A fabric quality online identification system and method

Abstract

The invention relates to the technical field of intelligent detection, and specifically discloses an online identification system and method for fabric quality. According to the printing quality deviation table value, fabric printed fabrics in a fabric printed fabric group are divided into surface deviation fabric printed fabrics and surface correct fabric printed fabrics. By comparing and processing the surface deviation fabric printed fabrics and the surface correct fabric printed fabrics, the state non-stable value of the fabric printed fabric group is obtained. Based on the state non-stable value, the state signal identification of the fabric printed fabric group is completed. According to the non-stable signal of the fabric printed fabric group, the surface quality of the fabric printed fabric in the processing process and the working condition of the printing equipment are traced and identified, and the control of the fabric printed fabric in the printing process is completed. The invention can timely and effectively identify the cause of abnormal printing when the printing of the fabric printed fabric is abnormal, facilitates the tracking and processing of the quality of the printing equipment, and improves the printing process fluency and printing quality of the fabric printed fabric.

Description

Technical Field

[0001] The present invention relates to the field of intelligent detection technology, and in particular to an online identification system and method for fabric quality. Background Art

[0002] Textile processing is the process of manipulating the material properties, surface texture and pattern of fabrics through processing technology. During the process, processing equipment is required to process the fabrics.

[0003] In the fabric processing technology, after the fabric is prepared, it needs to be quality inspected to determine the production quality of each batch of fabric, thereby reducing the circulation of defective products and improving the production and processing quality of fabric.

[0004] During the processing of existing fabrics, the printing quality is usually tested through manual sampling, which has poor accuracy. In addition, during the testing process, due to the sampling test, it is not possible to timely and effectively detect whether it is a single quality problem or a problem with the entire printing process. It has poor applicability and lacks multi-dimensional analysis of detection anomalies, and cannot effectively match the production speed of fabrics. Summary of the invention

[0005] In order to make up for the deficiencies of the prior art, the problem that the prior art cannot timely and effectively detect whether it is a single quality problem or an overall printing process problem during the sampling and testing process, and cannot effectively match the production speed of the fabric, is solved.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A method for online identification of fabric quality comprises the following steps:

[0008] In the process of fabric printing, a plurality of fabric printing fabrics are obtained to obtain a fabric printing fabric group, and each independent fabric printing image of the fabric printing fabric group is processed to obtain printing extraction data of the fabric printing image;

[0009] Based on the printing extraction data, the printing quality deviation value of the printed fabric is obtained;

[0010] According to the printing quality difference value, the printed fabrics in the printed fabric group are divided into the surface difference printed fabrics and the surface correct printed fabrics, and the state non-stable value of the printed fabric group is obtained by comparing the surface difference printed fabrics and the surface correct printed fabrics.

[0011] Based on the state non-stable value, the state signal recognition of the fabric printing fabric group is completed;

[0012] Wherein, the state signal of the fabric printing fabric group includes an unstable signal of the fabric printing fabric group and a stable signal of the fabric printing fabric group;

[0013] Based on the non-stationary signal of the fabric printing fabric group, the surface quality of the fabric during the processing and the working condition of the printing equipment are traced and identified to complete the regulation of the fabric during the printing process.

[0014] As a further solution of the present invention: the print extraction data of the fabric print image includes print color data and print surface data;

[0015] The printing color data is processed to obtain the saturation difference value, which is recorded as Zs;

[0016] The printed surface data is processed to obtain the surface quality difference value, which is recorded as Zb;

[0017] By formula The printing quality deviation value Zi of the printed fabric is calculated, where k is a preset proportional coefficient.

[0018] As a further solution of the present invention: the process of obtaining the saturation difference value is:

[0019] Dividing the fabric print image into a plurality of print image sub-units;

[0020] Convert each print image subunit into an adaptive color space and calculate the corresponding histogram so that the histogram of each color channel is between 0 and 1;

[0021] A saturation deviation ratio of each color channel is obtained based on the pixel value of each color channel;

[0022] Then, the saturation variance value and the saturation extreme deviation ratio of the fabric printing image are obtained according to the saturation deviation ratio of each color channel;

[0023] The saturation variance value is multiplied by the saturation extreme deviation ratio to obtain the saturation outlier of the fabric printing image.

[0024] As a further solution of the present invention: the process of obtaining the outlier value of the table quality is:

[0025] Recording the printed image subunit with printed fragments as a different image subunit;

[0026] Calculate the ratio of the number of different image sub-units to the number of printed image sub-units to obtain a different image unit ratio;

[0027] Obtaining the difference area of ​​the difference image sub-unit, calculating the ratio of the total area of ​​the difference area to the number of the print image sub-units, and obtaining the difference ratio of the difference area;

[0028] The surface quality difference value of the fabric printing image is obtained by multiplying the difference image unit ratio and the difference area surface difference ratio.

[0029] As a further solution of the present invention: the number of printed fabrics with different appearances is calculated by ratio to the total number of printed fabrics in the printed fabric group to obtain a ratio of different appearances of the printed fabric group, which is recorded as Ms;

[0030] The printing quality deviation value is calculated by taking the difference between the mean of the printing quality deviation values ​​of all the normal printed fabrics and the mean of the printing quality deviation values ​​of the abnormal printed fabrics;

[0031] The deviation value of printing quality difference is calculated by ratio with the mean value of printing quality difference of all printed fabrics, and the difference ratio of printed fabric group is obtained, which is recorded as Mc.

[0032] The variance of the printed fabric group printing quality difference table value is calculated to obtain the variance value of the printed fabric group, which is recorded as Mf;

[0033] By formula The state non-stable value Mi of the fabric printing fabric group is calculated, wherein a1 and a2 are both preset proportional coefficients.

[0034] As a further solution of the present invention: if the state instability value Mi of the textile printing fabric group is greater than or equal to the state instability threshold mi of the textile printing fabric group, an instability signal of the textile printing fabric group is generated;

[0035] If the state instability value Mi of the fabric printing fabric group is less than the state instability threshold mi of the fabric printing fabric group, a stable signal of the fabric printing fabric group is generated.

[0036] As a further solution of the present invention: based on the unstable signal of the fabric printing fabric group, the surface anomaly ratio of the fabric printing fabric group is analyzed:

[0037] If the surface anomaly ratio of the fabric printing fabric group is within a preset range of the surface anomaly ratio of the fabric printing fabric group, a fabric printing normal signal is generated;

[0038] If the surface difference number ratio of the fabric printing fabric group is outside the preset range of the surface difference number ratio of the fabric printing fabric group, a fabric printing validation signal is generated.

[0039] As a further solution of the present invention: based on the printing validation signal of fabric;

[0040] Check the appearance of all printed fabrics with different appearances in the printed fabric group;

[0041] Calculate the ratio of the number of non-standard fabrics to the number of printed fabrics with different appearances to obtain the non-standard ratio of fabrics;

[0042] If the non-standard ratio of the fabric is within a preset non-standard ratio range of the fabric, a printing device adjustment signal is generated;

[0043] If the non-standard ratio of the fabric is outside the preset non-standard ratio range of the fabric, a printing device tracking signal is generated;

[0044] According to the tracking signal of the printing equipment, the printing quality deviation values ​​of all the normal fabric printing fabrics in the fabric printing fabric group are processed, and the normal signal of the printing equipment and the adjustment signal of the printing equipment are obtained according to the changing trend of the printing quality deviation values ​​of all the normal fabric printing fabrics during the printing time.

[0045] As a further solution of the present invention: based on the printing device adjustment signal, the factors in the printing device that interfere with the printing quality of the fabric are tracked and identified.

[0046] As a further solution of the present invention: a fabric quality online identification system, comprising:

[0047] The printing data extraction module is used to obtain multiple fabric printing fabrics to obtain a fabric printing fabric group during the fabric printing process, process each independent fabric printing image of the fabric printing fabric group, obtain the printing extraction data of the fabric printing image, and upload the printing extraction data to the cloud management and control platform;

[0048] Printing quality characterization module, the printing quality characterization module receives the printing extraction data transmitted by the cloud management and control platform, obtains the printing quality deviation value of the fabric printing material, and uploads the printing quality deviation value to the cloud management and control platform;

[0049] A grouping state analysis module, the grouping state analysis module receives the printing quality deviation value transmitted by the cloud management and control platform, and the grouping state analysis module is used to divide the fabric printing fabrics in the fabric printing fabric group into the surface deviation fabric printing fabrics and the surface correct fabric printing fabrics, and obtain the state non-stable value of the fabric printing fabric group by comparing the surface deviation fabric printing fabrics and the surface correct fabric printing fabrics;

[0050] Complete the state signal recognition of the fabric printing fabric group based on the state non-stable value, and upload the state signal to the cloud management and control platform;

[0051] Wherein, the state signal of the fabric printing fabric group includes an unstable signal of the fabric printing fabric group and a stable signal of the fabric printing fabric group;

[0052] The traceability control module receives the unstable signal of the fabric printing fabric group transmitted by the cloud control platform. The traceability control module traces and identifies the surface quality of the fabric during the processing and the working condition of the printing equipment, and completes the control of the fabric during the printing process.

[0053] Beneficial effects of the present invention:

[0054] (1) In the continuous processing process of the fabric printing process, the present invention performs random inspection on the fabrics, and identifies the printing quality of the inspected fabric printing fabrics, that is, by identifying the color and appearance quality of the fabric printing image, the saturation difference value and the surface quality difference value of the fabric printing image are obtained, and the printing quality difference value of the fabric printing fabric is obtained according to the saturation difference value and the surface quality difference value. The larger the printing quality difference value, the more serious the color deviation and surface defects of the fabric printing fabric are, and the worse the printing processing quality of the fabric is, thereby completing the multi-dimensional identification of the fabric printing quality;

[0055] (2) The present invention identifies the quality of a fabric printing fabric group based on the print quality deviation value of the fabric printing fabric, that is, by obtaining the surface deviation fabric printing fabric and the surface correct fabric printing fabric of the fabric printing fabric group, processing the print quality deviation values ​​corresponding to the surface deviation fabric printing fabric and the surface correct fabric printing fabric, obtaining the surface deviation number ratio of the fabric printing fabric group and the surface deviation difference ratio of the fabric printing fabric group, and then combining the variance value of the fabric printing fabric group to obtain the state non-stable value of the fabric printing fabric group, and completing the overall state identification of the fabric printing fabric group through the state non-stable value of the fabric printing fabric group;

[0056] (3) The present invention checks the appearance quality of all abnormal fabrics printed on the basis of the unstable signal of the fabric printing fabric group, and judges whether the abnormality of the fabric during the printing process is caused by the quality problem of the fabric itself or the processing problem of the printing equipment. When the printing equipment adjustment signal is obtained, it indicates that the quality of the fabric itself is normal during the printing process, and the printing process of the printing equipment needs to be corrected and adjusted. The cause of the abnormal printing is identified in a timely and effective manner, which is convenient for tracking the quality of the printing equipment and improving the fluency and printing quality of the fabric printing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The present invention will be further described below in conjunction with the accompanying drawings.

[0058] Figure 1 It is a flow chart of a method for online identification of fabric quality according to an embodiment of the present invention;

[0059] Figure 2 It is a flow chart of fabric printing fabric identification in a fabric quality online identification method in an embodiment of the present invention;

[0060] Figure 3 It is a flow chart of fabric printing fabric group identification in a fabric fabric quality online identification method in an embodiment of the present invention;

[0061] Figure 4 It is a flowchart of a system for online identification of fabric quality according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example 1

[0063] See also Figure 1 As shown, the present invention is a method for online identification of fabric quality, comprising the following steps:

[0064] In the process of fabric printing, a plurality of fabric printing fabrics are obtained to obtain a fabric printing fabric group, and each independent fabric printing image of the fabric printing fabric group is processed to obtain printing extraction data of the fabric printing image;

[0065] Based on the printing extraction data, the printing quality deviation value of the printed fabric is obtained;

[0066] According to the printing quality difference value, the printed fabrics in the printed fabric group are divided into the surface difference printed fabrics and the surface correct printed fabrics, and the state non-stable value of the printed fabric group is obtained by comparing the surface difference printed fabrics and the surface correct printed fabrics.

[0067] Based on the state non-stable value, the state signal recognition of the fabric printing fabric group is completed;

[0068] Wherein, the state signal of the fabric printing fabric group includes an unstable signal of the fabric printing fabric group and a stable signal of the fabric printing fabric group;

[0069] Based on the non-stationary signal of the fabric printing fabric group, the surface quality of the fabric during the processing and the working condition of the printing equipment are traced and identified to complete the regulation of the fabric during the printing process.

[0070] Wherein, in the process of fabric printing, the printed fabric is recorded as fabric printing fabric;

[0071] Performing image acquisition on each fabric print fabric in the fabric print fabric group to obtain a fabric print image; performing feature extraction on the fabric print image to obtain print extraction data of the fabric print image;

[0072] It should be noted that the collection of fabric printing images is obtained by capturing the produced fabric printing images in real time with a high-speed camera;

[0073] In order to improve the quality of fabric printing images, remove noise, and enhance the useful information in fabric printing images, a denoising operation is performed on the fabric printing images to eliminate the noise in the fabric printing images. The denoising operation uses filters and other methods to make the fabric printing images smoother and clearer.

[0074] In order to highlight the key information in the fabric printing image, such as texture, color, etc., image enhancement is required. The fabric printing image is enhanced by methods such as histogram equalization and contrast stretching.

[0075] In order to simplify the image processing process, the grayscale image is converted into a binary image. By setting an appropriate threshold, the pixels in the fabric printing image are divided into foreground and background, which is convenient for subsequent feature extraction.

[0076] The print extraction data of the fabric print image includes print color data and print surface data;

[0077] Process the printing color data to obtain saturation values;

[0078] Process the printed surface data to obtain surface quality anomalies;

[0079] Dividing the fabric print image into a plurality of print image sub-units;

[0080] Convert each printed image subunit into a suitable color space and calculate the corresponding histogram;

[0081] Normalize the histogram of each color channel so that the histogram of each color channel is between 0 and 1;

[0082] Based on each color channel, the formula Calculate the saturation deviation ratio Si, where R, G, and B are the pixel values ​​of the red, green, and blue channels, respectively. is the maximum value among the three channels, is the minimum value among the three channels;

[0083] Obtaining the saturation deviation ratio of each color channel, averaging the saturation deviation ratios of each color channel, and obtaining the saturation deviation base of each printed image subunit;

[0084] Integrate the saturation deviation base number of all printed image units to obtain a saturation deviation base number array;

[0085] The saturation deviation base array is processed according to the variance calculation formula to obtain the saturation variance value;

[0086] The maximum saturation deviation base number and the minimum saturation deviation base number in the saturation deviation base number array are calculated to obtain the saturation deviation limit base number;

[0087] The saturation extreme deviation ratio is obtained by calculating the ratio of the saturation deviation limit base number to the minimum saturation deviation base number.

[0088] The saturation variance value is multiplied by the saturation extreme deviation ratio to obtain the saturation outlier of the fabric printing image.

[0089] Marking the printing fragments in the printing image subunit, and recording the printing image subunit with the printing fragments as the image subunit with difference;

[0090] The number of different image sub-units is obtained, and the ratio of the number of different image sub-units to the number of printed image sub-units is calculated to obtain a different image unit ratio;

[0091] Among them, printing fragments include surface defects and damaged points on the printed image;

[0092] Connect all the printed dots in the different image sub-unit end to end in a clockwise order to obtain the different area of ​​the different image sub-unit;

[0093] Obtaining the area value of the difference region, calculating the ratio of the area value of the difference region to the area value of the difference image sub-unit, and obtaining the area ratio of the difference region;

[0094] The difference area ratios of the plurality of difference image sub-units are summed to obtain a total difference area value;

[0095] The total area of ​​the difference region is calculated by ratio to the number of the sub-units of the printed image, so as to obtain the difference ratio of the difference region;

[0096] The surface quality difference value of the fabric printing image is obtained by multiplying the difference image unit ratio and the difference area surface difference ratio.

[0097] The saturation difference value of the fabric printing image is recorded as Zs;

[0098] The surface quality difference value of the fabric printing image is recorded as Zb;

[0099] By formula The printing quality difference value Zi of the printed fabric is calculated, where k is a preset proportional coefficient;

[0100] Among them, the process of obtaining k is: there are m groups of historical data, each group of historical data includes the saturation difference value Zs of the fabric printing image, the surface quality difference value Zb of the fabric printing image and the printing quality difference value Zi of the fabric printing image;

[0101] A linear model is used to fit m groups of historical data, and the prepared historical data is brought into the selected fitting model for fitting, and the mean of the fitting coefficients is obtained as the preset proportional coefficient k.

[0102] The technical scheme of the present invention is as follows: in the continuous processing process of the fabric printing process, the fabrics are sampled to obtain a fabric printing fabric group, and the printing quality of each fabric printing fabric in the fabric printing fabric group is identified, that is, by identifying the color and appearance quality of the fabric printing image, the saturation difference value and the surface quality difference value of the fabric printing image are obtained, and the printing quality difference value of the fabric printing fabric is obtained according to the saturation difference value and the surface quality difference value. The larger the printing quality difference value, the more serious the color deviation and surface defects of the fabric printing fabric are, and the worse the printing processing quality of the fabric is. Example 2

[0103] Based on the print quality deviation values ​​of the printed fabrics, the overall status of the printed fabric group is identified;

[0104] Comparing the printing quality deviation value of the fabric printed fabrics in the fabric printed fabric group with the printing quality deviation threshold value of the fabric printed fabrics;

[0105] If the printing quality deviation value of the fabric printed fabric is greater than or equal to the printing quality deviation threshold of the fabric printed fabric, the fabric printed fabric is recorded as a deviation fabric printed fabric;

[0106] If the printing quality deviation value of the fabric printed fabric is less than the printing quality deviation threshold of the fabric printed fabric, the fabric printed fabric is recorded as the normal fabric printed fabric;

[0107] The number of printed fabrics with different appearances is obtained, and the ratio of the number of printed fabrics with different appearances to the total number of printed fabrics in the printed fabric group is calculated to obtain the ratio of the number of printed fabrics with different appearances in the printed fabric group;

[0108] Obtain the surface-corrected fabric printed fabrics in the fabric printed fabric group, sum and average the printing quality deviation values ​​of all the surface-corrected fabric printed fabrics, and obtain the printing quality deviation-to-surface positive value of the surface-corrected fabric printed fabric;

[0109] Obtain the surface-different fabric printed fabrics in the fabric printed fabric group, sum and average the printing quality difference values ​​of all the surface-different fabric printed fabrics, and obtain the printing quality difference value of the surface-different fabric printed fabric;

[0110] The difference between the printing quality deviation value of the printed fabric with normal appearance and the printing quality deviation value of the printed fabric with abnormal appearance is calculated to obtain the printing quality deviation value;

[0111] The deviation value of printing quality difference is calculated by ratio with the positive value of printing quality difference, and the difference ratio of the printed fabric group is obtained;

[0112] According to the variance calculation formula, the variance value of the printing quality deviation value of the fabric printing fabric group is obtained, and recorded as the variance value of the fabric printing fabric group;

[0113] The surface difference ratio of the fabric printing fabric group is recorded as Ms;

[0114] The surface difference ratio of the fabric printing fabric group is recorded as Mc;

[0115] The variance value of the fabric printing fabric group is recorded as Mf;

[0116] By formula The state non-stable value Mi of the fabric printing fabric group is calculated, wherein a1 and a2 are both preset proportional coefficients, and a1 and a2 are both greater than zero;

[0117] Through the above formula for obtaining the state non-stable value Mi of the fabric printing fabric group, it can be known that:

[0118] The closer the surface difference ratio of the fabric printing fabric group is to 1, the greater the state instability of the fabric printing fabric group is, which means that the more the number of surface difference fabric printing fabrics in the fabric printing fabric group is, the worse the overall fabric printing quality of the fabric printing fabric group is.

[0119] The greater the surface difference of the fabric printing fabric group, the greater the difference in the printing quality surface difference values ​​between the surface-different fabric printing fabric and the surface-normal fabric printing fabric in the fabric printing fabric group, the greater the deviation of the printing quality surface difference values ​​of different fabric printing fabrics in the fabric printing fabric group, and the greater the state instability value of the fabric printing fabric group, indicating that the overall fabric printing quality deviation of the fabric printing fabric group is large;

[0120] The larger the variance value of the fabric printing fabric group is, the greater the fluctuation of the printing quality deviation value in the fabric printing fabric group is, and the fabric printing process is unstable. The larger the state instability value of the fabric printing fabric group is, the more stable the overall fabric printing quality of the fabric printing fabric group is.

[0121] Compare the state instability value Mi of the fabric printing fabric group with the state instability threshold mi of the fabric printing fabric group;

[0122] If the state instability value Mi of the fabric printing fabric group is greater than or equal to the state instability threshold mi of the fabric printing fabric group, it indicates that the overall printing quality of the fabric printing fabric group is poor, and an instability signal of the fabric printing fabric group is generated;

[0123] If the state instability value Mi of the fabric printing fabric group is less than the state instability threshold mi of the fabric printing fabric group, it indicates that the overall printing quality of the fabric printing fabric group is good, and a stable signal of the fabric printing fabric group is generated;

[0124] The technical scheme of the present invention is: based on the printing quality deviation value of the fabric printed fabric, the quality of the fabric printed fabric group is identified, that is, by obtaining the surface deviation fabric printed fabric and the surface correct fabric printed fabric of the fabric printed fabric group, the printing quality deviation value corresponding to the surface deviation fabric printed fabric and the surface correct fabric printed fabric is processed to obtain the surface deviation number ratio of the fabric printed fabric group and the surface deviation difference ratio of the fabric printed fabric group, and then combined with the variance value of the fabric printed fabric group, the state non-stable value of the fabric printed fabric group is obtained, and the overall state of the fabric printed fabric group is identified through the state non-stable value of the fabric printed fabric group. Example 3

[0125] Based on the non-stationary signals of the fabric printing fabric group, the quality of the fabric is identified;

[0126] If the surface difference ratio of the fabric printing fabric group is within the preset range of the surface difference ratio of the fabric printing fabric group, it means that the fabric printing is normal, and a fabric printing normal signal is generated;

[0127] If the surface anomaly ratio of the fabric printing fabric group is outside the preset surface anomaly ratio range of the fabric printing fabric group, it indicates that the fabric printing is abnormal, and a fabric printing validation signal is generated;

[0128] Based on the fabric printing effect signal

[0129] Check the appearance of all printed fabrics with different appearances in the printed fabric group;

[0130] The appearance inspection of printed fabrics with different appearance includes but is not limited to surface defects, damage, wrinkles and line marks;

[0131] The printed fabrics with different appearance and unqualified appearance are recorded as non-standard fabrics;

[0132] Obtaining the quantity of non-standard fabrics, calculating the ratio of the quantity of non-standard fabrics to the quantity of printed fabrics with different appearances, and obtaining the non-standard ratio of fabrics;

[0133] If the non-standard ratio of the fabric is within the preset non-standard ratio range of the fabric, it means that the proportion of the non-standard fabric in the surface-different fabric printing fabric group is normal, and the surface-different fabric printing fabric is mostly caused by the printing equipment, and a printing equipment adjustment signal is generated;

[0134] If the non-standard ratio of the fabric is outside the preset non-standard ratio range of the fabric, it means that the non-standard fabric accounts for a large proportion in the surface-different fabric printing fabric group, and a printing equipment tracking signal is generated.

[0135] Based on the printing equipment tracking signal

[0136] Obtaining the printing quality deviation values ​​of all the positive printed fabrics in the printed fabric group;

[0137] According to the order of the printing time of the fabrics, the printing quality deviation values ​​of the printed fabrics are sorted out;

[0138] If the printing quality deviation values ​​of all the printed fabrics of the surface-authentic fabrics fluctuate in a fluctuation chamber according to the order of printing time, it means that the printing quality deviation values ​​of the printed fabrics of the surface-authentic fabrics are in dynamic balance during the processing, and a normal signal of the printing equipment is generated;

[0139] Based on the normal signal of the printing equipment, it is shown that the printed fabrics with abnormal appearance are mostly caused by the appearance defects of the fabrics, and the quality of the fabrics in the printing process should be tested;

[0140] If the printing quality deviation values ​​of all genuine fabrics are decreasing according to the order of printing time, it means that the printing quality deviation values ​​of genuine fabrics are getting smaller and smaller, the printing processing quality of the fabric surface is getting better and better, and the printing equipment is generating a normal signal;

[0141] If the printing quality deviation values ​​of all genuine fabric printed fabrics show an upward trend according to the order of printing time, it means that the printing quality deviation values ​​of the genuine fabric printed fabrics are getting larger and larger, and the printing processing quality of the fabric surface is getting worse and worse, generating a printing equipment adjustment signal.

[0142] Based on the adjustment signal of the printing equipment, the printing equipment is regulated and processed to realize the normal processing of the printing process of the fabric;

[0143] During the printing process of fabrics, the printing pressure and speed of the printing equipment affect the printing quality of the fabrics. Therefore, the accuracy of the printing pressure and speed of the printing equipment is identified.

[0144] In a specific embodiment, the printing pressure values ​​of the printing equipment corresponding to all the surface positive fabric printed fabrics during the printing process are obtained, and the sum is averaged to obtain the surface positive pressure value of the surface positive fabric printed fabric;

[0145] Obtaining the printing pressure values ​​of the printing equipment corresponding to all the printed fabrics with different surfaces during the printing process, summing and averaging them, and obtaining the surface pressure value of the printed fabrics with different surfaces;

[0146] Among them, the printing pressure value of the printing equipment is the maximum value of the fabric during the printing process;

[0147] Performing difference processing on the surface positive pressure value of the surface positive fabric printed fabric and the surface differential pressure value of the surface differential fabric printed fabric to obtain the surface positive differential pressure difference of the fabric printed fabric;

[0148] If the surface positive differential pressure of the fabric printing material is within the preset surface positive differential pressure range of the fabric printing material, it means that the printing pressure of the printing equipment is not the cause of the abnormality of the fabric, and the other interference factors of the embossing equipment shall be processed;

[0149] If the surface positive differential pressure of the fabric printing material is outside the preset surface positive differential pressure range of the fabric printing material, it means that the printing pressure of the printing equipment is one of the reasons causing the abnormality of the fabric, and the printing pressure of the embossing equipment is adjusted.

[0150] In a specific embodiment, the printing temperature values ​​of the printing equipment corresponding to all the printed fabrics of the surface-corrected fabrics during the printing process are obtained, and the sum is averaged to obtain the surface-corrected temperature value of the printed fabrics of the surface-corrected fabrics;

[0151] Obtain the printing temperature values ​​of the printing equipment corresponding to all the printed fabrics with different surfaces during the printing process, sum and average them, and obtain the surface temperature value of the printed fabrics with different surfaces;

[0152] Among them, the printing temperature value of the printing equipment is the maximum value of the fabric during the printing process.

[0153] Performing difference processing on the surface positive temperature value of the surface positive fabric printed fabric and the surface differential temperature value of the surface differential fabric printed fabric to obtain the surface positive differential temperature difference of the fabric printed fabric;

[0154] If the surface temperature difference of the fabric printing material is within the preset surface temperature difference range of the fabric printing material, it means that the printing temperature of the printing equipment is not the cause of the abnormality of the fabric, and the other interference factors of the embossing equipment shall be processed;

[0155] If the surface temperature difference of the fabric printing material is outside the preset surface temperature difference range of the fabric printing material, it means that the printing temperature of the printing equipment is one of the reasons causing the abnormality of the fabric, and the printing temperature of the embossing equipment is adjusted.

[0156] In a specific embodiment, other interfering factors of the embossing equipment include, but are not limited to, identification of printing paste (viscosity, stability, and permeability of the paste), temperature and humidity of the printing environment, and drying status of the printed fabric.

[0157] The technical solution of the present invention is: according to the unstable signal of the fabric printing fabric group, the appearance quality of all the fabric printing fabrics with abnormal appearance is checked, and it is judged whether the cause of the abnormality of the fabric during the printing process is the quality problem of the fabric itself or the processing problem of the printing equipment. When the printing equipment adjustment signal is obtained, it means that the quality of the fabric itself is normal during the printing process, and the printing process of the printing equipment needs to be corrected and adjusted. The cause of the abnormal printing is identified in time and effectively, and at the same time, it is convenient to track the quality of the printing equipment. Example 4

[0158] See also Figure 4 As shown, the present invention is an online identification system for fabric quality, including a printing data extraction module, a printing quality characterization module, a grouping state analysis module, a traceability control module and a cloud management and control platform;

[0159] The printing data extraction module is used to obtain multiple fabric printing fabrics to obtain a fabric printing fabric group during the fabric printing process, process each independent fabric printing image of the fabric printing fabric group, obtain the printing extraction data of the fabric printing image, and upload the printing extraction data to the cloud management and control platform;

[0160] The printing quality characterization module receives the printing extraction data transmitted by the cloud management and control platform, obtains the printing quality deviation value of the fabric printing material, and uploads the printing quality deviation value to the cloud management and control platform;

[0161] The grouping state analysis module receives the printing quality deviation value transmitted by the cloud management and control platform. The grouping state analysis module is used to divide the fabric printed fabrics in the fabric printed fabric group into the surface deviation fabric printed fabrics and the surface correct fabric printed fabrics, and obtain the state non-stable value of the fabric printed fabric group by comparing the surface deviation fabric printed fabrics and the surface correct fabric printed fabrics.

[0162] Complete the state signal recognition of the fabric printing fabric group based on the state non-stable value, and upload the state signal to the cloud management and control platform;

[0163] Wherein, the state signal of the fabric printing fabric group includes an unstable signal of the fabric printing fabric group and a stable signal of the fabric printing fabric group;

[0164] The traceability control module receives the unstable signal of the fabric printing fabric group transmitted by the cloud control platform. The traceability control module traces and identifies the surface quality of the fabric during the processing and the working condition of the printing equipment, and completes the control of the fabric during the printing process.

[0165] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for online identification of fabric quality, It is characterized in that The following steps are involved: In the process of fabric printing, a plurality of fabric printing fabrics are obtained to obtain a fabric printing fabric group, and each independent fabric printing image of the fabric printing fabric group is processed to obtain printing extraction data of the fabric printing image; Based on the printing extraction data, the printing quality deviation value of the printed fabric is obtained; According to the printing quality difference value, the printed fabrics in the printed fabric group are divided into the surface difference printed fabrics and the surface correct printed fabrics, and the state non-stable value of the printed fabric group is obtained by comparing the surface difference printed fabrics and the surface correct printed fabrics. Based on the state non-stable value, the state signal recognition of the fabric printing fabric group is completed; Wherein, the state signal of the fabric printing fabric group includes an unstable signal of the fabric printing fabric group and a stable signal of the fabric printing fabric group; Based on the non-stationary signal of the fabric printing group, the surface quality of the fabric during the processing and the working condition of the printing equipment are traced and identified, and the control of the fabric during the printing process is completed; The print extraction data of the fabric print image includes print color data and print surface data; The printing color data is processed to obtain the saturation difference value, which is recorded as Zs; The printed surface data is processed to obtain the surface quality difference value, which is recorded as Zb; By formula The printing quality difference value Zi of the printed fabric is calculated, where k is a preset proportional coefficient; The process of obtaining saturation outliers is as follows: Dividing the fabric print image into a plurality of print image sub-units; Convert each print image subunit into an adaptive color space and calculate the corresponding histogram so that the histogram of each color channel is between 0 and 1; A saturation deviation ratio of each color channel is obtained based on the pixel value of each color channel; Then, the saturation variance value and the saturation extreme deviation ratio of the fabric printing image are obtained according to the saturation deviation ratio of each color channel; The saturation variance value is multiplied by the saturation extreme deviation ratio to obtain the saturation outlier of the fabric printing image. The process of obtaining the outlier value of the table quality is as follows: Recording the printed image subunit with printed fragments as a different image subunit; Calculate the ratio of the number of different image sub-units to the number of printed image sub-units to obtain a different image unit ratio; Obtaining the difference area of ​​the difference image sub-unit, calculating the ratio of the total area of ​​the difference area to the number of the print image sub-units, and obtaining the difference ratio of the difference area; The surface quality difference value of the fabric printing image is obtained by multiplying the difference image unit ratio and the difference area surface difference ratio. The ratio of the number of printed fabrics with different appearance to the total number of printed fabrics in the printed fabric group is calculated to obtain the ratio of the number of printed fabrics with different appearance, which is recorded as Ms. The printing quality deviation value is calculated by taking the difference between the mean of the printing quality deviation values ​​of all normal fabric printed fabrics and the mean of the printing quality deviation values ​​of the abnormal fabric printed fabrics; The deviation value of printing quality difference is calculated by ratio with the mean value of printing quality difference of all printed fabrics, and the difference ratio of printed fabric group is obtained, which is recorded as Mc. The variance of the printing quality variance of the fabric printing fabric group is calculated to obtain the variance value of the fabric printing fabric group, which is recorded as Mf; By formula The state non-stable value Mi of the fabric printing fabric group is calculated, where a1 and a2 are both preset proportional coefficients; Comparing the printing quality deviation value of the fabric printed fabrics in the fabric printed fabric group with the printing quality deviation threshold value of the fabric printed fabrics; If the printing quality deviation value of the fabric printed fabric is greater than or equal to the printing quality deviation threshold of the fabric printed fabric, the fabric printed fabric is recorded as a deviation fabric printed fabric; If the printing quality deviation value of the fabric printed fabric is less than the printing quality deviation threshold of the fabric printed fabric, the fabric printed fabric is recorded as the normal fabric printed fabric.

2. A method for online identification of fabric quality according to claim 1, It is characterized in that If the state instability value Mi of the fabric printing fabric group is greater than or equal to the state instability threshold mi of the fabric printing fabric group, an instability signal of the fabric printing fabric group is generated; If the state instability value Mi of the fabric printing fabric group is less than the state instability threshold mi of the fabric printing fabric group, a stable signal of the fabric printing fabric group is generated.

3. A method for online identification of fabric quality according to claim 2, It is characterized in that Based on the non-stationary signal of the fabric printing fabric group, the surface anomaly ratio of the fabric printing fabric group is analyzed: If the surface anomaly ratio of the fabric printing fabric group is within a preset range of the surface anomaly ratio of the fabric printing fabric group, a fabric printing normal signal is generated; If the surface difference number ratio of the fabric printing fabric group is outside the preset range of the surface difference number ratio of the fabric printing fabric group, a fabric printing validation signal is generated.

4. A method for online identification of fabric quality according to claim 3, It is characterized in that Based on the printing effect signal of fabric; Check the appearance of all printed fabrics in the printed fabric group; Calculate the ratio of the number of non-standard fabrics to the number of printed fabrics with different appearances to obtain the non-standard ratio of fabrics; If the non-standard ratio of the fabric is within a preset non-standard ratio range of the fabric, a printing device adjustment signal is generated; If the non-standard ratio of the fabric is outside the preset non-standard ratio range of the fabric, a printing device tracking signal is generated; According to the tracking signal of the printing equipment, the printing quality deviation values ​​of all the normal fabric printing fabrics in the fabric printing fabric group are processed, and the normal signal of the printing equipment and the adjustment signal of the printing equipment are obtained according to the changing trend of the printing quality deviation values ​​of all the normal fabric printing fabrics during the printing time.

5. A method for online identification of fabric quality according to claim 4, It is characterized in that Based on the adjustment signal of the printing equipment, the factors in the printing equipment that interfere with the printing quality of the fabric are tracked and identified.

6. An online identification system for fabric quality, It is characterized in that include: The printing data extraction module is used to obtain multiple fabric printing fabrics to obtain a fabric printing fabric group during the fabric printing process, process each independent fabric printing image of the fabric printing fabric group, obtain the printing extraction data of the fabric printing image, and upload the printing extraction data to the cloud management and control platform; Printing quality characterization module, the printing quality characterization module receives the printing extraction data transmitted by the cloud management and control platform, obtains the printing quality deviation value of the fabric printing material, and uploads the printing quality deviation value to the cloud management and control platform; A grouping state analysis module, the grouping state analysis module receives the printing quality deviation value transmitted by the cloud management and control platform, and the grouping state analysis module is used to divide the fabric printing fabrics in the fabric printing fabric group into the surface deviation fabric printing fabrics and the surface correct fabric printing fabrics, and obtain the state non-stable value of the fabric printing fabric group by comparing the surface deviation fabric printing fabrics and the surface correct fabric printing fabrics; Complete the state signal recognition of the fabric printing fabric group based on the state non-stable value, and upload the state signal to the cloud management and control platform; Wherein, the state signal of the fabric printing fabric group includes an unstable signal of the fabric printing fabric group and a stable signal of the fabric printing fabric group; The traceability control module receives the unstable signal of the fabric printing group transmitted by the cloud control platform. The traceability control module traces and identifies the surface quality of the fabric during the processing and the working condition of the printing equipment, and completes the control of the fabric during the printing process. The print extraction data of the fabric print image includes print color data and print surface data; The printing color data is processed to obtain the saturation difference value, which is recorded as Zs; The printed surface data is processed to obtain the surface quality difference value, which is recorded as Zb; By formula The printing quality difference value Zi of the printed fabric is calculated, where k is a preset proportional coefficient; The process of obtaining saturation outliers is as follows: Dividing the fabric print image into a plurality of print image sub-units; Convert each print image subunit into an adaptive color space and calculate the corresponding histogram so that the histogram of each color channel is between 0 and 1; A saturation deviation ratio of each color channel is obtained based on the pixel value of each color channel; Then, the saturation variance value and the saturation extreme deviation ratio of the fabric printing image are obtained according to the saturation deviation ratio of each color channel; The saturation variance value is multiplied by the saturation extreme deviation ratio to obtain the saturation outlier of the fabric printing image. The process of obtaining the outlier value of the table quality is as follows: Recording the printed image subunit with printed fragments as a different image subunit; Calculate the ratio of the number of different image sub-units to the number of printed image sub-units to obtain a different image unit ratio; Obtaining the difference area of ​​the difference image sub-unit, calculating the ratio of the total area of ​​the difference area to the number of the print image sub-units, and obtaining the difference ratio of the difference area; The surface quality difference value of the fabric printing image is obtained by multiplying the difference image unit ratio and the difference area surface difference ratio. The ratio of the number of printed fabrics with different appearance to the total number of printed fabrics in the printed fabric group is calculated to obtain the ratio of the number of printed fabrics with different appearance, which is recorded as Ms. The printing quality deviation value is calculated by taking the difference between the mean of the printing quality deviation values ​​of all normal fabric printed fabrics and the mean of the printing quality deviation values ​​of the abnormal fabric printed fabrics; The deviation value of printing quality difference is calculated by ratio with the mean value of printing quality difference of all printed fabrics, and the difference ratio of printed fabric group is obtained, which is recorded as Mc. The variance of the printed fabric group printing quality difference table value is calculated to obtain the variance value of the printed fabric group, which is recorded as Mf; By formula The state non-stable value Mi of the fabric printing fabric group is calculated, where a1 and a2 are both preset proportional coefficients; Comparing the printing quality deviation value of the fabric printed fabrics in the fabric printed fabric group with the printing quality deviation threshold value of the fabric printed fabrics; If the printing quality deviation value of the fabric printed fabric is greater than or equal to the printing quality deviation threshold of the fabric printed fabric, the fabric printed fabric is recorded as a deviation fabric printed fabric; If the printing quality deviation value of the fabric printed fabric is less than the printing quality deviation threshold of the fabric printed fabric, the fabric printed fabric is recorded as the normal fabric printed fabric.

Images