CN113411466B - Multicolor chromatic aberration intelligent correction method and system based on color printing production system - Google Patents
Multicolor chromatic aberration intelligent correction method and system based on color printing production system Download PDFInfo
- Publication number
- CN113411466B CN113411466B CN202110502558.XA CN202110502558A CN113411466B CN 113411466 B CN113411466 B CN 113411466B CN 202110502558 A CN202110502558 A CN 202110502558A CN 113411466 B CN113411466 B CN 113411466B
- Authority
- CN
- China
- Prior art keywords
- printing
- color difference
- color
- chromatic aberration
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/40—Picture signal circuits
- H04N1/409—Edge or detail enhancement; Noise or error suppression
- H04N1/4092—Edge or detail enhancement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention discloses a multicolor chromatic aberration intelligent correction method and a system based on a multicolor printing production system, wherein the method comprises the following steps: selecting optimal system parameters by using data in the experience database, and performing trial production on the printing sample; collecting pixel point data of a printing sample in real time, and correspondingly storing customer requirements, product types, optimal system parameters, design image pixel point data and the pixel point data of the printing sample; carrying out dimension reduction processing and color difference analysis on pixel point data in an experience database; carrying out intelligent chromatic aberration correction and printing chromatic aberration prediction on the design drawing by using a neural network model; and designing a template according to the corrected design drawing to print a sample, simultaneously acquiring the sample data and the weighted chromatic aberration of the printed sample in real time, and performing formal printing production if the production conditions are met. The invention not only effectively reduces the cost of manpower, materials and time brought by repeated plate making and manual adjustment, but also improves the production quality and efficiency of the printing system.
Description
Technical Field
The invention relates to the field of chromatic aberration correction of a color printing production system, in particular to a method and a system for intelligently correcting multi-color chromatic aberration based on the color printing production system.
Background
The quality of a color printed product mainly depends on the color precision, but is limited by factors such as printing process, equipment and materials, and the color of the printed product often has errors. In addition, the printing production process has the characteristics of less detection data, quick adjustment requirement and difficult transfer function, and the non-real-time and large-lag production system is very easily influenced by the environment so as to generate color difference. Aiming at color difference correction in a printing system, a method and a device (CN201810497360.5) for correcting the digital proofing color before printing the printing color of the environmental-friendly paper are provided, the steps of the method and the device for correcting the digital proofing color before printing the printing color of the environmental-friendly paper comprise: printing a first printing sample according to the original image, and comparing and adjusting the actually printed first printing sample with the original image to generate a first input image different from the original image, so that a second printing sample generated according to the first input image is closer to the original image than the original first printing sample. However, the color correction method related to the method and the device is realized by a linear method such as difference matching, and does not relate to a machine learning method such as a neural network model, and the color correction precision is difficult to guarantee.
In addition, in a method (CN201710546392.5) for implementing digital printing color correction by modifying a profile file, a method for implementing digital printing color correction by modifying a profile file is disclosed, which prints a corresponding characterization color patch image according to a source profile file of a digital file to be printed, measures a Lab value of each color patch, finds out a Lab value of a partial color patch having a smaller color difference with the color patch from the printed color patch, finds out a Lab value of a corresponding color patch in the source profile file, establishes a mapping relationship by using a polynomial least squares regression method, calculates a correction value of the corresponding color patch in the source profile file by using a target Lab value as an argument, and prints by using the modified profile file instead of the initial source profile file, so as to improve the accuracy of the printing color. However, in the method, a corresponding mapping relation needs to be established for each color block, and each color block also needs to be corrected one by one in the implementation process, so that the process is complicated and the efficiency is low.
In view of the above problems, the invention provides an intelligent correction method for multi-color and chromatic aberration based on a color printing production system, which utilizes the technology of combining an experience database and a BP neural network to replace experienced operators, reduces the times of repeated adjustment and plate making as much as possible, and greatly reduces the production cost of the color printing production system.
Disclosure of Invention
The invention aims to compensate chromatic aberration in the existing color printing production process, and provides a multicolor chromatic aberration intelligent correction method based on a color printing production system, the technology can compensate inherent chromatic aberration by automatically giving a design drawing through an experience database and a neural network, and the working principle is as follows: and respectively taking the printing sample in the experience database and the pixel data of the design drawing as the training input and output of the neural network, so that the neural network can carry out deep learning and remember the chromatic aberration in the printing process. When the design drawing is predicted by the trained neural network, the design drawing is taken as an ideal printing sample, so that the prediction result of the neural network is a new design drawing capable of offsetting the influence of the inherent chromatic aberration.
The invention is realized by at least one of the following technical schemes.
The intelligent correction method of multicolor aberration based on the color printing production system comprises the following steps:
selecting optimal system parameters by using data in an experience database according to customer requirements and product types, and performing trial production on a printing sample;
collecting pixel point data of a printing sample in real time, and correspondingly storing the customer requirements, the product type, the optimal system parameters and the design image pixel point data with the pixel point data of the printing sample;
thirdly, performing dimension reduction processing and color difference analysis on pixel point data in the experience database;
fourthly, the neural network model is utilized to carry out intelligent chromatic aberration correction and printing chromatic aberration prediction on the design drawing;
and fifthly, printing samples according to the corrected design drawing design template, simultaneously acquiring the weighted color difference between the sample data and the printed samples in real time, and if the weighted color difference meets the production conditions, performing formal printing production.
Preferably, in the step (i), the optimal system parameters include process parameters, control model parameters and equipment parameters;
the process parameters comprise printing area, printing thickness and printing grid density;
the control model parameters comprise an upper and lower allowable color difference limit, a weight-based color difference model weight and neural network model parameters;
the equipment parameters include printing speed, print cartridge type, and print ink density.
Preferably, the color difference analysis is to calculate the color difference generated by the primary printing sample by using a weighted color difference model, and the maximum value Δ E of the color difference is set max Comparing, and when the color difference is not less than the maximum value delta E of the color difference max Stopping to collect data and print; otherwise, entering a chromatic aberration correction link.
Preferably, the color difference analysis is performed by a weighted color difference model, wherein the weighted color difference model is as follows:
ΔE total ＝ω 1 ΔE color1 +ω 2 ΔE color2 +...+ω n ΔE colorn (3)
in the formula, delta E represents the distance between two pixel points, namely the color difference, delta L, delta a and delta b respectively represent the distance between the components of the two pixel points L, a and b, and delta E color RepresentTotal color difference of a certain color, m represents total number of pixel points of a certain color, and Δ E i The color difference of the ith pixel point is represented,
means, Δ E, of the mean value of the color difference of the color total Representing the total color difference, ω, of the entire picture n Represents the weight, Delta E, of a certain color difference in the color difference model colorn Representing the total color difference of the nth color.
Preferably, the step III is to decide the optimal neural network parameters according to the experience database, to perform dimension reduction and normalization processing on the pixel point data of the design drawing and the printing sample, to input the pixel point data into the neural network with the set optimal parameters for deep learning, to perform intelligent color difference correction and printing color difference prediction on the design drawing, and to perform intelligent color difference correction and printing color difference prediction on the design drawing when the predicted printing color difference is not greater than the minimum color difference delta E min Outputting the corrected design drawing to perform plate making and printing production; otherwise, executing step (iv) circularly.
Preferably, the normalization processing formula is as follows:
wherein x is data before normalization, y is data after normalization, and x max 、y max And x min 、y min Respectively corresponding to the maximum value and the minimum value in the data set;
preferably, in the step (iv), the specific step of the neural network for implementing chromatic aberration correction includes:
the optimal parameters of the neural network are decided, the neural network is set, the pixel point empirical data of the printing sample after the data dimension reduction and normalization processing are used as the input value of the neural network, the image pixel point empirical data are designed as the output value of the neural network, the neural network is enabled to carry out deep training learning, when the design drawing is input into the trained neural network, the neural network automatically carries out color compensation on the design drawing so as to counteract the inherent error possibly generated in the printing production process, and a new design drawing is output.
Preferably, in the step (iv), the specific step of the neural network implementing the color difference prediction function includes:
deciding the optimal parameters of the neural network, setting the neural network, taking the pixel point experience data of the design image after data processing as the input value of the neural network, taking the pixel point experience data of the printing sample as the output value of the neural network, leading the neural network to carry out deep training learning, leading the neural network to automatically carry out printing prediction on the design image when the design image after chromatic aberration correction is input into the trained neural network, predicting the printing chromatic aberration through a weight-sharing chromatic aberration model, and leading the neural network to predict the printing chromatic aberration not to be more than Delta E when the predicted printing chromatic aberration is not more than min Outputting the design drawing after chromatic aberration correction and printing; otherwise, circularly executing the chromatic aberration correction and the chromatic aberration prediction.
Preferably, the production conditions are the weighted color difference of the printed sample and the minimum value of the color difference Delta E min Comparing, and when the weighted color difference is not more than delta E min Then, formal printing production is carried out; otherwise, the system circularly executes the third step-the fifth step.
The system for realizing the intelligent correction method of the multicolor chromatic aberration based on the multicolor printing production system comprises a data acquisition module, an analysis decision module and a printing production module, wherein:
the data acquisition module comprises data acquisition equipment and an experience database, wherein the data acquisition equipment scans and prints production samples in real time and sends the production samples to the experience database in the workstation, so that the pixel point data of the printing samples are acquired and stored on line;
the analysis decision module decides out optimal system parameters by using data in the experience database and sends the optimal system parameters to the workstation and the printing production module, and meanwhile, the workstation predicts the actual printing effect of the design drawing through dimension reduction and normalization data processing, color difference calculation and analysis and a neural network and compensates the printing color difference;
the printing production module comprises a plate making machine and a printing machine, equipment parameters are set according to the optimal system parameters given by the analysis decision module, the design drawing subjected to color difference compensation is subjected to plate making and printing to obtain a sample, and meanwhile pixel point data of the printed sample is monitored in real time and fed back to the workstation.
It is emphasized that the emphasis of chromatic aberration correction and chromatic aberration prediction is on the continuous learning of empirical production data through neural networks to remember the inherent printing chromatic aberration generated during printing. In actual processing, due to the limitation of factors such as environment and equipment, inherent printing chromatic aberration often fluctuates to a certain extent, so that when the neural network is trained by using empirical data, the required chromatic accuracy requirement may not be met at one time, but when the empirical database is large enough, the required chromatic aberration correction and chromatic aberration prediction cycle times are less and less. In addition, the present invention is to reduce the time and cost for repeating plate making and manual adjustment to ensure the processing efficiency.
Compared with the prior art, the invention has the following beneficial effects:
1. compared with the traditional printing production system, the invention effectively reduces the chromatic aberration generated in the process of color printing, reduces the times of repeated plate making and manual adjustment when a printing enterprise corrects the chromatic aberration, and further saves a large amount of manpower, material resources and time cost.
2. By monitoring the printing sample in real time, comparing the printing sample with an ideal printing effect and setting a color difference threshold value, unqualified printing products can be screened, and the quality of factory products is improved.
3. Through the experience database, the calling and the analysis of past printing production data can be realized, and enterprise management and maintenance of printing equipment are facilitated.
Drawings
FIG. 1 is a schematic diagram of an intelligent correction method for multi-color and chromatic aberration of a color printing production system;
FIG. 2 is a flow chart of a multi-color aberration intelligent correction method for a color printing production system;
FIG. 3 is a data structure diagram of an experience database;
FIG. 4a is a design diagram of an embodiment of an intelligent correction method for multi-color and color difference in a color printing production system;
FIG. 4b is a scanned image of the color printing system before correction according to the embodiment of the intelligent correction method for multi-color aberration;
FIG. 4c is a scanned image of the color printing system after correcting one embodiment of the intelligent correction method for multi-color aberration;
FIG. 4d is a scanned image of the color printing system after correcting twice according to the embodiment of the intelligent correction method for multi-color aberration;
FIG. 5a is a scanned image of a multi-color aberration intelligent correction method printed sample of a color printing production system;
FIG. 5b is a diagram of the effect of the multicolor design drawing of the intelligent correction method for multicolor and color difference in the color printing production system.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited thereto.
The intelligent correction of multicolor block image colors in fig. 5a and 5b is taken as an example to explain the working principle of the intelligent correction method of multicolor color difference based on the color printing production system in detail, so as to verify the technical effect of the invention.
As shown in fig. 1, the system of the intelligent correction method for multi-color and color difference of a color printing production system includes a data acquisition module, an analysis decision module and a printing production module, and specifically includes the following steps:
the data acquisition module comprises an experience database and data acquisition equipment, wherein the original data in the experience database is historical reserved data of the printing production system, the historical reserved data can be updated in real time in the running process of the system subsequently to achieve higher chromatic aberration correction precision, and the data acquisition equipment (such as a CCD scanner) acquires the data of a printing sample in real time according to set instrument parameters and continuously updates and optimizes the experience database.
The analysis decision module comprises a workstation and a neural network model, wherein the workstation extracts a design drawing and corresponding sample data from an experience database to perform chromatic aberration calculation, and when chromatic aberration correction is determined to be needed, required original data can be extracted and input into the neural network model to perform deep learning, so that the neural network remembers the inherent error of the printing production system, the color data of the original design drawing is corrected, the final printing chromatic aberration is reduced, and the intelligent reconstruction of the printing production system is realized.
The neural network model can be various neural network models such as a BP neural network, an RBF neural network, an SAE neural network and a CNN neural network.
The printing production module comprises a plate making machine and a printing machine, the plate making machine is communicated with a workstation to obtain a new design drawing after intelligent correction of the neural network model, the design drawing in the original RGB format is converted into a CMYK format special for printing through image processing software such as CorelDRAW, Photoshop and the like, a printing plate is output to the printing machine, equipment parameters are set according to optimal system parameters given by the analysis decision module, the printing machine carries out production printing according to preset parameters such as ink density, printing speed and the like, a sample is obtained by plate making and printing of the design drawing after chromatic aberration compensation, and meanwhile pixel point data of the printed sample is monitored in real time and fed back to the workstation.
As shown in fig. 1 and fig. 2, an intelligent correction method for multi-color and chromatic aberration based on a color printing production system includes the following steps:
selecting system parameters: before color printing, an analysis decision module decides optimal system parameters according to customer requirements and an experience database, sets process and equipment parameters (printing area, printing thickness, printing grid density, printing speed, printing color box type, printing ink density and the like) of instruments such as a plate making machine, a printing machine, data acquisition equipment and the like in a plate making printing and data acquisition module, and sets control model parameters (upper and lower limits of allowed chromatic aberration, activation functions and neuron numbers of each layer of a neural network, weight, threshold, maximum iteration times, learning efficiency and the like) in the analysis decision module;
different customer requirements and product types can directly influence the target precision of printed products, the related printing process and printing equipment can be different, the required process parameters and equipment parameters can be changed, in addition, due to the fact that the process and the equipment are different, the inherent printing errors in the printing production process can also be changed, the control model parameters for error compensation can also be changed, and therefore the analysis decision module needs to select the printing production system parameters from the experience database in a preferred mode according to different customer requirements and product types.
Collecting and storing data: using high-precision image data acquisition equipment to acquire Lab data of a printing sample in real time, dividing the Lab data of the printing sample and the Lab data of a corresponding design drawing into three two-dimensional matrix data of L, a and b, and storing the three two-dimensional matrix data in an experience database, wherein the design drawing is required to be ensured to be in one-to-one correspondence with each pixel point data of the printing sample;
calculating and analyzing chromatic aberration: classifying pixel point data of a design drawing and a printing sample according to basic colors (red, yellow, green, blue, purple, black and white), respectively calculating color differences of different colors according to a weight-division color difference model, setting different color difference weights according to the stimulation degree of different colors to human eyes, calculating color difference delta E, comparing the color difference delta E with a preset color difference threshold value, and when delta E is more than or equal to delta E max When the system is abnormal in operation, the system gives an alarm and stops operating; when Δ E min ＜ΔE＜ΔE max When the total chromatic aberration is within the correctable range, the total chromatic aberration can enter an analysis decision module for intelligent chromatic aberration correction; when Delta E is less than or equal to Delta E min In the process, the total chromatic aberration does not influence the viewing of human eyes, chromatic aberration correction is not needed, and plate making and printing can be directly carried out; delta E max 、ΔE min Respectively representing maximum and minimum values of chromatic aberration, and minimum value of chromatic aberration Delta E min The analysis and decision module decides an optimal value according to the product type and the customer demand, and the value is delta E min At least to the extent that it cannot be easily recognized by the human eye.
The chromatic aberration is defined as the distance between two pixel points in a color gamut space, for a printed product, the most important is to meet the aesthetic requirements of customers, because the stimulation degrees of different colors to human eyes are different, in order to enable the calculated chromatic aberration to be closer to the chromatic aberration intuitively felt by the human eyes, a weight-divided chromatic aberration model is adopted, the chromatic aberration is respectively calculated for each color, and different weights are given according to the stimulation degrees of the color to the human eyes. Therefore, the weights of the color differences of different colors in the weighted color difference model are different, taking the Lab color gamut space as an example, the weighted color difference model is as follows:
ΔE total ＝ω 1 ΔE color1 +ω 2 ΔE color2 +...+ω n ΔE colorn (3)
in the formula, delta E represents the distance between two pixel points, namely the color difference, delta L, delta a and delta b respectively represent the distance between the components of the two pixel points L, a and b, and delta E color Representing the total color difference of a certain color, m representing the total number of pixels of a certain color, Δ E i The color difference of the ith pixel point is represented,
means, Δ E, of the mean value of the color difference of the color total Representing the total color difference, ω, of the entire picture n Represents the weight of a certain color difference in the color difference model, for example: the weight of the red color difference is 1, the weight of the green color difference is 0.3, and the weights of the other colors are assigned according to a similar rule; delta E colorn Representing the total color difference of the nth color.
Fourthly, intelligent chromatic aberration correction: after the data in the experience database is taken for dimension reduction and normalization, the data is input into a neural network with set parameters for deep learning so as to achieve the functions of intelligently correcting the color of the original design drawing and predicting the printing effect of a new design drawing, and when the color difference Delta E of the predicted printing effect is less than or equal to Delta E min When the process is carried out, the process of the step two to the step four can be repeated, and when delta E is larger than delta E, the process can be output min And repeating the intelligent chromatic aberration correction link until the predicted printing effect meets the target requirement.
The pixel point data of the design drawing and the printing sample are three-dimensional matrix data, can not be directly input into a neural network for training and learning, needs to be expanded into a two-dimensional matrix with L, a and b as column components through dimension reduction, and is subjected to normalization after dimension reduction, wherein the normalization processing formula is as follows:
wherein x is data before normalization, y is data after normalization, and x max 、y max And x min 、y min Respectively corresponding to the maximum value and the minimum value in the data set;
it should be noted that, the training data set is obtained by printing and accumulating the standard color correction card for multiple times, and the number of neuron points in the hidden layer of the BP neural network model is obtained according to an empirical formula:
h is the number of hidden layer neurons, m is the number of input layer neurons, n is the number of output layer neurons, a is an adjustment constant within 1-10, through data analysis in the existing experience database, the input and output layer neurons of the neural network model are set to be 3, the hidden layer neurons are set to be 9, the learning efficiency is 0.1, the error precision is 0.01, and the maximum iteration number is 100.
It should be noted that, when the color composition of the printed product changes, the weight values of the weighted color difference model and the upper and lower limits of the allowable color difference may also change, but the overall structural form does not change. Through the analysis of the color stimulation degree, the weight of each color difference in the calculation model is set as: red: 1.0, yellow: 0.8, black: 0.7, purple: 0.6, blue: 0.5, green: 0.3, white: 0.2, etc., and furthermore, the color difference limit value delta E is adjusted according to the actual requirements of enterprises max Set to 100, color difference threshold Δ E min Is set to 30.
The printing production module designs a template and prints samples according to the design drawing intelligently corrected by the analysis decision module, and the data acquisition module acquires data in real timeCollecting sample data and feeding the data back to an analysis decision module, and calculating the weighted color difference of the printing sample and the minimum value delta E of the color difference by the analysis decision module min Comparing, and when the weighted color difference is not more than delta E min The printing production module can carry out formal printing production; otherwise, the system circularly executes the third step-the fifth step.
The invention utilizes the neural network model, the data dimension reduction, the normalization processing and the like to fully utilize mass production experience data of the printing production line, realizes the functions of intelligent chromatic aberration correction, printing effect prediction and the like of the printing products, and improves the phenomenon that the existing printing production line excessively depends on manual operation when correcting the printing chromatic aberration.
It should be emphasized that the premise for realizing the intelligent correction of the multicolor and the chromatic aberration of the color printing production system is to solve the correlation problem between the equipment and between the equipment and the PC, which is not to simply stack the existing software and hardware, and needs to cross a certain technical barrier (such as compatibility between different types of equipment, data interaction between the equipment and the PC, and the like) to integrate the existing software and hardware. In addition, the weight-sharing color difference model and the neural network model are established by analyzing inherent errors accumulated by various errors such as identification, color gamut conversion, printing and the like generated in the whole process from design to production of the printing production system, and the intelligent decision process of the process, equipment and control model parameters is designed based on customer requirements and product types and needs to establish a corresponding experience database and perform preferential selection.
The experience database of this embodiment is shown in fig. 3. The experience database is mainly divided into three categories of evaluation indexes, target values and process parameters. Wherein, the evaluation index comprises: product type (M), customer requirements (N), optimal system process parameters (a) op ) Upper and lower limits of allowable chromatic aberration (Δ E) max(min) ) Color difference model weight (omega) and BP neural network model parameter (K); the target value should be set according to the product type (M), including: print quality grade (I), print thickness (h), print area(s) c ) (ii) a The process parameters are set according to the product type (M) and the customer requirement (N), and comprise: process parameters (A), control model parameters (P) and equipment parameters (a).
The following describes how to achieve intelligent correction of colors by using an empirical database and a neural network algorithm in the intelligent correction method for multi-color and chromatic aberration based on a color printing production system according to a further embodiment.
An experience database is built in a workstation, and a data acquisition module, an analysis decision module and a printing production module of the multicolor color difference intelligent correction method based on the color printing production system are built through communication among devices such as a CCD scanner (MICROTEK Phantom 9900XL) -workstation, a workstation-CTP platemaker (FUSHENG 800), a CTP platemaker-printer (LS 640) and the like.
Before intelligent correction of colors, selecting optimal system parameters according to past production data in an experience database, setting main working parameters of equipment such as a CCD scanner, a plate making machine, a printing machine and the like, inputting relevant parameters of a neural network model into an analysis decision module, and carrying out chromatic aberration correction and printing effect prediction on an original design drawing through a neural network by the analysis decision module according to the set system parameters; in addition, weights of different color differences in the weighted color difference model are set according to the stimulation degree of different colors to human eyes, the total color difference model is used for predicting the printing color difference delta E of the corrected design drawing, and when delta E is less than or equal to delta E min (color difference threshold), outputting the design drawing to print the sample, sending the printed sample into the CCD scanner to scan to obtain the actual printed image data, calculating the actual color difference by using the total color difference model again, and when the delta E is more than or equal to the delta E max In time, a printing production system may have mechanical faults, and the system automatically alarms; when Δ E min ＜ΔE＜ΔE max When the printed sample does not reach the set chromatic aberration threshold value, the chromatic aberration correction and the printed sample prediction step need to be repeated; when Delta E is less than or equal to Delta E min And (4) the printing sample meets the production condition, and formal production printing can be performed.
In order to verify the chromatic aberration correction effect of the intelligent chromatic aberration correction method for color printing in a color printing production system, taking an original design drawing (as shown in fig. 4a) as an example, a scanning drawing (as shown in fig. 4b) is obtained by scanning a printing sample which is not subjected to correction processing, and an initial chromatic aberration delta E is calculated by using a chromatic aberration calculation model and is 47.6989.
The analysis and decision module uses a neural network and an experience database to correct the chromatic aberration of the original design drawing according to the set system parameters, predicts the printing chromatic aberration delta E through a chromatic aberration calculation model, and compares the predicted printing chromatic aberration delta E with a preset chromatic aberration limit value delta E max And a color difference threshold Δ E min And (6) comparing and analyzing. Since the initial color difference is less than delta E max So that the printing production system does not generate mechanical failure, but Δ E > Δ E min Therefore, chromatic aberration correction is required.
The analysis decision module performs chromatic aberration correction and printing effect prediction on the original design drawing through a neural network and an experience database with set parameters, the printing effect prediction after primary chromatic aberration correction is shown in figure 4c, the calculated chromatic aberration value is 40.0064, the condition of exiting the analysis decision module is not met, the printing effect prediction after secondary chromatic aberration correction is shown in figure 4d, the calculated chromatic aberration value is 25.1058, the condition of being smaller than the chromatic aberration threshold is met, printing sample production can be performed, the obtained original drawing and the scanning drawing of the printing sample are respectively shown in figures 4a and 4b, the actual chromatic aberration is 26.6255 < 30 through the calculation of a chromatic aberration model, and the production requirements of enterprises are met.
In conclusion, the color difference of the printed product can reach the set threshold value through the intelligent prediction and correction functions of the experience database and the neural network, the whole process does not need human intervention, the labor cost, the material cost and the time cost caused by repeated plate making and human adjustment are effectively avoided, and the production quality and the efficiency of the printing system are greatly improved.
The present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents and are included in the scope of the present invention.
Claims (8)
1. The intelligent correction method of the multicolor chromatic aberration based on the color printing production system is characterized by comprising the following steps:
selecting optimal system parameters by using data in an experience database according to customer requirements and product types, and performing trial production on a printing sample;
collecting pixel point data of a printing sample in real time, and correspondingly storing the pixel point data of the printing sample, the customer requirements, the product type, the optimal system parameters and the design image;
thirdly, performing dimensionality reduction processing and color difference analysis by using pixel point data in the experience database;
fourthly, the neural network model is utilized to carry out intelligent chromatic aberration correction and printing chromatic aberration prediction on the design drawing; the specific steps of utilizing the neural network model to realize chromatic aberration correction comprise:
deciding the optimal parameters of a neural network model, setting a neural network, taking pixel point empirical data of a printing sample after data dimension reduction and normalization processing as an input value of the neural network, designing image pixel point empirical data as an output value of the neural network, enabling the neural network to carry out deep training learning, and when a design drawing is input into the trained neural network, automatically carrying out color compensation on the design drawing by the neural network so as to counteract inherent errors possibly generated in the printing production process and output a new design drawing;
the specific steps of realizing the color difference prediction function by utilizing the neural network model comprise:
deciding the optimal parameters of the neural network, setting the neural network, taking the pixel point experience data of the design image after data processing as the input value of the neural network, taking the pixel point experience data of the printing sample as the output value of the neural network, leading the neural network to carry out deep training learning, leading the neural network to automatically carry out printing prediction on the design image when the design image after chromatic aberration correction is input into the trained neural network, predicting the printing chromatic aberration through a weight-sharing chromatic aberration model, and when the predicted weight-sharing chromatic aberration is not more than Delta E min Outputting the design drawing after chromatic aberration correction and printing; otherwise, circularly executing chromatic aberration correction and chromatic aberration prediction circulation;
and fifthly, printing samples according to the corrected design drawing design template, simultaneously collecting sample data and the weighted color difference of the printed samples in real time, if the weighted color difference is smaller than a set color difference threshold value, performing formal printing production, and otherwise, performing the third step-the fifth step in a system circulation mode.
2. The intelligent correction method for multicolor color difference based on the color printing production system as claimed in claim 1, characterized in that in step I, the optimal system parameters comprise process parameters, control model parameters and equipment parameters;
the process parameters comprise printing area, printing thickness and printing grid density;
the control model parameters comprise an upper and lower allowable color difference limit, a weight of a weight color difference model and neural network model parameters;
the equipment parameters include printing speed, print cartridge type, and print ink density.
3. The intelligent correction method for multicolor color difference based on color printing production system as claimed in claim 2, wherein the color difference analysis is performed by calculating the color difference generated by the primary printing sample by using a weighted color difference model, and comparing the calculated color difference with a set maximum value of color difference Δ E max Comparing, and when the weighted color difference is not less than the maximum value of the color difference Delta E max Stopping to collect data and print; otherwise, entering a chromatic aberration correction link.
4. The intelligent correction method for multicolor chromatic aberration based on color printing production system according to claim 3, characterized in that chromatic aberration analysis is performed by a weighted chromatic aberration model, the weighted chromatic aberration model is:
ΔE total ＝ω 1 ΔE color1 +ω 2 ΔE color2 +...+ω n ΔE colorn (3)
in the formula, delta E represents the distance between two pixel points, namely the color difference, delta L, delta a and delta b respectively represent the distance between the components of the two pixel points L, a and b, and delta E color Representing the total color difference of a certain color, m representing the total number of pixels of a certain color, Δ E i The color difference of the ith pixel point is represented,
the average value of the color difference, Delta E, of the color total Representing the total color difference, ω, of the entire picture n Represents the weight, Delta E, of a certain color difference in the color difference model colorn Representing the total color difference of the nth color.
5. The intelligent correction method of multicolor color difference based on color printing production system as claimed in claim 4, wherein the step three is to decide the optimal neural network parameters according to the experience database, to perform dimension reduction and normalization on the pixel point data of the design drawing and the printing sample, to input the neural network with the set optimal parameters for deep learning, and to perform intelligent color difference correction and printing color difference prediction on the design drawing, when the predicted weighted color difference is not greater than the weighted color difference minimum value Δ E min Outputting the corrected design drawing to perform plate making and printing production; otherwise, executing step (iv) circularly.
6. The intelligent correction method for multicolor aberration based on color printing production system according to claim 5, wherein the normalization processing formula is as follows:
wherein x is data before normalization, y is data after normalization, and x max 、y max And x min 、y min Respectively corresponding to the maximum value and the minimum value in the data set。
7. The intelligent correction method for multicolor color difference based on color printing production system according to any of claims 1 to 6, wherein if the weighted color difference is smaller than the set color difference threshold, performing formal printing production includes: dividing the weighted color difference of the printing sample into the minimum value delta E of the color difference min Comparing, when the weighted color difference is less than the minimum value delta E of the color difference min Then, formal printing production is carried out.
8. The system for realizing the intelligent multi-color and color difference correction method based on the color printing production system is characterized by comprising a data acquisition module, an analysis decision module and a printing production module, wherein:
the data acquisition module comprises data acquisition equipment and an experience database, wherein the data acquisition equipment scans and prints production samples in real time and sends the production samples to the experience database in the workstation to realize online acquisition and storage of pixel point data of the printing samples;
the analysis decision module decides out optimal system parameters by using data in the experience database and sends the optimal system parameters to the workstation and the printing production module, and meanwhile, the workstation predicts the actual printing effect of the design drawing through dimension reduction and normalization data processing, color difference calculation and analysis and a neural network and compensates the printing color difference;
the printing production module comprises a plate making machine and a printing machine, equipment parameters are set according to the optimal system parameters given by the analysis decision module, the design drawing subjected to color difference compensation is subjected to plate making and printing to obtain a sample, and meanwhile pixel point data of the printed sample is monitored in real time and fed back to the workstation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110502558.XA CN113411466B (en) | 2021-05-08 | 2021-05-08 | Multicolor chromatic aberration intelligent correction method and system based on color printing production system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110502558.XA CN113411466B (en) | 2021-05-08 | 2021-05-08 | Multicolor chromatic aberration intelligent correction method and system based on color printing production system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113411466A CN113411466A (en) | 2021-09-17 |
| CN113411466B true CN113411466B (en) | 2022-09-20 |
Family
ID=77678123
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110502558.XA Active CN113411466B (en) | 2021-05-08 | 2021-05-08 | Multicolor chromatic aberration intelligent correction method and system based on color printing production system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113411466B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114638786B (en) * | 2022-02-21 | 2022-12-13 | 杭州印鸽科技有限公司 | Picture printing effect simulation system and color matching method for different processes |
| CN114633565B (en) * | 2022-03-31 | 2023-06-30 | 浙江普崎数码科技有限公司 | Automatic compensation method for density of inkjet printing nozzle |
| CN115179667B (en) * | 2022-06-14 | 2023-10-31 | 深圳达捷科技有限公司 | Automatic control method and system for spray printing of multi-product production line |
| CN114757954B (en) * | 2022-06-16 | 2022-08-26 | 南通人民彩印有限公司 | Carton printing color difference defect detection method based on artificial intelligence system |
| CN116843770B (en) * | 2023-07-19 | 2024-01-09 | 广东兴艺数字印刷股份有限公司 | Color detection method, device and equipment for printed image and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5162899A (en) * | 1989-03-15 | 1992-11-10 | Matsushita Electric Industrial Co., Ltd. | Color data correction apparatus ultilizing neural network |
| US5189511A (en) * | 1990-03-19 | 1993-02-23 | Eastman Kodak Company | Method and apparatus for improving the color rendition of hardcopy images from electronic cameras |
| US5774230A (en) * | 1995-06-21 | 1998-06-30 | Sharp Kabushiki Kaisha | Color image processing apparatus for color-correcting input image data using neural network |
| US5929906A (en) * | 1995-12-07 | 1999-07-27 | Shiro Usui | Color correcting method and apparatus |
| US6100999A (en) * | 1997-01-14 | 2000-08-08 | Fuji Xerox Co., Ltd. | Method for predicting color transfer characteristics |
| WO2003096675A1 (en) * | 2002-05-14 | 2003-11-20 | Fuji Xerox Co., Ltd. | Color image creating device and color image creating method |
| JP2010232842A (en) * | 2009-03-26 | 2010-10-14 | Fuji Xerox Co Ltd | Color processing apparatus, and color processing program |
| CN107395919A (en) * | 2017-07-06 | 2017-11-24 | 杭州电子科技大学 | A kind of method for realizing digital printing color correction by changing characteristic file |
| CN108819492A (en) * | 2018-05-22 | 2018-11-16 | 中煤航测遥感集团有限公司 | Digital sample printing color correcting method and device before environment protecting paper process color color printing |
| CN109992928A (en) * | 2019-04-30 | 2019-07-09 | 南京林业大学 | A method of establishing china picture-character paper output quality prediction model |
| CN110475043A (en) * | 2019-07-31 | 2019-11-19 | 西安工程大学 | A kind of conversion method of CMYK to Lab color space |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7777916B2 (en) * | 2006-04-13 | 2010-08-17 | E. I. Du Pont De Nemours And Company | Method for producing a table of predicted reflectances under target operating conditions and data structure and printing system incorporating the table |
| CN103873736B (en) * | 2012-12-17 | 2017-03-22 | 北大方正集团有限公司 | Color closed-loop correction method and system based on multi-dimensional table |
| CN103152511B (en) * | 2013-01-17 | 2015-03-25 | 武汉大学 | Digital image printing color correction method for printing publishing |
| CN103870689B (en) * | 2014-03-11 | 2016-08-31 | 武汉大学 | A kind of printing print system Forecast of Spectra method |
| JP2019220828A (en) * | 2018-06-19 | 2019-12-26 | 株式会社リコー | Image processing apparatus, image processing method, and program |
| US11023791B2 (en) * | 2019-10-30 | 2021-06-01 | Kyocera Document Solutions Inc. | Color conversion using neural networks |
-
2021
- 2021-05-08 CN CN202110502558.XA patent/CN113411466B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5162899A (en) * | 1989-03-15 | 1992-11-10 | Matsushita Electric Industrial Co., Ltd. | Color data correction apparatus ultilizing neural network |
| US5189511A (en) * | 1990-03-19 | 1993-02-23 | Eastman Kodak Company | Method and apparatus for improving the color rendition of hardcopy images from electronic cameras |
| US5774230A (en) * | 1995-06-21 | 1998-06-30 | Sharp Kabushiki Kaisha | Color image processing apparatus for color-correcting input image data using neural network |
| US5929906A (en) * | 1995-12-07 | 1999-07-27 | Shiro Usui | Color correcting method and apparatus |
| US6100999A (en) * | 1997-01-14 | 2000-08-08 | Fuji Xerox Co., Ltd. | Method for predicting color transfer characteristics |
| WO2003096675A1 (en) * | 2002-05-14 | 2003-11-20 | Fuji Xerox Co., Ltd. | Color image creating device and color image creating method |
| JP2010232842A (en) * | 2009-03-26 | 2010-10-14 | Fuji Xerox Co Ltd | Color processing apparatus, and color processing program |
| CN107395919A (en) * | 2017-07-06 | 2017-11-24 | 杭州电子科技大学 | A kind of method for realizing digital printing color correction by changing characteristic file |
| CN108819492A (en) * | 2018-05-22 | 2018-11-16 | 中煤航测遥感集团有限公司 | Digital sample printing color correcting method and device before environment protecting paper process color color printing |
| CN109992928A (en) * | 2019-04-30 | 2019-07-09 | 南京林业大学 | A method of establishing china picture-character paper output quality prediction model |
| CN110475043A (en) * | 2019-07-31 | 2019-11-19 | 西安工程大学 | A kind of conversion method of CMYK to Lab color space |
Non-Patent Citations (3)
| Title |
|---|
| High quality color correction method combining neural networks with genetic algorithms;Tatsumi Watanabe et al.;《Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205)》;20011010;553-556 * |
| 喷墨打印纸印刷适性预测模型;洪亮 等;《包装工程》;20160831;第37卷(第5期);194-198 * |
| 基于RBF神经网络的曲面加工误差补偿;邹明山 等;《新技术新工艺》;20090228;44-47 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113411466A (en) | 2021-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113411466B (en) | Multicolor chromatic aberration intelligent correction method and system based on color printing production system | |
| JP4595734B2 (en) | Profile creation method, profile creation device, profile creation program, print control method, print control device, and print control program | |
| US8693050B2 (en) | Controlling process color in a color adjustment system | |
| JP5310298B2 (en) | Image processing apparatus, image forming system, and program | |
| US7307755B2 (en) | Method for correcting unadjusted printing data on the basis of colormetrically measured reference sheet | |
| EP0479312B1 (en) | Method of correcting setup parameter decision characteristics and automatic setup apparatus | |
| JP4235837B2 (en) | Graininess prediction, profile creation and printing device | |
| CN106464775B (en) | Color model | |
| US20070242293A1 (en) | Method for creating a color transform relating color reflectances produced under reference and target operating conditions and data structure incorporating the same | |
| US20050270600A1 (en) | Increasing profile accuracy and accelerating profile creation | |
| CN101325648B (en) | Color processing apparatus, color processing method | |
| JP6349741B2 (en) | Look-up table generation method and color conversion device | |
| CN112537148A (en) | Method for matching colors of digital printing and flexography | |
| US8335017B1 (en) | Spot color rendering via feedback-based multi-objective optimization | |
| JP2011077975A (en) | Granularity predicting device, method of predicting granularity, method of creating profile, and printer | |
| JP2003298862A (en) | Method and device for color matching | |
| JP4275716B2 (en) | Color tone monitoring apparatus, color tone management system, color tone monitoring method, and color tone management method | |
| EP1433605B1 (en) | Method and apparatus for controlling the Ink feeding rate | |
| US20070242296A1 (en) | Method for producing a table of predicted reflectances under target operating conditions and data structure and pringting system incorporating the table | |
| US20220131998A1 (en) | Color prediction | |
| US7782489B2 (en) | Method for selecting a sample set useful in relating color reflectances producible under reference and target operating conditions and the sample set producing thereby | |
| US20210227095A1 (en) | Modeling a printed halftone image | |
| US20090109452A1 (en) | Color conversion table generation apparatus and color conversion table generation method | |
| TWI777776B (en) | A method for predicting the color measurement value of printed matter by image | |
| CN101237517A (en) | Image processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |