A digital printing quality analysis system based on machine vision

Abstract

The present invention relates to the field of digital printing technology, and particularly relates to a digital printing quality analysis system based on machine vision, which includes constructing an inkjet concentration model based on the color features and historical inkjet concentrations in the sample images of historical digital printing, inputting the color features of the real-time image into the inkjet concentration model, and outputting the real-time inkjet concentration of each sub-region of the real-time image; judging whether the inkjet concentration in the sub-region is spatially uniform according to the real-time inkjet concentration of the sub-region, if not, identifying the degree of uneven distribution of the spatial inkjet concentration; if the degree of spatial inkjet non-uniformity is high, integrating the inkjet concentrations of all sub-regions according to the spatial position into a sub-region inkjet concentration empty sequence, analyzing the real-time inkjet concentration of the sub-regions in the inkjet concentration empty sequence, and determining the spatial inkjet power compensation value. The present invention can specifically adjust the inkjet power of the digital printing machine, improve the problem of uneven inkjet, and enhance the printing quality of digital printed items.

Description

Technical Field

[0001] The present invention relates to the technical field of digital printing, and particularly relates to a digital printing quality analysis system based on machine vision. Background Art

[0002] Digital printing has been widely used in the printing industry due to its advantages such as no need for plate making and the ability to achieve personalized printing. As a modern printing method, it has significant advantages in terms of printing efficiency and flexibility.

[0003] However, in the actual production process, various quality problems are likely to occur in digital printing. Among them, the influence of inkjet concentration on digital printing is particularly prominent. The existing technologies have a single detection dimension in detecting the inkjet concentration during the digital printing process, and thus cannot evaluate the digital printing quality. In the spatial dimension, the inkjet concentration distribution and uniformity of each sub-region cannot be accurately analyzed; in the time dimension, the stability change of the inkjet concentration over time is ignored, which leads to the inability to detect and solve in time the printing quality fluctuation problems caused by the change of the inkjet concentration.

[0004] Therefore, we propose a digital printing quality analysis system based on machine vision. Summary of the Invention

[0005] The purpose of the present invention is to provide a digital printing quality analysis system based on machine vision to solve at least one of the above-mentioned problems of the existing technologies.

[0006] The present invention provides a digital printing quality analysis system based on machine vision, including:

[0007] A data acquisition module: Based on the color features in the sample images of historical digital printing and the historical inkjet concentration, an inkjet concentration model is constructed. The color features of the real-time image are input into the inkjet concentration model, and the real-time inkjet concentration of each sub-region of the real-time image is output.

[0008] An inkjet uniformity judgment module: According to the real-time inkjet concentration of the sub-region, it is judged whether the inkjet concentration in the sub-region is spatially uniform. If it is not uniform, the degree of non-uniform distribution of the spatial inkjet concentration is identified.

[0009] A spatial dimension power compensation module: If the degree of spatial inkjet non-uniformity is high, the inkjet concentrations of all sub-regions are integrated into a sub-region inkjet concentration empty sequence according to the spatial position. The real-time inkjet concentration of the sub-regions in the inkjet concentration empty sequence is analyzed to determine the spatial inkjet power compensation value, and the inkjet power during the inkjet of the digital printer is regulated.

[0010] As a further solution of the present invention: it further includes:

[0011] Inkjet Stability Analysis Module: If the degree of non-uniformity of spatial inkjet is low, identify whether the inkjet concentration in the sub-region is stable during the analysis period;

[0012] Time Dimension Power Compensation Module: If it is unstable, determine the time inkjet power compensation value according to the inkjet concentration in the sub-region during the analysis period, and regulate the inkjet power when the digital printer performs inkjet.

[0013] As a further solution of the present invention: The process of obtaining the real-time inkjet concentration of the sub-region is as follows:

[0014] Obtain the color features of the sample image of historical digital printing and the historical inkjet concentration. Among them, the color features include saturation and brightness, and build a model;

[0015] Divide the collected real-time image into sub-regions, obtain all the saturation and brightness within the sub-region of the real-time image, and perform summation and averaging processing respectively to obtain the average saturation and average brightness of the current sub-region, and input them into the model to output the real-time inkjet concentration of the sub-region.

[0016] As a further solution of the present invention: The specific process of judging whether the inkjet concentration of the sub-region is spatially uniform is as follows:

[0017] Obtain the inkjet concentration of all sub-regions in the real-time image, calculate the inkjet uniformity judgment value through the variance processing formula. If the inkjet uniformity judgment value ≥ the inkjet uniformity judgment threshold, generate an inkjet non-uniform signal.

[0018] As a further solution of the present invention: The specific process of identifying the degree of non-uniform distribution of spatial inkjet concentration is as follows:

[0019] Analyze the inkjet concentration of the sub-regions in the sub-region inkjet concentration sequence to obtain the adjacent concentration change rate and the reference concentration change rate. Perform data processing on the adjacent concentration change rate and the reference concentration change rate to obtain the concentration change rate deviation value and the asynchronous concentration change rate;

[0020] Calculate the proportion of the number of asynchronous concentration change rates, extract the concentration change rate deviation values corresponding to all asynchronous concentration change rates, and perform summation and averaging to obtain the average deviation of asynchronous concentration change rates;

[0021] Multiply the average deviation of asynchronous concentration change rates by the proportion of the number of asynchronous concentration change rates to obtain the spatial concentration distribution value; If the spatial concentration distribution value < the spatial concentration distribution threshold, it indicates that the degree of non-uniformity of spatial inkjet is low; otherwise, it indicates that the degree of non-uniformity of spatial inkjet is high.

[0022] As a further solution of the present invention: The process of obtaining the concentration change rate deviation value and the asynchronous concentration change rate is as follows:

[0023] Take the absolute value after subtracting the adjacent concentration change rate from the reference concentration change rate to obtain the concentration change rate deviation value. If the concentration change rate deviation value ≥ the concentration change rate deviation threshold, record the adjacent concentration change rate corresponding to the concentration change rate deviation value as the asynchronous concentration change rate.

[0024] As a further solution of the present invention: the process of obtaining the reference concentration change rate is as follows:

[0025] Extract the inkjet concentration of the sub-region ranked first in order and the inkjet concentration of the sub-region ranked first in reverse order in the sub-region inkjet concentration empty sequence; and take the absolute value after subtraction to obtain the peak-valley inkjet concentration change value. Measure the distance between the sub-regions between the first in order and the first in reverse order to obtain the peak-valley distance difference; calculate the ratio of the peak-valley inkjet concentration change value to the peak-valley distance difference to obtain the reference concentration change rate.

[0026] As a further solution of the present invention: the process of obtaining the adjacent concentration change rate is as follows:

[0027] Extract the inkjet concentration of the sub-region in the previous sequence and the inkjet concentration of the sub-region in the next sequence in the sub-region inkjet concentration empty sequence; and take the absolute value after subtraction to obtain the adjacent inkjet concentration change value. Record the distance between the sub-regions between the previous sequence and the next sequence as the adjacent distance difference; calculate the ratio of the adjacent inkjet concentration change value to the adjacent distance difference to obtain the adjacent concentration change rate.

[0028] As a further solution of the present invention: the method of obtaining the spatial inkjet power compensation value is as follows:

[0029] Extract the inkjet concentration of the sub-region in the inkjet concentration empty sequence, subtract it from the standard value of the inkjet concentration of the sub-region, and then take the absolute value to obtain the spatial inkjet concentration deviation value. Construct a fitting relationship model between the inkjet power and the inkjet concentration, and calculate the spatial inkjet power compensation value when the digital printer jets ink according to the fitting relationship model.

[0030] As a further solution of the present invention: the process of identifying whether the inkjet concentration of the sub-region is stable during the analysis period is as follows;

[0031] Analyze the inkjet concentration at all time points during the sub-region analysis period to obtain the inkjet concentration stability value; if the inkjet concentration stability value ≥ the inkjet concentration stability threshold, it indicates that the inkjet concentration in the inkjet space fluctuates in the time dimension during the analysis period.

[0032] As a further solution of the present invention: the process of obtaining the inkjet concentration stability value is as follows:

[0033] Obtain the inkjet concentrations at all time points within the analysis period of the sub-region, integrate them into the time series of inkjet concentrations, extract all the inkjet concentrations in the time series of inkjet concentrations and calculate the variance to obtain the variance of the inkjet concentration for each sub-region.

[0034] Sum up and take the average of the variances of the inkjet concentrations of all sub-regions to obtain the inkjet concentration stability value.

[0035] As a further solution of the present invention: the process of obtaining the time inkjet power compensation value is as follows:

[0036] Extract the inkjet concentration of the sub-region in the time series of inkjet concentrations, subtract it from the standard value of the inkjet concentration of the sub-region, then take the absolute value to obtain the time inkjet concentration deviation value, input it into the fitting relationship model between the inkjet power and the inkjet concentration, and calculate the time inkjet power compensation value when the digital printer performs inkjetting.

[0037] Advantages of the present invention:

[0038] 1. Based on the color features in the sample images of historical digital printing and the historical inkjet concentrations, the present invention constructs an inkjet concentration model, inputs the color features of the real-time image into the inkjet concentration model, and outputs the real-time inkjet concentration of each sub-region of the real-time image; according to the real-time inkjet concentration of the sub-region, determine whether the inkjet concentration in the space of the sub-region is uniform. If it is not uniform, identify the degree of non-uniform distribution of the spatial inkjet concentration; if the degree of spatial inkjet non-uniformity is high, integrate the inkjet concentrations of all sub-regions according to the spatial position into the spatial sequence of sub-region inkjet concentrations, analyze the real-time inkjet concentration of the sub-region in the spatial sequence of inkjet concentrations, determine the spatial inkjet power compensation value, and regulate the inkjet power when the digital printer performs inkjetting; the present invention calculates the inkjet concentration through the average value of the sub-region color features, realizes the output of the inkjet concentration of the sub-region according to the image color features, and then judges whether the inkjet concentration distribution in the space is consistent. For the case where the concentration distribution is inconsistent, evaluate the linear change degree of the inkjet concentration of each sub-region with the spatial distance, identify the degree of inkjet non-uniformity in the spatial dimension, provide a basis for compensation regulation, and in the case of high inkjet non-uniformity, by analyzing the inkjet concentration of the sub-region in the spatial sequence of inkjet concentrations, use the pre-constructed fitting relationship model between the inkjet power and the inkjet concentration to calculate the spatial inkjet power compensation value and send it to the inkjet device for regulation, which can specifically adjust the inkjet power of the digital printer, improve the problem of inkjet non-uniformity, and enhance the printing quality of digital printed items.

[0039] 2. If the degree of non-uniformity of spatial inkjet in the present invention is low, it is necessary to identify and analyze whether the inkjet concentration in the sub-region is stable during the analysis period. If it is not stable, the time inkjet power compensation value is determined according to the inkjet concentration in the sub-region during the analysis period, and the inkjet power during inkjet of the digital printer is regulated. The present invention can carry out stability analysis on the inkjet concentration of each sub-region during a specific analysis period when the degree of non-uniformity of spatial inkjet is low, measure the fluctuation of the inkjet concentration in the time dimension, timely warn of the problem of inconsistent inkjet in the time series during the printing process, and compensate for the fluctuation of the inkjet concentration in the time dimension to ensure that the inkjet concentration of the digital printer tends to be stable in the time series, further improving the printing quality of digital printed items and reducing printing quality problems caused by inkjet concentration factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0041] Figure 1 It is a system block diagram of a digital printing quality analysis system based on machine vision according to an embodiment of the present invention;

[0042] Figure 2 It is a flowchart of a digital printing quality analysis method based on machine vision according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0044] Embodiment 1

[0045] As Figure 1 shown, a digital printing quality analysis system based on machine vision provided by an embodiment of the present invention specifically includes:

[0046] Data acquisition module: Based on the color features in the sample images of historical digital printing and the historical inkjet concentration, an inkjet concentration model is constructed, and the color features of the real-time image are input into the inkjet concentration model to output the real-time inkjet concentration of each sub-region of the real-time image;

[0047] In some embodiments, a machine vision system is used to obtain sample images printed at different historical inkjet concentrations and convert them into the HSV color space;

[0048] Extract the color features and historical inkjet concentrations in the sample images of historical digital printing, where the color features include saturation S and brightness V;

[0049] Use the least squares method to perform linear regression model fitting and construct an error function, specifically:

[0050]

[0051] where is the inkjet concentration of the i-th sample, is the saturation of the i-th sample, is the brightness of the i-th sample, and n is the number of samples;

[0052] Verify the linear regression model according to the error function. If the verification is qualified, by taking the partial derivative of the error function after verification and setting it to zero, the values of coefficients a, b, and c are solved, and then a linear regression model between the inkjet concentration C, saturation S, and brightness V is constructed, specifically:

[0053]

[0054] If the verification is unqualified, use other models to map the relationship between the inkjet concentration C, saturation S, and brightness V. Other models include but are not limited to random forest models and deep learning models;

[0055] Taking the initial inkjet position of the digital printing machine as the origin, divide the image into multiple juxtaposed sub-regions with the same volume along the inkjet direction;

[0056] Based on any sub-region, obtain all the saturations and brightnesses within the sub-region, and perform summation and averaging respectively to obtain the average saturation and average brightness of the current sub-region;

[0057] Input the average saturation and average brightness of the sub-region into the model, and output the real-time inkjet concentration of the sub-region;

[0058] Inkjet uniformity judgment module: According to the real-time inkjet concentration of the sub-region, judge whether the inkjet concentration in the sub-region is spatially uniform. If it is not uniform, identify the degree of non-uniform distribution of the spatial inkjet concentration;

[0059] In some embodiments, extract the inkjet concentrations of all sub-regions in the real-time image, and calculate the inkjet uniformity judgment value through the variance processing formula for the inkjet concentrations of all sub-regions in the real-time image;

[0060] Compare the inkjet uniformity judgment value with the inkjet uniformity judgment threshold. The specific comparison process is as follows:

[0061] If the inkjet uniformity judgment value is greater than or equal to the inkjet uniformity judgment threshold, it indicates that the inkjet concentrations in each sub-region are inconsistent, and further indicates that the inkjet concentration distribution in the spatial dimension is inconsistent, generating an inkjet non-uniform signal;

[0062] If the inkjet uniformity judgment value is less than the inkjet uniformity judgment threshold, it indicates that the inkjet concentrations in each sub-region are consistent, and further indicates that the inkjet concentration distribution in the spatial dimension is consistent, generating an inkjet uniform signal;

[0063] It should be noted that the inkjet uniformity judgment value is used to characterize the consistency degree of the inkjet concentrations in each sub-region of the real-time image. Generating an inkjet non-uniform signal indicates that the inkjet concentrations in each sub-region are inconsistent in the spatial dimension; generating an inkjet uniform signal indicates that the inkjet concentrations in each sub-region are relatively consistent in the spatial dimension, and the inkjet uniform signal can reflect that the inkjet concentrations in each sub-region will not affect the printing quality of the digital printed article;

[0064] Based on the inkjet non-uniform signal, integrate the inkjet concentrations of all sub-regions according to their spatial positions into the sub-region inkjet concentration spatial sequence, where the spatial position is the distance between the center point of the sub-region and the digital printed article;

[0065] Extract the inkjet concentration of the sub-region ranked first in the sub-region inkjet concentration spatial sequence and the inkjet concentration of the sub-region ranked first in the reverse order; perform a difference process on the inkjet concentration of the sub-region ranked first and the inkjet concentration of the sub-region ranked first in the reverse order, and take the absolute value to obtain the peak-valley inkjet concentration change value. Measure the distance between the sub-regions ranked first and the sub-region ranked first in the reverse order to obtain the peak-valley distance difference; perform a ratio calculation on the peak-valley inkjet concentration change value and the peak-valley distance difference to obtain the reference concentration change rate;

[0066] Extract the inkjet concentration of the sub-region in the previous sequence and the inkjet concentration of the sub-region in the next sequence in the sub-region inkjet concentration spatial sequence; perform a difference process on the inkjet concentration of the sub-region in the previous sequence and the inkjet concentration of the sub-region in the next sequence, and take the absolute value to obtain the adjacent inkjet concentration change value. Measure the distance between the sub-regions in the previous sequence and the next sequence, denoted as the adjacent distance difference; perform a ratio calculation on the adjacent inkjet concentration change value and the adjacent distance difference to obtain the adjacent concentration change rate;

[0067] Based on any adjacent concentration change rate, perform a difference process on the adjacent concentration change rate and the reference concentration change rate, and take the absolute value of the difference to obtain the concentration change rate deviation value;

[0068] Compare the concentration change rate deviation value with the concentration change rate deviation threshold. The specific comparison process is as follows:

[0069] If the concentration change rate deviation value is greater than or equal to the concentration change rate deviation threshold, it indicates that the inkjet concentration change trend in the adjacent area corresponding to the current adjacent concentration change rate is quite different from the inkjet concentration change trend corresponding to the reference concentration change rate. Record the adjacent concentration change rate corresponding to the concentration change rate deviation value as the asynchronous concentration change rate.

[0070] If the concentration change rate deviation value is less than the concentration change rate deviation threshold, it indicates that the inkjet concentration change trend in the adjacent area corresponding to the current adjacent concentration change rate is relatively similar to the inkjet concentration change trend corresponding to the reference concentration change rate. Record the adjacent concentration change rate corresponding to the concentration change rate deviation value as the synchronous concentration change rate.

[0071] Count the number of asynchronous concentration change rates, calculate the proportion of the number of asynchronous concentration change rates, extract the concentration change rate deviation values corresponding to all asynchronous concentration change rates, and sum and average them to obtain the average asynchronous concentration change rate deviation.

[0072] Multiply the average asynchronous concentration change rate deviation by the proportion of the number of asynchronous concentration change rates to obtain the spatial concentration distribution value.

[0073] Compare the spatial concentration distribution value with the spatial concentration distribution threshold. The specific comparison process is as follows:

[0074] If the spatial concentration distribution value is less than the spatial concentration distribution threshold, it indicates that the inkjet concentration of each sub-region changes linearly with the change of spatial distance to a high degree, and the spatial inkjet non-uniformity is low.

[0075] If the spatial concentration distribution value is greater than or equal to the spatial concentration distribution threshold, it indicates that the inkjet concentration of each sub-region changes linearly with the change of spatial distance to a low degree, and the spatial inkjet non-uniformity is high.

[0076] It should be noted that the spatial concentration distribution value is obtained by processing the data of the average asynchronous concentration change rate deviation and the proportion of the number of asynchronous concentration change rates. The larger the average asynchronous concentration change rate deviation, the lower the degree of linear change of the inkjet concentration of each sub-region with the change of distance in space. The larger the proportion of the number of asynchronous concentration change rates, the lower the degree of linear change of the inkjet concentration of each sub-region with the change of distance in space. The smaller the proportion of the number of asynchronous concentration change rates, the higher the degree of linear change of the inkjet concentration of each sub-region with the change of spatial distance, indicating that the inkjet concentration in the digital printing is relatively uniform in the spatial dimension and has little impact on the printing quality of digital printed items.

[0077] Spatial Dimension Power Compensation Module: If the non-uniformity of spatial inkjet is high, integrate the inkjet concentrations of all sub-regions according to their spatial positions into a spatial sequence of sub-region inkjet concentrations, analyze the real-time inkjet concentration of each sub-region in the spatial sequence of inkjet concentrations, determine the spatial inkjet power compensation value, and regulate the inkjet power during inkjetting of the digital printer;

[0078] In some embodiments, extract the inkjet concentration of each sub-region in the spatial sequence of inkjet concentrations, subtract it from the standard value of the inkjet concentration of the sub-region, and then take the absolute value of the difference to obtain the spatial inkjet concentration deviation value , construct a fitting relationship model between the inkjet power and the inkjet concentration, and calculate the spatial inkjet power compensation value KP during inkjetting of the digital printer according to the fitting relationship model. The specific formula is:

[0079]

[0080] where k is the proportionality coefficient between the inkjet power and the inkjet concentration determined through experiments in the fitting relationship model;

[0081] It should be noted that the standard value of the inkjet concentration of the sub-region is set by the personnel in this field according to historical experience;

[0082] Send the spatial inkjet power compensation value to the inkjet device of the digital printer to regulate the inkjet power during inkjetting of the digital printer;

[0083] The technical solution of this embodiment of the present invention is as follows: Based on the color features in the sample images of historical digital printing and the historical inkjet concentration, an inkjet concentration model is constructed. The color features of the real-time image are input into the inkjet concentration model to output the real-time inkjet concentration of each sub-region of the real-time image. According to the real-time inkjet concentration of the sub-region, it is judged whether the inkjet in the space of the sub-region is uniform. If it is not uniform, the degree of non-uniform distribution of the spatial inkjet concentration is identified. If the degree of spatial inkjet non-uniformity is high, the inkjet concentrations of all sub-regions are integrated into a sub-region inkjet concentration spatial sequence according to the spatial position, the real-time inkjet concentration of the sub-region in the inkjet concentration spatial sequence is analyzed, the spatial inkjet power compensation value is determined, and the inkjet power during digital printing is regulated. Through calculating the inkjet concentration by the mean value of the sub-region color features, the present invention realizes outputting the inkjet concentration of the sub-region according to the image color features, and then judges whether the inkjet concentration distribution in the space is consistent. For the case where the concentration distribution is inconsistent, the degree of linear change of the inkjet concentration of each sub-region with the spatial distance is evaluated, the degree of inkjet non-uniformity in the spatial dimension is identified, providing a basis for compensation regulation. In the case of a high degree of inkjet non-uniformity, by analyzing the inkjet concentration of the sub-region in the inkjet concentration spatial sequence, using the pre-constructed fitting relationship model between the inkjet power and the inkjet concentration, the spatial inkjet power compensation value is calculated and sent to the inkjet device for regulation, which can specifically adjust the inkjet power of the digital printer, improve the problem of non-uniform inkjet, and enhance the printing quality of digital printed products.

[0084] Embodiment 2

[0085] As Figure 1 shown, a digital printing quality analysis system based on machine vision provided by an embodiment of the present invention specifically includes:

[0086] Inkjet stability analysis module: If the degree of spatial inkjet non-uniformity is low, it identifies whether the inkjet concentration of the sub-region is stable during the analysis period;

[0087] In some implementation schemes, based on the fact that the inkjet concentration of each sub-region changes linearly with the spatial distance to a relatively high degree, a preset analysis period is set. The duration of the analysis period is T, the current time is used as the end time of the analysis period, and the current time is denoted as TE. The time interval corresponding to the analysis period is [TE - T, TE];

[0088] It can be understood that the value of the duration T of the analysis period is set in advance by those skilled in the art;

[0089] Obtain the inkjet concentrations of all time points of the sub-region and integrate them into an inkjet concentration time sequence, extract all the inkjet concentrations in the inkjet concentration time sequence and calculate the variance to obtain the inkjet concentration variance of each sub-region;

[0090] It is understandable that the time intervals for continuously collecting digital printing images during the analysis period are the same. The analysis period is divided into several time points with equal time intervals, and a digital printing image is collected at each time point.

[0091] Sum up and average the variance of the inkjet concentration of all sub-regions to obtain the stable value of the inkjet concentration.

[0092] Compare the stable value of the inkjet concentration with the stable threshold of the inkjet concentration. The specific comparison process is as follows:

[0093] If the stable value of the inkjet concentration is greater than or equal to the stable threshold of the inkjet concentration, it indicates that the inkjet concentration in the inkjet space fluctuates in the time dimension during the analysis period.

[0094] If the stable value of the inkjet concentration is less than the stable threshold of the inkjet concentration, it indicates that the inkjet concentration in the inkjet space tends to be stable in the time dimension during the analysis period.

[0095] It should be noted that the stable value of the inkjet concentration is used to characterize the fluctuation of the inkjet concentration in the inkjet space in the time dimension under the condition of low inkjet unevenness. The fluctuation of the inkjet concentration in the inkjet space in the time dimension during the analysis period indicates the inconsistency of inkjet in the time series during the printing process, which helps to judge the printing quality of digital printed products. The fact that the inkjet concentration in the inkjet space tends to be stable in the time dimension during the analysis period indicates that the inkjet concentration is stable in both the time and space series during the printing process, which further indicates that the inkjet concentration has little impact on the printing quality of digital printed products.

[0096] Time dimension power compensation module: If it is unstable, determine the time inkjet power compensation value according to the inkjet concentration of the sub-region during the analysis period, and adjust the inkjet power when the digital printer performs inkjet.

[0097] In some embodiments, extract the inkjet concentration of the sub-region in the inkjet concentration time series, subtract it from the standard value of the inkjet concentration of the sub-region, and then take the absolute value of the difference to obtain the time inkjet concentration deviation value. Input it into the fitting relationship model between the input inkjet power and the inkjet concentration, and calculate the time inkjet power compensation value SP when the digital printer performs inkjet. The specific formula is:

[0098]

[0099] Send the time inkjet power compensation value to the inkjet device of the digital printer to adjust the inkjet power when the digital printer performs inkjet.

[0100] The technical solution of this embodiment is as follows: If the degree of unevenness of spatial inkjet is low, it is necessary to identify and analyze whether the inkjet concentration in the sub-region is stable during the analysis period. If it is not stable, according to the inkjet concentration in the sub-region during the analysis period, determine the time inkjet power compensation value, and regulate the inkjet power when the digital printer performs inkjet. The present invention can, when the degree of uneven spatial inkjet is low, carry out stability analysis on the inkjet concentration of each sub-region during a specific analysis period, measure the fluctuation of the inkjet concentration in the time dimension, promptly warn of the problem of inconsistent inkjet in the time series during the printing process, and compensate for the fluctuation of the inkjet concentration in the time dimension to ensure that the inkjet concentration of the digital printer tends to be stable in the time series, further improving the printing quality of digital printed items and reducing printing quality problems caused by inkjet concentration factors.

[0101] Embodiment III

[0102] As Figure 2 shown, a digital printing quality analysis method based on machine vision provided by an embodiment of the present invention specifically includes the following steps:

[0103] Step 1: Based on the color features and historical inkjet concentration in the sample images of historical digital printing, construct an inkjet concentration model, input the color features of the real-time image into the inkjet concentration model, and output the real-time inkjet concentration of each sub-region of the real-time image;

[0104] Step 2: According to the real-time inkjet concentration of the sub-region, determine whether the spatial inkjet of the inkjet concentration in the sub-region is uniform. If it is not uniform, identify the degree of uneven distribution of the spatial inkjet concentration;

[0105] Step 3: If the degree of uneven spatial inkjet is high, integrate the inkjet concentrations of all sub-regions into an inkjet concentration spatial sequence according to the spatial position, analyze the real-time inkjet concentration of the sub-regions in the inkjet concentration spatial sequence, determine the spatial inkjet power compensation value, and regulate the inkjet power when the digital printer performs inkjet.

[0106] Step 4: If the degree of uneven spatial inkjet is low, identify whether the inkjet concentration in the sub-region is stable during the analysis period;

[0107] Step 5: If it is not stable, according to the inkjet concentration in the sub-region during the analysis period, determine the time inkjet power compensation value, and regulate the inkjet power when the digital printer performs inkjet.

[0108] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0109] The above has described in detail an embodiment of the present invention, but the above content is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. A digital printing quality analysis system based on machine vision, characterized in that, Including: Data acquisition module: Based on the color features in the sample images of historical digital printing and the historical inkjet concentration, an inkjet concentration model is constructed. The color features of the real-time image are input into the inkjet concentration model, and the real-time inkjet concentration of each sub-region of the real-time image is output; Inkjet uniformity judgment module: According to the real-time inkjet concentration of the sub-region, judge whether the spatial inkjet of the inkjet concentration in the sub-region is uniform. If it is not uniform, identify the uneven distribution degree of the spatial inkjet concentration; Integrate the inkjet concentrations of all sub-regions according to their spatial positions to form an empty sequence of sub-region inkjet concentrations; Analyze the inkjet concentrations of the sub-regions in the empty sequence of sub-region inkjet concentrations to obtain the adjacent concentration change rate and the reference concentration change rate. Perform data processing on the adjacent concentration change rate and the reference concentration change rate to obtain the concentration change rate deviation value and the asynchronous concentration change rate; The adjacent concentration change rate is the ratio of the difference in inkjet concentration between adjacent sub-regions to the difference in distance between adjacent sub-regions; The reference concentration change rate is the ratio of the difference in inkjet concentration between the first sub-region in sequence and the first sub-region in reverse order to the distance between the sub-regions between the first in sequence and the first in reverse order; The concentration change rate deviation is the absolute value of the difference between the adjacent concentration change rate and the reference concentration change rate; If the concentration change rate deviation value is greater than or equal to the concentration change rate deviation threshold, record the adjacent concentration change rate corresponding to the concentration change rate deviation value as the asynchronous concentration change rate; Calculate the proportion of the number of asynchronous concentration change rates, extract the concentration change rate deviation values corresponding to all asynchronous concentration change rates, and perform summation and averaging to obtain the average asynchronous concentration change rate deviation; Perform a product process on the average asynchronous concentration change rate deviation and the proportion of the number of asynchronous concentration change rates to obtain the spatial concentration distribution value. If the spatial concentration distribution value < the spatial concentration distribution threshold, it indicates that the degree of spatial inkjet non-uniformity is low; otherwise, it indicates that the degree of spatial inkjet non-uniformity is high; Spatial dimension power compensation module: If the degree of spatial inkjet non-uniformity is high, integrate the inkjet concentrations of all sub-regions according to their spatial positions to form an empty sequence of sub-region inkjet concentrations, analyze the real-time inkjet concentration of the sub-regions in the empty sequence of inkjet concentrations, determine the spatial inkjet power compensation value, and regulate the inkjet power when the digital printer jets ink.

2. The digital printing quality analysis system based on machine vision according to claim 1, wherein Also including: Inkjet stability analysis module: If the degree of spatial inkjet non-uniformity is low, identify whether the inkjet concentration of the sub-region is stable during the analysis period; Time dimension power compensation module: If it is unstable, determine the time inkjet power compensation value according to the inkjet concentration of the sub-region during the analysis period, and regulate the inkjet power when the digital printer jets ink.

3. A digital printing quality analysis system based on machine vision according to claim 1, characterized in that, The process of obtaining the real-time inkjet concentration of the sub-region is as follows: Obtain the color features of the sample images of historical digital printing and the historical inkjet concentration. Among them, the color features include saturation and brightness, and construct a model; Divide the collected real-time image into sub-regions, obtain all the saturation and brightness within the sub-region of the real-time image, and perform summation and averaging respectively to obtain the average saturation and average brightness of the current sub-region, and input them into the model to output the real-time inkjet concentration of the sub-region.

4. The digital printing quality analysis system based on machine vision according to claim 3, characterized in that, The specific process of judging whether the spatial inkjet of the inkjet concentration in the sub-region is uniform is as follows: Obtain the inkjet concentration of all sub-regions in the real-time image, calculate the inkjet uniformity judgment value through the variance processing formula. If the inkjet uniformity judgment value ≥ the inkjet uniformity judgment threshold, generate an inkjet non-uniform signal.

5. A digital printing quality analysis system based on machine vision according to claim 1, wherein The process of obtaining the concentration change rate deviation value and the asynchronous concentration change rate is as follows: Perform a difference operation on the adjacent concentration change rate and the reference concentration change rate and then take the absolute value to obtain the concentration change rate deviation value. If the concentration change rate deviation value ≥ the concentration change rate deviation threshold, record the adjacent concentration change rate corresponding to the concentration change rate deviation value as the asynchronous concentration change rate.

6. A digital printing quality analysis system based on machine vision according to claim 1, characterized in that, The process of obtaining the reference concentration change rate is as follows: Extract the inkjet concentration of the sub-region ranked first in order and the inkjet concentration of the sub-region ranked first in reverse order in the sub-region inkjet concentration null sequence; perform a difference operation and then take the absolute value to obtain the peak-valley inkjet concentration change value. Measure the distance between the sub-regions ranked first in order and first in reverse order to obtain the peak-valley distance difference; perform a ratio calculation on the peak-valley inkjet concentration change value and the peak-valley distance difference to obtain the reference concentration change rate.

7. The digital printing quality analysis system based on machine vision according to claim 1, characterized in that, The process of obtaining the adjacent concentration change rate is as follows: Extract the inkjet concentration of the sub-region in the previous sequence and the inkjet concentration of the sub-region in the next sequence in the sub-region inkjet concentration null sequence; perform a difference operation and then take the absolute value to obtain the adjacent inkjet concentration change value. Record the distance between the previous sequence and the next sequence as the adjacent distance difference; perform a ratio calculation on the adjacent inkjet concentration change value and the adjacent distance difference to obtain the adjacent concentration change rate.

8. A digital printing quality analysis system based on machine vision according to claim 1, characterized in that, The method for obtaining the spatial inkjet power compensation value is as follows: Extract the inkjet concentration of the sub-region in the inkjet concentration null sequence, perform a subtraction operation with the standard value of the inkjet concentration of the sub-region, and then take the absolute value to obtain the spatial inkjet concentration deviation value. Construct a fitting relationship model between the inkjet power and the inkjet concentration, and calculate the spatial inkjet power compensation value when the digital printing machine performs inkjet according to the fitting relationship model.

9. The digital printing quality analysis system based on machine vision according to claim 2, characterized in that, The process of identifying whether the inkjet concentration of the sub-region is stable during the analysis period is as follows; Analyze the inkjet concentration at all time points within the sub-region analysis period to obtain the inkjet concentration stability value; if the inkjet concentration stability value ≥ the inkjet concentration stability threshold, it indicates that the inkjet concentration in the inkjet space fluctuates in the time dimension during the analysis period.

10. A digital printing quality analysis system based on machine vision according to claim 9, characterized in that, The process of obtaining the inkjet concentration stability value is as follows: Obtain the inkjet concentration at all time points within the sub-region analysis period, integrate it into an inkjet concentration time series, extract all the inkjet concentrations in the inkjet concentration time series and perform variance calculation to obtain the inkjet concentration variance of each sub-region; Sum up and take the average of all the sub-region inkjet concentration variances to obtain the inkjet concentration stability value.

11. A digital printing quality analysis system based on machine vision according to claim 10, characterized in that, The process of obtaining the time inkjet power compensation value is as follows: Extract the inkjet concentration of the sub-region in the inkjet concentration time series, perform a subtraction operation with the standard value of the inkjet concentration of the sub-region, and then take the absolute value to obtain the time inkjet concentration deviation value. Input it into the fitting relationship model between the inkjet power and the inkjet concentration, and calculate the time inkjet power compensation value when the digital printing machine performs inkjet.

Images