A method and system for monitoring the production quality of home textile fabrics based on industrial data

Abstract

The present invention relates to the technical field of production quality monitoring of home textile fabrics, and specifically to a method and system for monitoring the production quality of home textile fabrics based on industrial data. It includes: First, by obtaining the planned finished product information, select a home textile fabric template from the template library and calculate the difference degree between it and the planned finished product. According to the difference degree, determine the initial equipment parameters and use these parameters to set the equipment. Next, during the production process, use a home textile fabric quality monitoring model to collect, process, and analyze images of the target home textile fabric, identify and separate the bottom layer and the top layer in the image. Quality monitoring includes feature analysis of the bottom layer and top layer images, judging the qualification degree, and generating a qualification degree vector. If there are quality problems in the monitoring results, adjust the equipment parameters and continuously conduct quality monitoring. Finally, update the equipment parameters according to the unqualified data through the deduction model to ensure the production quality and reduce the number of unqualified products.

Description

Technical Field

[0001] The present invention relates to the technical field of home textile fabric production quality monitoring, and specifically provides a method and system for home textile fabric production quality monitoring based on industrial data. Background Art

[0002] The production of home textile fabrics is a highly complex systematic project involving multiple links, usually including multiple complex processes such as spinning, weaving, and dyeing and finishing. In these links, various parameters such as temperature and humidity, equipment operating status, and raw material quality will significantly affect the quality of the final fabric. Therefore, strict quality monitoring must be implemented during the production process to ensure that each link strictly meets the predetermined quality standards.

[0003] However, traditional quality monitoring methods mainly focus on the inspection of the produced finished products, and this approach has certain limitations. For example, existing methods are difficult to detect potential problems in a timely manner during the production process and usually rely only on the later inspection of finished products to evaluate quality. Although this method helps to identify defects, it cannot detect problems in the early stage of production, thus unable to effectively guarantee the overall quality of home textile fabrics and resulting in an increased production rate of unqualified products.

[0004] To solve this problem, the present invention proposes a method and system for home textile fabric production quality monitoring based on industrial data, aiming to improve the quality control ability during the production process and the qualified rate of products. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for home textile fabric production quality monitoring based on industrial data, which is used to monitor the quality during the production process of home textile fabrics. The present invention mainly starts from the following aspects: First, by obtaining the planned finished product information and automatically adjusting the equipment parameters according to their difference degrees, the initial equipment settings are ensured to be accurate, thereby improving the production efficiency; Second, a home textile fabric quality monitoring model is used for real-time image processing and quality evaluation, including the separation and feature extraction of the bottom layer and the top layer, to ensure the quality control of each production link; Third, the fabric is carefully inspected through multi-level quality monitoring sub-layers (bottom layer, top layer, and combined quality monitoring), and unqualified items are timely discovered and marked to ensure the quality consistency of the final product; Finally, when problems are found in the monitoring results, the system can intelligently adjust the parameters of the production equipment for continuous quality monitoring to achieve the overall optimization and quality improvement of the home textile fabric production process.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for home textile fabric production quality monitoring based on industrial data, comprising:

[0008] Obtain the planned finished product information of the home textile fabric to be processed;

[0009] Furthermore, according to the planned finished product information, obtain the initial equipment parameters; wherein, the obtaining process of the initial equipment parameters includes:

[0010] Select a home textile fabric template from the template library according to the planned finished product information;

[0011] Furthermore, calculate the difference degree between the planned finished product information and the home textile fabric template;

[0012] When the difference degree is 0, use the standard equipment parameters of the home textile fabric template as the initial equipment parameters of the home textile fabric to be processed;

[0013] When the difference degree is not 0, calculate the deviation value between the home textile fabric to be processed and the home textile fabric template, and construct a home textile fabric deviation matrix;

[0014] According to the deviation values in the home textile fabric deviation matrix, obtain the equipment attributes of each production step;

[0015] Furthermore, establish an equipment-parameter mapping between the equipment attributes of the home textile fabric template and the standard equipment parameters;

[0016] Furthermore, obtain the standard equipment correction parameters through an equipment parameter correction function; wherein, the equipment parameter correction function is: ; wherein, is the standard equipment correction parameter; is the standard equipment parameter; is the mapping matrix corresponding to the equipment-parameter mapping; is the home textile fabric deviation matrix; is a non-linear correction function; is the production environment data;

[0017] Furthermore, use the standard equipment correction parameters as the initial equipment parameters.

[0018] Furthermore, set the equipment using the initial equipment parameters;

[0019] Furthermore, use the home textile fabric to be processed as the target object for production;

[0020] Furthermore, use a home textile fabric quality monitoring model to monitor the quality of the target object, including: an acquisition layer for acquiring images of the target object during the monitoring process;

[0021] The preprocessing layer is used to perform image processing on the target object image to obtain a standard target object image. Specifically, the preprocessing layer includes: segmenting the background of the target object image to obtain a first target object image; denoising the first target object image to obtain a second target object image; smoothing the second target object image to obtain a third target object image; performing edge detection on the third target object image to obtain a fourth target object image; and enhancing the fourth target object image to obtain the standard target object image.

[0022] The recognition and separation layer is used to recognize and separate the bottom layer and the top layer of the standard target object image.

[0023] The specific process includes: starting from the upper left corner of the standard target object image, moving horizontally through the recognition frame to obtain the recognition frame image and the recognition frame coordinates. The recognition frame coordinates use the center point of the standard target object image as the origin, and the coordinates are recorded as (0, 0).

[0024] Furthermore, extract the features of the recognition frame image to obtain a first feature vector.

[0025] Furthermore, input the first feature vector into the classification model to obtain a classification result. The classification result includes two classification types: the bottom layer and the top layer.

[0026] Furthermore, crop the standard target object image according to the classification result to obtain the bottom layer image and the top layer image.

[0027] The quality monitoring layer is used to monitor the quality of the bottom layer image and the top layer image. The quality monitoring layer includes: the bottom layer quality monitoring sub-layer, the top layer quality monitoring sub-layer, and the joint quality monitoring sub-layer. The specific process of the quality monitoring layer includes:

[0028] Input the bottom layer image into the bottom layer quality monitoring sub-layer to obtain the bottom layer features of the target object. The bottom layer features of the target object include: bottom layer specification features, bottom layer structure features, bottom layer color features, and bottom layer surface features.

[0029] Perform bottom layer quality analysis on the bottom layer features of the target object to obtain a bottom layer fabric qualification vector. The bottom layer fabric qualification vector includes: bottom layer specification qualification, bottom layer structure qualification, bottom layer color qualification, and bottom layer defect qualification. When any one of the qualification degrees in the bottom layer fabric qualification vector does not meet the bottom layer standard, mark the bottom layer unqualified type and obtain the corresponding unqualified bottom layer data.

[0030] Input the top-layer image into the top-layer quality monitoring sub-layer to obtain the top-layer features of the target object; wherein, the top-layer features of the target object include: top-layer shape features, top-layer color features, top-layer texture features, and top-layer integrity features; perform top-layer quality analysis on the top-layer features of the target object to obtain the top-layer fabric qualification vector; wherein, the top-layer fabric qualification vector includes: top-layer shape qualification, top-layer color qualification, top-layer texture qualification, and top-layer integrity qualification; when any one of the qualification degrees in the top-layer fabric qualification vector does not meet the top-layer standard, mark the top-layer unqualified type and obtain the corresponding unqualified top-layer data;

[0031] Input the bottom-layer image and the top-layer image into the joint quality monitoring sub-layer, and perform image superposition according to the standard target object image to obtain the joint image of the target object; extract the joint features of the joint image of the target object; wherein, the joint features include: position features, size features, and proportion features;

[0032] Calculate the expected error between the joint features and the current production expectation, and construct a production error vector; when the comprehensive error value of the production error vector exceeds the error tolerance, mark the joint unqualified type and obtain the corresponding unqualified joint data.

[0033] Output layer, used to output the monitoring results of the quality monitoring layer;

[0034] Further, when there is no quality problem in the monitoring result, use the initial equipment parameters of the current production step; when there is a quality problem in the monitoring result, adjust the initial equipment parameters of the current production step as equipment adjustment parameters and continuously perform quality monitoring; the specific process includes:

[0035] Obtain the unqualified type and the corresponding unqualified data of the target object;

[0036] Further, characterize the unqualified type and the unqualified data to obtain an unqualified feature vector;

[0037] Further, input the unqualified feature vector and the initial equipment parameters into a random forest model to obtain an equipment parameter influence vector;

[0038] Further, perform correlation analysis on the equipment parameter influence vector to obtain an equipment parameter correlation matrix;

[0039] Further, obtain the current production expectation, and obtain the current production expectation feature vector through characterization;

[0040] Further, input the unqualified feature vector, the initial equipment parameters, and the equipment parameter correlation matrix into the deduction model; when the error between the deduced unqualified feature vector and the current production expected feature vector is minimized, the initial equipment parameters are updated, and the equipment adjustment parameters for the current production step are obtained.

[0041] A home textile fabric production quality monitoring system based on industrial data, the system comprising:

[0042] A home textile fabric production plan generation unit for generating a production plan for the home textile fabric to be processed;

[0043] A template unit for storing templates of various home textile fabrics, as well as corresponding template information and standard equipment parameters;

[0044] An initial equipment parameter generation unit for generating initial parameters of each device in the production process of the home textile fabric to be processed; wherein, the initial equipment parameter generation unit:

[0045] Select a home textile fabric template from the template library according to the planned finished product information; calculate the difference degree between the planned finished product information and the home textile fabric template;

[0046] When the difference degree is 0, use the standard equipment parameters of the home textile fabric template as the initial equipment parameters of the home textile fabric to be processed;

[0047] When the difference degree is not 0, calculate the deviation value between the home textile fabric to be processed and the home textile fabric template, and construct a home textile fabric deviation matrix;

[0048] Further, obtain the device attributes of each production step according to the deviation values in the home textile fabric deviation matrix;

[0049] Further, establish a device-parameter mapping between the device attributes of the home textile fabric template and the standard equipment parameters;

[0050] Further, obtain standard equipment correction parameters through an equipment parameter correction function; wherein, the equipment parameter correction function is: ; wherein, is the standard equipment correction parameter; is the standard equipment parameter; is the mapping matrix corresponding to the device-parameter mapping; is the home textile fabric deviation matrix; is a non-linear correction function; is production environment data; use the standard equipment correction parameter as the initial equipment parameter;

[0051] A production unit for producing the home textile fabric to be processed;

[0052] A home textile fabric quality monitoring unit for monitoring the quality of the home textile fabric during the production process; wherein, the home textile fabric quality monitoring unit uses a home textile fabric quality monitoring model for quality monitoring, including:

[0053] Further, the home textile fabric to be processed in the production unit is used as the target object;

[0054] Further, an acquisition layer is used to obtain an image of the target object;

[0055] Further, the image of the target object is input into a preprocessing layer to obtain a standard target object image; wherein, the preprocessing layer includes: performing background segmentation on the image of the target object to obtain a first target object image; performing denoising on the first target object image to obtain a second target object image; performing smoothing processing on the second target object image to obtain a third target object image; performing edge detection on the third target object image to obtain a fourth target object image; performing image enhancement on the fourth target object image to obtain the standard target object image.

[0056] Further, the standard target object image is input into an identification and separation layer to obtain a bottom layer image and a top layer image; wherein, the identification and separation layer includes: starting from the upper left corner of the standard target object image with an identification frame and moving horizontally to obtain an identification frame image and identification frame coordinates; extracting the features of the identification frame image to obtain a first feature vector; inputting the first feature vector into a classification model to obtain a classification result; and cropping the standard target object image according to the classification result to obtain the bottom layer image and the top layer image.

[0057] Further, the bottom layer image and the top layer image are input into a quality monitoring layer to obtain a quality monitoring result; wherein, the specific process of the quality monitoring layer includes:

[0058] Inputting the bottom layer image into the bottom layer quality monitoring sublayer to obtain the bottom layer features of the target object; wherein, the bottom layer features of the target object include: bottom layer specification features, bottom layer structure features, bottom layer color features, and bottom layer surface features;

[0059] Performing bottom layer quality analysis on the bottom layer features of the target object to obtain a bottom layer qualification vector of the fabric; wherein, the bottom layer qualification vector of the fabric includes: bottom layer specification qualification, bottom layer structure qualification, bottom layer color qualification, and bottom layer defect qualification; when any one of the qualification degrees in the bottom layer qualification vector of the fabric does not meet the bottom layer standard, marking the bottom layer unqualified type and obtaining the corresponding unqualified bottom layer data;

[0060] Input the top layer image to the top layer quality monitoring sublayer to obtain the top layer features of the target object; wherein the top layer features of the target object include: top layer shape features, top layer color features, top layer texture features and top layer integrity features;

[0061] Performing a top-level quality analysis on the top-level features of the target object to obtain a top-level quality vector of the fabric; wherein the top-level quality vector of the fabric includes: top-level shape quality, top-level color quality, top-level texture quality and top-level integrity quality; when any quality item in the top-level quality vector of the fabric does not meet the top-level standard, marking the top-level unqualified type and obtaining the corresponding unqualified top-level data;

[0062] Inputting the bottom layer image and the top layer image into the joint quality monitoring sublayer, and performing image superposition according to the standard target object image to obtain a target object joint image;

[0063] Extracting joint features of the joint image of the target object; wherein the joint features include: position features, size features and scale features;

[0064] The expected error between the joint feature and the current production expectation is calculated to construct a production error vector; when the comprehensive error value of the production error vector exceeds the error tolerance, the joint unqualified type is marked, and the corresponding unqualified joint data is obtained.

[0065] Furthermore, the quality monitoring result is outputted using an output layer.

[0066] Equipment parameter optimization unit, used to optimize the initial parameters of the equipment;

[0067] The feedback unit is used to provide real-time feedback on the quality monitoring results of the home textile fabrics to be processed and the equipment parameter adjustment status.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] 1. This method selects a suitable home textile fabric template from the template library, calculates the difference degree and deviation matrix according to the planned finished product information, and precisely adjusts the production equipment by combining equipment parameter mapping and correction functions, so as to achieve precise control of equipment parameters, optimize the production process, and ensure the production quality of home textile fabrics. When the difference degree is 0, the standard equipment parameters of the home textile fabric template are directly adopted; when the difference degree is not 0, the deviation value is calculated and a home textile fabric deviation matrix is constructed, and the equipment parameters are further adjusted through equipment-parameter mapping and correction functions to ensure that the equipment can adapt to the production requirements of different fabrics. In addition, production environment data (such as temperature, humidity, etc.) as part of the correction function can further improve the accuracy of parameter adjustment, ensure the stability and efficiency of the equipment in a changing environment, and thus effectively improve the quality and qualification rate of home textile fabrics.

[0070] 2. The present invention proposes a quality monitoring model for home textile fabrics, which is used to monitor the quality in the production process of home textile fabrics. This model generates a standardized target image through multi-level image processing and feature extraction; classifies the image through an identification separation layer, extracts bottom-layer and top-layer features, and combines bottom-layer, top-layer, and joint quality monitoring sub-layers to comprehensively analyze key indicators such as the specifications, structure, color, surface features, shape, texture, and integrity of the fabric. Finally, according to the monitoring results, the quality qualification vector and production error of the fabric are output, unqualified items are identified in a timely manner, and their types are marked, so as to achieve precise quality monitoring of home textile fabrics and provide strong analysis support for quality control in subsequent production processes and the production of high-quality home textile fabrics.

[0071] 3. The present invention proposes a method for adjusting equipment parameters in the production process of home textile fabrics according to quality monitoring results. Through a series of precise analysis steps, this method first obtains the unqualified type and data of the target object, characterizes them to generate an unqualified feature vector; then, in the random forest model, the unqualified feature vector and initial equipment parameters are input to obtain an equipment parameter influence vector; then, a correlation analysis ， is performed to generate an equipment parameter correlation matrix; after the current production expectation is characterized, it is used as an optimization target to input into the deduction model to update the initial equipment parameters, thereby significantly improving the product quality of home textile fabrics and reducing the defective rate. Description of the Drawings

[0072] Figure 1 is a flowchart of a method for monitoring the production quality of home textile fabrics based on industrial data provided by an embodiment of the present invention;

[0073] Figure 2 is a structural diagram of a system for monitoring the production quality of home textile fabrics based on industrial data provided by an embodiment of the present invention;

[0074] Figure 3 Flow chart for obtaining initial device parameters provided by an embodiment of the present invention;

[0075] Figure 4 Schematic diagram of an identification box for acquiring an image provided by an embodiment of the present invention. Detailed implementation manners

[0076] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0077] Home textile fabrics are fabrics specifically used for manufacturing household textiles (such as bedding, curtains, sofa covers, carpets, etc.). With the continuous improvement of consumers' requirements for the comfort, functionality, and aesthetics of the home environment, the demand for home textile fabrics and their quality standards have also increased accordingly.

[0078] With the increasingly strict requirements for the production process and quality control of home textile fabrics, quality control has become a crucial link, which directly affects the performance and reliability of products. Therefore, in the production process of home textile fabrics, every link must be strictly controlled. However, traditional quality monitoring methods mainly focus on finished product inspection, and this approach has certain limitations. Existing methods are difficult to detect potential problems in a timely manner and usually rely on post-production inspection to judge quality. Although this method can identify defects, it cannot detect problems in advance during the production process, resulting in an increase in the production rate of defective products.

[0079] To solve the deficiencies of the existing technology, the present invention proposes a method and system for monitoring the production quality of home textile fabrics based on industrial data. The method and system aim to achieve early warning of the quality of home textile fabrics and optimize device parameters by real-time monitoring of various data during the production process, thereby improving the production quality.

[0080] The following will detail the specific applications of the present invention through two embodiments.

[0081] Embodiment 1:

[0082] In the embodiment of the present application, the production quality of home textile fabric A is monitored; among them, home textile fabric A is a pure cotton bedding fabric; the method and system of the present invention are used to monitor home textile fabric A; as Figure 1 shown, in Figure 1The specific implementation process of the method of the present invention is given in [reference], including: S10. Obtaining the planned finished product information of the home textile fabric to be processed; S20. Obtaining the initial equipment parameters according to the planned finished product information; S30. Setting the equipment using the initial equipment parameters; S40. Using the home textile fabric to be processed as the target object for production; S50. Monitoring the quality of the target object using the home textile fabric quality monitoring model; S60. Judging whether there are quality problems in the monitoring results; In Figure 2 The system structure of the present invention is given in [reference], including: a home textile fabric production plan generation unit, a template unit, an initial equipment parameter generation unit, a production unit, a home textile fabric quality monitoring unit, an equipment parameter optimization unit, and a feedback unit; The examples of the present application will be described in detail in combination with Figure 1 and Figure 2 ;

[0083] According to the home textile fabric production plan generation unit of the system, the planned finished product information of home textile fabric A is obtained, corresponding to Figure 1 step S10 in [reference]; Among them, the planned finished product information includes: production specifications, models, quantities, batches, and process requirements, etc.;

[0084] Furthermore, using the initial equipment parameter generation unit of the system, according to the planned finished product information, the initial equipment parameters of each device in the production process of home textile fabric A are obtained, corresponding to step S20; Among them, referring to Figure 3 , the specific implementation process of the initial equipment parameter generation unit includes:

[0085] According to the planned finished product information, the corresponding home textile fabric template is matched from the template library managed by the template unit; Among them, the selection of the home textile fabric template is mainly through keyword matching of the planned finished product information and the template finished product information, and at least includes: keywords such as models and process requirements during the matching process;

[0086] Furthermore, calculate the difference degree between the planned finished product information and the home textile fabric template; Among them, the difference degree refers to the difference value of the production specifications in the planned finished product information of home textile fabric A and the matching home textile fabric template in terms of size and production process; By comparing the difference degree of home textile fabric A and the home textile fabric template regarding size;

[0087] For example, the calculation formula of the size difference degree is: ; Among them, is the planned finished product size; is the template size; The calculation formula of the production process difference degree is: ; Among them, is the total number of production process items; is the th production process vector of the planned finished product; is the a production process vector;

[0088] When the difference degree is 0, the standard equipment parameters corresponding to the selected home textile fabric template are used as the initial equipment parameters for the production of home textile fabric A;

[0089] When the difference degree is not 0, calculate the deviation value between home textile fabric A and the home textile fabric template, and construct a home textile fabric deviation matrix; wherein, the deviation value refers to the deviation value of each item between the planned finished product information and the template finished product information, including: size deviation value, production process deviation value, fabric property deviation value, etc.; the deviation values listed above are only partial descriptions of the embodiments of the present application, and the deviation calculation can be specifically carried out according to the content of the finished product information.

[0090] Further, according to the deviation values in the home textile fabric deviation matrix, obtain the equipment attributes of each production step; wherein, the equipment attribute refers to the equipment name;

[0091] Further, establish an equipment-parameter mapping relationship between the equipment attributes in the home textile fabric template and the standard equipment parameters. For the specific situation of the equipment-parameter mapping, please refer to the partial mapping column description given in Table 1.

[0092] Table 1 Partial Examples of Equipment-Parameter Mapping

[0093]

[0094] Further, input the home textile fabric deviation matrix and the equipment-parameter mapping into the equipment parameter correction function; wherein, the equipment parameter correction function is: ; wherein, is the standard equipment correction parameter; is the standard equipment parameter; is the mapping matrix corresponding to the equipment-parameter mapping; is the home textile fabric deviation matrix; is the non-linear correction function; is the production environment data;

[0095] Among them, the non-linear correction function is expressed as: ; wherein, is the non-linear relationship between the equipment parameter and the fabric deviation; is the non-linear relationship between the equipment parameter and the fabric deviation;

[0096] In the linear correction function ; wherein, is the standardization operation; and are constants used to adjust the strength of the non-linear relationship; ; wherein, , and is a constant used to adjust the impact of production environment data on the correction process;

[0097] Among them, 、 、 、 、 and can be obtained by fitting historical data;

[0098] Furthermore, the standard equipment correction parameters are used as the initial equipment parameters.

[0099] Taking the content of Table 1 as an example, after being corrected by the equipment parameter correction function, the weaving density deviation of the loom is +2 pieces / cm. According to the equipment-parameter mapping relationship, the standard weaving density of the loom is 80 pieces / cm. After a deviation of +2 pieces / cm, the adjusted weaving density should be 82 pieces / cm. The dye concentration deviation of the dyeing machine is +1%. The original standard dye concentration is 5%, and it should be adjusted to 6% after correction.

[0100] The embodiment of the present application adopts a method to set the initial parameter settings for each production equipment of home textile fabric A; this method selects a home textile fabric template from the template library and adjusts the equipment parameters according to the difference degree between the planned finished product information and the template. When the difference degree is 0, the standard equipment parameters of the template are directly adopted; when the difference degree is not 0, the deviation value is calculated and a deviation matrix is constructed, and the adjustment is carried out through the equipment parameter correction function, and finally the initial equipment parameters suitable for home textile fabric A are obtained. It can dynamically adjust the equipment parameters according to specific production requirements, improve the accuracy and efficiency of the production process, and ensure the consistency of the final product quality.

[0101] Furthermore, according to the content of step S30, set the initial equipment parameters of each equipment on the production line;

[0102] Furthermore, according to the content of step S40, take home textile fabric A as the target object and produce it by the production unit of the system;

[0103] Furthermore, during the production process of the production unit, use the home textile fabric quality monitoring unit of the system for monitoring, corresponding to the content of S50; among them, the home textile fabric quality monitoring unit uses a home textile fabric quality monitoring model for quality monitoring. The specific process includes: using the image of the target object currently being produced after the acquisition layer; among them, the acquisition layer uses high-definition monitoring equipment to collect images of the production process of home textile fabric A in real time;

[0104] Further, input the target object image into the preprocessing layer to obtain a standard target object image; wherein, the preprocessing layer includes: performing background segmentation on the target object image to obtain a first target object image; denoising the first target object image to obtain a second target object image; smoothing the second target object image to obtain a third target object image; performing edge detection on the third target object image to obtain a fourth target object image; and enhancing the fourth target object image to obtain a standard target object image.

[0105] In the embodiment of the present application, the preprocessing layer is used to process the acquired image, including: background segmentation, denoising, smoothing, edge detection, and image enhancement; through the above methods, noise can be removed, image details can be enhanced, key features can be extracted, and the accuracy of the monitoring model can be improved, so as to more accurately identify unqualified products.

[0106] Further, input the standard target object image into the recognition and separation layer to obtain a bottom layer image and a top layer image; refer to Figure 4 , wherein, the recognition and separation layer includes: starting from the upper left corner of the standard target object image with a recognition frame, moving horizontally to obtain a recognition frame image and recognition frame coordinates; wherein, the recognition frame adopts the sliding window technique; in the embodiment of the present application, the size of the recognition frame is set to 64×64 pixels, the moving interval each time is set to 32 pixels, and it moves every 2 frames;

[0107] Extract the features of the recognition frame image to obtain a first feature vector;

[0108] Input the first feature vector into the classification model to obtain a classification result; wherein, the classification model is a trained convolutional neural network (CNN), which is used to perform binary classification operations on the images in the recognition frame, and its classification results include: two types, namely the bottom layer and the top layer;

[0109] Crop the standard target object image according to the classification result to obtain a bottom layer image and a top layer image.

[0110] In the embodiment of the present application, the classification results of multiple rounds show that the parameters set for the recognition frame above are the optimal values; wherein, during this process, the present invention also sets multiple recognition frame parameters, such as: (1) recognition frame size: 64×64 pixels, moving interval each time: 32 pixels, update frequency: update per frame; (2) recognition frame size: 64×64 pixels, moving interval each time: 64 pixels, update frequency: update every 2 frames; (3) recognition frame size: 64×64 pixels, moving interval each time: 128 pixels, update frequency: update every 3 frames; (4) recognition frame size: 128×128 pixels, moving interval each time: 32 pixels, update frequency: update per frame.

[0111] Referring to Table 2, according to the parameters set for the recognition frame, the comparison of classification accuracies obtained by the classification model is as follows:

[0112] Table 2 Comparison of Classification Accuracies for Different Recognition Frame Parameters

[0113]

[0114] Recognition frame size: 64×64 pixels, moving interval 64 pixels, updated every 2 frames (accuracy: 88.5%). This is the setting with the highest classification accuracy in the table. Compared with the first setting, the moving interval is increased to 64 pixels and updated every 2 frames, which means that the step size of each slide increases, but the update frequency decreases. The advantage of this is that it reduces the computational burden and can still capture the key features of the image well. The results show that this setting achieves a classification accuracy of 88.5%, which is considered the best parameter combination.

[0115] The last setting uses a larger recognition frame (128×128 pixels) and adopts per-frame update and a smaller moving interval (32 pixels). Although increasing the size of the recognition frame helps to capture more global information, the corresponding computational amount also increases significantly. A larger recognition frame will result in higher computational and memory consumption, making the model processing speed slower, and the high frequency of per-frame update fails to improve the classification accuracy, ultimately leading to the accuracy dropping to 80.7%.

[0116] Furthermore, the bottom-layer image and the top-layer image are input into the quality monitoring layer to obtain the quality monitoring result; among them, the quality monitoring layer includes: a bottom-layer quality monitoring sub-layer, a top-layer quality monitoring sub-layer, and a combined quality monitoring sub-layer; among them, the specific process of the quality monitoring layer includes: inputting the bottom-layer image into the bottom-layer quality monitoring sub-layer to obtain the bottom-layer features of the target object; among them, the bottom-layer features of the target object include: bottom-layer specification features, bottom-layer structure features, bottom-layer color features, and bottom-layer surface features;

[0117] Conduct bottom-layer quality analysis on the bottom-layer features of the target object to obtain the bottom-layer fabric qualification vector; among them, the bottom-layer fabric qualification vector includes: bottom-layer specification qualification, bottom-layer structure qualification, bottom-layer color qualification, and bottom-layer defect qualification; when any one of the qualification degrees in the bottom-layer fabric qualification vector does not meet the bottom-layer standard, mark the bottom-layer unqualified type and obtain the corresponding unqualified bottom-layer data;

[0118] Input the top-layer image into the top-layer quality monitoring sub-layer to obtain the top-layer features of the target object; among them, the top-layer features of the target object include: top-layer shape features, top-layer color features, top-layer texture features, and top-layer integrity features;

[0119] Perform top - level quality analysis on the top - level features of the target object to obtain the fabric top - level qualification vector. Among them, the fabric top - level qualification vector includes: top - level shape qualification, top - level color qualification, top - level texture qualification, and top - level integrity qualification. When any one of the qualification degrees in the fabric top - level qualification vector does not meet the top - level standard, mark the top - level unqualified type and obtain the corresponding unqualified top - level data.

[0120] Input the bottom - layer image and the top - layer image into the joint quality monitoring sub - layer, and perform image superposition according to the standard target object image to obtain the target object joint image.

[0121] Extract the joint features of the target object joint image. Among them, the joint features include: position features, size features, and proportional features.

[0122] Calculate the expected error between the joint features and the current production expectation to construct a production error vector. When the comprehensive error value of the production error vector exceeds the error tolerance, mark the joint unqualified type and obtain the corresponding unqualified joint data. Among them, by analyzing the fluctuation range and distribution of errors in historical data, a common error range can be obtained. This range can be used to set the error tolerance.

[0123] In the embodiment of the present application, by using the quality monitoring layer to perform multi - level quality monitoring on the bottom - layer, top - layer, and joint images of the home textile fabric A, the specifications, structures, colors, textures, and other features of the fabric can be comprehensively evaluated, unqualified items can be detected in a timely manner, and detailed unqualified parameters can be obtained, which helps to ensure that the product meets the production standards and improve the accuracy of the production process and product quality.

[0124] Furthermore, use the output layer to output the quality monitoring results. Among them, based on the monitoring results of the quality monitoring layer, "bottom - layer structure unqualified" and "top - layer color unqualified" are obtained.

[0125] Furthermore, use the device parameter optimization unit to optimize the initial device parameters of the current production step, corresponding to step S60. During the optimization process, it is necessary to judge whether optimization is required according to the monitoring results. When there are no quality problems in the monitoring results, the initial device parameters of the current production step remain unchanged; when there are quality problems in the monitoring results, adjust the initial device parameters of the current production step as device adjustment parameters and continuously perform quality monitoring.

[0126] Adjust the initial device parameters of the current production step according to the results of the home textile fabric quality monitoring model. The specific process is as follows: obtain the unqualified type of the target object and the corresponding unqualified data; refer to Table 3, and the corresponding data description is given in Table 3.

[0127] Table 3 Explanation of Quality Monitoring Results

[0128]

[0129] Further, characterize the unqualified type and the unqualified data to obtain an unqualified feature vector;

[0130] Further, input the unqualified feature vector and the initial device parameters into a random forest model to obtain a device parameter influence vector;

[0131] Further, perform a correlation analysis on the device parameter influence vector to obtain a device parameter correlation matrix; wherein, the device parameter correlation matrix is calculated using the Pearson correlation coefficient analysis method;

[0132] Further, obtain the current production expectation and obtain a current production expectation feature vector through characterization;

[0133] Further, input the unqualified feature vector, the initial device parameters, and the device parameter correlation matrix into a deduction model; wherein, the deduction model is a machine learning algorithm that automatically identifies the complex relationship between device parameters and product quality based on a large amount of historical production data and device operation data. The model can analyze the gap between the unqualified feature vector and the production expectation in real time and adjust the initial device settings according to the device parameter correlation matrix. By continuously optimizing these input parameters, the deduction model can predict and optimize the device parameters;

[0134] Define the function of the deduction model as: ; wherein, is the deduction model function; is the initial device parameter of the current production step ; is the unqualified feature vector; is the device parameter correlation matrix; is the current production step Deduce the unqualified feature vector; is the current production expectation feature vector.

[0135] When the error between the deduced unqualified feature vector and the current production expectation feature vector is minimized, the update of the initial device parameters is completed, and the device adjustment parameters for the current production step are obtained.

[0136] In the embodiment of the present application, the device parameters of the current production step are adjusted according to the quality monitoring results. By analyzing the unqualified data, extracting the unqualified feature vectors, and inputting them together with the initial device parameters into the random forest model, the device parameter influence vectors are obtained. Then, the device parameter correlation analysis is carried out to construct the device parameter correlation matrix. By performing a characteristic comparison with the current production expectations, the deduction model optimizes the device parameters until the error between the unqualified feature vectors and the expected feature vectors is minimized, thereby obtaining the updated device adjustment parameters, which can greatly improve the quality, stability, and efficiency of production.

[0137] Further, the feedback unit of the system is used to feedback the monitoring results and the device parameter adjustment situation during the production process.

[0138] In the embodiment of the present application, the device parameters in the production process of home textile fabric A are dynamically adjusted and quality monitored; specifically including: First, by selecting a suitable home textile fabric template from the template library and correcting the production parameters according to the difference degree and deviation matrix, the production requirements can be accurately matched with the device configuration, avoiding production errors caused by improper device parameter settings, and ensuring that each link in the production process operates under the best parameters. Second, the quality monitoring model carefully analyzes the underlying and top-layer features of the home textile fabric through image processing and machine learning algorithms, so as to accurately identify possible quality problems in the production process. By performing quality analysis on multi-dimensional features (such as color, structure, shape, etc.) of the underlying and top layers, defects can be detected in the early stage of production, and corrected by adjusting the device parameters, avoiding the production of a large number of defective products, and improving the consistency and reliability of the products. In addition, the adaptive capabilities of the random forest model and the deduction model in the embodiment of the present application under quality anomalies can flexibly adjust the device parameters according to the deviation information in the actual production process to meet different production requirements. When there are quality problems in the monitoring results, the model automatically optimizes the device parameters by analyzing the correlation between the unqualified feature vectors and the device parameters, and feeds the optimized parameters back to the production system, thereby realizing continuous quality monitoring and adjustment, forming a closed-loop control.

[0139] Embodiment 2:

[0140] In Embodiment 1, the quality monitoring and the control of the production device parameters of home textile fabric A during the production process are realized; in order to illustrate that the present invention has good adaptability to the production monitoring of different types of home textile fabrics; the embodiment of the present application will describe the process of home textile fabric B as follows:

[0141] Obtain the planned finished product information of home textile fabric B;

[0142] Further, according to the planned finished product information, obtain the initial equipment parameters; wherein, the difference degree of selecting the home textile fabric template from the template library according to the finished product information is 0; therefore, use the standard equipment parameters of the home textile fabric template as the initial equipment parameters of home textile fabric B.

[0143] Further, use the home textile fabric quality monitoring model to monitor the quality of the target object; according to the quality monitoring results, it can be known that there is a quality problem of "inconsistent thickness" in the weaving step of home textile fabric B, and obtain the unqualified thickness data of home textile fabric B in this step.

[0144] Further, define the unqualified type of home textile fabric B as "inconsistent thickness" and the corresponding unqualified data.

[0145] Further, characterize the unqualified type and unqualified data to obtain an unqualified feature vector.

[0146] Further, input the unqualified feature vector and the initial equipment parameters into the random forest model to obtain an equipment parameter influence vector.

[0147] Further, conduct a correlation analysis on the equipment parameter influence vector to obtain an equipment parameter correlation matrix.

[0148] Further, obtain the current production expectation, and obtain the current production expectation feature vector through characterization.

[0149] Further, input the unqualified feature vector, the initial equipment parameters and the equipment parameter correlation matrix into the deduction model; when the error between the unqualified feature vector and the current production expectation feature vector is the smallest, the update of the initial equipment parameters is completed, and the equipment adjustment parameters for the current production step are obtained.

[0150] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirits of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring the production quality of home textile fabrics based on industrial data, characterized in that, Including: Obtain the planned finished product information of the home textile fabric to be processed; Obtain the initial equipment parameters according to the planned finished product information; Set the equipment using the initial equipment parameters; Use the home textile fabric to be processed as the target object for production; Adopt a home textile fabric quality monitoring model to monitor the quality of the target object, including: an acquisition layer for acquiring the target object image during the monitoring process; a preprocessing layer for performing image processing on the target object image to obtain a standard target object image; An identification and separation layer for identifying and separating the bottom layer and the top layer of the standard target object image. The specific process includes: starting from the upper left corner of the standard target object image with an identification frame, moving horizontally to obtain the identification frame image and the identification frame coordinates; extracting the features of the identification frame image to obtain a first feature vector; inputting the first feature vector into a classification model to obtain a classification result; cropping the standard target object image according to the classification result to obtain the bottom layer image and the top layer image; A quality monitoring layer: used to monitor the quality of the bottom layer image and the top layer image; An output layer for outputting the monitoring results of the quality monitoring layer; When there is no quality problem in the monitoring result, keep the initial equipment parameters of the current production step unchanged; when there is a quality problem in the monitoring result, adjust the initial equipment parameters of the current production step as equipment adjustment parameters and continuously perform quality monitoring; The process of obtaining the initial equipment parameters includes: selecting a home textile fabric template from the template library according to the planned finished product information; calculating the difference degree between the planned finished product information and the home textile fabric template; When the difference degree is 0, use the standard equipment parameters of the home textile fabric template as the initial equipment parameters of the home textile fabric to be processed; When the difference degree is not 0, calculate the deviation value between the home textile fabric to be processed and the home textile fabric template, and construct a home textile fabric deviation matrix; obtain the equipment attributes of each production step according to the deviation values in the home textile fabric deviation matrix; Establish the equipment-parameter mapping between the equipment attributes of the home textile fabric template and the standard equipment parameters; obtain the standard equipment correction parameters through the equipment parameter correction function; The quality monitoring layer includes: a bottom layer quality monitoring sub-layer, a top layer quality monitoring sub-layer and a joint quality monitoring sub-layer; among them, the specific process of the quality monitoring layer includes: inputting the bottom layer image into the bottom layer quality monitoring sub-layer to obtain the bottom layer features of the target object; performing bottom layer quality analysis on the bottom layer features of the target object to obtain a fabric bottom layer qualification vector; when any one of the qualification degrees in the fabric bottom layer qualification vector does not meet the bottom layer standard, mark the bottom layer unqualified type and obtain the corresponding unqualified bottom layer data; Input the top layer image into the top layer quality monitoring sub-layer to obtain the top layer features of the target object; perform top layer quality analysis on the top layer features of the target object to obtain a fabric top layer qualification vector; when any one of the qualification degrees in the fabric top layer qualification vector does not meet the top layer standard, mark the top layer unqualified type and obtain the corresponding unqualified top layer data; Input the underlying image and the top - layer image into the joint quality monitoring sub - layer, perform image superposition according to the standard target object image to obtain the target object joint image; extract the joint features of the target object joint image; calculate the expected error between the joint features and the current production expectation to construct a production error vector; when the comprehensive error value of the production error vector exceeds the error tolerance, mark the joint non - qualified type and obtain the corresponding non - qualified joint data.

2. The method for monitoring the production quality of home textile fabrics based on industrial data according to claim 1, wherein, The device parameter correction function is as follows: Among them, is the standard device correction parameter; is the standard device parameter; is the mapping matrix corresponding to the device - parameter mapping; is the home textile fabric deviation matrix; is the non - linear correction function; is the production environment data; Take the standard device correction parameter as the initial device parameter.

3. A method for monitoring the production quality of home textile fabrics based on industrial data according to claim 1, characterized in that, The pre - processing layer includes: performing background segmentation on the target object image to obtain the first target object image; denoising the first target object image to obtain the second target object image; smoothing the second target object image to obtain the third target object image; performing edge detection on the third target object image to obtain the fourth target object image; performing image enhancement on the fourth target object image to obtain the standard target object image.

4. A method for monitoring the production quality of home textile fabrics based on industrial data according to claim 1, characterized in that, When there are quality problems in the monitoring results, the adjustment process of the initial equipment parameters for the current production step includes: Obtain the non - qualified type and the corresponding non - qualified data of the target object; characterize the non - qualified type and the non - qualified data to obtain a non - qualified feature vector; input the non - qualified feature vector and the initial equipment parameters into a random forest model to obtain an equipment parameter influence vector; perform correlation analysis on the equipment parameter influence vector to obtain an equipment parameter correlation matrix; obtain the current production expectation and obtain the current production expectation feature vector through characterization; input the non - qualified feature vector, the initial equipment parameters and the equipment parameter correlation matrix into a deduction model; when the error between the deduced non - qualified feature vector and the current production expectation feature vector is the smallest, the update of the initial equipment parameters is completed, and the equipment adjustment parameters for the current production step are obtained.

5. A home textile fabric production quality monitoring system based on industrial data, which adopts a home textile fabric production quality monitoring method based on industrial data as described in claim 1, is characterized in that, The system includes: a home textile fabric production plan generation unit for generating a production plan for the to - be - processed home textile fabric; a template unit for storing templates of various home textile fabrics, as well as corresponding template information and standard equipment parameters; an initial equipment parameter generation unit for generating initial parameters of each device during the production of the to - be - processed home textile fabric; a production unit for producing the to - be - processed home textile fabric; a home textile fabric quality monitoring unit for monitoring the quality during the production of the to - be - processed home textile fabric; an equipment parameter optimization unit for optimizing the initial parameters of the equipment; and a feedback unit for real - time feedback of the quality monitoring results and equipment parameter adjustment conditions of the to - be - processed home textile fabric.

Images