Online dyeing detection system for textiles based on machine vision

Abstract

The present invention discloses a textile online dyeing detection system and method based on machine vision, which belongs to the field of image processing technology, and includes: a scanner pre-verification module, a pre-verification data analysis module, a textile image acquisition module, a dyeing quality determination module, and a dyed textile drying execution determination module. The textile online dyeing detection method based on machine vision provided by the present invention receives the textile dyeing detection signal through a cloud computing center and synchronously starts the scanner for pre-verification, thereby evaluating the feedback parameters of the scanner startup performance and the image quality, and adjusting the scanner, instead of directly detecting and processing the textile image, thereby effectively improving the accuracy of textile dyeing detection and solving the problem of inaccurate textile dyeing quality control.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to an online textile dyeing detection system based on machine vision. Background Art

[0002] With the rapid development of the textile industry and consumers' increasing requirements for product quality, the importance of textile quality control has become increasingly prominent. Therefore, in order to meet the growing demand for textile quality control, it has become necessary to conduct rigorous testing of textiles.

[0003] Existing textile dyeing detection mainly relies on manual visual inspection, that is, workers carefully observe the color uniformity, color difference, color spots and other potential defects of each piece of dyed cloth with their naked eyes. This method is not only inefficient, but also has problems such as strong subjectivity and poor consistency.

[0004] The prior art, for example, is a quilt cover dyeing quality assessment method based on machine vision disclosed in the invention patent announcement with announcement number: CN115880297B, which includes: obtaining a surface image of a quilt cover to be detected and a surface image of a standard quilt cover, and obtaining a surface grayscale image of the quilt cover to be detected and a grayscale image of the standard quilt cover; obtaining a first color difference similarity, and normalizing the sum of the first color difference similarities of all pixels to obtain the quilt cover color difference between the quilt cover to be detected and the standard quilt cover; obtaining the proportion of each chromaticity value of each superpixel block; obtaining the quilt cover color distance between the first superpixel block and the second superpixel block; obtaining a surface dyeing significance value of each superpixel block; obtaining a dyeing quality assessment value based on the surface dyeing significance values ​​of all superpixel blocks, and judging the quality of the quilt cover surface dyeing based on the size of the assessment value.

[0005] The prior art, for example, is a method for detecting textile defects based on machine vision, as disclosed in the invention patent announcement with announcement number CN115272321B, which includes: dividing the surface grayscale image of the textile to be inspected into pixel blocks; screening out suspected defective pixel blocks from all pixel blocks to obtain suspected defective areas; smoothing the grayscale histogram of the suspected defective area by a linear interpolation algorithm to obtain a grayscale curve; and determining a strategy for determining a segmentation threshold of the suspected defective area based on the peak characteristics of the grayscale curve to obtain a segmentation threshold, and performing threshold segmentation on the suspected defective area to obtain a defective area.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the prior art, direct detection and processing are usually performed on textile images. In actual use, the acquisition of textile images mostly depends on the acquisition equipment, which will cause uneven lighting, color distortion, noise interference and other effects on the images. Therefore, if the acquired textile images are directly processed and the quality is judged, it will cause the negative phenomenon of low accuracy of textile dyeing quality detection. Therefore, the prior art has the problem of inaccurate textile dyeing quality control. Summary of the invention

[0008] The embodiment of the present application provides a textile online dyeing detection system based on machine vision, including: a scanner pre-verification module, a pre-verification data analysis module, a textile image acquisition module and a dyeing quality judgment module; the scanner pre-verification module is used to receive a textile dyeing detection signal through a cloud computing center, and synchronously start the scanner for pre-verification to obtain the pre-verification data of the scanner; the pre-verification data analysis module is used to obtain the pre-verification judgment result of the scanner according to the pre-verification data of the scanner and through analysis, and synchronously perform adaptive adjustment and control on the scanner based on the pre-verification judgment result, and record the scanner after adaptive adjustment and control as an available scanner; the textile image acquisition module is used to collect a plane scanned image of the textile after dyeing through an available scanner, and compare it with the textile plane dyeing effect diagram to extract a textile dyeing difference data set; the dyeing quality judgment module is used to analyze the textile dyeing difference data set, obtain the quality judgment result of the textile dyeing, and transmit it to the cloud computing center for prompting.

[0009] Furthermore, the synchronously starting the scanner for pre-verification to obtain pre-verification data of the scanner includes: starting the scanner for pre-verification, and obtaining the scanner's startup performance feedback parameters and startup capture image, wherein the startup performance feedback parameters include startup response delay and scanning execution rate; based on the scanner's startup capture image, and inputting it into the image processor to identify and extract the imaging data of the startup capture image, specifically including imaging resolution, imaging signal-to-noise ratio, imaging contrast and imaging saturation; and jointly marking the scanner's startup performance feedback parameters and the imaging data of the startup capture image as the scanner's pre-verification data.

[0010] Furthermore, the pre-verification judgment result of the scanner is obtained through analysis, and the steps include: extracting the initial archive verification image of the scanner from the database, and extracting the imaging data of the initial archive verification image through the image processor, and marking it as the imaging control data of the scanner, wherein the imaging control data includes the initial imaging resolution, the initial imaging signal-to-noise ratio, the initial imaging contrast and the initial imaging saturation; based on the imaging control data of the scanner, and comparing it with the imaging data of the startup captured image, and combining the startup performance feedback parameters of the scanner, a first pre-verification evaluation value of the scanner is obtained through comprehensive processing; a first pre-verification evaluation threshold is obtained, and compared with the first pre-verification evaluation value of the scanner, if the first pre-verification evaluation value of the scanner is above the first pre-verification evaluation threshold, the pre-verification judgment result of the scanner is defined as the scanner pre-verification is normal, and if the first pre-verification evaluation value of the scanner is less than the first pre-verification evaluation threshold, the pre-verification judgment result of the scanner is defined as the scanner pre-verification is abnormal.

[0011] Furthermore, the adaptive adjustment control of the scanner is performed based on the pre-verification judgment result, and the specific process is as follows: based on the pre-verification judgment result, if the pre-verification judgment result is that the pre-verification of the scanner is normal, there is no need to perform adaptive adjustment control on the scanner; if the pre-verification judgment result is that the pre-verification of the scanner is abnormal, the preset adaptive adjustment parameters are extracted from the database, and the scanner is adaptively adjusted for the first time based on the preset adaptive adjustment parameters, and the scanner after the first adaptive adjustment control is recorded as the scanner after the initial adjustment; the scanner is pre-verified for a second time after the initial adjustment, and a second pre-verification evaluation value of the scanner after the initial adjustment is obtained by analysis; the second pre-verification evaluation value of the scanner after the initial adjustment is compared with the first pre-verification evaluation value of the scanner, and the scanner after the initial adjustment is extracted. The difference between the second pre-verification evaluation value of the scanner and the first pre-verification evaluation value of the scanner is recorded as the pre-verification adjustment benchmark value; if the pre-verification adjustment benchmark value is below zero, a warning message is generated and uploaded to the cloud computing center; if the pre-verification adjustment benchmark value is greater than zero, the pre-verification adjustment benchmark value is compared with the pre-verification adjustment threshold in the database, if the pre-verification adjustment benchmark value is greater than the pre-verification adjustment threshold, the scanner after the initial adjustment is marked as an available scanner, if the pre-verification adjustment benchmark value is below the pre-verification adjustment threshold, the scanner after the initial adjustment is adaptively adjusted again with the preset adaptive adjustment parameters until the pre-verification adjustment benchmark value is greater than the pre-verification adjustment threshold, and when the pre-verification adjustment benchmark value is greater than the pre-verification adjustment threshold, the current scanner is recorded as an available scanner.

[0012] Furthermore, the planar scanned image of the textile after dyeing is collected by an available scanner, and compared with the planar dyeing effect diagram of the textile to extract the textile dyeing difference data set. The specific comparison process includes: collecting the planar scanned image of the textile after dyeing by an available scanner; extracting the planar dyeing effect diagram of the textile, and dividing the planar dyeing effect diagram of the textile into various dyeing sub-regions according to a preset division area, and synchronously extracting the scanned image corresponding to each dyeing sub-region in the planar scanned image of the textile after dyeing according to the distribution position of each dyeing sub-region, and recording it as the actual dyeing plane image of each dyeing sub-region; recording the image corresponding to each dyeing sub-region in the planar dyeing effect diagram of the textile as the ideal dyeing effect diagram of each dyeing sub-region; comparing the ideal dyeing effect diagram of each dyeing sub-region with the actual dyeing plane image of each dyeing sub-region, to obtain the textile dyeing difference data set.

[0013] Furthermore, the textile dyeing difference data set specifically compares the following processes: dyeing pattern area difference of each dyeing sub-region, dyeing pattern edge contour line length difference, dyeing pattern porosity, and color difference of each dyeing sub-region at each screening point.

[0014] Furthermore, the textile dyeing difference data set is analyzed to obtain the quality judgment result of the textile dyeing, specifically including: obtaining a preset textile dyeing difference control data set, and comparing it with the textile dyeing difference data set to obtain a textile dyeing difference evaluation value; obtaining a textile dyeing difference evaluation threshold, and comparing it with the textile dyeing difference evaluation value, if the textile dyeing difference evaluation value is above the dyeing difference evaluation threshold, then the quality judgment result of the textile dyeing is invalid dyeing, if the textile dyeing difference evaluation value is less than the dyeing difference evaluation threshold, then the quality judgment result of the textile dyeing is valid dyeing.

[0015] Furthermore, the textile dyeing difference evaluation value is calculated by the formula:

[0016]

[0017]

[0018] In the formula, Q 1 represents the textile dyeing difference evaluation value, i represents the number of the dyeing sub-region, n represents the number of the dyeing sub-region, D i represents the difference in the dyeing pattern area of ​​the i-th dyeing sub-region, D 0 Indicates the preset allowable area difference of the dyeing pattern, L i represents the length difference of the edge contour line of the dyeing pattern of the i-th dyeing sub-region, L 0 Indicates the preset allowable edge contour line allowable length difference, P irepresents the saturation difference of the center point of the dyeing pattern of the i-th dyeing sub-region, P 0 Indicates the preset permitted center point saturation difference, C i represents the color difference of the i-th dyeing subregion at the screening point, C 0 Indicates the preset permissible color difference, ω 1 represents the influence weight of the dyeing pattern area difference, ω 2 Indicates the influence weight of the length difference of the edge contour line of the dyeing pattern, ω 3 Indicates the influence weight of the center point saturation difference, ω 4 Indicates that color difference affects the weight.

[0019] Furthermore, it also includes a drying execution judgment module for dyed textiles, which is used to perform drying execution judgment on the dyed textiles based on the quality judgment result of textile dyeing, specifically including: based on the quality judgment result of textile dyeing, if the quality judgment result is invalid dyeing, no drying treatment is required, and drying impact prompt information is generated to the cloud computing center for prompting; if the quality judgment result is effective dyeing, the drying impact parameters corresponding to the dyed textiles are collected, and the drying feasibility judgment value of the dyed textiles is obtained through processing, and compared with the textile drying feasibility judgment threshold, if the drying feasibility judgment value of the dyed textiles is above the textile drying feasibility judgment threshold, it is determined that the dyed textiles can be dried, and if the drying feasibility judgment value of the dyed textiles is less than the textile drying feasibility judgment threshold, it is determined that the dyed textiles need to be pre-drying, thereby completing the drying execution judgment of the dyed textiles.

[0020] Furthermore, the online dyeing detection method for textiles based on machine vision is characterized in that it includes the following steps: receiving textile dyeing detection signals through a cloud computing center, and synchronously starting a scanner for pre-verification to obtain pre-verification data of the scanner; according to the pre-verification data of the scanner, and obtaining the pre-verification judgment result of the scanner through analysis, synchronously adaptively adjusting and controlling the scanner based on the pre-verification judgment result, and recording the scanner after adaptive adjustment and control as an available scanner; collecting a plane scan image of the textile after dyeing through an available scanner, and comparing it with a plane dyeing effect diagram of the textile to extract a textile dyeing difference data set; analyzing the textile dyeing difference data set to obtain a quality judgment result of the textile dyeing, and transmitting it to the cloud computing center for prompting.

[0021] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0022] 1. The online textile dyeing detection method based on machine vision provided by the present invention receives the textile dyeing detection signal through the cloud computing center and synchronously starts the scanner for pre-verification, thereby evaluating the feedback parameters and image quality of the scanner startup performance and adjusting the scanner, instead of directly detecting and processing the textile image, thereby effectively improving the accuracy of textile dyeing detection and solving the problem of inaccurate textile dyeing quality control.

[0023] 2. The present invention collects a plane scan image of the textile after dyeing and compares it with the ideal dyeing effect diagram, thereby extracting a textile dyeing difference data set and analyzing the textile dyeing difference evaluation value, making the dyeing quality analysis more accurate.

[0024] 3. The present invention also provides a drying execution judgment module for dyed textiles, which performs drying execution judgment on the dyed textiles based on the quality judgment result of textile dyeing, thereby obtaining a drying feasibility judgment value for the dyed textiles, and compares it with the textile drying feasibility judgment threshold, thereby realizing a secondary drying judgment, ensuring accurate judgment of the drying process, and avoiding waste of resources caused by drying invalid dyed textiles. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of a module of a machine vision-based textile online dyeing detection system provided in an embodiment of the present application;

[0026] Figure 2 A flowchart of a method for online dyeing detection of textiles based on machine vision provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] The embodiment of the present application solves the problem of inaccurate quality control of textile dyeing in the prior art by providing an online textile dyeing detection system based on machine vision. The textile dyeing detection signal is received through a cloud computing center, and a scanner is simultaneously started for pre-verification to obtain pre-verification data of the scanner. According to the pre-verification data of the scanner, a pre-verification judgment result of the scanner is obtained through analysis, and the scanner is adaptively adjusted and controlled based on the pre-verification judgment result, and the scanner after the adaptive adjustment control is recorded as an available scanner; a plane scan image of the textile after dyeing is collected through the available scanner, and compared with a plane dyeing effect diagram of the textile to extract a textile dyeing difference data set; the textile dyeing difference data set is analyzed to obtain a quality judgment result of the textile dyeing, and the result is transmitted to the cloud computing center for prompting, thereby realizing accurate identification of whether the color distribution of the textile is uniform.

[0028] The technical solution in the embodiment of the present application is to solve the problem of inaccurate quality control of textile dyeing mentioned above, and the overall idea is as follows: receiving textile dyeing detection signals through a cloud computing center, and synchronously starting a scanner for pre-verification to obtain pre-verification data of the scanner, and obtaining a pre-verification judgment result of the scanner based on the pre-verification data of the scanner and through analysis, and synchronously performing adaptive adjustment control on the scanner based on the pre-verification judgment result, and recording the scanner after adaptive adjustment control as an available scanner; collecting a plane scan image of the textile after dyeing through an available scanner, and comparing it with a plane dyeing effect diagram of the textile to extract a textile dyeing difference data set; analyzing the textile dyeing difference data set to obtain a quality judgment result of the textile dyeing, and transmitting it to the cloud computing center for prompting, so as to accurately identify whether the color distribution of the textile is uniform.

[0029] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0030] like Figure 1 As shown, it is a structural schematic diagram of a textile online dyeing detection system based on machine vision provided in an embodiment of the present application. The textile online dyeing detection system based on machine vision provided in an embodiment of the present application includes: a scanner pre-verification module, a pre-verification data analysis module, a textile image acquisition module and a dyeing quality judgment module; the scanner pre-verification module is used to receive a textile dyeing detection signal through a cloud computing center, and synchronously start the scanner for pre-verification to obtain the pre-verification data of the scanner; the pre-verification data analysis module is used to obtain the pre-verification judgment result of the scanner according to the pre-verification data of the scanner and through analysis, and synchronously perform adaptive adjustment and control on the scanner based on the pre-verification judgment result, and record the scanner after adaptive adjustment and control as an available scanner; the textile image acquisition module is used to collect a plane scanned image of the textile after dyeing through an available scanner, and compare it with the textile plane dyeing effect diagram to extract a textile dyeing difference data set; the dyeing quality judgment module is used to analyze the textile dyeing difference data set to obtain the quality judgment result of the textile dyeing, and transmit it to the cloud computing center for prompting.

[0031] In this embodiment, it should be noted that the textile is dyed on a specific production line, and the production line will be equipped with a sensor or control unit to monitor the state of the dyeing process. When the textile reaches the preset dyeing stage, a textile dyeing detection signal will be sent to the cloud computing center. At this time, the cloud computing center receives the textile dyeing detection signal and synchronously starts the scanner for pre-verification. The cloud computing center refers to an institution or unit that provides computing resources, storage resources and other services to the outside world through the Internet based on a supercomputer system. A large amount of computing resources, storage resources and network resources can be concentrated. The cloud computing center in this embodiment is responsible for receiving the textile dyeing detection signal. The scanner pre-verification module is mainly responsible for performing a preliminary verification of the performance of the scanner before the detection begins to ensure that the scanner can work in the best state. This pre-verification process determines whether the scanner meets the use requirements by obtaining and analyzing the startup parameters and imaging data of the scanner. If the pre-verification result is normal, the scanner can be put into use; otherwise, the system will automatically adjust the parameters of the scanner to ensure that the subsequent dyeing detection can obtain accurate and reliable data. The pre-verification data analysis module determines the actual working state of the scanner through in-depth analysis of the pre-verification data, and makes necessary adaptive adjustments based on the judgment results. This process helps to ensure the stability and consistency of the scanner in different environments and conditions, and improves the reliability of dyeing detection. The textile image acquisition module uses the available scanner after pre-calibration to collect plane images of the dyed textiles, and extracts the difference data set of textile dyeing by comparing with the standard dyeing image. The difference data set of textile dyeing includes color difference, brightness difference, etc., which can provide detailed basis for subsequent quality judgment. The dyeing quality judgment module analyzes the collected dyeing difference data and finally obtains the dyeing quality judgment result of the textile. The dyeing quality judgment result can intuitively reflect whether the dyeing effect of the textile meets the expected standard.

[0032] It should also be noted that the quality determination result of textile dyeing is transmitted to the cloud computing center for prompting, and the prompt text is, for example: the quality determination result of textile dyeing is invalid dyeing, to ensure effective drying, please reprocess it.

[0033] The flat dyeing effect diagram of textiles can be obtained through the database. The dyeing effect diagram can intuitively display the ideal visual effects such as color, pattern and texture of textiles. It usually represents the best dyeing state that can be achieved by textile dyeing. It is usually uploaded and stored in the database directly by designers.

[0034] Furthermore, the scanner is synchronously started for pre-verification to obtain pre-verification data of the scanner, including: starting the scanner for pre-verification, and obtaining the startup performance feedback parameters of the scanner and the startup captured image, wherein the startup performance feedback parameters include the startup response delay and the scanning execution rate; based on the startup captured image of the scanner, the imaging data of the startup captured image is input into the image processor to identify and extract the imaging data, specifically including the imaging resolution, imaging signal-to-noise ratio, imaging contrast and imaging saturation; the startup performance feedback parameters of the scanner and the imaging data of the startup captured image are jointly marked as the pre-verification data of the scanner.

[0035] In this embodiment, it should be noted that after receiving the textile dyeing detection signal, the system will automatically start the scanner for pre-verification. The pre-verification process includes the hardware performance check of the scanner and the status monitoring during the startup process. Through the pre-verification, the abnormality of the scanner during the startup process can be detected, so as to ensure that the scanner can perform image acquisition of textiles in the best state. The startup response delay refers to the time interval from the scanner receiving the startup instruction to the actual start of work. The scanning execution rate is the speed at which the scanner completes the scanning operation per unit time. These parameters reflect the startup efficiency and working ability of the scanner, and are important indicators for evaluating whether the scanner is in a normal working state. During the startup process of the scanner, the scanner will collect a startup shooting image and input the image into the image processor for processing. The image processor will extract imaging data from the image, and the imaging data specifically includes imaging resolution, imaging signal-to-noise ratio, imaging contrast and imaging saturation.

[0036] Furthermore, the pre-verification judgment result of the scanner is obtained through analysis, and the steps include: extracting the initial archive verification image of the scanner from the database, and extracting the imaging data of the initial archive verification image through the image processor, and marking it as the imaging control data of the scanner, wherein the imaging control data includes the initial imaging resolution, the initial imaging signal-to-noise ratio, the initial imaging contrast and the initial imaging saturation; based on the imaging control data of the scanner, and comparing it with the imaging data of the startup captured image, and combining the startup performance feedback parameters of the scanner, a first pre-verification evaluation value of the scanner is obtained through comprehensive processing; a first pre-verification evaluation threshold is obtained, and compared with the first pre-verification evaluation value of the scanner, if the first pre-verification evaluation value of the scanner is above the first pre-verification evaluation threshold, the pre-verification judgment result of the scanner is defined as the scanner pre-verification is normal, and if the first pre-verification evaluation value of the scanner is less than the first pre-verification evaluation threshold, the pre-verification judgment result of the scanner is defined as the scanner pre-verification is abnormal.

[0037] In this embodiment, the initial archived verification image is a standard image generated by the scanner when it leaves the factory or is used for the first time, and is used for performance testing, calibration and long-term maintenance of the device. No matter how long the device has been used, performance checks can be performed using these archived images to ensure the accuracy of the scanning results. When the scanner fails or the image quality deteriorates, the user can quickly find the problem by comparing it with the initial archived verification image.

[0038] The calculation formula of the first pre-calibration evaluation value of the scanner is:

[0039]

[0040] In the formula, E 1 represents the first pre-calibration evaluation value of the scanner, R 1 Represents the imaging resolution of the scanner, R 0 represents the initial imaging resolution, σ R Indicates the image resolution calibration reference value, S 1 represents the imaging signal-to-noise ratio, S 0 represents the initial imaging signal-to-noise ratio, σ S Indicates the image signal-to-noise ratio calibration reference value, A 1 A represents the image contrast. 0 represents the initial contrast, σ A Indicates the image contrast calibration reference value, H 1 Indicates image saturation, H 0 represents the initial image saturation, σ H Indicates the image saturation calibration reference value, T 0 represents the standard startup response delay in the database, T r Indicates the startup response delay, V 0 represents the standard scan execution rate in the database, V r represents the scan execution rate, α 1 represents the weight of image resolution, α 2 represents the imaging signal-to-noise ratio influence weight, α 3 Represents the image contrast influence weight, α 4 Indicates the image saturation influence weight.

[0041] It should be noted that the normalization factor of the startup response delay and the normalization factor of the scanning execution rate can be obtained through a database. The imaging data of the startup image can be obtained by scanning with a scanner and then transmitting it to an image processor for processing. The image processor is an electronic device for processing the image data and obtaining relevant parameters of the image, such as a GPU (graphics processor) and a CPU (central processing unit).

[0042] The imaging signal-to-noise ratio calibration reference value, the imaging contrast calibration reference value and the imaging saturation calibration reference value can be obtained through the database.

[0043] The imaging resolution verification reference value, imaging signal-to-noise ratio verification reference value, imaging contrast verification reference value and imaging saturation verification reference value represent the lowest reference standard of the scanner under ideal conditions, i.e., the lowest reference value of the ideal scanner. If the value after the scanner imaging comparison data is calculated with the imaging data of the start-up image and the comparison value of the lowest reference value of the ideal scanner are at a higher value level, it indicates that the first pre-verification evaluation value of the scanner is good and suitable for image scanning. The initial archive verification image can be obtained through the database and used as the basis of data parameters to facilitate subsequent comparison with the imaging data of the start-up image.

[0044] It should be noted that the imaging resolution influence weight, imaging signal-to-noise ratio influence weight, imaging contrast influence weight and imaging saturation influence weight respectively represent the numerical values ​​of the influence of the imaging resolution, imaging signal-to-noise ratio, imaging contrast and imaging saturation on the first pre-verification evaluation value. When used, the influence weights corresponding to the imaging resolution, imaging signal-to-noise ratio, imaging contrast and imaging saturation can be directly obtained from the database, and the corresponding relationship can be a pre-set mapping relationship. For example, the imaging resolution, imaging signal-to-noise ratio, imaging contrast and imaging saturation form a mapping set with the imaging resolution influence weight, imaging signal-to-noise ratio influence weight, imaging contrast influence weight and imaging saturation influence weight. The real-time imaging resolution, imaging signal-to-noise ratio, imaging contrast and imaging saturation are input into the mapping set to obtain the influence weights corresponding to the imaging resolution, imaging signal-to-noise ratio, imaging contrast and imaging saturation. The mapping relationship can be a one-to-one correspondence or a many-to-one relationship. For example, the imaging resolution corresponds to the imaging resolution influence weight, the imaging signal-to-noise ratio corresponds to the imaging signal-to-noise ratio influence weight, the imaging contrast corresponds to the imaging contrast influence weight, and the imaging saturation corresponds to the imaging saturation influence weight. The first pre-verification evaluation threshold can be obtained through a database, and the first pre-verification evaluation threshold is used as the basic data for the first pre-verification evaluation, and the first pre-verification evaluation threshold is used to determine whether the scanner pre-verification is normal.

[0045] It should also be noted that in this embodiment, the first pre-verification evaluation value of the scanner is obtained by processing according to the imaging resolution of the scanner, the imaging signal-to-noise ratio of the scanner, the imaging contrast of the scanner, and the imaging saturation of the scanner, taking into account the mutual influence between these parameters, for example: the imaging resolution represents the clarity of the image, the higher the resolution, the clearer the image, the higher the signal-to-noise ratio, the better the image quality, the higher the contrast, the stronger the layering of the image, the higher the saturation, the brighter the color of the image. A high-resolution image can contain more details, that is, at the same noise level, a high-resolution image may look more "noisy". By improving the signal-to-noise ratio during image acquisition and processing, noise can be reduced while maintaining high resolution, thereby improving image quality. High contrast can make details more vivid and further enhance the clarity and layering of the image. Resolution mainly affects the detail performance of the image, and saturation focuses on the vividness of the color. In a high-resolution image, richer color details and higher saturation can enhance each other. The startup response delay has an important impact on imaging. If the startup response delay is too long, it will affect the synchronization and scanning accuracy of the scanner, especially in application scenarios that require high precision and fast response. The delay may cause incomplete or misaligned image capture, thereby affecting the quality of the final imaging. The startup response delay and the scanning execution rate are closely related in the imaging process of the scanner. Scanners with shorter startup response delays can enter the working state faster and can cooperate with higher scanning execution rates to achieve fast and efficient image acquisition. Therefore, when designing and debugging scanners, it is necessary to balance the relationship between the two to ensure that the scanner can both respond quickly to the startup command and achieve high-quality image presentation at an appropriate rate.

[0046] Furthermore, the scanner is adaptively adjusted and controlled based on the pre-verification judgment result, and the specific process is: based on the pre-verification judgment result, if the pre-verification judgment result is that the pre-verification of the scanner is normal, there is no need to adaptively adjust and control the scanner; if the pre-verification judgment result is that the pre-verification of the scanner is abnormal, the preset adaptive adjustment parameters are extracted from the database, and the scanner is adaptively adjusted and controlled for the first time based on the preset adaptive adjustment parameters, and the scanner after the first adaptive adjustment control is recorded as the scanner after the initial adjustment; the scanner after the initial adjustment is pre-verified for a second time, and the second pre-verification evaluation value of the scanner after the initial adjustment is obtained by analysis; the second pre-verification evaluation value of the scanner after the initial adjustment is compared with the first pre-verification evaluation value of the scanner, and the first pre-verification evaluation value of the scanner after the initial adjustment is extracted. The difference between the second pre-verification evaluation value and the first pre-verification evaluation value of the scanner is recorded as the pre-verification adjustment benchmark value; if the pre-verification adjustment benchmark value is below zero, a warning message is generated and uploaded to the cloud computing center; if the pre-verification adjustment benchmark value is greater than zero, the pre-verification adjustment benchmark value is compared with the pre-verification adjustment threshold in the database, if the pre-verification adjustment benchmark value is greater than the pre-verification adjustment threshold, the scanner after the initial adjustment is marked as an available scanner, if the pre-verification adjustment benchmark value is below the pre-verification adjustment threshold, the scanner after the initial adjustment is adaptively adjusted again with the preset adaptive adjustment parameters until the pre-verification adjustment benchmark value is greater than the pre-verification adjustment threshold, and when the pre-verification adjustment benchmark value is greater than the pre-verification adjustment threshold, the current scanner is recorded as an available scanner.

[0047] In the present embodiment, it should be noted that the preset adaptive adjustment parameters include: resolution adjustment supplementary value, color mode, brightness adjustment supplementary value and contrast adjustment supplementary value. The adaptive adjustment parameters of the scanner are adjusted through the internal image sensor, optical system, light source and related electronic control and processing chips. These hardware components cooperate with the internal software algorithm to optimize the working parameters of the scanner. The prompt text of the early warning information is, for example: Scanner adjustment failed, please deal with it in time. The method and steps for performing a second pre-verification on the scanner after the initial adjustment are the same as the method and steps for pre-verification of the scanner when it is first started, and the pre-verification judgment result of the scanner is obtained by analysis, which will not be repeated in the following text.

[0048] The scanner after the initial adjustment is pre-calibrated for the second time, and the second pre-calibration evaluation value of the scanner after the initial adjustment is obtained by analysis. The calculation formula of the second pre-calibration evaluation value of the scanner after the initial adjustment is:

[0049]

[0050] In the formula, E 2 represents the second pre-calibration evaluation value of the scanner after the initial adjustment, R 2 Represents the imaging resolution of the scanner after the initial adjustment, R0 represents the initial imaging resolution, σ R Indicates the image resolution calibration reference value, S 2 It represents the imaging signal-to-noise ratio of the scanner after the initial adjustment, S 0 represents the initial imaging signal-to-noise ratio, σ S Indicates the image signal-to-noise ratio calibration reference value, A 2 Indicates the imaging contrast of the scanner after the initial adjustment, A 0 represents the initial contrast, σ A Indicates the image contrast calibration reference value, H 2 Indicates the image saturation of the scanner after the initial adjustment, H 0 represents the initial image saturation, σ H Indicates the image saturation calibration reference value, T 0 represents the standard startup response delay in the database, T r represents the startup response delay of the scanner after the initial adjustment, V 0 represents the standard scan execution rate in the database, V r represents the scanning execution rate of the scanner after the initial adjustment, α 1 represents the weight of image resolution, α 2 represents the imaging signal-to-noise ratio influence weight, α 3 Represents the image contrast influence weight, α 4 Indicates the image saturation influence weight.

[0051] It should be noted that the pre-calibration data of the scanner after the initial adjustment includes: the imaging signal-to-noise ratio, imaging signal-to-noise ratio, imaging contrast and imaging saturation of the scanner after the initial adjustment, and startup performance feedback parameters.

[0052] It should be understood that the pre-verification adjustment reference value is the difference between the second pre-verification evaluation value of the scanner after the initial adjustment and the first pre-verification evaluation value of the scanner, specifically the second pre-verification evaluation value of the scanner after the initial adjustment minus the first pre-verification evaluation value of the scanner, and the result can be a positive number, a negative number or zero.

[0053] It is also necessary to understand that the number of times of adaptive adjustment and control of the scanner based on the pre-verification judgment result is not unique, and the imaging resolution, imaging signal-to-noise ratio, imaging contrast and imaging saturation obtained after each adaptive adjustment can be used as the imaging data of the startup image captured by the scanner after the initial adjustment and substituted into the second pre-verification evaluation value calculation formula of the scanner after the initial adjustment to calculate and update the second pre-verification evaluation value of the scanner after the initial adjustment. Since the processing method of the subsequent steps is unified, it will not be repeated here.

[0054] Furthermore, a plane scanned image of the textile after dyeing is collected by an available scanner, and the image is compared with a plane dyeing effect diagram of the textile to extract a textile dyeing difference data set. The specific comparison process includes: collecting a plane scanned image of the textile after dyeing by an available scanner; extracting the plane dyeing effect diagram of the textile, and dividing the plane dyeing effect diagram of the textile into various dyeing sub-regions according to a preset division area, and synchronously extracting a scanned image corresponding to each dyeing sub-region in the plane scanned image of the textile after dyeing according to the distribution position of each dyeing sub-region, and recording it as an actual dyeing plane image of each dyeing sub-region; recording the image corresponding to each dyeing sub-region in the plane dyeing effect diagram of the textile as an ideal dyeing effect diagram of each dyeing sub-region; comparing the ideal dyeing effect diagram of each dyeing sub-region with the actual dyeing plane image of each dyeing sub-region, to obtain a textile dyeing difference data set.

[0055] In this embodiment, the step of collecting the plane scanned image of the textile after dyeing ensures that the surface information of the textile after dyeing can be completely and accurately collected, providing basic data for subsequent image processing. Through the pre-designed dyeing scheme or template, the plane dyeing effect diagram of the textile is extracted. The effect diagram is an ideal image set in the dyeing process, reflecting the expected result of the dyeing process, and ensuring that each dyeing sub-area can accurately correspond to the dyeing part on the actual textile.

[0056] Furthermore, the textile dyeing difference dataset, the specific comparison process includes: the dyeing pattern area difference of each dyeing sub-region, the dyeing pattern edge contour line length difference, the dyeing pattern porosity and the color difference of each dyeing sub-region at each screening point.

[0057] In this embodiment, it should be noted that the textile dyeing difference dataset can be obtained by scanning with a scanner and then transmitting it to an image processor for processing, such as a GPU (graphics processing unit) and a CPU (central processing unit).

[0058] Furthermore, the textile dyeing difference data set is analyzed to obtain the quality judgment result of the textile dyeing, specifically including: obtaining a preset textile dyeing difference control data set, and comparing it with the textile dyeing difference data set to obtain a textile dyeing difference evaluation value; obtaining a textile dyeing difference evaluation threshold, and comparing it with the textile dyeing difference evaluation value; if the textile dyeing difference evaluation value is above the dyeing difference evaluation threshold, then the quality judgment result of the textile dyeing is invalid dyeing; if the textile dyeing difference evaluation value is less than the dyeing difference evaluation threshold, then the quality judgment result of the textile dyeing is valid dyeing.

[0059] In this embodiment, it should be understood that the preset textile dyeing difference comparison data set includes: preset dyeing pattern reference area difference of each dyeing sub-region, preset dyeing pattern edge contour line reference length difference, preset dyeing pattern reference porosity, and preset reference color difference of each dyeing sub-region at each screening point. The preset textile dyeing difference comparison data set can be obtained through a database.

[0060] Furthermore, the textile dyeing difference evaluation value is calculated as follows:

[0061]

[0062] In the formula, Q 1 represents the textile dyeing difference evaluation value, i represents the number of the dyeing sub-region, n represents the number of the dyeing sub-region, D i represents the difference in the dyeing pattern area of ​​the i-th dyeing sub-region, D 0 Indicates the preset allowable area difference of the dyeing pattern, L i represents the length difference of the edge contour line of the dyeing pattern of the i-th dyeing sub-region, L 0 Indicates the preset allowable edge contour line allowable length difference, P i represents the saturation difference of the center point of the dyeing pattern of the i-th dyeing sub-region, P 0 Indicates the preset permitted center point saturation difference, C i represents the color difference of the i-th dyeing subregion at the screening point, C 0 Indicates the preset permissible color difference, ω 1 represents the influence weight of the dyeing pattern area difference, ω 2 Indicates the influence weight of the length difference of the edge contour line of the dyeing pattern, ω 3 Indicates the influence weight of the center point saturation difference, ω 4 Indicates that color difference affects the weight.

[0063] In this embodiment, it should be understood that the preset allowable area difference of the dyeing pattern, the preset allowable length difference of the edge contour line, the preset center point saturation difference and the preset allowable color difference represent the minimum reference standard of textile dyeing differences under ideal conditions, that is, the minimum reference value of textile dyeing differences under ideal conditions. The purpose of comparing the textile dyeing difference data set with the ideal minimum reference value of textile dyeing differences is to take into account that if the textile dyeing difference control data set is at a lower numerical level compared to these ideal minimum reference values ​​of textile dyeing differences, the quality of the textile dyeing is better.

[0064] It should be noted that the influence weight of the dyeing pattern area difference, the influence weight of the length difference of the edge contour line of the dyeing pattern, the influence weight of the saturation difference of the center point of the dyeing pattern and the influence weight of the color difference respectively represent the numerical values ​​of the influence degree of the dyeing pattern area difference, the length difference of the edge contour line of the dyeing pattern, the saturation difference of the center point of the dyeing pattern and the color difference on the evaluation value of the dyeing difference of the textile. When used, the influence weights corresponding to the dyeing pattern area difference, the length difference of the edge contour line of the dyeing pattern, the saturation difference of the center point of the dyeing pattern and the color difference can be directly obtained from the database, and the corresponding relationship can be a pre-set mapping relationship, for example, the dyeing pattern area difference, the length difference of the edge contour line of the dyeing pattern, the saturation difference of the center point of the dyeing pattern and the color difference and the dyeing pattern area difference influence weight, the dyeing pattern edge contour line length difference influence weight, the dyeing pattern porosity influence weight and the color difference influence weight form a mapping set, and the real-time dyeing pattern area difference, the dyeing pattern edge contour line length difference, the dyeing pattern center point saturation difference and the color difference are input into the mapping set to obtain the influence weights corresponding to the dyeing pattern area difference, the dyeing pattern edge contour line length difference, the dyeing pattern center point saturation difference and the color difference, wherein the mapping relationship can be a one-to-one correspondence or a many-to-one relationship. For example: the area difference of the dyeing pattern corresponds to the influence weight of the area difference of the dyeing pattern, the length difference of the edge contour line of the dyeing pattern corresponds to the influence weight of the edge contour line of the dyeing pattern, the saturation difference of the center point of the dyeing pattern corresponds to the influence weight of the saturation difference of the center point of the dyeing pattern, and the color difference corresponds to the influence weight of the color difference.

[0065] It should also be noted that in this embodiment, the textile dyeing difference evaluation value is obtained according to the difference in the dyeing pattern area of ​​each dyeing sub-region, the difference in the length of the edge contour line of the dyeing pattern, the difference in the saturation of the center point of the dyeing pattern, and the color difference of each dyeing sub-region at each screening point. This is to take into account the mutual influence between these parameters. For example, the expansion of the area of ​​the dyeing pattern may cause a corresponding increase in the length of its edge contour line. This change may further indirectly affect the apparent presentation of porosity through the distribution and penetration of the dye, thereby reducing the number or area of ​​pores. At the same time, the change in color difference directly reflects the stability of the dyeing process and the uniformity of the dye. In addition, the color difference is also closely related to the clarity of the edge contour, and together constitutes an important dimension for evaluating the quality of the dyeing pattern. In summary, the interaction between these parameters jointly shapes the final quality of the dyeing pattern. Pre-drying includes: leaving the dyed textile to stand and using infrared rays to directly heat the moisture on the surface of the textile for microwave pre-drying.

[0066] Furthermore, the online dyeing detection system for textiles based on machine vision is characterized in that it also includes a drying execution judgment module for dyed textiles, which is used to perform drying execution judgment on the dyed textiles based on the quality judgment result of textile dyeing, specifically including: based on the quality judgment result of textile dyeing, if the quality judgment result is invalid dyeing, no drying treatment is required, and drying impact prompt information is generated to the cloud computing center for prompting; if the quality judgment result is effective dyeing, the drying impact parameters corresponding to the dyed textile are collected, and the drying feasibility judgment value of the dyed textile is obtained through processing, and compared with the textile drying feasibility judgment threshold, if the drying feasibility judgment value of the dyed textile is above the textile drying feasibility judgment threshold, it is determined that the dyed textile can be dried, and if the drying feasibility judgment value of the dyed textile is less than the textile drying feasibility judgment threshold, it is determined that the dyed textile needs to be pre-drying, thereby completing the drying execution judgment of the dyed textile.

[0067] In this embodiment, the drying influencing parameters corresponding to the dyed textile include: the average surface humidity of the dyed textile, the porosity of the dyed textile, the average surface temperature of the dyed textile, and the specific surface area of ​​the dyed textile. The calculation formula for the determination value of the drying feasibility of the dyed textile is:

[0068]

[0069] In the formula, G 1 represents the drying feasibility judgment value of dyed textiles, B represents the average surface humidity of dyed textiles, and B 0 represents the preset permissible surface average humidity, K represents the porosity of the dyed textile, K 0 represents the preset permissible porosity, F represents the average surface temperature of the textile after dyeing, F 0 represents the preset permissible surface average temperature, M represents the specific surface area of ​​the dyed textile, M 0 represents the preset allowable specific surface area, γ 1 represents the average surface humidity influence weight of the dyed textile, γ 2 represents the porosity influence weight of the dyed textile, γ 3 Represents the weight of the average surface temperature of the textile after dyeing.

[0070] It should be understood that the preset permissible porosity, the preset permissible surface average temperature and the preset permissible specific surface area represent the lowest reference standard for drying under ideal conditions, i.e., the lowest reference value for ideal drying. By comparing the drying influence parameters corresponding to the dyed textile with the lowest reference value for ideal drying, the purpose is to consider that if the drying influence parameters corresponding to the dyed textile are at a higher numerical level than these lowest reference values ​​for ideal drying, it indicates that the drying feasibility judgment value of the dyed textile is better and more suitable for drying treatment. The preset permissible specific surface area represents the highest reference standard for specific surface area under ideal conditions, i.e., the highest reference value for ideal specific surface area. By comparing the specific surface area of ​​the dyed textile with the highest reference value for ideal specific surface area, the purpose is to consider that if the specific surface area of ​​the dyed textile is at a more dispersed numerical level than the highest reference value for ideal specific surface area, it indicates that the drying feasibility judgment value of the dyed textile is better and more suitable for drying treatment.

[0071] It should be noted that the preset permitted surface average humidity, the preset permitted surface average temperature, the preset permitted porosity and the preset permitted specific surface area can be obtained by the database. The surface average humidity of the dyed textile can be obtained by a humidity sensor, the porosity of the dyed textile can be obtained by collecting the surface image of the textile by a scanner and combining the image processing software (such as ImageJ), the surface average temperature of the dyed textile can be obtained by measuring the temperature sensor, and the specific surface area of ​​the dyed textile can be obtained by measuring the fully automatic specific surface and porosity analyzer. In the present embodiment, the textile drying feasibility determination value after dyeing is obtained by processing the surface average humidity of the dyed textile, the porosity of the dyed textile, the surface average temperature of the dyed textile and the specific surface area of ​​the dyed textile, considering the mutual influence between these parameters, for example: porosity enhances air permeability, promotes surface humidity evaporation, reduces thermal conductivity, and high specific surface area increases the contact point between fiber and environment, improves hygroscopicity and heat exchange efficiency. Surface humidity affects the water distribution in the pores and indirectly changes thermal conductivity. Thermal conductivity is regulated by the pore structure, and the pore size and shape distribution have a significant impact.

[0072] It should also be noted that the influence weight of the average surface humidity of the dyed textile, the influence weight of the porosity of the dyed textile and the influence weight of the average surface temperature of the dyed textile respectively represent the numerical values ​​of the influence degree of the average surface humidity of the dyed textile, the porosity of the dyed textile and the average surface temperature of the dyed textile on the textile dyeing difference assessment value. When used, the influence weights corresponding to the average surface humidity of the dyed textile, the porosity of the dyed textile and the average surface temperature of the dyed textile can be directly obtained from the database, and the corresponding relationship can be a pre-set mapping relationship. For example, the average surface humidity of the dyed textile, the porosity of the dyed textile and the average surface temperature of the dyed textile form a mapping set with the influence weight of the average surface humidity of the dyed textile, the influence weight of the porosity of the dyed textile and the influence weight of the average surface temperature of the dyed textile, and the real-time average surface humidity of the dyed textile, the porosity of the dyed textile and the average surface temperature of the dyed textile are input into the mapping set to obtain the influence weights corresponding to the average surface humidity of the dyed textile, the porosity of the dyed textile and the average surface temperature of the dyed textile, wherein the mapping relationship can be a one-to-one correspondence or a many-to-one relationship. For example: the average surface humidity of the dyed textile and the influence weight of the average surface humidity of the dyed textile, the porosity of the dyed textile and the influence weight of the porosity of the dyed textile, and the average surface temperature of the dyed textile and the influence weight of the average surface temperature of the dyed textile correspond one to one.

[0073] Furthermore, a method for online dyeing detection of textiles based on machine vision is characterized in that it includes the following steps: receiving a textile dyeing detection signal through a cloud computing center, and synchronously starting a scanner for pre-verification to obtain pre-verification data of the scanner; according to the pre-verification data of the scanner, and obtaining a pre-verification judgment result of the scanner through analysis, synchronously adaptively adjusting and controlling the scanner based on the pre-verification judgment result, and recording the scanner after adaptive adjustment and control as an available scanner; collecting a plane scan image of the textile after dyeing through an available scanner, and comparing it with a plane dyeing effect diagram of the textile to extract a textile dyeing difference data set; analyzing the textile dyeing difference data set to obtain a quality judgment result of the textile dyeing, and transmitting it to the cloud computing center for prompting.

[0074] In this embodiment, the cloud computing center receives the dyeing detection signal and starts the scanner for pre-verification synchronously, which can significantly improve the automation and accuracy of the detection process. The scanner is adaptively adjusted and controlled based on the pre-verification result to ensure that the scanner is always in the best working state, thereby improving the reliability and consistency of textile dyeing quality detection.

[0075] To summarize, this embodiment receives the textile dyeing detection signal through the cloud computing center and synchronously starts the scanner for pre-verification, thereby evaluating the feedback parameters and image quality of the scanner startup performance, and then accurately identifying whether the color distribution of the textile is uniform, effectively solving the problem of inaccurate textile dyeing quality control in the prior art.

[0076] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0077] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0078] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0079] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0080] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0081] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. The online dyeing detection system for textiles based on machine vision is characterized by: include: Scanner pre-verification module, pre-verification data analysis module, textile image acquisition module and dyeing quality determination module; The scanner pre-verification module is used to receive the textile dyeing detection signal through the cloud computing center, and synchronously start the scanner to perform pre-verification to obtain the scanner pre-verification data; The pre-verification data analysis module is used to obtain the pre-verification determination result of the scanner through analysis according to the pre-verification data of the scanner, synchronously perform adaptive adjustment control on the scanner based on the pre-verification determination result, and record the scanner after adaptive adjustment control as an available scanner; The textile image acquisition module is used to acquire a plane scanned image of the textile after dyeing through an available scanner, and compare it with the textile plane dyeing effect diagram to extract a textile dyeing difference data set; The dyeing quality determination module is used to analyze the textile dyeing difference data set, obtain the textile dyeing quality determination result, and transmit it to the cloud computing center for prompting.

2. The online dyeing detection system for textiles based on machine vision as claimed in claim 1, characterized in that: The synchronously starting the scanner to perform pre-verification and obtain the pre-verification data of the scanner includes: Start the scanner for pre-checking, and obtain startup performance feedback parameters of the scanner and startup shooting images, wherein the startup performance feedback parameters include startup response delay and scanning execution rate; According to the startup image of the scanner, the image is input into the image processor to identify and extract the imaging data of the startup image, including imaging resolution, imaging signal-to-noise ratio, imaging contrast and imaging saturation; The startup performance feedback parameters of the scanner and the imaging data of the startup captured image are jointly marked as the pre-verification data of the scanner.

3. The online dyeing detection system for textiles based on machine vision as claimed in claim 2, characterized in that: The steps of obtaining the pre-verification result of the scanner by analysis include: Extracting an initial archive verification image of the scanner from a database, and extracting imaging data of the initial archive verification image through an image processor, and marking the imaging data as imaging control data of the scanner, wherein the imaging control data includes initial imaging resolution, initial imaging signal-to-noise ratio, initial imaging contrast, and initial imaging saturation; Based on the scanner's imaging control data, the imaging data of the startup image is compared with the imaging data of the startup image, and the startup performance feedback parameters of the scanner are combined to obtain a first pre-verification evaluation value of the scanner; A first pre-verification evaluation threshold is obtained and compared with the first pre-verification evaluation value of the scanner. If the first pre-verification evaluation value of the scanner is above the first pre-verification evaluation threshold, the pre-verification judgment result of the scanner is defined as normal scanner pre-verification; if the first pre-verification evaluation value of the scanner is less than the first pre-verification evaluation threshold, the pre-verification judgment result of the scanner is defined as abnormal scanner pre-verification.

4. The online dyeing detection system for textiles based on machine vision as claimed in claim 3, characterized in that: The specific process of adaptively adjusting and controlling the scanner based on the pre-verification result is as follows: Based on the pre-verification determination result, if the pre-verification determination result is that the scanner pre-verification is normal, there is no need to perform adaptive adjustment control on the scanner; If the pre-verification determination result is that the scanner pre-verification is abnormal, the preset adaptive adjustment parameters are extracted from the database, and the scanner is adaptively adjusted and controlled for the first time based on the preset adaptive adjustment parameters, and the scanner after the first adaptive adjustment control is recorded as the scanner after the initial adjustment; Performing a second pre-calibration on the scanner after the initial adjustment, and analyzing to obtain a second pre-calibration evaluation value of the scanner after the initial adjustment; Compare the second pre-calibration evaluation value of the scanner after the initial adjustment with the first pre-calibration evaluation value of the scanner, and extract the difference between the second pre-calibration evaluation value of the scanner after the initial adjustment and the first pre-calibration evaluation value of the scanner, and record it as the pre-calibration adjustment reference value; If the pre-check adjustment reference value is below zero, a warning message is generated and uploaded to the cloud computing center; If the pre-verification adjustment baseline value is greater than zero, the pre-verification adjustment baseline value is compared with the pre-verification adjustment threshold in the database. If the pre-verification adjustment baseline value is greater than the pre-verification adjustment threshold, the scanner after the initial adjustment is marked as an available scanner. If the pre-verification adjustment baseline value is below the pre-verification adjustment threshold, the scanner after the initial adjustment is adaptively adjusted again with the preset adaptive adjustment parameters until the pre-verification adjustment baseline value is greater than the pre-verification adjustment threshold. When the pre-verification adjustment baseline value is greater than the pre-verification adjustment threshold, the current scanner is marked as an available scanner.

5. The online textile dyeing detection system based on machine vision as claimed in claim 1, characterized in that: The method uses a scanner to collect a plane scanned image of the textile after dyeing, and compares it with the textile plane dyeing effect diagram to extract a textile dyeing difference data set. The specific comparison process includes: Acquiring flat scan images of textiles after dyeing by using available scanners; Extracting a textile plane dyeing effect image, dividing the textile plane dyeing effect image into various dyeing sub-regions according to a preset division area, and synchronously extracting a scanning image corresponding to each dyeing sub-region from a plane scanning image after the textile is dyed according to the distribution position of each dyeing sub-region, and recording it as an actual dyeing plane image of each dyeing sub-region; The image corresponding to each dyeing sub-region in the textile plane dyeing effect diagram is recorded as the ideal dyeing effect diagram of each dyeing sub-region; The ideal dyeing effect image of each dyeing sub-region is compared with the actual dyeing plane image of each dyeing sub-region to obtain the textile dyeing difference dataset.

6. The online textile dyeing detection system based on machine vision as claimed in claim 5, characterized in that: The textile dyeing difference data set specifically includes the comparison process of: the dyeing pattern area difference of each dyeing sub-region, the length difference of the edge contour line of the dyeing pattern, the porosity of the dyeing pattern, and the color difference of each dyeing sub-region at each screening point.

7. The online textile dyeing detection system based on machine vision as claimed in claim 6, characterized in that: The textile dyeing difference data set is analyzed to obtain the textile dyeing quality determination result, specifically including: Obtaining a preset textile dyeing difference control data set, and comparing it with the textile dyeing difference data set to obtain a textile dyeing difference evaluation value; The textile dyeing difference assessment threshold is obtained and compared with the textile dyeing difference assessment value. If the textile dyeing difference assessment value is above the dyeing difference assessment threshold, the quality judgment result of the textile dyeing is invalid dyeing. If the textile dyeing difference assessment value is less than the dyeing difference assessment threshold, the quality judgment result of the textile dyeing is valid dyeing.

8. The online textile dyeing detection system based on machine vision as claimed in claim 7, characterized in that: The textile dyeing difference evaluation value is calculated by the formula: Where Q1 represents the textile dyeing difference evaluation value, i represents the number of the dyeing sub-region, n represents the number of the dyeing sub-region, and D i represents the area difference of the dyeing pattern of the i-th dyeing sub-region, D0 represents the preset allowable area difference of the dyeing pattern, and L i represents the length difference of the edge contour line of the dyeing pattern of the i-th dyeing sub-region, L0 represents the preset allowable edge contour line allowable length difference, P i represents the saturation difference of the center point of the dyeing pattern of the i-th dyeing sub-region, P0 represents the preset permitted center point saturation difference, C i represents the color difference of the i-th dyeing sub-region at the screening point, C0 represents the preset permitted color difference, ω1 represents the influence weight of the dyeing pattern area difference, ω2 represents the influence weight of the dyeing pattern edge contour line length difference, ω3 represents the influence weight of the center point saturation difference, and ω4 represents the influence weight of the color difference.

9. The online textile dyeing detection system based on machine vision as claimed in claim 1, characterized in that: It also includes a post-dyed textile drying execution determination module, which is used to perform a drying execution determination on the dyed textile based on the quality determination result of the textile dyeing, and specifically includes: Based on the quality judgment result of textile dyeing, if the quality judgment result is invalid dyeing, there is no need to perform drying treatment, and a drying impact prompt information is generated and sent to the cloud computing center for prompting; If the quality judgment result is effective dyeing, the drying influencing parameters corresponding to the dyed textile are collected, and the drying feasibility judgment value of the dyed textile is obtained through processing, and compared with the textile drying feasibility judgment threshold. If the drying feasibility judgment value of the dyed textile is above the textile drying feasibility judgment threshold, it is judged that the dyed textile can be dried; if the drying feasibility judgment value of the dyed textile is less than the textile drying feasibility judgment threshold, it is judged that the dyed textile needs to be pre-drying, thereby completing the drying execution judgment of the dyed textile.

10. The method for online dyeing detection of textiles based on machine vision according to claim 1, characterized in that: The following steps are involved: The textile dyeing detection signal is received through the cloud computing center, and the scanner is started simultaneously for pre-calibration to obtain the pre-calibration data of the scanner; According to the pre-verification data of the scanner, a pre-verification determination result of the scanner is obtained by analysis, and the scanner is adaptively adjusted and controlled based on the pre-verification determination result, and the scanner after the adaptive adjustment control is recorded as an available scanner; The flat scanned images of textiles after dyeing are collected by using available scanners, and compared with the flat dyeing effect images of textiles to extract the textile dyeing difference dataset; The textile dyeing difference data set is analyzed to obtain the quality judgment results of textile dyeing, and then transmitted to the cloud computing center for prompting.

Images