Digital printing color management optimization method based on artificial intelligence

Abstract

The invention discloses a digital printing color management optimization method based on artificial intelligence, and relates to the technical field of digital printing. Comprising the following steps: collecting the spectral reflectance of a plurality of fabrics made of single fibers, and detecting the fiber content ratio in the printed fabric made of various fibers to obtain fabric index data; the method for obtaining the index data of the printed cloth comprises the following steps of: 1, obtaining the proportion of each fiber in the printed cloth through detection or through a printed cloth component table; the method comprises the following steps: collecting cloth index data, carrying out permeability test on each verified cloth sample, optimizing an ICC curve in digital printing equipment, establishing a relationship between ink penetration depths and printing color values of different types of cloth, establishing a judgment model, and optimizing the ICC curve stored in the printing equipment to reduce chromatic aberration of a finished product. And artificial intelligence color control of printing is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of digital printing, and specifically to an optimized method for digital printing color management based on artificial intelligence. Background Art

[0002] Digital printing is a modern printing technology that uses a digital printer and special inks to directly print patterns onto textiles. Compared with traditional printing processes, digital printing has many advantages. First of all, digital printing has a high pattern reduction ability. Through a digital printer, high-resolution images can be accurately transferred onto textiles, achieving fine patterns and details, without being limited by the number and complexity of colors. Secondly, digital printing has the characteristics of flexibility and personalization. Digital printing can be customized according to customer needs, and personalized patterns and designs can be easily achieved to meet the needs of different customers. At the same time, digital printing also supports rapid design modification and small-batch production, which is conducive to quickly responding to market demands.

[0003] The Chinese invention patent, application publication number CN116587759A, discloses an intuitive and efficient digital printing color correction management method and system, which can optimize and correct the parameters of digital printing equipment and inks, and adjust the equipment control parameters and ink configuration parameters according to the color development law to improve color performance and reduce color differences.

[0004] Another example is that the Chinese invention patent, application publication number CN110418030A, discloses a color consistency mapping method for textile inkjet printing based on image color blocks. Through the established conversion relationship, any RGB picture can be accurately converted into a corresponding CMYK color value picture for inkjet printing of textile pattern drafts.

[0005] The above-mentioned solutions have all solved the technical problems they each proposed. However, in practical applications, the above-mentioned solutions still have certain defects. For example:

[0006] 1. During the printing process, the ink sprayed by the nozzle of the digital printing equipment is printed on the fabric, and the ink will penetrate into the fabric. As the penetration depth increases, color differences will occur between the obtained fabric print and the sample provided by the customer. The above-mentioned solutions only adjust the equipment control parameters and ink configuration parameters by summarizing the color development law, without considering the above-mentioned ink penetration factor, which may still cause color difference problems;

[0007] 2. The prior art generally eliminates the color difference problems caused by the ink penetration factor by increasing the ink volume and adjusting the ink ratio. However, the ink is expensive, which will inevitably increase the production cost;

[0008] 3. For different fabrics, due to different penetration depths, the L values of the prints vary. The existing technology is not convenient for quickly adjusting the L values of different fabrics on digital printing equipment, reducing production efficiency. Summary of the Invention

[0009] The purpose of the present invention is to provide an optimized method for digital printing color management based on artificial intelligence to solve the problems raised in the above background technology.

[0010] To achieve the above purpose, the present invention provides the following technical solution: An optimized method for digital printing color management based on artificial intelligence, including collecting the spectral reflectance of several fabrics made of a single fiber, detecting the proportion of fiber content in the printed fabrics made of multiple fibers to obtain fabric index data;

[0011] The method for obtaining fabric index data is as follows: First step, obtain the proportion of each fiber in the printed fabric by detection or through the printed fabric ingredient list;

[0012] Second step, multiply the proportion of each fiber by the spectral reflectance of the corresponding fiber;

[0013] Third step, divide the obtained product by the number of fiber types in the printed fabric to obtain fabric index data;

[0014] Collect several verification fabric samples, conduct permeability tests on each verification fabric sample, establish a correspondence table between fabric index data and the average value of ink penetration depth, thereby establishing a linear relationship between fabric index data and the average value of ink penetration depth;

[0015] Obtain verification samples, analyze the verification samples and the printed verification fabric samples, and optimize the ICC curve in the digital printing equipment;

[0016] Establish a first prediction model for predicting the lowest heating temperature of the hot plate within the color difference threshold range for different fabrics to be printed;

[0017] Establish a second prediction model for predicting the L value of the print outside the color difference threshold range for different fabrics to be printed;

[0018] Establish an artificial intelligence model to train the relationship between the ink penetration depth and the printed color value of different types of fabrics, and optimize the ICC curve stored in the printing equipment; train the first prediction model and the second prediction model, which are respectively used to predict the lowest heating temperature of the hot plate within the color difference threshold range for different fabrics to be printed, and to predict the L value outside the color difference threshold range for different fabrics to be printed.

[0019] Further, the specific method for performing a permeability test on each of the verification fabric samples and establishing a correspondence table between fabric index data and the average value of ink penetration depth is as follows:

[0020] First step: At room temperature, perform single-color printing on the verification fabric samples through a digital printing device;

[0021] Second step: Cut the verification fabric samples that have undergone single-color printing, measure the ink penetration depth through an instrument, collect the ink penetration depth at the printing position, and calculate the average value of the ink penetration depth;

[0022] Third step: Establish a correspondence table between fabric index data and the average value of ink penetration depth.

[0023] Further, the method for obtaining verification samples, analyzing the verification samples and the printed verification fabric samples, and optimizing the ICC curve in the digital printing device is as follows:

[0024] First step: Obtain verification samples, read the LAB color values of the verification samples, and convert the LAB color values of the verification samples into CMYK values of the digital printing device through color space conversion;

[0025] Second step: Adjust the hot plate of the digital printing device to different temperatures. After the hot plate temperature is constant at different temperatures, the digital printing device performs printing on the verification fabric samples to obtain multiple verification fabric samples printed at different hot plate temperatures;

[0026] Third step: Use a spectrophotometer to sequentially measure the L values of the printed verification fabric samples;

[0027] Fourth step: Establish a judgment model to optimize the ICC curve stored in the printing device;

[0028] Fifth step: Use image recognition technology and sensor technology to real-time detect the L values of the verification fabric samples, automatically adjust the CMYK values of the digital printing device according to the differences, and ensure that the difference between the L value of the printed verification fabric samples and the L value of the verification samples is within the error range.

[0029] Further, the method for establishing the judgment model is as follows:

[0030] First step: Calculate the color difference between the verification samples and the printed verification fabric samples at different hot plate temperatures. If the color difference is less than or equal to the threshold, record the minimum hot plate temperature of the verification fabric samples;

[0031] Second step: If the color difference is greater than the threshold, record the maximum hot plate temperature, and input the detected L value into the color management software for optimizing the ICC curve stored in the printing device, thereby changing the CMYK values of the digital printing device.

[0032] Furthermore, the method for establishing the first prediction model is:

[0033] Step 1: When the minimum hot plate temperature of the fabric sample within the threshold range is less than the maximum heating temperature of the hot plate itself, obtain the hot plate temperature value. The hot plate temperature value set is (ai, bi, ci, ..., ni), where ai is the temperature of the hot plate at room temperature;

[0034] Step 2: Obtain the color value of the verification fabric sample at the corresponding hot plate temperature, and the color value set of the verification fabric sample is (Ai, Bi, Ci, ..., Ni);

[0035] Step 3: Fit several sets of hot plate temperature value sets and several sets of verification fabric sample color value sets to obtain a first prediction model.

[0036] Furthermore, the method for predicting the minimum heating temperature of the hot plate within the color difference threshold range for different fabrics to be printed includes:

[0037] Step 1: Obtain the average ink penetration depth of the fabric to be printed and the printing color value of the sample fabric provided by the customer;

[0038] Step 2: Calculate the difference between the printed color value and the color difference threshold in the sample fabric provided by the customer to obtain the color value of the fabric to be printed, input the color value of the fabric to be printed into the first prediction model, and calculate the minimum heating temperature of the hot plate under the average value of the ink penetration depth of the fabric to be printed.

[0039] Furthermore, the method for establishing the second prediction model is:

[0040] Step 1: Based on the color value of the verification sample, print and dye different verification fabric samples at the highest hot plate temperature, measure the L value of the verification fabric samples respectively, and set the L value set to (L1, L2, L3, ..., Ln);

[0041] Step 2: Cut the printed verification fabric sample, measure and calculate the average dyeing depth, and set the average dyeing depth set to (P1, P2, P3, ..., Pn);

[0042] Step 3: Obtain the average value of the ink penetration depth of each verification fabric sample at room temperature, and set the average value set of the ink penetration depth of the verification fabric samples to (Q1, Q2, Q3, ..., Qn);

[0043] Step 4: Fitting the ink penetration depth average value set and the dyeing depth average value set of the verification fabric samples to obtain a first linear regression equation, and fitting the dyeing depth average value set and the L value set of the verification fabric samples to obtain a second linear regression equation;

[0044] Step 5: Establish a second prediction model based on the first linear regression equation and the second linear regression equation.

[0045] Furthermore, the method for predicting the printing L value of different fabrics to be printed outside the color difference threshold range is as follows:

[0046] Obtain the average value of the ink penetration depth of the fabric to be printed, input the average value of the ink penetration depth into the first linear regression equation to obtain the average value of the dyeing depth of the fabric to be printed, and finally input the average value of the dyeing depth of the fabric to be printed into the second linear regression equation to obtain the L value of the fabric to be printed at the maximum hot plate temperature.

[0047] Compared with the prior art, the beneficial effects of the present invention are:

[0048] The digital printing process optimization method and system based on artificial intelligence color management collect fabric index data, perform permeability tests on each verified fabric sample, and optimize the ICC curve in the digital printing equipment, establish the relationship between the ink penetration depth of different types of fabrics and the printing color value, establish a judgment model, optimize the ICC curve stored in the printing equipment to reduce the color difference of the finished product, and realize the artificial intelligence color control of printing.

[0049] At the same time, a first prediction model is established. When batch printing fabrics to be printed with the same batch and specification, by controlling the hot plate temperature, the printing of the finished product can meet the color standards of customers. While reducing production costs, the production process adjustment time is reduced by predicting the hot plate temperature through the first prediction model, improving production efficiency.

[0050] In addition, a second prediction model is established. The fabric to be printed is heated by the maximum hot plate temperature to reduce its ink penetration depth, which is beneficial to saving production costs. Predict the L value of the printed fabric at the minimum penetration depth, and adjust the CMYK value and process parameters of the digital printing equipment through artificial intelligence color management to adapt to the L value of the printed fabric, further improving production efficiency while reducing production costs. Description of the Drawings

[0051] Figure 1 is the general flow chart of the present invention;

[0052] Figure 2 is the working state diagram among the fabric, the hot plate and the printing head of the present invention;

[0053] Figure 3 is the relationship diagram between the ink penetration depth and the color difference value of the same type of fabric of the present invention;

[0054] Figure 4This is a graph showing the relationship between the ink penetration depth and the hot plate temperature for the same type of fabric in the present invention. Detailed implementation manners

[0055] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0056] Such as Figure 1 This is the overall flowchart of the present invention; Figure 2 This is the working state diagram among the fabric, the hot plate and the printing head of the present invention; Figure 3 This is the graph showing the relationship between the ink penetration depth and the color difference value for the same type of fabric in the present invention and Figure 4 This is the graph showing the relationship between the ink penetration depth and the hot plate temperature for the same type of fabric in the present invention. The present invention provides a technical solution: an optimization method for digital printing color management based on artificial intelligence, including:

[0057] Collect the spectral reflectance of several fabrics made of a single fiber. Among them, the spectral band used for testing the spectral reflectance in this solution is 600nm. The specific detection method is a mature existing technology and will not be elaborated in this solution. Obtain several printed fabrics obtained on the market or customized by manufacturers, and detect the fiber content ratio in the printed fabrics made of multiple fibers to obtain fabric index data. Among them, the printed fabrics cover the main types and qualities circulating on the market.

[0058] The method for obtaining the fabric index data is as follows: First step, obtain the proportion of each fiber in the printed fabric through detection or through the ingredient list of the printed fabric. Second step, multiply the proportion of each fiber by the spectral reflectance of the corresponding fiber. Third step, divide the obtained product by the number of fiber types in the printed fabric to obtain the fabric index data. The advantages of such fabric index data are: combining the fiber composition ratio with the optical characteristics (spectral reflectance), and comprehensively reflecting the physical properties and optical performance of the fabric through weighted calculation; because the fiber with a higher proportion has a greater impact on the overall index, it conforms to the dominant effect in practical applications. By dividing by the number of fiber types, the numerical deviation caused by the complexity of the composition of different fabrics is eliminated, thus facilitating the standardized quality assessment.

[0059] Collect several verification fabric samples, conduct permeability tests on each verification fabric sample, and establish a corresponding table between the fabric index data and the average value of the ink penetration depth. The specific method is:

[0060] Step 1: At room temperature, perform single-color printing on the verified fabric sample using a digital printing device. Since the ink is single-color and the influence of the ink on the penetration depth of the mixed color can be ignored, to reduce costs (select inexpensive single-color ink) and improve detection efficiency, use single-color to print on the verified fabric sample.

[0061] Step 2: Slice the verified fabric sample that has undergone single-color printing, measure the ink penetration depth through an instrument, collect the ink penetration depth at the printing position, and calculate the average value of the ink penetration depth. To improve data accuracy, 3 - 5 slicing points can be selected at the edge of the printing position, and then 3 - 5 slicing points can be selected at regular intervals from the edge of the printing position towards the center of the printing position. The number of selected slicing points should be no less than 12.

[0062] Step 3: Establish a correspondence table between the fabric index data and the average value of the ink penetration depth.

[0063] Through the correspondence table between the fabric index data and the average value of the ink penetration depth, establish a linear relationship between the fabric index data and the average value of the ink penetration depth, that is, inputting a fabric index data can obtain the average value of the ink penetration depth of the fabric.

[0064] Obtain a verification sample, analyze the verification sample and the printed verified fabric sample, and optimize the ICC curve in the digital printing device.

[0065] Specifically, the method for obtaining a verification sample, analyzing the verification sample and the printed verified fabric sample, and optimizing the ICC curve in the digital printing device is as follows:

[0066] Step 1: Obtain a verification sample, read the LAB color value of the verification sample, and convert the LAB color value of the verification sample into the CMYK value of the digital printing device through color space conversion.

[0067] Step 2: Adjust the hot plate of the digital printing device to different temperatures. After the hot plate temperature is constant at different temperatures, the digital printing device prints on the verified fabric sample to obtain multiple verified fabric samples printed at different hot plate temperatures. The temperature range of the hot plate is from room temperature to 200 °C. To improve the production efficiency of the samples, the interval temperature is 20 °C, that is, room temperature is 20 °C, and ten verified fabric samples of the same type and specification need to be prepared.

[0068] Step 3: Use a spectrophotometer to measure the L value of the printed verification fabric sample in sequence.

[0069] Step 4: Establish a judgment model to optimize the ICC curve stored in the printing device.

[0070] Step 5: Use image recognition technology and sensor technology to detect and verify the L value of the fabric sample in real time. Color management software and algorithms can be used to precisely adjust the CMYK values of the digital printing equipment according to the change of the L value, so as to ensure that the difference between the L value of the printed fabric sample and the L value of the verification sample is within the error range. The corresponding numerical adjustment coefficients are stored in the system.

[0071] In this way, based on the continuous adjustment of the difference between the L value of the printed fabric sample and the L value of the verification sample, the color management system establishes the matching relationship between the verification fabric sample under the fabric index data, the hot plate heating temperature during printing, and the CMYK values of the digital printing equipment for the artificial intelligence color management of the system.

[0072] The method for establishing the judgment model is as follows:

[0073] Step 1: Calculate the color difference between the verification sample and the printed verification fabric sample at different hot plate temperatures. If the color difference is less than or equal to the threshold value, record the minimum hot plate temperature of the verification fabric sample. For example, assume that the verification fabric sample is heated by hot plates at temperatures of 100, 120, 140, 160, and 180 degrees Celsius. The color difference between the verification sample and the printed verification fabric sample at a hot plate temperature of 120 degrees Celsius is equal to or less than and closest to the threshold value. Since the smaller the ink penetration depth, the smaller the color difference, and the color differences between the verification sample and the printed verification fabric samples heated by hot plates at temperatures of 140, 160, and 180 degrees Celsius are all less than the threshold value. At this time, record the minimum hot plate temperature of this verification fabric sample as 120 degrees.

[0074] Step 2: If the color difference is greater than the threshold value, record the maximum hot plate temperature, and input the detected L value into the color management software to optimize the ICC curve stored in the printing equipment, thereby changing the CMYK values of the digital printing equipment. For example, assume that the verification fabric sample is heated by hot plates at temperatures of 100, 120, 140, 160, 180, and 200 degrees Celsius. The color difference between the verification sample and the printed verification fabric sample at a hot plate temperature of 200 degrees Celsius is greater than the threshold value. Then record the minimum hot plate temperature of this verification fabric sample as 200 degrees.

[0075] The purpose of the setting is that in the cost control of the printing technology, the cost generated by the hot plate heating is less than the cost of the increased ink volume. If the color difference can be effectively reduced by changing the hot plate heating temperature, the production cost can be effectively reduced.

[0076] Establish the first prediction model to predict the minimum heating temperature of the hot plate within the color difference threshold range for different fabrics to be printed.

[0077] The method for establishing the first prediction model is as follows:

[0078] Step 1: When the minimum hot plate temperature of the fabric sample within the threshold range is less than the maximum heating temperature of the hot plate itself, obtain the hot plate temperature value. The hot plate temperature value set is (ai, bi, ci, ..., ni), where ai is the temperature of the hot plate at room temperature;

[0079] Step 2: Obtain the color value of the verification fabric sample at the corresponding hot plate temperature, and the color value set of the verification fabric sample is (Ai, Bi, Ci, ..., Ni);

[0080] Step 3: Pair several sets of hot plate temperature value sets with several sets of verification fabric sample color value sets to construct a data set: D={(ai,Ai),(bi,Bi),...,(ni,Ni)}, fit the data set, and obtain the first prediction model.

[0081] The method for predicting the minimum heating temperature of the hot plate within the color difference threshold range for different printed fabrics includes:

[0082] Step 1: Obtain the average ink penetration depth of the fabric to be printed and the printing color value of the sample fabric provided by the customer. The average ink penetration depth of the fabric to be printed is obtained through the linear relationship between the fabric index data and the average ink penetration depth;

[0083] Step 2: Calculate the difference between the printed color value and the color difference threshold in the sample fabric provided by the customer to obtain the color value of the fabric to be printed, input the color value of the fabric to be printed into the first prediction model, and calculate the minimum heating temperature of the hot plate under the average value of the ink penetration depth of the fabric to be printed.

[0084] In this way, when batch printing is carried out on the same batch of fabrics with the same specifications to be printed, the temperature of the hot plate can be controlled to ensure that the printed products meet the customer's color standards. While reducing production costs, the first prediction model predicts the temperature of the hot plate, which reduces the production process adjustment time and improves production efficiency.

[0085] A second prediction model is established to predict the L value of printing of different fabrics to be printed outside the color difference threshold range.

[0086] Specifically, the method for establishing the second prediction model is:

[0087] Step 1: Based on the color value of the verification sample, print and dye different verification fabric samples at the highest hot plate temperature, measure the L value of the verification fabric samples respectively, and set the L value set to (L1, L2, L3, ..., Ln);

[0088] Step 2: Cut the printed verification fabric sample, measure and calculate the average dyeing depth, and set the average dyeing depth set to (P1, P2, P3, ..., Pn);

[0089] Step 3: Obtain the average value of the ink penetration depth of each verified fabric sample at room temperature, and set the set of average values of the ink penetration depth of the verified fabric samples as (Q1, Q2, Q3, ..., Qn);

[0090] Step 4: Fit the set of average values of the ink penetration depth of the verified fabric samples and the set of average values of the dyeing depth to obtain the first linear regression equation, and fit the set of average values of the dyeing depth of the verified fabric samples and the set of L values of the verified fabric samples to obtain the second linear regression equation;

[0091] Step 5: Establish a second prediction model based on the first linear regression equation and the second linear regression equation.

[0092] In addition, the method for predicting the printing L value of different fabrics to be printed outside the color difference threshold range is as follows:

[0093] Obtain the average value of the ink penetration depth of the fabric to be printed based on the linear relationship between the fabric index data of the fabric to be printed and the average value of the ink penetration depth, input the average value of the ink penetration depth into the first linear regression equation to obtain the average value of the dyeing depth of the fabric to be printed, and finally input the average value of the dyeing depth of the fabric to be printed into the second linear regression equation to obtain the L value of the fabric to be printed at the maximum hot plate temperature.

[0094] The fabric to be printed is heated by the maximum hot plate temperature to reduce its ink penetration depth, which is beneficial to saving production costs. Predict the L value of the printed fabric at the minimum penetration depth, and adjust the CMYK values and process parameters of the digital printing equipment through artificial intelligence color management to adapt to the L value of the printed fabric, which improves production efficiency while reducing production costs.

[0095] Establish an artificial intelligence model to train the relationship between the ink penetration depth of different types of fabrics and the printing color values, and realize the optimization of the ICC curve stored in the printing equipment; train the first prediction model and the second prediction model, which are respectively used to predict the minimum heating temperature of the hot plate for different fabrics to be printed within the color difference threshold range, and to predict the L values of different fabrics to be printed outside the color difference threshold range.

[0096] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended embodiments and their equivalents.

Claims

1. A digital printing color management optimization method based on artificial intelligence, characterized in that: Collect the spectral reflectance of several fabrics made of a single fiber, detect the fiber content ratio in printed fabrics made of multiple fibers, and obtain fabric index data; The method for obtaining the index data of fabrics is as follows: the first step is to obtain the proportion of each fiber in the printed fabrics through testing or through the printed fabric composition table; In the second step, the proportion of each fiber is multiplied by the spectral reflectance of the corresponding fiber; The third step is to divide the obtained product by the number of fiber types of the printed fabric to obtain the fabric index data; Collect a number of verification fabric samples, perform a permeability test on each verification fabric sample, establish a correspondence table between fabric index data and average ink penetration depth, and thus establish a linear relationship between fabric index data and average ink penetration depth; Obtain verification samples, analyze verification samples and printed verification fabric samples, and optimize the ICC curve in the digital printing equipment; Establishing a first prediction model for predicting the minimum heating temperature of the hot plate within a color difference threshold range for different printed fabrics; Establishing a second prediction model for predicting the L value of printing on different printed fabrics outside the color difference threshold range; An artificial intelligence model is established to train the relationship between the ink penetration depth and printing color value of different types of fabrics, so as to optimize the ICC curve stored in the printing equipment; the first prediction model and the second prediction model are trained to predict the minimum heating temperature of the hot plate within the color difference threshold range for different fabrics to be printed, and to predict the L value of different fabrics to be printed outside the color difference threshold range.

2. The method for color management optimization of digital printing based on artificial intelligence according to claim 1, characterized in that: The specific method of conducting a permeability test on each verification fabric sample and establishing a corresponding table between fabric index data and average ink penetration depth is as follows: Step 1: Perform single-color printing on the verification fabric sample using digital printing equipment at room temperature; Step 2: Slice the verification fabric sample that has been printed in single color, measure the ink penetration depth through the instrument, collect the ink penetration depth at the printing position, and calculate the average ink penetration depth; Step 3: Establish a correspondence table between fabric index data and average ink penetration depth.

3. The method for color management optimization of digital printing based on artificial intelligence according to claim 2, characterized in that: The method of obtaining the verification sample, analyzing the verification sample and the printed verification fabric sample, and optimizing the ICC curve in the digital printing device is: Step 1: Get the verification sample, read the LAB color value of the verification sample, and convert the LAB color value of the verification sample into the CMYK value of the digital printing equipment through color space conversion; Step 2: The hot plate of the digital printing equipment is adjusted to different temperatures. After the hot plate temperature at different temperatures is constant, the digital printing equipment prints the verification fabric sample to obtain multiple verification fabric samples after printing at different hot plate temperatures; Step 3: Use a spectrophotometer to measure and verify the L value of the printed fabric samples in turn; Step 4: Establish a judgment model to optimize the ICC curve stored in the printing equipment; Step 5: Use image recognition technology and sensor technology to detect and verify the L value of the fabric sample in real time, and automatically adjust the CMYK value of the digital printing equipment according to the difference to ensure that the difference between the L value of the printed fabric sample and the L value of the verification sample is within the error range.

4. The method for color management optimization of digital printing based on artificial intelligence according to claim 3, characterized in that: The method for establishing the judgment model is: Step 1: Calculate the color difference between the verification sample and the verification fabric sample printed at different hot plate temperatures. If the color difference is less than or equal to the threshold, record the minimum hot plate temperature of the verification fabric sample. Step 2: If the color difference value is greater than the threshold, record the maximum hot plate temperature and input the detected L value into the color management software to optimize the ICC curve stored in the printing equipment, thereby changing the CMYK value of the digital printing equipment.

5. The method for color management optimization of digital printing based on artificial intelligence according to claim 4, characterized in that: The method for establishing the first prediction model is: Step 1: When the minimum hot plate temperature of the fabric sample within the threshold range is less than the maximum heating temperature of the hot plate itself, obtain the hot plate temperature value. The hot plate temperature value set is (ai, bi, ci, ..., ni) where: ai is the temperature of the hot plate at room temperature; Step 2: Obtain the color value of the verification fabric sample at the corresponding hot plate temperature, and the color value set of the verification fabric sample is (Ai, Bi, Ci, ..., Ni); Step 3: Fit several sets of hot plate temperature value sets and several sets of verification fabric sample color value sets to obtain a first prediction model.

6. The method for color management optimization of digital printing based on artificial intelligence according to claim 5, characterized in that: The method for predicting the lowest heating temperature of the hot plate within the color difference threshold range for different printed fabrics comprises: Step 1: Obtain the average ink penetration depth of the fabric to be printed and the printing color value of the sample fabric provided by the customer; Step 2: Calculate the difference between the printed color value and the color difference threshold in the sample fabric provided by the customer to obtain the color value of the fabric to be printed, input the color value of the fabric to be printed into the first prediction model, and calculate the minimum heating temperature of the hot plate under the average value of the ink penetration depth of the fabric to be printed.

7. The method for color management optimization of digital printing based on artificial intelligence according to claim 6, characterized in that: The method for establishing the second prediction model is: Step 1: Based on the color value of the verification sample, print and dye different verification fabric samples at the highest hot plate temperature, measure the L value of the verification fabric samples respectively, and set the L value set to (L1, L2, L3, ..., Ln); Step 2: Cut the printed verification fabric sample, measure and calculate the average dyeing depth, and set the average dyeing depth set to (P1, P2, P3, ..., Pn); Step 3: Obtain the average value of the ink penetration depth of each verification fabric sample at room temperature, and set the average value set of the ink penetration depth of the verification fabric samples to (Q1, Q2, Q3, ..., Qn); Step 4: Fitting the ink penetration depth average value set and the dyeing depth average value set of the verification fabric samples to obtain a first linear regression equation, and fitting the dyeing depth average value set and the L value set of the verification fabric samples to obtain a second linear regression equation; Step 5: Establish a second prediction model based on the first linear regression equation and the second linear regression equation.

8. The method for color management optimization of digital printing based on artificial intelligence according to claim 7, characterized in that: The method for predicting the printing L value of different printed fabrics outside the color difference threshold range is: The average value of the ink penetration depth of the fabric to be printed is obtained, and the average value of the ink penetration depth is input into the first linear regression equation to obtain the average value of the dyeing depth of the fabric to be printed. Finally, the average value of the dyeing depth of the fabric to be printed is input into the second linear regression equation to obtain the L value of the fabric to be printed at the maximum hot plate temperature.

Images