A printing and dyeing detection method for textile products

Abstract

The present invention discloses a printing and dyeing detection method for textile products, and relates to the technical field of textile detection. The method comprises: mining spectral features of multiple dyeing solutions, building a printing and dyeing feature library; reading the printing and dyeing data stream of the target printing and dyeing product; pre-processing the printing and dyeing data stream, transmitting it to the printing and dyeing detection module, performing parallel independent analysis of blocks and lateral interactive verification, integrating and determining a printing and dyeing detection column; traversing the printing and dyeing detection column, performing defect coverage and defect generation identification and location tracing under composite printing and dyeing, integrating and determining a composite point detection column; integrating and mapping the printing and dyeing detection column and the composite point detection column, and generating a printing and dyeing detection result. The present invention solves the technical problem that the accuracy of the detection result is low due to the limitations of the existing printing and dyeing detection methods, and achieves the technical effect of combining spectral data and process data for parallel defect detection, accurately identifying and locating printing and dyeing defects, and improving the efficiency and accuracy of product quality detection.

Description

Technical Field

[0001] The invention relates to the technical field of textile detection, and in particular to a printing and dyeing detection method for textile products. Background Art

[0002] The printing and dyeing quality of textiles is directly related to the aesthetics, durability and market acceptance of the products. As consumers' requirements for textile quality increase, printing and dyeing testing has become a key link in ensuring the quality of textiles. Commonly used testing methods include spectrophotometry, electrochemical method, chromatography, gas chromatography-mass spectrometry, etc. However, the existing printing and dyeing testing methods all have their own limitations. For example, the chromatography equipment is relatively complex, the technical requirements for the testing personnel are high, the testing time is long, and the spectrophotometry method is easily interfered with and affects the accuracy, which leads to a decrease in the accuracy of the test results. Summary of the invention

[0003] The present application provides a printing and dyeing detection method for textile products, which is used to solve the technical problems that the existing printing and dyeing detection methods have their own limitations and the accuracy of the detection results is low.

[0004] In a first aspect of the present application, a method for detecting printing and dyeing of textile products is provided, the method comprising: mining spectral features of a plurality of dyeing solutions, building a printing and dyeing feature library, the spectral features comprising a single dyeing solution dimension and a mixed dyeing solution dimension; reading a printing and dyeing data stream of a target printing and dyeing product, wherein the printing and dyeing method of the target printing and dyeing product is single-layer printing and dyeing or composite printing and dyeing, and the printing and dyeing data stream comprises a spectral data stream and a dynamic process data stream based on spectrometer detection; building a printing and dyeing detection module, the printing and dyeing detection module comprising a spectral analysis block and a process analysis block configured in parallel, and a defect location unit connected to a rear end, the spectral analysis block being embedded with the printing and dyeing feature library; pre-processing the printing and dyeing data stream, transmitting the spectral analysis block to the printing and dyeing detection module, performing block parallel independent analysis and lateral interactive verification, locating and verifying printing and dyeing defect points that meet the verification standards, and integrating and determining a printing and dyeing detection single column; traversing the printing and dyeing detection single column, performing defect coverage and defect generation identification and location tracing under composite printing and dyeing, and integrating and determining a composite point detection single column; integrating and mapping the printing and dyeing detection single column and the composite point detection single column, and generating a printing and dyeing detection result.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0006] The present application provides a printing and dyeing detection method for textile products, which relates to the technical field of textile detection. Through spectral feature mining, the printing and dyeing data stream of the target printing and dyeing product is read, and parallel independent analysis and lateral interactive verification of blocks are performed, and a printing and dyeing detection column is integrated to determine the defect coverage and defect generation identification and location tracing under composite printing and dyeing, and a composite point detection column is integrated and determined. The printing and dyeing detection column and the composite point detection column are integrated and mapped to generate printing and dyeing detection results, which solves the technical problems that the existing printing and dyeing detection methods have their own limitations and the accuracy of the detection results is low. By combining spectral data and process data for parallel defect detection, comprehensive detection of the printing and dyeing quality of textile products is achieved, and the technical effect of improving the efficiency and accuracy of product quality detection is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 A schematic diagram of a process flow of a printing and dyeing detection method for a textile product provided in an embodiment of the present application;

[0009] Figure 2 A schematic diagram of a process for determining effective printing and dyeing data in a printing and dyeing detection method for a textile product provided in an embodiment of the present application;

[0010] Figure 3 A schematic diagram of a flow chart for determining a composite point detection column in a printing and dyeing detection method for a textile product provided in an embodiment of the present application. DETAILED DESCRIPTION

[0011] The present application provides a printing and dyeing detection method for textile products, which is used to solve the technical problems that the existing printing and dyeing detection methods have their own limitations and the accuracy of the detection results is low.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Embodiment 1

[0014] like Figure 1 As shown, the present application provides a printing and dyeing detection method for textile products, the method comprising:

[0015] P10: Spectral feature mining of various dye solutions is performed to build a dyeing and printing feature library. The spectral features include single dye solution dimension and mixed dye solution dimension.

[0016] Furthermore, step P10 of the embodiment of the present application further includes:

[0017] P11: Using the plurality of dye solutions as indexes, an engine search is performed to determine dye solution printing and dyeing records;

[0018] P12: Based on the dyeing records, mining spectral features and superimposed dyeing features;

[0019] P13: Based on the dyeing record, mining the implicit relationship of features, wherein the implicit relationship of features is the relative relationship between spectral features and printing and dyeing features;

[0020] P14: Integrate the spectral features and associate them with the implicit relationship of the features and the superimposed printing and dyeing features to construct the printing and dyeing feature library.

[0021] It should be understood that by mining and analyzing the spectral characteristics of multiple dyeing solutions, a printing and dyeing feature library including the spectral characteristics of single dyeing solutions and the spectral characteristics of mixed dyeing solutions is established. Specifically, an internal search engine or similar search tool of a printing and dyeing enterprise is used, and multiple dyeing solutions are used as keywords or indexes to perform engine search to obtain printing and dyeing records of multiple dyeing solutions, and the dyeing and dyeing records should include detailed information of the dyeing solutions, such as type, concentration, manufacturer, etc., as well as printing and dyeing process parameters, printing and dyeing results, etc.

[0022] Further, based on the dyeing records, the spectral characteristics and superimposed dyeing characteristics of multiple dyeing liquids are excavated, wherein the spectral characteristics are such as absorption peaks, emission peaks, spectral shapes, etc., and the superimposed dyeing characteristics refer to the dyeing characteristics of the mixed dyeing liquid, that is, when multiple dyeing liquids are mixed, the interaction between them affects the dyeing effect. Further, according to the dyeing records, the implicit relationship of the characteristics is excavated, that is, the relative relationship between the spectral characteristics and the dyeing characteristics, including analyzing the correlation between the intensity of the spectral characteristics and the dyeing effect, the influence of different spectral feature combinations on the dyeing effect, etc. For example, by measuring the characteristics of the dyeing liquid at different wavelengths, such as light absorption or reflection intensity, presenting a spectrum, analyzing the dyeing condition of the dyeing liquid, such as the relative content of various dyes in the mixed dyeing liquid, the dyeing interaction between dyes, the combination mode between dyes and textile product fibers, etc., to obtain the implicit relationship of the characteristics.

[0023] Furthermore, the spectral features, the feature implicit relationships, and the superimposed printing and dyeing features are associated and integrated to establish the printing and dyeing feature library. The printing and dyeing feature library not only includes the spectral characteristics of various dye solutions, but also includes the associations between different spectral characteristics and printing and dyeing effects, which can provide important reference and basis for subsequent printing and dyeing detection.

[0024] P20: Read the printing and dyeing data stream of the target printing and dyeing product, wherein the printing and dyeing method of the target printing and dyeing product is single-layer printing and dyeing or composite printing and dyeing, and the printing and dyeing data stream includes a spectral data stream based on spectrometer detection and a dynamic process data stream.

[0025] Optionally, the printing and dyeing data stream of the target printing and dyeing product, that is, the printing and dyeing related data of the target printing and dyeing product, is read. The target printing and dyeing product refers to a specific product that needs to be tested for printing and dyeing, such as various textiles, such as clothing, cloth, curtains, etc. In addition, the printing and dyeing method of the target printing and dyeing product is single-layer printing and dyeing or composite printing and dyeing. Single-layer printing and dyeing refers to only a single-layer printing and dyeing process, while the composite printing and dyeing may include a multi-layer printing and dyeing process, each layer is printed and dyed based on a single dye liquid or a mixed dye liquid, and the hierarchical printing and dyeing may be different areas or overlapping printing and dyeing, involving the use of multiple dye liquids, and through different printing and dyeing processes and technologies, complex patterns and colors are formed on textiles.

[0026] Furthermore, the printing and dyeing data stream includes a spectral data stream based on spectrometer detection and a dynamic process data stream. The spectral data stream is data obtained by spectrometer detection. The spectrometer can measure and analyze the spectral characteristics of dyeing liquid and printing and dyeing products, such as absorption, reflection, transmission, etc., and can provide information on dyeing liquid composition, concentration, printing and dyeing effects, etc. The dynamic process data stream refers to various process parameters and states in the printing and dyeing process, such as temperature, pressure, speed, concentration, etc., which can reflect the dynamic changes in the printing and dyeing process and help analyze and optimize the printing and dyeing process.

[0027] P30: Build a printing and dyeing detection module, which includes a spectral analysis block and a process analysis block configured in parallel, and a defect positioning unit connected to the rear, and the spectral analysis block is embedded with the printing and dyeing feature library.

[0028] Specifically, a printing and dyeing inspection module is constructed to analyze and evaluate the spectral data and process data of printing and dyeing products. The printing and dyeing inspection module includes a spectral analysis block and a process analysis block configured in parallel, and a post-connected defect location unit.

[0029] Among them, the spectral analysis block is mainly responsible for processing and analyzing the spectral data stream of printing and dyeing products, which can be obtained by collecting sample dye spectral data for deep learning. And the spectral analysis block is embedded with the printing and dyeing feature library, so that the block can quickly and accurately identify the key features in the spectral data, and compare and match it with the information in the printing and dyeing feature library. Through spectral analysis, detailed information on the composition, concentration, printing and dyeing effect of the dye solution can be obtained, providing an important basis for subsequent printing and dyeing detection.

[0030] The process analysis block is configured in parallel with the spectrum analysis block, and the process analysis block is used to process and analyze the dynamic process data flow of printing and dyeing products, which can be obtained by collecting sample process data for deep learning. It can monitor and analyze various process parameters and states in the printing and dyeing process in real time, such as temperature, pressure, speed, concentration, etc. Through process analysis, dynamic changes and potential problems in the printing and dyeing process can be revealed, providing strong support for optimizing process parameters and ensuring printing and dyeing quality.

[0031] The defect location unit, as the post-connection part of the printing and dyeing detection module, is responsible for locating defects in printing and dyeing products based on the results of spectral analysis and process analysis. Once a printing and dyeing defect or process defect is detected in a printing and dyeing product, the defect location unit can quickly determine the location and nature of the defect, and generate a corresponding test report and early warning information to promptly discover and correct the problem and ensure the quality and consistency of the printing and dyeing products.

[0032] P40: pre-process the printing and dyeing data stream, transmit it to the printing and dyeing detection module, perform parallel independent analysis of blocks and lateral interactive verification, locate the printing and dyeing defect points that meet the verification standards, and integrate and determine the printing and dyeing detection single column;

[0033] The lateral interactive verification is a continuous interactive verification accompanied by block analysis.

[0034] Furthermore, step P40 of the embodiment of the present application also includes:

[0035] P41: construct a preprocessing model, the preprocessing model includes a parallel spectrum processing layer and a data processing layer, the data processing layer includes a heterogeneous conversion unit, a data fusion unit and a redundant processing unit;

[0036] P42: Based on the preprocessing model, the printing and dyeing data stream is divided and processed within a layer to determine effective printing and dyeing data.

[0037] It should be understood that the printing and dyeing data stream is preprocessed. Specifically, first, deep learning is performed by acquiring sample data to build a preprocessing model. The preprocessing model includes a parallel spectral processing layer and a data processing layer. The spectral processing layer is used to preprocess the spectral data stream, such as strengthening demand features and weakening invalid features, while the data processing layer includes a heterogeneous conversion unit, a data fusion unit and a redundant processing unit, which are used to perform heterogeneous conversion, data fusion and redundancy removal on the dynamic process data stream to ensure the accuracy and effectiveness of the data.

[0038] Furthermore, based on the preprocessing model, the printing and dyeing data stream is divided and processed within the layer, that is, the printing and dyeing data stream is divided into a spectral data stream and a dynamic process data stream, and the spectral data stream and the dynamic process data stream are respectively input into the spectral processing layer and the data processing layer for processing to determine the effective printing and dyeing data.

[0039] Furthermore, the valid printing and dyeing data is transmitted to the printing and dyeing detection module, and the spectral analysis block and the process analysis block of the printing and dyeing detection module are used for parallel and independent analysis, that is, defect analysis is performed from the spectral perspective and the process perspective respectively, and the analysis results are input into the post-connected defect location unit for lateral interactive verification, that is, the spectral analysis results are compared with the process analysis results for consistency, and the defect identification results that meet the verification results are used for defect location, and multiple printing and dyeing defect points are obtained, and the multiple printing and dyeing defect points are integrated to determine the printing and dyeing detection column.

[0040] Further, such as Figure 2 As shown, step P42 of the embodiment of the present application also includes:

[0041] P42-1: extracting the spectral data stream, combining the spectral processing layer to perform spectrum processing guided by identification requirements, and determining the first effective printing and dyeing data;

[0042] P42-2: extract the dynamic process data flow and determine the main data structure, where the main data structure is any data structure in the dynamic process data flow;

[0043] P42-3: Based on the main data structure, the dynamic process data flow is subjected to heterogeneous conversion processing, the homologous data is merged, and redundant data is eliminated based on repeated information and reactive information, and the second effective printing and dyeing data is determined by integration.

[0044] Optionally, based on the preprocessing model, the printing and dyeing data stream is divided and processed within the layer. First, the spectral data stream is extracted from the printing and dyeing data stream, and the spectral processing layer is combined to perform spectral processing oriented to the identification requirements, that is, the features to be extracted are processed in a recognizable manner, including strengthening the required features, weakening the invalid features, etc. For example, the signal-to-noise ratio and resolution of the spectral data are improved by filtering, noise reduction, baseline correction, feature extraction, etc., so as to more accurately identify the key spectral features. After the spectral processing, the obtained spectral data will be used as the first effective printing and dyeing data.

[0045] Furthermore, a dynamic process data stream is extracted from the printing and dyeing data stream. There are multiple different data structures in the dynamic process data stream, such as temperature data, pressure data, speed data, etc. In order to facilitate analysis, it is necessary to determine a main data structure as a benchmark. The main data structure can be any representative or important data structure in the dynamic process data stream.

[0046] Furthermore, based on the master data structure, the dynamic process data stream is subjected to heterogeneous conversion processing, that is, the dynamic process data stream is unified into the same data structure and format, and then the homologous data after heterogeneous conversion is fused. The homologous data refers to data from the same data source or with the same physical meaning. Through data fusion, the information of multiple homologous data can be integrated to improve the integrity and accuracy of the data. At the same time, based on repeated information and reactive information, redundant data is eliminated. The repeated information refers to information that appears multiple times in the data set and has no added value. Reactive information refers to information that has no contribution or little contribution to printing and dyeing detection and analysis. By eliminating redundant data, the size of the data set can be reduced and the efficiency of the data can be improved.

[0047] After heterogeneous conversion, data fusion and redundant data elimination, the obtained process data will be used as the second effective printing and dyeing data. Together with the first effective printing and dyeing data obtained by the spectral processing layer, it can provide comprehensive and accurate data support for subsequent printing and dyeing detection and analysis.

[0048] Furthermore, step P40 of the embodiment of the present application also includes:

[0049] P43: Read the output of the first block and the output of the second block, and transfer them to the defect location unit synchronously;

[0050] P44: Using the defect trend as the first verification feature and the defect level as the second verification feature, verify the first block output and the second block output, perform hierarchical binarization, and determine a hierarchical verification mark. The defect trend is determined based on the defect type and defect direction, wherein the same direction and meeting the preset level interval are binarized positive directions;

[0051] P45: Identify the hierarchical verification mark, integrate the marked part and the unmarked part, and determine the printing and dyeing detection single column.

[0052] Specifically, after completing the parallel independent analysis of the spectral analysis block and the process analysis block, the output results of the two blocks are read, that is, the output of the first block and the output of the second block are read, including the printing and dyeing defects, key parameters, analysis results, etc. detected by each block, and the reading results are synchronously transferred to the defect location unit.

[0053] In the defect location unit, the first block output and the second block output are verified. First, the defect trend is used as the first verification feature. The defect trend can be determined based on the type of defect (such as uneven color, spots, stains, etc.) and the direction of the defect (such as horizontal, vertical, oblique, etc.). By comparing whether the defect trends of the two block outputs are consistent, the accuracy of the analysis results can be preliminarily judged. Next, the defect level is used as the second verification feature. The defect level can be divided according to factors such as the severity, scope, and impact of the defect. By comparing whether the defect levels of the two block outputs are similar or meet the preset level range, the reliability of the analysis results can be further verified.

[0054] If the outputs of the two blocks meet the verification conditions in terms of defect trend and defect level, that is, they are in the same direction and meet the preset level range, they are regarded as binary positive, that is, the output results of the two blocks are considered to be consistent and reliable. In addition, during the verification process, hierarchical binarization processing is performed, that is, different binarization thresholds or processing strategies are used for defects or features of different levels or different degrees of importance to ensure that the system can more accurately identify and process printing and dyeing defects of different levels.

[0055] After completing the hierarchical binarization process, the hierarchical check marks output by each block are identified. These marks may include different colors, symbols, numbers, etc., which are used to represent defects or features of different levels or different degrees of importance. Further, the hierarchical check marks are identified, and the marked parts are integrated. The marked parts are for checking the accuracy and the printing and dyeing defects corresponding to the final printing and dyeing results. The results with the same or similar check marks output by each block are merged and integrated to form a more complete and accurate printing and dyeing detection column. At the same time, the non-marked part is integrated, and the non-marked part includes detection errors or non-mapping conditions, that is, defects within the steps in the dynamic process. By integrating and analyzing the non-marked part, the content of the printing and dyeing detection column can be further enriched and improved. The printing and dyeing detection column contains complete defect results and defect evolution links for printing and dyeing products, which can provide important references for subsequent decision-making.

[0056] P50: traverse the printing and dyeing inspection single column, carry out defect coverage and defect generation identification and location tracing under composite printing and dyeing, and integrate and determine the composite point inspection single column;

[0057] Further, such as Figure 3 As shown, step P50 of the embodiment of the present application also includes:

[0058] P51: traverse the printing and dyeing detection single column, cluster each printing and dyeing layer, and determine the hierarchical printing and dyeing single column;

[0059] P52: performing defect mapping between adjacent layers for the layer printing and dyeing single column, and determining an inter-layer mapping group;

[0060] P53: Traverse the inter-layer mapping group, measure the defect variables, and perform direction marking to generate the composite point detection single column, wherein the direction marking includes the positive direction of defect coverage and the negative direction of defect generation.

[0061] In a possible embodiment of the present application, during the printing and dyeing inspection process, when multiple layers or composite printing and dyeing are involved, it is necessary to analyze the printing and dyeing conditions of each layer separately and consider the mutual influence between different layers. Therefore, it is necessary to traverse the printing and dyeing inspection columns, perform defect coverage and defect generation identification and location tracing under composite printing and dyeing, and integrate and determine the composite point inspection columns.

[0062] Specifically, the printing and dyeing detection single column is traversed, and according to the hierarchical structure of printing and dyeing, the information in the printing and dyeing detection single column is clustered according to different printing and dyeing layers to generate a series of hierarchical printing and dyeing single columns, each of which corresponds to a specific printing and dyeing layer. Further, the mutual influence between different layers is analyzed, and the defect mapping between adjacent layers is performed on the hierarchical printing and dyeing single column to determine the correlation and transmission of defects between different layers. For example, through mapping, it is identified which layer the defect is generated from and transmitted to other layers, or the defect that exists independently in a certain layer is determined. Then a series of inter-layer mapping groups are generated, each of which contains the mapping relationship between defects between adjacent layers.

[0063] Furthermore, the inter-layer mapping group is traversed to measure defect variables, which include the size, shape, color, position and other characteristics of the defect. By measuring the defect variables, the defects can be described and identified more accurately. While measuring the defect variables, the defects are marked with directions, including the positive direction of defect coverage and the negative direction of defect generation. The positive direction indicates that the defect is transmitted from the previous layer to the current layer, and the negative direction indicates that the defect is independently generated in the current layer. Through the direction marking, the source and propagation path of the defect can be clearly identified, providing an important basis for subsequent defect repair and process optimization.

[0064] Finally, a composite point inspection column is generated based on the measurement results and direction marks. This column contains information and analysis results of all key defects in composite printing and dyeing, which can provide strong support for subsequent decision-making.

[0065] Furthermore, step P50 of the embodiment of the present application also includes:

[0066] P54: Determine a defect vector, wherein the defect vector identifies a location where the defect occurs in time and space;

[0067] P55: Based on the defect vector, the defect occurrence is traced, the minimum cause unit is determined, and the frequency is determined to determine the traceability result, wherein the traceability result includes accidental causes and inevitable causes;

[0068] P56: Map and integrate the traceability result and the defect vector to generate the composite point detection single column.

[0069] Specifically, a more in-depth analysis is conducted on the defects under composite printing and dyeing to determine the defect vector. The defect vector is a multi-dimensional data representation that identifies the location of the defect in time and space, which helps to more accurately describe the nature and location of the defect. Based on the defect vector, the defect is traced to determine the minimum causal unit that causes the defect, that is, the smallest and most direct factor or unit that causes the defect, which may be a specific process parameter, equipment status, material properties, etc. And a frequency determination is performed, that is, an assessment of the frequency of occurrence of the minimum causal unit to determine the traceability results. The traceability results include accidental inducements and inevitable inducements. Accidental inducements refer to inducements that occur occasionally and are difficult to predict and control, while inevitable inducements refer to those inducements that occur frequently and have a certain regularity and predictability.

[0070] Furthermore, the traceability results are mapped and integrated with the defect vector, including associating and integrating the minimum causal unit, frequency judgment and other information in the traceability results with the spatiotemporal position, characteristic attributes and other information in the defect vector, so as to generate the final composite point detection column. The composite point detection column will contain more complete and accurate defect information and traceability results, which can provide strong data support for subsequent defect repair, process optimization, quality control, etc.

[0071] P60: Integrate and map the printing and dyeing detection column and the composite point detection column to generate a printing and dyeing detection result.

[0072] Optionally, the printing and dyeing inspection single column and the compound point inspection single column are integrated and mapped to generate the final printing and dyeing inspection result, which contains comprehensive information of the printing and dyeing products, including printing and dyeing process parameters, material properties, spectral characteristics, defect conditions, traceability results, etc. This information is presented in an easy-to-understand and easy-to-use form, such as tables, charts, reports, etc., so that relevant personnel can quickly and accurately understand the overall situation of the printing and dyeing products, providing strong support for subsequent quality control and process improvements.

[0073] In summary, the embodiments of the present application have at least the following technical effects:

[0074] This application uses spectral feature mining to read the printing and dyeing data stream of the target printing and dyeing product, conducts parallel independent analysis of blocks and lateral interactive verification, integrates and determines the printing and dyeing detection column, identifies and locates the defect coverage and defect generation under composite printing and dyeing, integrates and determines the composite point detection column, integrates and maps the printing and dyeing detection column and the composite point detection column, and generates the printing and dyeing detection results.

[0075] The technical effect of combining spectral data and process data for parallel defect detection, accurately identifying and locating printing and dyeing defects, and improving the efficiency and accuracy of product quality detection has been achieved.

[0076] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0077] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0078] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A printing and dyeing detection method for textile products, characterized in that: The method comprises: P10: Spectral feature mining of various dye solutions to build a dyeing and printing feature library, where the spectral features include single dye solution dimension and mixed dye solution dimension; P20: reading the printing and dyeing data stream of the target printing and dyeing product, wherein the printing and dyeing method of the target printing and dyeing product is single-layer printing and dyeing or composite printing and dyeing, and the printing and dyeing data stream includes a spectral data stream based on spectrometer detection and a dynamic process data stream; P30: Building a printing and dyeing inspection module, the printing and dyeing inspection module includes a spectral analysis block and a process analysis block configured in parallel, and a defect location unit connected to the back, the spectral analysis block is embedded with the printing and dyeing feature library; P40: pre-process the printing and dyeing data stream, transmit it to the printing and dyeing detection module, perform parallel independent analysis of blocks and lateral interactive verification, locate the printing and dyeing defect points that meet the verification standards, and integrate and determine the printing and dyeing detection single column; P50: traverse the printing and dyeing inspection single column, carry out defect coverage and defect generation identification and location tracing under composite printing and dyeing, and integrate and determine the composite point inspection single column; P60: Integrate and map the printing and dyeing detection column and the composite point detection column to generate a printing and dyeing detection result; Wherein, step P10 comprises: P11: using the internal search engine or search tool of the printing and dyeing enterprise, taking the multiple dyeing liquids as keywords or indexes, performing engine search, and determining the dyeing liquid printing and dyeing records, wherein the dyeing liquid printing and dyeing records contain detailed information of the dyeing liquids, including type, concentration, manufacturer, printing and dyeing process parameters, and printing and dyeing results; P12: based on the dyeing liquid printing and dyeing records, mining the spectral characteristics and superimposed printing and dyeing characteristics of the multiple dyeing liquids, wherein the spectral characteristics include absorption peaks, emission peaks, and spectral shapes, and the superimposed printing and dyeing characteristics refer to the printing and dyeing characteristics of the mixed dyeing liquid. When multiple dyeing liquids are mixed and used When used, the interaction between multiple dyeing liquids affects the printing and dyeing effect; P13: Based on the dyeing and printing records, the feature implicit relationship is mined, and the feature implicit relationship is the relative relationship between the spectral feature and the printing and dyeing feature, including analyzing the correlation between the intensity of the spectral feature and the printing and dyeing effect, and the influence of different spectral feature combinations on the printing and dyeing effect; P14: Integrate the spectral features, and associate them with the feature implicit relationship and the superimposed printing and dyeing features to construct the printing and dyeing feature library, which contains the spectral characteristics of various dyeing liquids, and the correlation between different spectral characteristics and printing and dyeing effects; Step P20 includes: reading a printing and dyeing data stream of a target printing and dyeing product, the printing and dyeing data stream refers to printing and dyeing related data of the target printing and dyeing product; wherein the target printing and dyeing product refers to a specific product that needs to be tested for printing and dyeing, and the printing and dyeing method of the target printing and dyeing product is single-layer printing and dyeing or composite printing and dyeing; the printing and dyeing data stream includes a spectral data stream based on spectrometer detection and a dynamic process data stream, the spectral data stream is data obtained by spectrometer detection, and the spectrometer is used to measure and analyze the spectral characteristics of dyeing liquid and printing and dyeing products; the dynamic process data stream refers to various process parameters and states in the printing and dyeing process; Step P30 includes: building a printing and dyeing detection module for analyzing and evaluating the spectral data and process data of the printing and dyeing products, the printing and dyeing detection module includes a spectral analysis block and a process analysis block configured in parallel, and a post-connected defect location unit; the spectral analysis block is mainly responsible for processing and analyzing the spectral data stream of the printing and dyeing products, which is obtained by collecting sample dye spectral data for deep learning; the spectral analysis block is embedded with the printing and dyeing feature library, which is used to identify key features in the spectral data and compare and match it with the information in the printing and dyeing feature library; the process analysis block is configured in parallel with the spectral analysis block, the process analysis block is used to process and analyze the dynamic process data stream of the printing and dyeing products, which is obtained by collecting sample process data for deep learning; the defect location unit, as the post-connected part of the printing and dyeing detection module, is used to locate defects in the printing and dyeing products according to the results of spectral analysis and process analysis; when it is detected that the printing and dyeing product has a printing and dyeing defect or a process defect, the defect location unit is used to determine the location and nature of the defect, and generate a corresponding detection report and early warning information; Step P40 includes: P41: constructing a preprocessing model, the preprocessing model includes a parallel spectral processing layer and a data processing layer, the data processing layer includes a heterogeneous conversion unit, a data fusion unit and a redundant processing unit; P42: based on the preprocessing model, dividing and processing the printing and dyeing data stream within the layer to determine the effective printing and dyeing data; P43: reading the first block output and the second block output, and synchronously transferring them to the defect positioning unit; P44: using the defect trend as the first verification feature and the defect level as the second verification feature, verifying the first block output and the second block output, performing hierarchical binarization processing, and determining the hierarchical verification mark, the defect trend is determined based on the defect type and the defect direction, wherein the same direction and meeting the preset level interval are the positive direction of binarization; P45: identifying the hierarchical verification mark, integrating the marked part and the non-marked part, and determining the printing and dyeing detection single column; Step P42 also includes: P42-1: extracting the spectral data stream, combining the spectral processing layer to perform spectrum processing guided by identification requirements, and determining the first effective printing and dyeing data; P42-2: extracting the dynamic process data stream, and determining the main data structure, the main data structure is any data structure in the dynamic process data stream; P42-3: based on the main data structure, performing heterogeneous conversion processing on the dynamic process data stream, fusing homologous data, removing redundant data based on repeated information and reactive information, and integrating to determine the second effective printing and dyeing data; Step P50 includes: P51: traversing the printing and dyeing detection column, clustering each printing and dyeing layer, and determining the hierarchical printing and dyeing column; P52: performing defect mapping between adjacent layers of the hierarchical printing and dyeing column to determine the inter-layer mapping group; P53: traversing the inter-layer mapping group, measuring the defect variables, and performing direction marking to generate the composite point detection column, wherein the direction marking includes the positive direction of defect coverage and the negative direction of defect generation; P54: determining the defect vector, wherein the defect vector identifies the time and space occurrence location; P55: based on the defect vector, tracing the defect occurrence, determining the minimum inducement unit and performing frequency judgment, and determining the tracing result, wherein the tracing result includes accidental inducements and inevitable inducements; P56: mapping and integrating the tracing result and the defect vector to generate the composite point detection column.

2. The method according to claim 1, characterized in that The lateral cross-verification is a continuous cross-verification accompanied by block analysis.

Images