A decorative paper color matching optimization method and system based on color management system

Abstract

The present invention discloses a method and system for optimizing color matching of decorative paper based on a color management system, and relates to the technical field of printing color matching, including: obtaining the dot-ink matching ratio of decorative base paper according to a dot-ink matching ratio prediction model, and generating a reference sample paper by proofing; establishing a standard sample paper color matching model, obtaining color data of the reference sample paper, and inputting the data into the standard sample paper color matching model to obtain a standard sample paper for printing production; obtaining a decorative paper color matching model by training a color difference data set and a process parameter data set through multi-agent reinforcement learning; obtaining the color data of the decorative paper of a printing detection station at the current moment, and obtaining the optimization behavior of the printing process parameters at the current moment based on the decorative paper color matching model to complete the optimization of the color matching of the decorative paper; the method in the present application can perform adaptive color matching before the design of decorative paper and during the printing process, so as to reduce the color error in the production of decorative paper.

Description

Technical Field

[0001] The present invention relates to the field of printing color matching technology, and in particular to a decorative paper color matching optimization method and system based on a color management system. Background Art

[0002] Decorative paper is a special paper made by printing decorative base paper, impregnating it with resin and drying it to form a film paper. It is pressed on the surface of the board under high temperature conditions to decorate and protect the artificial board. At present, the printing of decorative paper is mainly gravure printing. In the actual production process of decorative paper printing, due to design errors, process flow, personnel operation and environmental control, some printing defects and color difference problems often occur, resulting in an increase in the scrap rate and affecting the company's benefits.

[0003] A Chinese patent application with application announcement number CN103347141B discloses a CMYK four-color printing color difference positioning method, including: acquiring the RGB color values ​​and CMYK color values ​​of a color card, establishing a mapping relationship between the RGB values ​​and the CMYK values, and then converting the RGB color values ​​of a reference printed image and the RGB color values ​​of a printed image to be detected into corresponding CMYK values, and finally calculating the average color difference of the two images in the CMYK color space, thereby completing the color difference positioning on the four CMYK color plates. This method solves the problem that the existing color difference analysis method cannot reflect the color difference value of each color plate, but fails to solve the color matching problem in the design and printing process.

[0004] To this end, the present invention proposes a decorative paper color matching optimization method and system based on a color management system. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a decorative paper color matching optimization method and system based on a color management system, which can solve the color matching problem of decorative paper in the design of reference sample paper and standard sample paper and in the printing process of decorative paper.

[0006] In view of the technical problems existing in the prior art, the present invention provides a method for optimizing color matching of decorative paper based on a color management system, and the specific steps are as follows:

[0007] Acquire color information of dots on decorative paper and ink proportion of the dots, and establish a dot-ink proportion data set;

[0008] According to the halftone ink ratio prediction model, the halftone ink ratio of the decorative base paper is obtained, and a reference sample paper is generated by proofing; the halftone ink ratio prediction model is obtained by training the halftone ink ratio data set;

[0009] Establishing a standard sample paper color matching model, obtaining color data of a reference sample paper, and inputting the color data into the standard sample paper color matching model to obtain a standard sample paper for printing production;

[0010] Obtain the color data difference between the decorative paper of the printing test table and the standard sample paper, establish a color difference data set, obtain printing process parameters, construct a process parameter data set, and train the color difference data set and the process parameter data set through multi-agent reinforcement learning to obtain a decorative paper color matching model;

[0011] The color data of the decorative paper of the printing detection table at the current moment is obtained, and based on the decorative paper color matching model, the optimization behavior of the printing process parameters at the current moment is obtained to complete the decorative paper color matching optimization.

[0012] In particular, the color information of the dots of the decorative paper and the ink proportion of the dots are obtained, wherein:

[0013] The color information of the decorative paper dots is obtained by using a spectrophotometer;

[0014] The color information includes Lab color space value and CMYK color mode value;

[0015] The ink ratio of the dots refers to the ratio concentration of the basic color ink.

[0016] In particular, the steps of obtaining the dot-ink ratio of the decorative base paper according to the dot-ink ratio prediction model and generating a reference sample paper by proofing are as follows:

[0017] Step S11: training a dot ink ratio prediction model;

[0018] Step S12: inputting the color information of the reference sample paper into the dot-ink ratio prediction model to obtain the dot-ink ratio of the reference sample paper;

[0019] Step S13: Proofing according to the dot-inking ratio to generate a first reference sample paper;

[0020] Step S14: Use a spectrophotometer to measure the color information of the first reference sample paper and perform color difference value calculation with the color information of the reference sample paper to set a first threshold value:

[0021] If the proofing color difference value is less than or equal to the first threshold, re-proofing is rejected and the first reference sample paper is updated as the reference sample paper;

[0022] If the color difference value of the proof is greater than the first threshold, proof again to generate the Y i Reference sample paper, for the Y i The reference paper is again subjected to color difference value judgment until the Yth iIf the color difference value of the reference sample paper is less than or equal to the first threshold, i The reference sample paper is updated to the reference sample paper, where i=2, ...n;

[0023] Step S15: The Yth i The color information of the reference sample paper and the corresponding dot-ink matching ratio are updated to a dot-ink matching ratio data set, where i=1, 2, ...n;

[0024] The halftone dot ink ratio prediction model is obtained by training the halftone dot ink ratio data set through a BP neural network model.

[0025] In particular, the steps of establishing the standard sample paper color matching model are:

[0026] Step S21: obtaining color data of a standard sample paper, wherein the color data includes five index parameters: dot gain, color difference, overprint rate, relative contrast and overprint error;

[0027] Step S22: Using the analytic hierarchy process to classify the color data of the standard paper and calculate the weight w of the index parameter a , where a=1, 2, …5;

[0028] Step S23: Calculate the weight w of the index parameter in the color data of the standard sample paper using the entropy method b , where b = 1, 2, ... 5;

[0029] Step S24: Calculate the weight w of the index parameter in the color data of the standard sample paper c ,

[0030]

[0031] Wherein, c = 1, 2, ... 5;

[0032] Step S25: normalizing the numerical value of the index parameter to obtain the index standardized value;

[0033] Step S26: Compare the index normalization value and the weight w of the index parameter c Multiply them together to get the comprehensive score;

[0034] Step S27: grading the comprehensive scores, setting a second threshold, establishing a color evaluation system, and obtaining a standard sample paper color matching model.

[0035] In particular, the steps of obtaining the color data of the reference sample paper and inputting it into the standard sample paper color matching model to obtain the standard sample paper for printing production are:

[0036] Step S31: obtaining color data of a reference sample paper;

[0037] Step S32: inputting into the standard paper color matching model to calculate the comprehensive score;

[0038] Step S33: The color data comprehensive score of the reference sample paper is judged against a second threshold to obtain a standard sample paper for printing production;

[0039] If the comprehensive score of the color data of the reference paper sample is greater than or equal to the second threshold, the reference paper sample is updated to the standard paper sample for printing production;

[0040] If the color data comprehensive score of the reference sample paper is less than the second threshold, proofing is performed again to generate the Yth j Reference sample paper, calculate the Y j The color data comprehensive score of the reference sample paper reference sample paper until the Yth j The color data comprehensive score of the reference sample paper is less than the second threshold value, and the Y j The reference sample paper is updated to the standard sample paper used for printing production, where j=2, ...n.

[0041] In particular, the step of training the color difference data set and the process parameter data set through multi-agent reinforcement learning to obtain a decorative paper color matching model is:

[0042] Step 41: obtaining the color data difference between the decorative paper of the printing test table and the standard sample paper, establishing a color difference data set, obtaining printing process parameters, and constructing a process parameter data set;

[0043] Step 42: Based on the color difference data set and the process parameter data set, construct an expert experience sequence A(s t , a t , r, s t+1 ); the t is the color difference state set, a t is the set of actions to optimize process parameters, r is the set of reward values ​​for operation effects, s t+1 is the color difference state set for the next moment;

[0044] Step S43: constructing an expert experience pool based on the expert experience sequence;

[0045] Step S44: using the expert experience pool to train a multi-agent reinforcement learning model to obtain a decorative paper color matching model;

[0046] The color data difference is the color difference and the overprint rate difference between the decorative paper of the printing test table and the standard sample paper;

[0047] The printing process parameters include printing pressure, surface finish of the rubber roller, temperature and printing speed.

[0048] Furthermore, the steps of using the expert experience pool to train the multi-agent reinforcement learning model to obtain the decorative paper color matching model are:

[0049] Step S51: constructing a prediction network and a target network using a BP neural network;

[0050] The BP neural network consists of an input layer, a hidden layer and an output layer;

[0051] The number of neurons in the input layer is equal to the number of states in the color difference state set;

[0052] The number of neurons in the output layer is equal to the number of actions in the set of actions for optimizing process parameters;

[0053] Step S52: constructing N agents based on the prediction network, the target network and the experience pool, where N is a variable and N≥1;

[0054] Step S53: Initialize the experience pool as an expert experience pool; the experience pool has a capacity of M;

[0055] Step S54: Initialize the prediction network using random prediction network weights θ and use random target network weights θ * =θ initialize the target network and initialize the parameter state set s t , batch n and step size m;

[0056] Step S55: Randomly obtain expert experience sequence A(s) in batches n from the experience pool t , a t , r, s t+1 ) for training, where t is a variable and n<M;

[0057] Predict the network input color difference state set s t and optimize the process parameter action set a t , output prediction value function Q N (s t , a t |θ) and the next moment optimization process parameter action set a t+1 * ;

[0058] The target network inputs the color difference state set s at the next moment t+1 , output target value function F N , F N The calculation formula is:

[0059]

[0060] In the formula, is the color difference state set s t When optimizing the process parameter action set a t The operation effect reward value, γ is the discount factor and γ∈(0,1), max Q N (s t+1 , a t+1 * |θ * ) is the color difference state set s at the next moment t+1 When the next moment is selected, the optimal process parameter action set a is selected t+1 * The maximum predicted value function obtained;

[0061] Step S56: Every m steps, the weight parameter θ of the prediction network is copied to the weight parameter θ of the target network. * ;

[0062] Step S57: Calculate the joint prediction value function Q of multiple agents TN (s t , a t |θ), the calculation formula is:

[0063] Q TN (s t , a t |θ)=f(Q1(s t , a t |θ), Q2(s t , a t |θ),…Q N (s t , a t |θ));

[0064] Calculate the joint objective value function F of multiple agents N , the calculation formula is:

[0065] F TN =g(F1, F2, ... F N );

[0066] In the formula, the f function and g function are parameterized as BP neural network model calculation;

[0067] The multi-agent loss function is updated to Loss N , the calculation formula is:

[0068] Loss N =F TN -Q TN (s t , a t |θ);

[0069] Step S58: Use gradient descent method to minimize the MSE loss function Loss N , update the network model parameters θ and θ * Until convergence, the decorative paper color matching model is obtained.

[0070] Another aspect of the present application provides a decorative paper color matching optimization system based on a color management system, the system comprising:

[0071] Dot data acquisition module: used to acquire the color information of the dots on the decorative paper and the ink ratio of the dots, and establish a dot-ink ratio data set;

[0072] Ink dot ratio prediction module: used to obtain the dot ink dot ratio of the decorative base paper according to the dot ink dot ratio prediction model, and to generate a reference sample paper by proofing; the dot ink dot ratio prediction model is trained by the dot ink dot ratio data set;

[0073] Color matching model module: used to establish the color matching model of standard sample paper;

[0074] Standard sample paper generation module: used to obtain the color data of the reference sample paper, input it into the standard sample paper color matching model, and obtain the standard sample paper for printing production;

[0075] Color matching model module: used to obtain the color data difference between the decorative paper of the printing test table and the standard sample paper, establish a color difference data set, obtain printing process parameters, construct a process parameter data set, and train the color difference data set and the process parameter data set through multi-agent reinforcement learning to obtain a decorative paper color matching model;

[0076] Color matching optimization module: used to obtain the color data of the decorative paper of the printing detection table at the current moment, obtain the optimization behavior of the printing process parameters at the current moment based on the decorative paper color matching model, and complete the decorative paper color matching optimization.

[0077] Another aspect of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the above-mentioned method for optimizing color matching of decorative paper based on a color management system are implemented.

[0078] Another aspect of the present application provides a readable storage medium storing a computer program, wherein the computer program is suitable for being loaded by a processor to implement the steps of the above-mentioned method for optimizing color matching of decorative paper based on a color management system.

[0079] Beneficial technical effects of the present invention:

[0080] The present invention uses the color information and ink matching ratio of the decorative paper dots to train the dot-ink matching ratio prediction model. Based on the dot-ink matching ratio prediction model, the color matching ratio of the dots can be predicted in the design of the reference sample paper, and proofing can be performed according to the color matching ratio, thereby reducing the trial and error cost and proofing error of manual proofing, improving the design level, and reducing the color difference problem in the early design of decorative paper. A verification and judgment mechanism is added after the dot-ink matching ratio prediction model, so that even samples with incorrect predictions can be retained and passed to the training set of the prediction model, which enriches the training data set and improves the robustness of the training model.

[0081] The present invention establishes a color evaluation system, obtains a standard sample paper color matching model, evaluates the color quality of the reference sample paper, further selects qualified reference sample paper as the standard sample paper for printing production, and reduces the color difference problem in the design of decorative paper. The decorative paper color matching model is obtained by training the color difference data set and the process parameter data set through multi-agent reinforcement learning. The process parameters can be adjusted according to the color error in real time during the printing process, the printing process can be controlled in real time, the color error problem in printing production can be reduced, and the color matching adaptive optimization in the decorative paper printing process can be completed. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 This is a flow chart of a decorative paper color matching optimization method based on a color management system according to Embodiment 1 of the present invention.

[0083] Figure 2 This is a step of establishing a standard sample paper color matching model in embodiment 1 of the present invention.

[0084] Figure 3 This is a step of training a multi-agent reinforcement learning model in Example 1 of the present invention.

[0085] Figure 4 A decorative paper color matching optimization system based on a color management system is provided in Example 2 of the present invention.

[0086] Figure 5 This is a schematic diagram of the structure of an electronic device provided in Example 3 of the present invention.

[0087] Figure 6 It is a schematic diagram of the structure of a computer-readable storage medium provided in Example 4 of the present invention. DETAILED DESCRIPTION

[0088] In order to better understand the present application, a more detailed description will be made of various aspects of the present application with reference to the accompanying drawings. It should be understood that these detailed descriptions are only descriptions of exemplary embodiments of the present application, and are not intended to limit the scope of the present application in any way. Throughout the specification, the same reference numerals refer to the same elements. The expression "and / or" includes any and all combinations of one or more of the associated listed items.

[0089] In the accompanying drawings, the size, dimensions and shapes of the elements have been slightly adjusted for ease of illustration. The drawings are for illustration only and are not strictly drawn to scale. As used herein, the terms "substantially", "approximately" and similar terms are used as terms of approximation, not as terms of degree, and are intended to illustrate the inherent deviations in measurements or calculations that will be recognized by those of ordinary skill in the art. In addition, in this application, the order in which the steps are described does not necessarily represent the order in which these processes occur in actual operation, unless otherwise specified or can be derived from the context.

[0090] It should also be understood that expressions such as "include", "including", "have", "contain" and / or "comprising" are open rather than closed expressions in this specification, which indicate the presence of the stated features, elements and / or components, but do not exclude the presence of one or more other features, elements, components and / or combinations thereof. In addition, when expressions such as "at least one of..." appear after a list of listed features, they modify the entire list of features rather than just the individual elements in the list. In addition, when describing embodiments of the present application, "may" is used to mean "one or more embodiments of the present application". And, the term "exemplary" is intended to refer to an example or illustration.

[0091] Unless otherwise specified, all words (including engineering terms and scientific and technological terms) used in this article have the same meaning as those commonly understood by ordinary technicians in the field to which this application belongs. It should also be understood that, unless clearly stated in this application, words defined in common dictionaries should be interpreted as having the same meaning as their meaning in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense.

[0092] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0093] Example 1

[0094] like Figure 1 to Figure 3 As shown, this embodiment provides a decorative paper color matching optimization method based on a color management system;

[0095] like Figure 1 As shown, the steps of the decorative paper color matching optimization method based on the color management system include:

[0096] S1. Acquire color information of dots on decorative paper and ink ratio of the dots, and establish a dot-ink ratio data set;

[0097] In particular, the color information of the dots of the decorative paper and the ink proportion of the dots are obtained, wherein:

[0098] The color information of the decorative paper dots is obtained by using a spectrophotometer;

[0099] The color information includes Lab color space value and CMYK color mode value;

[0100] The ink ratio of the dots refers to the ratio concentration of the basic color ink.

[0101] S2. According to the dot-ink ratio prediction model, the dot-ink ratio of the decorative base paper is obtained, and a reference sample paper is generated by proofing; the dot-ink ratio prediction model is obtained by training the dot-ink ratio data set;

[0102] In particular, the steps of obtaining the dot-ink ratio of the decorative base paper according to the dot-ink ratio prediction model and generating a reference sample paper by proofing are as follows:

[0103] Step S11: training a dot ink ratio prediction model;

[0104] Step S12: inputting the color information of the reference sample paper into the dot-ink ratio prediction model to obtain the dot-ink ratio of the reference sample paper;

[0105] Step S13: Proofing according to the dot-inking ratio to generate a first reference sample paper;

[0106] Step S14: Use a spectrophotometer to measure the color information of the first reference sample paper and perform color difference value calculation with the color information of the reference sample paper to set a first threshold value:

[0107] If the proofing color difference value is less than or equal to the first threshold, re-proofing is rejected and the first reference sample paper is updated as the reference sample paper;

[0108] If the color difference value of the proof is greater than the first threshold, proof again to generate the Y i Reference sample paper, for the Y i The reference paper is again subjected to color difference value judgment until the Yth i If the color difference value of the reference sample paper is less than or equal to the first threshold, i The reference sample paper is updated to the reference sample paper, where i=2, ...n;

[0109] Step S15: The Yth i The color information of the reference sample paper and the corresponding dot-ink matching ratio are updated to a dot-ink matching ratio data set, where i=1, 2, ...n;

[0110] The halftone dot ink ratio prediction model is obtained by training the halftone dot ink ratio data set through a BP neural network model.

[0111] The dot-ink ratio prediction model is trained using the color information and ink ratio of the decorative paper dots. Based on the dot-ink ratio prediction model, the color ratio of the dots can be predicted in the design of the reference sample paper, and proofing can be performed according to the color ratio, thereby reducing the trial and error cost and proofing error of manual proofing, improving the design level, and reducing the color difference problem in the early design of decorative paper.

[0112] A verification and judgment mechanism is added after the dot-ink ratio prediction model. Even samples with incorrect predictions can be retained and passed to the training set of the prediction model, which enriches the training data set and improves the robustness of the training model.

[0113] S3, establishing a standard sample paper color matching model, obtaining color data of the reference sample paper, and inputting the color data into the standard sample paper color matching model to obtain a standard sample paper for printing production;

[0114] In particular, see Figure 2 The steps of establishing the standard sample paper color matching model are:

[0115] Step S21: obtaining color data of a standard sample paper, wherein the color data includes five index parameters: dot gain, color difference, overprint rate, relative contrast and overprint error;

[0116] In a preferred embodiment,

[0117] The dot gain is obtained by measuring the result with a spectrophotometer.

[0118] Chromatic Aberration The method of obtaining is: select a batch of samples, use a spectrophotometer to measure the L of the CIEL*a*b* uniform color space at the same position of all samples * 、a * and b * The color difference is calculated using the CIE1976 color difference formula. When calculating the color difference of samples in the same batch, the maximum value is taken as the color difference of the same color in the same batch of samples;

[0119]

[0120] Where, L * Indicates the value of a * and b* Refers to the chromaticity value.

[0121] Overprint rate T 1+2 The method of obtaining is: using a spectrophotometer to measure the density values ​​of the first spot color, the second spot color and the overprint of the two spot colors on the measurement and control strip in sequence, and the overprint rate is calculated using the Preucil formula;

[0122] T 1+2 =(D 1+2 -D1) / D2;

[0123] Where D 1+2 It is the overprint density minus the paper density, D1 is the first printed ink density minus the paper density, and D2 is the second printed ink density minus the paper density.

[0124] The relative contrast K is obtained by measuring the density of the solid color block and the dot area ratio color block using a spectrophotometer;

[0125]

[0126] Where D R D is the integrated density of the dot area at a certain point between the middle tone and the dark tone on the screen. V It is the solid density value on the printed product.

[0127] The method for obtaining the overprint error is: place the sample flatly under the light source of the test bench, and use a 20x scale microscope with an accuracy of 0.01mm to measure the overprint error between any two colors in the main and secondary parts of the sample at 3 points, and take the maximum value and record it, which is the overprint error of the sample.

[0128] Step S22: Using the analytic hierarchy process to classify the color data of the standard paper and calculate the weight w of the index parameter a , where a=1,2,…5;

[0129] Step S23: Calculate the weight w of the index parameter in the color data of the standard sample paper using the entropy method b , where b = 1, 2, ... 5;

[0130] Step S24: Calculate the weight w of the index parameter in the color data of the standard sample paper c ,

[0131]

[0132] Where c = 1, 2, ... 5;

[0133] Step S25: normalizing the numerical value of the index parameter to obtain the index standardized value;

[0134] Step S26: Compare the index normalization value and the weight w of the index parameter c Multiply them together to get the comprehensive score;

[0135] Step S27: grading the comprehensive scores, setting a second threshold, establishing a color evaluation system, and obtaining a standard sample paper color matching model.

[0136] In particular, the steps of obtaining the color data of the reference sample paper and inputting it into the standard sample paper color matching model to obtain the standard sample paper for printing production are:

[0137] Step S31: obtaining color data of a reference sample paper;

[0138] Step S32: inputting into the standard paper color matching model to calculate the comprehensive score;

[0139] Step S33: The color data comprehensive score of the reference sample paper is judged against a second threshold to obtain a standard sample paper for printing production;

[0140] If the comprehensive score of the color data of the reference paper sample is greater than or equal to the second threshold, the reference paper sample is updated to the standard paper sample for printing production;

[0141] If the color data comprehensive score of the reference sample paper is less than the second threshold, proofing is performed again to generate the Yth j Reference sample paper, calculate the Y j The color data comprehensive score of the reference sample paper reference sample paper until the Yth j The color data comprehensive score of the reference sample paper is less than the second threshold value, and the Y j The reference sample paper is updated to the standard sample paper used for printing production, where j=2,…n.

[0142] Furthermore, the steps of AHP are:

[0143] (1) For each level of indicators, the relative importance of each other is compared, and the comparison results are assigned using the 1-9 scaling method to construct a judgment matrix A. For multiple indicators at the same level, the pairwise comparison judgment matrix A can be obtained as (a ij ), the values ​​in matrix A must satisfy a ij >0, a ii =1; The following table shows the scales of the judgment matrix 1-9 and their meanings;

[0144]

[0145] (2) Calculate the index weight W i :

[0146]

[0147]

[0148] Where M i is the product of the values ​​of each row of the judgment matrix, a ij is the element in the judgment matrix;

[0149] (3) Perform consistency test. First, calculate the consistency index C I :

[0150]

[0151]

[0152]

[0153] In the formula, λ max is the maximum eigenvalue of the judgment matrix;

[0154] Query the average random consistency index R I , as shown in the table:

[0155]

[0156] Finally, calculate the consistency ratio C R :

[0157]

[0158] Verification results:

[0159] If C R =0, the judgment matrix is ​​completely consistent;

[0160] If C R When <0.1, the judgment matrix has satisfactory consistency;

[0161] If C R When ≥0.1, the judgment matrix has unsatisfactory consistency and should be adjusted.

[0162] Furthermore, the weight calculation steps of the entropy method are:

[0163] (1) Input a value

[0164] Select n samples and m indicators, then the value of the jth indicator of the i-th sample, where i = 1, 2, ..., n; j = 1, 2, ..., m;

[0165] (2) Normalization of indicators

[0166] When j is a positive indicator:

[0167]

[0168] When j is a negative indicator:

[0169]

[0170] (3) Calculate specific gravity p ij

[0171] Calculate the proportion of the i-th sample under the j-th indicator:

[0172]

[0173] Where, i = 1, ..., n; j = 1, ..., m;

[0174] (4) Calculate the entropy value e of the jth indicator j :

[0175]

[0176] (5) Calculate the information entropy redundancy d j :

[0177] d j =1-e j ;

[0178] (6) Calculate the weight w of each indicator j :

[0179]

[0180] S4, obtaining the color data difference between the decorative paper of the printing test table and the standard sample paper, establishing a color difference data set, obtaining printing process parameters, constructing a process parameter data set, and training the color difference data set and the process parameter data set through multi-agent reinforcement learning to obtain a decorative paper color matching model;

[0181] In particular, the steps of obtaining a decorative paper color matching model by training the color difference data set and the process parameter data set through multi-agent reinforcement learning are:

[0182] Step 41: obtaining the color data difference between the decorative paper of the printing test table and the standard sample paper, establishing a color difference data set, obtaining printing process parameters, and constructing a process parameter data set;

[0183] Step 42: Based on the color difference data set and the process parameter data set, construct an expert experience sequence A(s t , a t , r, s t+1 ); the t is the color difference state set, a tis the set of actions to optimize process parameters, r is the set of reward values ​​for operation effects, s t+1 is the color difference state set for the next moment;

[0184] Step S43: constructing an expert experience pool based on the expert experience sequence;

[0185] Step S44: using the expert experience pool to train a multi-agent reinforcement learning model to obtain a decorative paper color matching model;

[0186] The color data difference is the color difference and the overprint rate difference between the decorative paper of the printing test table and the standard sample paper;

[0187] The printing process parameters include printing pressure, surface finish of the rubber roller, temperature and printing speed.

[0188] Further, see Figure 3 The steps of using the expert experience pool to train the multi-agent reinforcement learning model to obtain the decorative paper color matching model are:

[0189] Step S51: constructing a prediction network and a target network using a BP neural network;

[0190] The BP neural network consists of an input layer, a hidden layer and an output layer;

[0191] The number of neurons in the input layer is equal to the number of states in the color difference state set;

[0192] The number of neurons in the output layer is equal to the number of actions in the set of actions for optimizing process parameters;

[0193] Step S52: constructing N agents based on the prediction network, the target network and the experience pool, where N is a variable and N≥1;

[0194] Step S53: Initialize the experience pool as an expert experience pool; the experience pool has a capacity of M;

[0195] Step S54: Initialize the prediction network using random prediction network weights θ and use random target network weights θ * =θ initialize the target network and initialize the parameter state set s t , batch n and step size m;

[0196] Step S55: Randomly obtain expert experience sequence A(s) in batches n from the experience pool t , a t , r, s t+1 ) for training, where t is a variable and n<M;

[0197] Predict the network input color difference state set s tand optimize the process parameter action set a t , output prediction value function Q N (s t , a t |θ) and the next moment optimization process parameter action set a t+1 * ;

[0198] The target network inputs the color difference state set s at the next moment t+1 , output target value function F N , F N The calculation formula is:

[0199]

[0200] In the formula, is the color difference state set s t When optimizing the process parameter action set a t The operation effect reward value, γ is the discount factor and γ∈(0,1), max Q N (s t+1 , a t+1 * |θ * ) is the color difference state set s at the next moment t+1 When the next moment is selected, the optimal process parameter action set a is selected t+1 * The maximum predicted value function obtained;

[0201] Step S56: Every m steps, the weight parameter θ of the prediction network is copied to the weight parameter θ of the target network. * ;

[0202] Step S57: Calculate the joint prediction value function Q of multiple agents TN (s t , a t |θ), the calculation formula is:

[0203] Q TN (s t , a t |θ)=f(Q1(s t , a t |θ), Q2(s t , a t |θ),…Q N (s t , a t |θ));

[0204] Calculate the joint objective value function F of multiple agents N , the calculation formula is:

[0205] FTN =g(F1, F2, ... F N );

[0206] In the formula, the f function and g function are parameterized as BP neural network model calculation;

[0207] The multi-agent loss function is updated to Loss N , the calculation formula is:

[0208] Loss N =F TN -Q TN (s t , a t |θ);

[0209] Step S58: Use gradient descent method to minimize the MSE loss function Loss N , update the network model parameters θ and θ * Until convergence, the decorative paper color matching model is obtained.

[0210] S5. Acquire the color data of the decorative paper of the printing inspection table at the current moment, obtain the optimization behavior of the printing process parameters at the current moment based on the decorative paper color matching model, and complete the decorative paper color matching optimization.

[0211] A color evaluation system was established to obtain a standard sample paper color matching model, the color quality of the reference sample paper was evaluated, and qualified reference sample paper was further screened out as the standard sample paper for printing production, thereby achieving the purpose of reducing the color difference problem in decorative paper design.

[0212] The decorative paper color matching model is obtained by training the color difference dataset and the process parameter dataset through multi-agent reinforcement learning. The process parameters can be adjusted adaptively according to the color error in real time during the printing process, the printing process can be controlled in real time, the color error problem in printing production can be reduced, and the color matching of decorative paper in the printing process can be adaptively optimized.

[0213] Example 2

[0214] Figure 4 An embodiment of the present application provides a decorative paper color matching optimization system based on a color management system. The system as a whole can be divided into:

[0215] Dot data acquisition module: used to acquire the color information of the dots on the decorative paper and the ink ratio of the dots, and establish a dot-ink ratio data set;

[0216] Ink dot ratio prediction module: used to obtain the dot ink dot ratio of the decorative base paper according to the dot ink dot ratio prediction model, and to generate a reference sample paper by proofing; the dot ink dot ratio prediction model is trained by the dot ink dot ratio data set;

[0217] Color matching model module: used to establish the color matching model of standard sample paper;

[0218] Standard sample paper generation module: used to obtain the color data of the reference sample paper, input it into the standard sample paper color matching model, and obtain the standard sample paper for printing production;

[0219] Color matching model module: used to obtain the color data difference between the decorative paper of the printing test table and the standard sample paper, establish a color difference data set, obtain printing process parameters, construct a process parameter data set, and train the color difference data set and the process parameter data set through multi-agent reinforcement learning to obtain a decorative paper color matching model;

[0220] Color matching optimization module: used to obtain the color data of the decorative paper of the printing detection table at the current moment, obtain the optimization behavior of the printing process parameters at the current moment based on the decorative paper color matching model, and complete the decorative paper color matching optimization.

[0221] Example 3

[0222] Figure 5 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 5 As shown, according to another aspect of the present application, an electronic device 100 is provided. The electronic device 100 may include one or more processors and one or more memories. The memories store computer readable codes, and when the computer readable codes are executed by one or more processors, a decorative paper color matching optimization method based on a color management system as described above may be executed.

[0223] The method or device according to the embodiment of the present application can also be used by Figure 5 The electronic device architecture shown in FIG. Figure 5As shown, the electronic device 100 may include a bus 101, one or more CPUs 102, a read-only memory (ROM) 103, a random access memory (RAM) 104, a communication port 105 connected to a network, an input / output component 106, a hard disk 107, etc. The storage device in the electronic device 100, such as the ROM 103 or the hard disk 107, may store a decorative paper color matching optimization method based on a color management system provided in the present application. A decorative paper color matching optimization method based on a color management system may, for example, include the following steps: obtaining color information of decorative paper dots and the ink matching ratio of the dots, and establishing a dot ink matching ratio data set; obtaining the dot ink matching ratio of decorative base paper according to a dot ink matching ratio prediction model, and generating a reference sample paper by proofing; the dot ink matching ratio prediction model is trained by the dot ink matching ratio data set; establishing a standard sample paper color matching model, obtaining color data of the reference sample paper, and inputting it into the standard sample paper color matching model to obtain a standard sample paper for printing production; obtaining the color data difference between the decorative paper of a printing test station and the standard sample paper, establishing a color difference data set, obtaining printing process parameters, constructing a process parameter data set, and training the color difference data set and the process parameter data set through multi-agent reinforcement learning to obtain a decorative paper color matching model; obtaining the color data of the decorative paper of the printing test station at the current moment, and obtaining the optimization behavior of the printing process parameters at the current moment based on the decorative paper color matching model to complete the decorative paper color matching optimization.

[0224] Furthermore, the electronic device 100 may further include a user interface 108. Figure 5 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 5 One or more components of an electronic device are shown.

[0225] Example 4

[0226] Figure 6 Schematic diagram of a computer-readable storage medium structure provided by an embodiment of the present application. Figure 6 As shown, a computer-readable storage medium 200 according to an embodiment of the present application is shown. Computer-readable instructions are stored on the computer-readable storage medium 200. When the computer-readable instructions are executed by the processor, a method for optimizing color matching of decorative paper based on a color management system according to an embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium 200 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0227] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, the non-transitory machine-readable storage medium stores machine-readable instructions, the machine-readable instructions can be run by a processor to execute instructions corresponding to the method steps provided by the present application, and when the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0228] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0229] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0230] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0231] The above preset parameters or preset thresholds are all set by technicians in this field according to actual conditions or obtained through large-scale data simulation.

[0232] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A decorative paper color matching optimization method based on a color management system, characterized in that: The following steps are involved: Acquire color information of decorative paper dots and ink ratio of the dots, and establish a dot-ink ratio data set; According to the halftone ink ratio prediction model, the halftone ink ratio of the decorative base paper is obtained, and a reference sample paper is generated by proofing; the halftone ink ratio prediction model is obtained by training the halftone ink ratio data set; Establishing a standard sample paper color matching model, obtaining color data of a reference sample paper, and inputting the color data into the standard sample paper color matching model to obtain a standard sample paper for printing production; Obtain the color data difference between the decorative paper of the printing test table and the standard sample paper, establish a color difference data set, obtain printing process parameters, construct a process parameter data set, and train the color difference data set and the process parameter data set through multi-agent reinforcement learning to obtain a decorative paper color matching model; The steps to obtain the decorative paper color matching model are: Step S51: constructing a prediction network and a target network using a BP neural network; The BP neural network consists of an input layer, a hidden layer and an output layer; The number of neurons in the input layer is equal to the number of states in the color difference state set; The number of neurons in the output layer is equal to the number of actions in the set of actions for optimizing process parameters; Step S52: construct N agents based on the prediction network, the target network and the experience pool, where is a variable and ; Step S53: Initialize the experience pool to an expert experience pool; the capacity of the expert experience pool is M; Step S54: Use random prediction network weights Initialize the prediction network with random target network weights Initialize the target network and initialize the parameter state set , batch n and step size m; Step S55: Randomly obtain expert experience sequences from the experience pool in batches n Conduct training, including ; Predict the network input color difference state set and optimize the process parameter action set , output prediction value function and the next moment optimization process parameter action set ; The target network inputs the color difference state set at the next moment , output target value function , The calculation formula is: ； In the formula, is the color difference state set Optimize the process parameter action set The operation effect reward value, is the discount factor and , The color difference state set for the next moment The next moment optimization process parameter action set is selected The maximum predicted value function obtained; Step S56: Every m steps, the weight parameters of the prediction network are Weight parameters copied to the target network ; Step S57: Calculate the joint prediction value function of multiple agents , the calculation formula is: ； Calculate the joint objective value function of multiple agents , the calculation formula is: ； In the formula, The function and g function are parameterized as BP neural network model calculation; The multi-agent loss function is updated as , the calculation formula is: ； Step S58: Use gradient descent method to minimize the MSE loss function , update the network model parameters and Until convergence, the decorative paper color matching model is obtained; The color data of the decorative paper of the printing detection table at the current moment is obtained, and based on the decorative paper color matching model, the optimization behavior of the printing process parameters at the current moment is obtained to complete the decorative paper color matching optimization.

2. The method for optimizing color matching of decorative paper based on a color management system according to claim 1, characterized in that: The step of obtaining the color information of the dots on the decorative paper and the ink proportion of the dots comprises: The color information of the decorative paper dots is obtained by using a spectrophotometer; The color information includes Lab color space value and CMYK color mode value; The ink ratio of the dots refers to the ratio concentration of the basic color ink.

3. The method for optimizing color matching of decorative paper based on a color management system according to claim 2, characterized in that: The steps of obtaining the dot-ink ratio of the decorative base paper according to the dot-ink ratio prediction model and generating a reference sample paper by proofing are as follows: Step S11: training a dot ink ratio prediction model; Step S12: inputting the color information of the reference sample paper into the dot-ink ratio prediction model to obtain the dot-ink ratio of the reference sample paper; Step S13: Proofing according to the dot-inking ratio to generate a first reference sample paper; Step S14: Use a spectrophotometer to measure the color information of the first reference sample paper and perform color difference value calculation with the color information of the reference sample paper to set a first threshold value: If the proofing color difference value is less than or equal to the first threshold, re-proofing is rejected and the first reference sample paper is updated as the reference sample paper; If the color difference value of the proof is greater than the first threshold, proofing is performed again to generate the Reference sample paper, for the The reference paper is again subjected to color difference value judgment until the If the color difference value of the reference sample paper is less than or equal to the first threshold, The reference sample paper is updated to the reference sample paper, which ; Step S15: The first The color information of the reference sample paper and the corresponding dot-ink matching ratio are updated to the dot-ink matching ratio data set, where ; The halftone dot ink ratio prediction model is obtained by training the halftone dot ink ratio data set through a BP neural network model.

4. The method for optimizing color matching of decorative paper based on a color management system according to claim 3, characterized in that: The steps of establishing the standard sample paper color matching model are: Step S21: obtaining color data of a standard sample paper, wherein the color data includes five index parameters: dot gain, color difference, overprint rate, relative contrast and overprint error; Step S22: Using the analytic hierarchy process to classify the color data of the standard paper and calculate the weight of the index parameters ,in ; Step S23: Calculate the weight of the index parameter in the color data of the standard sample paper using the entropy method ,in ; Step S24: Calculate the weight of the index parameter in the color data of the standard sample paper , ； In the formula, ; Step S25: normalizing the numerical value of the index parameter to obtain the index standardized value; Step S26: Compare the index normalization value with the weight of the index parameter Multiply them together to get the comprehensive score; Step S27: grading the comprehensive scores, setting a second threshold, establishing a color evaluation system, and obtaining a standard sample paper color matching model.

5. The method for optimizing color matching of decorative paper based on a color management system according to claim 4, characterized in that: The steps of obtaining the color data of the reference sample paper and inputting it into the standard sample paper color matching model to obtain the standard sample paper for printing production are: Step S31: obtaining color data of a reference sample paper; Step S32: inputting into the standard paper color matching model to calculate the comprehensive score; Step S33: The color data comprehensive score of the reference sample paper is judged against a second threshold to obtain a standard sample paper for printing production; If the comprehensive score of the color data of the reference paper sample is greater than or equal to the second threshold, the reference paper sample is updated to the standard paper sample for printing production; If the color data comprehensive score of the reference sample paper is less than the second threshold, proofing is performed again to generate the first Base sample paper, calculate the The color data of the reference sample paper is comprehensively scored until the The color data comprehensive score of the reference sample paper is less than the second threshold value, The reference paper is updated to the standard paper used for printing production, including .

6. The method for optimizing color matching of decorative paper based on a color management system according to claim 5, characterized in that: The step of obtaining a decorative paper color matching model by training the color difference data set and the process parameter data set through multi-agent reinforcement learning is: Step 41: obtaining the color data difference between the decorative paper of the printing test table and the standard sample paper, establishing a color difference data set, obtaining printing process parameters, and constructing a process parameter data set; Step 42: Construct an expert experience sequence based on the color difference data set and the process parameter data set ; is the color difference state set, To optimize the process parameter action set, is the set of reward values ​​for the operation effect, is the color difference state set for the next moment; Step S43: constructing an expert experience pool based on the expert experience sequence; Step S44: using the expert experience pool to train a multi-agent reinforcement learning model to obtain a decorative paper color matching model; The color data difference is the color difference and the overprint rate difference between the decorative paper of the printing test table and the standard sample paper; The printing process parameters include printing pressure, surface finish of the rubber roller, temperature and printing speed.

7. A decorative paper color matching optimization system based on a color management system, which is based on a decorative paper color matching optimization method based on a color management system according to any one of claims 1 to 6, characterized in that: The system comprises: Dot data acquisition module: used to acquire the color information of the dots on the decorative paper and the ink ratio of the dots, and establish a dot ink ratio data set; Ink dot ratio prediction module: used to obtain the dot ink dot ratio of the decorative base paper according to the dot ink dot ratio prediction model, and to generate a reference sample paper by proofing; the dot ink dot ratio prediction model is trained by the dot ink dot ratio data set; Color matching model module: used to establish the color matching model of standard sample paper; Standard sample paper generation module: used to obtain the color data of the reference sample paper, input it into the standard sample paper color matching model, and obtain the standard sample paper for printing production; Color matching model module: used to obtain the color data difference between the decorative paper of the printing test table and the standard sample paper, establish a color difference data set, obtain printing process parameters, construct a process parameter data set, and train the color difference data set and the process parameter data set through multi-agent reinforcement learning to obtain a decorative paper color matching model; Color matching optimization module: used to obtain the color data of the decorative paper of the printing detection table at the current moment, obtain the optimization behavior of the printing process parameters at the current moment based on the decorative paper color matching model, and complete the decorative paper color matching optimization.

8. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the method for optimizing color matching of decorative paper based on a color management system as claimed in any one of claims 1 to 6 are implemented.

9. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor to execute the steps in the method for optimizing color matching of decorative paper based on a color management system according to any one of claims 1 to 6.

Images