforcement instruction is issued; if the distance between the two abnormal points is not less than L0, a primary reinforcement instruction is issued to each of the two abnormal points;

[0112] When the number of abnormal points exceeds 2, a third-level reinforcement instruction is issued;

[0113] Among them, the suture equipment adopts an automated intelligent suture machine, which uses a pressure plate to push the semi-finished product to change its position to ensure Figure 2 Each seam area in the system can be sewn, and after each seam area is completely sewn, a preliminary finished product can be obtained. The clothing and accessories targeted by this system are mainly underwear and tights, which are easy to cut and assemble.

[0114] When identifying whether there is an abnormality at each pin:

[0115] Use a microscope or high-resolution camera to place the seam area under the microscope or high-resolution camera, adjust the focus to clearly display each stitch, and ensure that the lighting conditions are sufficient and uniform during visual inspection to avoid shadows or reflections affecting observation; when using a high-resolution camera for point-by-point evaluation, use the established normal template to compare with the real-time image at each point. If there is a difference with the normal template, it means that there is an abnormality; otherwise, it does not exist;

[0116] Among them, a normal template means: it presents uniform, continuous and tight arrangement, with clear stitches and no broken thread or skipped stitches;

[0117] Then, for abnormal identification, at least the following are included:

[0118] Skipped stitches: manifested as discontinuous breakpoints in the stitches, i.e. the thread does not pass through the fabric in the expected path; Loose thread ends: Part of the stitches appear loose and do not stick to the surface of the fabric, which may be caused by uneven thread tension or failure to tighten the thread during sewing; Skewed stitches: The stitches are not arranged in a straight line, but are skewed or zigzag, which may affect the strength of the stitching;

[0119] Reference Figure 2 As shown, the seam areas on the same side are on the same straight line, for example, the seam areas on the same side of the outer trousers are the seam areas on the same straight line;

[0120] When the abnormal points given above are 0, 1, 2 or above, they all represent the seam area on the same side. The reinforcement instructions issued subsequently are only for the seam area on this side. For other seam areas, the corresponding number or type of instructions will be issued for subsequent reinforcement operations and responses.

[0121] When the number of outliers is 2 and the distance between two outliers is less than the preset L0, the number of outliers on the same side is 2. Figure 2 As shown, when the distance between the two abnormal points is less than L0, the seam overlap area will appear in the subsequent reinforcement action, avoiding too many overlapping sewing lines and affecting the subsequent wearing effect; and when the distance between the two abnormal points is not less than the preset L0, there will be no seam overlap area;

[0122] Issue a first-level reinforcement command:

[0123] The suture equipment performs Xcm reinforcement suture operation with the current abnormal point as the center;

[0124] Issue a secondary reinforcement command:

[0125] The suture device takes the midpoint of the line segment formed by the two abnormal points as the center and performs a 2Xcm reinforcement suture operation;

[0126] Issue a three-level reinforcement command:

[0127] The suture equipment performs a whole-section reinforcement suture operation on the cutting area where more than two abnormal points are located;

[0128] The value of X is 1 to 5, which is set according to actual needs. For example, if the target clothing is children's clothing, the value of X is 1, and if it is ready-made clothing, the value of X can be 2. Regardless of whether the reinforced seams are Xcm or 2Xcm, the midpoint of the reinforced area is the corresponding center.

[0129] By adopting the above technical solutions, refined supervision and intelligent reinforcement of the sewing process are achieved. By evaluating the stitch quality of the sewing area point by point, the system can accurately identify and mark abnormal points, such as skipped stitches, loose threads, skewed stitches, etc., ensuring the fine quality of sewn products;

[0130] According to the number and distribution of abnormal points, the system intelligently executes analysis and control strategies, issues reinforcement instructions of different levels, and accurately reinforces the problem areas, which not only avoids excessive sewing that affects the wearing effect, but also ensures the strength and quality of the seams; especially for underwear, tights and other clothing, the system significantly improves the sewing efficiency and product qualification rate by combining automated intelligent sewing machines with sophisticated reinforcement strategies; in addition, the system also takes into account the different requirements of different types of clothing (such as children's clothing and ready-made clothing) for reinforcement width, and achieves flexible response by adjusting the X value;

[0131] In summary, this technical solution effectively solves the problem of identifying and handling abnormal points in the sewing process and realizes intelligent control of the quality of sewn products.

[0132] Test optimization module:

[0133] After executing the analysis and control strategy, a tensile test is performed on the preliminary finished product. After the tensile test, the sewing supervision mechanism is triggered for the second time. If an abnormal point is also identified and marked, the evaluation data set is collected according to the preset constraints, and the constructed linear adjustment model or exponential adjustment model is triggered to generate a suture tension estimate to guide the suture equipment to improve the sewing process, that is, adjust the suture tension to achieve the purpose of reducing the occurrence of abnormal points;

[0134] Among them, when performing tensile testing, you can refer to Figure 2 As shown, the direction of the pulling force is Figure 2 The direction indicated by the arrow;

[0135] The solid arrow indicates: it is located above the preliminary finished product;

[0136] The dotted arrow indicates: it is located below the preliminary finished product;

[0137] Regardless of whether the arrow direction of the solid line or the dotted line is vertically distributed with respect to the seam area of ​​the corresponding side, it is ensured that the tension can act on the seam area, thereby realizing the tension test operation on the seam area. When performing the tension test, a strip-shaped grab plate can be fixed on the surface of the preliminary finished product, and the position of the strip-shaped grab plate coincides with the position of the line segment; wherein, the position of the line segment is: a line segment formed by connecting the points where the two corresponding tension arrows act on the preliminary finished product; the purpose of such design is: to ensure the stability of the grab position and to ensure that the tension can completely cover the seam area;

[0138] The preset constraints are:

[0139] The number of abnormal points Nt obtained by the secondary triggering seam supervision mechanism is compared with the preset threshold Mo, which is a fixed value in the initial state and can be set independently according to the actual situation;

[0140] When Nt≤Mo, the linear adjustment model is triggered;

[0141] When Nt>Mo, the index adjustment model is triggered;

[0142] Regardless of the constraints, an evaluation data set needs to be collected, and the evaluation data set at least includes: the number of abnormal points N0 obtained by the initial triggering of the seam supervision mechanism, the number of abnormal points Nt obtained by the secondary triggering of the seam supervision mechanism, the tension value and its standard value set when performing the tension test, and the initial set suture tension; before using the evaluation data set for subsequent calculations, each data in it needs to be dimensionless processed to ensure that subsequent calculations and processing are in the same dimension;

[0143] The linear adjustment model is constructed based on the following formula:

[0144] ;

[0145] In the formula, Indicates the estimated suture tension. Both represent weight coefficients, and their value ranges are [0, 1]. Fs represents the tension value set when performing the tension test. Indicates the standard value corresponding to the tension, Indicates the initial set suture tension;

[0146] Logical explanation: The change in the number of abnormal points and the difference between the tension value and the standard value are taken into account, and the suture tension is adjusted by linearly combining these two factors. It reflects the influence of the change in the number of abnormal points on the suture tension. If the number of abnormal points in the second time is more than the first time, it means that the suture tension needs to be increased; otherwise, it should be reduced. It reflects the influence of the difference in tension value on the suture tension. If the actual tension value is lower than the standard value, it means that the suture tension needs to be increased to improve the strength; otherwise, it can be considered to be reduced.

[0147] The weight coefficient is determined by the coefficient of variation method, which is a method of assigning weights to each indicator based on the degree of variation between the current value of each evaluation indicator and the target value. If the numerical difference of a certain indicator is large, it can clearly distinguish the evaluated objects, indicating that the indicator has rich information for distinguishing, and thus the indicator should be given a larger weight. On the contrary, if the numerical difference of each evaluated object on a certain indicator is small, then the ability of this indicator to distinguish the evaluated objects is weak, and thus the indicator should be given a smaller weight. This method directly uses the information contained in each indicator to obtain the weight of the indicator through calculation, so it is objective. The same principle can also be used to determine the adjustment coefficient.

[0148] The formula used to construct the index adjustment model is as follows:

[0149] ;

[0150] In the formula, Both represent adjustment coefficients, and their value ranges are [0, 1];

[0151] Logical explanation: An exponential function is used to reflect the nonlinear effect of the change in the number of abnormal points on the suture tension. When the number of abnormal points changes greatly, the adjustment range of the suture tension will also increase accordingly; The exponential adjustment degree of suture tension to the change of the number of abnormal points is jointly controlled, which makes the adjustment of suture tension more sensitive and accurate when the number of abnormal points changes greatly; This reflects the linear effect of the difference in pulling force on suture tension, which is consistent with the linear adjustment model. similar;

[0152] In summary, the two formulas use linear adjustment and exponential adjustment to calculate the estimated suture tension value after the equipment is adjusted. The linear adjustment model is suitable for situations where the number of abnormal points changes and the difference in tension value is small; the exponential adjustment model is more complex and accurate and is suitable for situations that require more precise adjustments. In practical applications, the appropriate formula can be selected for calculation according to the specific situation. Through the combination of exponential adjustment and linear adjustment, the two formulas can more accurately reflect the changing needs of suture tension and optimize the adjustment effect to a certain extent.

[0153] By adopting the above technical solutions, in-depth optimization and intelligent control of sewing product quality are achieved;

[0154] Through secondary seam supervision after the tensile test, the system can accurately identify and process the remaining abnormal points to ensure continuous improvement of product quality; combining the linear adjustment model with the exponential adjustment model, the system intelligently estimates and adjusts the suture tension according to the number of abnormal points and the tensile test results, which not only solves the quality problems caused by improper tension during the seam process, but also avoids the efficiency loss caused by excessive adjustment; especially for different numbers of abnormal points, the system flexibly selects the adjustment model to achieve fine control of the suture tension;

[0155] In summary, this technical solution effectively solves the problem of seam quality control and significantly improves the pass rate and overall quality of sewn products through an intelligent and refined adjustment mechanism.

[0156] Feedback adjustment module:

[0157] Continuously monitor the effect of adjusting the suture equipment according to the estimated suture tension value, obtain the total number of abnormal points obtained by the initial and secondary triggering of the suture supervision mechanism, and establish a curve chart according to the changes in the total number of abnormal points each time the preliminary finished product processing is completed, analyze the trend of the curve chart, and when the curve chart has an upward trend, execute the dynamic threshold adjustment strategy, feed back the results of the dynamic threshold adjustment strategy to the constraint conditions, and execute the corrective action;

[0158] The X-axis in the curve graph represents the number of times the preliminary finished product is processed, and the Y-axis represents the total number of abnormal points under the corresponding number, thereby obtaining a curve graph of changes in different processing times;

[0159] To analyze the trend of the curve graph, the tools that can be used include: Excel, Python's Matplotlib and Seaborn libraries, and any of the R languages. These tools can intuitively display the data trend. When there is an upward trend in the curve graph, it can be quickly identified by observing the shape and direction of the curve. The upward trend is usually manifested as a continuous increase in data points, forming an obvious upward straight line or curve.

[0160] The process of implementing a dynamic threshold adjustment policy is as follows:

[0161] Two types of adjustment models are selected at the same time to make adjustments based on the two suture tension estimates calculated, and the effects of the two adjustments are compared, that is, the total number of two abnormal points is compared;

[0162] If the total number of outliers under the linear adjustment model is higher than that under the exponential adjustment model, it means that the exponential adjustment model is more effective, and the first-level threshold correction action is performed, which means:

[0163] The results of the dynamic threshold adjustment strategy and the corrective actions to be performed;

[0164] If the total number of outliers under the exponential adjustment model is higher than that under the linear adjustment model, it means that the linear adjustment model is more effective, and the secondary threshold correction action is performed, which means:

[0165] The results of the dynamic threshold adjustment strategy and the corrective actions to be performed;

[0166] In addition, there may be a rare situation where the total number of outliers under the exponential adjustment model is equal to the total number of outliers under the linear adjustment model. If this happens, just maintain the original adjustment model.

[0167] The contents of the first-level threshold correction action are as follows:

[0168] The threshold Mo in the preset constraint condition is adjusted, and clothing optimization processing is performed according to the modified threshold Mox, wherein the modified threshold Mox is limited by the following method:

[0169] ;

[0170] In the formula, q represents the adjustment fixed value, q>0, and the value of q is usually 1;

[0171] The contents of executing the secondary threshold correction action are as follows:

[0172] The threshold Mo in the preset constraint condition is adjusted, and clothing optimization processing is performed according to the modified threshold Mox, wherein the modified threshold Mox is limited by the following method:

[0173] ;

[0174] In the formula, the value of q is not only greater than 0, but also less than Mo. As in the above formula, its value is usually 1;

[0175] Specifically, by setting a threshold and selecting different formulas according to the change in the number of abnormal points, we can achieve a balance between the flexibility and accuracy of suture tension adjustment; in the actual production process, the number of abnormal points may be affected by many factors, including suture material, sewing process, operator skills, etc.; therefore, it is reasonable and necessary to select different adjustment formulas according to the change in the number of abnormal points; in addition, this conditional selection method can also help production personnel better understand and implement the suture tension adjustment process, improve production efficiency and product quality;

[0176] Specifically, by continuously monitoring the adjustment effect of the suture equipment and establishing a curve chart of the total number of abnormal points, the system can automatically identify and respond to the upward trend and execute a dynamic threshold adjustment strategy. This strategy compares the effects of the linear and exponential adjustment models, intelligently selects a better model, and accordingly amends the defined threshold to achieve dynamic adjustment of the constraint conditions.

[0177] This technical solution effectively solves the quality problems caused by improper thread tension during the sewing process, improves the flexibility and accuracy of the adjustment strategy, and through the dynamic feedback mechanism, the system can continuously optimize the thread tension according to actual conditions, reduce the number of abnormal points, and improve the overall quality of sewn products. Therefore, this solution provides the garment manufacturing industry with an intelligent and refined thread tension adjustment solution, effectively improving the quality and production efficiency of sewn products.

[0178] Embodiment 2:

[0179] See also Figure 3 Based on Example 1, this embodiment further provides a method for optimizing the production of clothing and accessories, comprising the following specific steps:

[0180] S1. Using RFID technology, a unique identification is assigned to each piece of fabric, the fabric information is recorded in the identification, and the fabric is cut to obtain a semi-finished product to be sewn;

[0181] S2. Perform defect detection on the semi-finished products to be sewn;

[0182] If a defective point is detected, the material replacement instruction will be triggered until no defective point is detected;

[0183] If no defective points are detected, the edge detection mechanism is executed, and the edge features of the semi-finished product are extracted through the image processing algorithm, and compared with the preset cutting template to determine whether the cutting is complete; if it is incomplete, the secondary fabric cutting action is performed until the cutting is complete, and the sewing start signal is output;

[0184] S3. Under the condition of receiving the sewing start signal, the corresponding sewing equipment performs sewing processing according to the preset sewing area position on the semi-finished product to obtain a preliminary finished product, triggering the sewing supervision mechanism, and executing the analysis control strategy according to the result of the sewing supervision mechanism;

[0185] S4. Perform a tensile test on the preliminary finished product and trigger the suture supervision mechanism for the second time. If an abnormal point is identified and marked, collect the evaluation data set according to the preset constraint conditions, and trigger the constructed linear adjustment model or exponential adjustment model to generate a suture tension estimate to guide the suture equipment to adjust the suture tension.

[0186] S5. Continuously monitor the effect of adjusting the suture equipment based on the estimated suture tension value, obtain the total number of abnormal points obtained by the initial and secondary triggering of the suture supervision mechanism, and establish a curve chart according to the changes in each preliminary finished product processing and the total number of abnormal points, analyze the changing trend of the curve chart, and choose whether to implement the dynamic threshold adjustment strategy based on the analysis results.

[0187] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0188] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0189] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A clothing and accessories production optimization system, characterized in that: The system includes: The cutting and identification module uses RFID technology to assign a unique identification to each piece of fabric, record the fabric information in the identification, and cut the fabric to obtain the semi-finished product to be sewn; The status monitoring module performs defect detection on the semi-finished products to be sewn; If a defective point is detected, the material replacement instruction will be triggered until no defective point is detected; If no defective points are detected, the edge detection mechanism is executed, and the edge features of the semi-finished product are extracted through the image processing algorithm, and compared with the preset cutting template to determine whether the cutting is complete; if it is incomplete, the secondary fabric cutting action is performed until the cutting is complete, and the sewing start signal is output; The stitching supervision module, under the condition of receiving the stitching start signal, the corresponding stitching equipment performs stitching processing according to the preset stitching area position on the semi-finished product to obtain a preliminary finished product, triggers the stitching supervision mechanism, and executes the analysis and control strategy according to the result of the stitching supervision mechanism; The test optimization module performs a tensile test on the preliminary finished product and triggers the suture supervision mechanism for the second time. If an abnormal point is identified and marked, the evaluation data set is collected according to the preset constraints, and the constructed linear adjustment model or exponential adjustment model is triggered to generate a suture tension estimate to guide the suture equipment to adjust the suture tension. The feedback adjustment module continuously monitors the effect of adjusting the suture equipment based on the estimated suture tension value, obtains the total number of abnormal points obtained by the initial and secondary triggering of the suture supervision mechanism, and establishes a curve chart according to the changes in each preliminary finished product processing and the total number of abnormal points, analyzes the changing trend of the curve chart, and chooses whether to implement the dynamic threshold adjustment strategy based on the analysis results.

2. The clothing and accessories production optimization system according to claim 1, characterized in that: The fabric information recorded in the label includes at least: fabric type, batch, production date and size.

3. The clothing and accessories production optimization system according to claim 1, characterized in that: The process of defect detection is as follows: the semi-finished product to be sewn is laid on a light-transmitting plate, and the image of light passing through the semi-finished product is detected from above using optical principles, and image processing software is used to analyze it to identify whether a defect point is detected; Among them, the image processing software analyzes the following contents: Whether the image of the light passing through the semi-finished product is different from the pre-set normal area image, if there is a different area, it means that a defect point is detected; otherwise, it means that no defect point is detected.

4. The clothing and accessories production optimization system according to claim 1, characterized in that: The process of triggering the seam supervision mechanism is: evaluating each stitch on each seam area point by point; identifying whether each stitch has abnormalities; If it exists, it is marked as an outlier; If it does not exist, it will not be marked.

5. The clothing and accessories production optimization system according to claim 4, characterized in that: The process of executing the analysis control strategy is as follows: When any stitch is identified to be abnormal, the number of abnormal points in the seam area on the same side is obtained by counting; When the number of abnormal points is 0, no response action is taken; When the number of abnormal points is 1, a first-level reinforcement instruction is issued; When the number of abnormal points is 2 and the distance between the two abnormal points is less than the preset L0, a secondary reinforcement instruction is issued; if the distance between the two abnormal points is not less than L0, a primary reinforcement instruction is issued to each of the two abnormal points; When the number of abnormal points exceeds 2, a third-level reinforcement instruction is issued; Among them, a first-level reinforcement instruction is issued: the suture equipment performs Xcm reinforcement suture operation with the current abnormal point as the center; a second-level reinforcement instruction is issued: the suture equipment performs 2Xcm reinforcement suture operation with the midpoint of the line segment formed by the two abnormal points as the center; a third-level reinforcement instruction is issued: the suture equipment performs a whole-segment reinforcement suture operation on the cutting area where more than two abnormal points are located; the value of X is 1~5.

6. The clothing and accessories production optimization system according to claim 1, characterized in that: The preset constraint condition is: compare the number of abnormal points Nt obtained by the secondary triggering seam supervision mechanism with the preset delimitation threshold Mo; When Nt≤Mo, the linear adjustment model is triggered; When Nt>Mo, the index adjustment model is triggered; Among them, the evaluation data set includes at least: the number of abnormal points N0 obtained by the initial triggering of the seam supervision mechanism, the number of abnormal points Nt obtained by the secondary triggering of the seam supervision mechanism, the tension value set when performing the tension test and its standard value, and the initially set suture tension.

7. The clothing and accessories production optimization system according to claim 6, characterized in that: The linear adjustment model is constructed based on the following formula: ； In the formula, Indicates the estimated suture tension. Both represent weight coefficients, and their value ranges are [0, 1]. Fs represents the tension value set when performing the tension test. Indicates the standard value corresponding to the tension, Indicates the initial set suture tension; The formula used to construct the index adjustment model is as follows: ； In the formula, Both represent adjustment coefficients, and their value ranges are [0, 1].

8. The clothing and accessories production optimization system according to claim 6, characterized in that: The process of choosing whether to implement the dynamic threshold adjustment strategy based on the analysis results is as follows: Two types of adjustment models are selected at the same time to make adjustments based on the two suture tension estimates calculated, and the effects of the two adjustments are compared, that is, the total number of two abnormal points is compared; If the total number of outliers corresponding to the linear adjustment model is higher than the total number of outliers corresponding to the exponential adjustment model, the first-level threshold correction action is performed; If the total number of outliers corresponding to the exponential adjustment model is higher than the total number of outliers corresponding to the linear adjustment model, the secondary threshold correction action is performed; Among them, executing the first-level threshold correction action means: The threshold Mo in the preset constraint condition is adjusted, and the clothing optimization process is performed according to the modified threshold Mox, and the modified threshold Mox is limited by the following method: ； In the formula, q represents the adjustment fixed value, and q is greater than 0; Executing the secondary threshold correction action means: The threshold Mo in the preset constraint condition is adjusted, and clothing optimization processing is performed according to the modified threshold Mox, wherein the modified threshold Mox is limited by the following method: ； In the formula, the value of q is not only greater than 0, but also less than Mo.

9. A method for optimizing the production of clothing and accessories, using any system described in claims 1 to 8, characterized in that: The steps include: S1. Using RFID technology, a unique identification is assigned to each piece of fabric, the fabric information is recorded in the identification, and the fabric is cut to obtain a semi-finished product to be sewn; S2. Perform defect detection on the semi-finished products to be sewn; If a defective point is detected, the material replacement instruction will be triggered until no defective point is detected; If no defective points are detected, the edge detection mechanism is executed, and the edge features of the semi-finished product are extracted through the image processing algorithm, and compared with the preset cutting template to determine whether the cutting is complete; if it is incomplete, the secondary fabric cutting action is performed until the cutting is complete, and the sewing start signal is output; S3. Under the condition of receiving the sewing start signal, the corresponding sewing equipment performs sewing processing according to the preset sewing area position on the semi-finished product to obtain a preliminary finished product, triggering the sewing supervision mechanism, and executing the analysis control strategy according to the result of the sewing supervision mechanism; S4. Perform a tensile test on the preliminary finished product and trigger the suture supervision mechanism for the second time. If an abnormal point is identified and marked, collect the evaluation data set according to the preset constraint conditions, and trigger the constructed linear adjustment model or exponential adjustment model to generate a suture tension estimate to guide the suture equipment to adjust the suture tension. S5. Continuously monitor the effect of adjusting the suture equipment based on the estimated suture tension value, obtain the total number of abnormal points obtained by the initial and secondary triggering of the suture supervision mechanism, and establish a curve chart according to the changes in each preliminary finished product processing and the total number of abnormal points, analyze the changing trend of the curve chart, and choose whether to implement the dynamic threshold adjustment strategy based on the analysis results.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to claim 9 is implemented.

Images