Digital printing online color correction method based on machine vision

Abstract

The invention discloses a digital printing online color correction method based on machine vision, relates to the technical field of printing color correction, and mainly aims to solve the technical problems that an effect detection link after color correction is incomplete, and key parameters in the correction process are not monitored in real time and subjected to feedback adjustment. According to the method, full-process, multi-dimensional and high-precision online color correction is achieved, the printing color rendition degree and consistency are guaranteed, accurate recognition and efficient correction of color deviation in the digital printing process are achieved through the full-process design of multi-dimensional color information recognition, a targeted correction strategy, real-time correction effect detection and dynamic feedback adjustment, and the printing quality is improved. The method is particularly suitable for complex printing stock scenes such as concave-convex texture fabric, the printing color stability and the production efficiency are remarkably improved, and the product rejection rate is reduced.

Description

Technical Field

[0001] This invention relates to the field of printing color correction technology, specifically to a digital printing online color correction method based on machine vision. Background Technology

[0002] In the digital printing industry, color reproduction and printing consistency are core indicators for measuring product quality. As the requirements for printing precision in application scenarios such as textile fabrics and packaging materials continue to increase, traditional digital printing color correction technology can no longer meet actual production needs.

[0003] Currently, the mainstream color correction methods in the industry are mostly offline or semi-online correction. Offline correction requires calibration equipment through standard color cards before printing, and cannot respond in real time to dynamic deviations during the printing process, such as color deviations caused by differences in substrate texture, ink characteristic drift, and fluctuations in environmental temperature and humidity. Problems are often only discovered after printing is completed, leading to the scrapping of batch products and low production efficiency.

[0004] While semi-online calibration can achieve some real-time monitoring, it mostly only detects single color parameters (such as brightness and color difference) and lacks comprehensive recognition of multi-dimensional parameters such as color space, color level, and color matching degree. Especially when dealing with special substrates such as textured fabrics, it is easy to cause sampling deviation due to texture shadows and uneven reflection, resulting in insufficient calibration accuracy.

[0005] Meanwhile, current correction schemes for printing and coating anomalies lack specificity, have not established a cross-device color reference library and a three-dimensional depth image acquisition mechanism, making it difficult to accurately locate the source of color deviation and texture area differences. The effect detection process after correction is imperfect, and key parameters in the correction process (such as jet pressure compensation, ink drying speed, ink absorption speed, etc.) are not monitored and adjusted in real time. This often results in inefficient correction or the working status of adjacent printheads being affected after correction, further reducing the stability and consistency of printed colors.

[0006] To address the aforementioned technical shortcomings, a machine vision-based online color correction method for digital printing is proposed. This method forms a complete online color correction solution encompassing "multi-dimensional anomaly identification, targeted correction, real-time detection of correction effects, and dynamic feedback adjustment," thereby promoting the development of the digital printing industry towards higher precision, higher efficiency, and higher stability. Summary of the Invention

[0007] The purpose of this invention is to solve the problems mentioned above by proposing a machine vision-based online color correction method for digital printing.

[0008] The objective of this invention can be achieved through the following technical solution: a machine vision-based online color correction method for digital printing, comprising the following steps:

[0009] Step 1: Printing color recognition and detection. The required printing image is extracted from the substrate using computer image processing technology, and the printing image is divided into sub-regions. Based on the sub-region recognition and detection, the printing process is classified and correction decisions are made.

[0010] Step 2: Printing process identification and correction; Identify and correct colors according to the different time periods in the printing process, specifically including spraying anomaly correction and printing anomaly correction;

[0011] Step 3: Printing anomaly correction and detection. During the printing anomaly correction stage, detection and adjustments are performed.

[0012] Step 4: Spraying Anomaly Correction and Detection. During the spraying anomaly correction stage, detection and control measures are implemented.

[0013] Furthermore, color information includes color space, color hierarchy, and color matching degree; color space is represented by the number and distribution range of colors; color hierarchy is represented by the distribution of color depth and brightness in each area, with brightness as the quantitative data in actual scenarios; color matching degree is represented by the matching degree between the digital printing machine printhead and the ink, with the specific quantitative data being the deviation between the color setting range and the actual range.

[0014] Furthermore, the process of print color recognition and detection is as follows:

[0015] The deviation between the number of colors in the color space of the substrate and the set number is obtained during the printing process. At the same time, the ratio of the number of areas with abnormal color distribution to the number of printed sub-areas is obtained.

[0016] If the deviation between the number of colors in the color space of the substrate and the set number exceeds the number deviation threshold during the printing process, or if the ratio of the number of areas with abnormal color layer distribution to the number of completed printing sub-areas exceeds the number ratio threshold, it is inferred that the color spraying of the substrate is abnormal during the printing process, and the current printing process is marked as a spraying abnormal process.

[0017] If the deviation between the number of colors in the color space of the substrate and the set number does not exceed the number deviation threshold during the printing process, and the ratio of the number of areas with abnormal color layer distribution to the number of completed printing sub-areas does not exceed the number ratio threshold, then it is inferred that the color spraying of the substrate is normal during the printing process, and the current printing process is marked as a normal spraying process.

[0018] Furthermore, the quantitative data of color matching degree during the printing process is acquired simultaneously, and the changes in the quantitative data are recorded. If the peak value corresponding to the quantitative data of color matching degree exceeds the range area peak value, or the fluctuation frequency of the quantitative data of color matching degree exceeds the fluctuation frequency threshold, it is inferred that the printing process is abnormal, and the corresponding printing process is marked as an abnormal printing process. If the peak value corresponding to the quantitative data of color matching degree does not exceed the range area peak value, and the fluctuation frequency of the quantitative data of color matching degree does not exceed the fluctuation frequency threshold, it is inferred that the printing process is normal, and the corresponding printing process is marked as a normal printing process.

[0019] Furthermore, color correction is performed based on the various time periods within the printing process; abnormal spraying processes and abnormal printing processes are identified, and color correction is performed based on the corresponding time periods.

[0020] Furthermore, the printing anomaly correction and detection process is as follows:

[0021] The printing anomaly correction is detected. During the printing anomaly correction execution phase, the deviation value between the corresponding jet pressure adjustment amount and the actual jet pressure compensation amount of the printing equipment printhead is obtained. At the same time, the deviation value between the drying speed of the ink characteristic parameters equipped with the printing equipment printhead and the drying speed required for printing on the substrate is obtained.

[0022] If the deviation between the spray pressure adjustment amount corresponding to the print head of the printing equipment and the actual spray pressure compensation amount exceeds the pressure deviation threshold, or if the deviation between the drying speed in the ink characteristic parameters of the print head of the printing equipment and the drying speed required for printing on the substrate exceeds the drying speed deviation threshold, then the current printing abnormality correction process will be marked as an inefficient printing correction stage.

[0023] If the deviation between the adjusted spray pressure and the actual compensation of the spray pressure of the print head does not exceed the pressure deviation threshold, and the deviation between the drying speed and the required drying speed of the substrate in the ink characteristic parameters of the print head does not exceed the drying speed deviation threshold, then the current printing anomaly correction process is marked as the high-efficiency printing correction stage.

[0024] Furthermore, the process for correcting and detecting abnormalities in spraying is as follows:

[0025] During the abnormal spraying correction stage, the maximum deviation of the ink absorption speed of the substrate when receiving ink of different colors is obtained. At the same time, when the ink corresponding to the maximum deviation of the ink absorption speed is successively sprayed, the overlap time between the abnormal fluctuation stage of the real-time printing environment temperature and the ink spraying stage is obtained.

[0026] If the maximum deviation of the ink absorption speed when the substrate receives ink of different colors exceeds the maximum deviation threshold, or if the overlap time between the abnormal temperature fluctuation stage of real-time printing environment and the ink spraying stage exceeds the overlap time threshold, then the current spraying abnormal correction will be marked as the spraying correction inefficient stage.

[0027] If the maximum deviation of the ink absorption speed when the substrate receives ink of different colors does not exceed the maximum deviation threshold, and the overlap time between the abnormal temperature fluctuation stage of real-time printing environment and the ink spraying stage does not exceed the overlap time threshold, then the current spraying abnormal correction is marked as the high-efficiency spraying correction stage.

[0028] Compared with the prior art, the beneficial effects of the present invention are:

[0029] 1. This invention enables comprehensive identification and accurate judgment of multi-dimensional color information. By collecting three core types of information—color space, color level, and color matching degree—and combining them with multi-threshold comparison, it significantly reduces the misjudgment rate and missed judgment rate of spraying abnormalities and printing mismatches. This solves the problem of incomplete identification of single parameters in traditional technologies and lays the foundation for subsequent targeted correction.

[0030] 2. This invention also improves the targeting and adaptability of color correction. It designs a spraying anomaly correction scheme based on the joint acquisition of two-dimensional and three-dimensional images for textured fabrics, which effectively solves the sampling deviation caused by texture shadows and uneven reflection. It establishes a cross-device color reference library for printing anomalies, so as to achieve accurate positioning and targeted adjustment of the source of color deviation, and adapt to different substrates, inks and equipment scenarios.

[0031] 3. The present invention also constructs a real-time detection and dynamic feedback adjustment mechanism for the correction effect. By monitoring the deviation of key parameters in the printing and coating correction process, the correction execution efficiency is accurately judged, targeted parameter adjustment is carried out for inefficient stages, and the working status of adjacent nozzles is taken into account to avoid cross interference and ensure the stability and continuity of the correction effect.

[0032] 4. This invention significantly improves the color reproduction and consistency of digital printing, reduces the product scrap rate caused by color deviation, and enables online calibration throughout the entire process without interrupting production for offline calibration. This greatly improves production efficiency, reduces production costs, and meets the industry's development needs for high precision, high efficiency, and high stability, thus having broad application prospects. Attached Figure Description

[0033] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0034] Figure 1 This is a block diagram illustrating the overall principle of the method of the present invention;

[0035] Figure 2This is a flowchart of the printing color recognition and detection method in this invention. Detailed Implementation

[0036] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0038] Please see Figures 1-2 As shown, the online color correction method for digital printing based on machine vision specifically includes the following steps:

[0039] Step 1: Print color recognition and detection;

[0040] The required printing image is extracted from the substrate using computer image processing technology, and the printing image is divided into several sub-regions. The color information of each sub-region is recorded. The color information includes color space, color level, and color matching degree. Color space is represented by the number and distribution range of colors. Color level is represented by the distribution of color depth and brightness in each region, with brightness as the quantitative data in actual scenarios. Color matching degree is represented by the matching degree between the digital printing machine printhead and the ink, and the specific quantitative data is the deviation between the set color range and the actual range.

[0041] The deviation between the number of colors in the color space of the substrate and the set number is obtained during the printing process. At the same time, the ratio of the number of areas with abnormal color distribution to the number of printed sub-areas is also obtained. It should be explained that abnormal color distribution means that the distribution of color depth and brightness in any sub-area deviates from the set distribution position.

[0042] If the deviation between the number of colors in the color space of the substrate and the set number exceeds the number deviation threshold during the printing process, or if the ratio of the number of areas with abnormal color layer distribution to the number of completed printing sub-areas exceeds the number ratio threshold, it is inferred that the color spraying of the substrate is abnormal during the printing process, and the current printing process is marked as a spraying abnormal process.

[0043] If the deviation between the number of colors in the color space of the substrate and the set number does not exceed the number deviation threshold during the printing process, and the ratio of the number of areas with abnormal color layer distribution to the number of completed printing sub-areas does not exceed the number ratio threshold, it is inferred that the color spraying of the substrate is normal during the printing process, and the current printing process is marked as a normal spraying process.

[0044] Simultaneously, quantitative data of color matching degree during the printing process is acquired and the changes of quantitative data are recorded. If the peak value of the quantitative data of color matching degree exceeds the range area peak value, or the fluctuation frequency of the quantitative data of color matching degree exceeds the fluctuation frequency threshold, it is inferred that the printing process is abnormal and the corresponding printing process is marked as an abnormal printing process.

[0045] If the peak value of the quantitative data of color matching degree does not exceed the peak value of the range area, and the fluctuation frequency of the quantitative data of color matching degree does not exceed the fluctuation frequency threshold, it is inferred that the printing process is normal and the corresponding printing process is marked as a normal printing process.

[0046] It needs to be explained that the print color recognition and detection process involves three core thresholds:

[0047] First, the quantity deviation threshold (the deviation threshold between the number of colors in the color space and the set quantity) is obtained as follows: Firstly, based on the limitations on color quantity deviation in digital printing as specified in industry standards such as GB / T17001.1-1997 "Anti-counterfeiting Printed Products Part 1: General Technical Requirements" and ISO12647-2:2013 "Printing Technology - Process Control of Halftone Printed Matter - Part 2: Offset Printing", a basic threshold range is determined. Then, 10 commonly used typical substrates (including cotton, linen, polyester, and textured fabrics), 5 mainstream ink types, and 3 different models of digital printing equipment are selected to conduct multiple orthogonal experiments. Corresponding data on color quantity deviation and printed color reproduction are collected under each experimental scenario. Linear regression analysis is used to determine the optimal deviation value for different scenarios. Finally, considering the different color accuracy requirements of customers in actual production (such as high-end textile printing and ordinary packaging printing), the basic threshold is fine-tuned to form a quantity deviation threshold adapted to different scenarios. A dynamic threshold update mechanism is established and regularly optimized based on subsequent production data accumulation.

[0048] Second, the quantity ratio threshold (the ratio of the number of abnormal color layer distribution areas to the number of completed printed sub-areas) is obtained as follows: Select printed patterns with different color layer complexity (from simple monochrome patterns to complex gradient color patterns), conduct batch experiments at different printing speeds and under different printhead conditions, and record the correspondence between the proportion of abnormal color layer distribution areas and the printing visual effect score (blindly evaluated by 5 senior industry technicians and the average score is taken). The proportion when the visual effect score is lower than 80 points (out of 100 points) is the critical threshold. Then, combined with the machine vision recognition accuracy (excluding false positive anomalies caused by recognition errors), the critical threshold is corrected to determine the final quantity ratio threshold. At the same time, for special substrates such as textured fabrics, separate experimental calibration is performed to adjust the threshold to adapt to the slight layer distribution deviation caused by their texture characteristics.

[0049] Thirdly, the peak range (peak threshold of color matching measurement data) and the fluctuation frequency threshold (fluctuation frequency threshold of color matching measurement data) are obtained as follows: First, a cross-device color matching database is established, and historical color matching data of different brands and models of printheads and ink combinations are collected. The peak range and fluctuation frequency range of color matching measurement data under normal printing conditions are statistically analyzed. Then, through fault simulation experiments, fault scenarios such as printhead clogging, ink deterioration, and pressure fluctuations are artificially set, and the critical values ​​when the peak color matching value exceeds the normal range and the fluctuation frequency exceeds the normal range are recorded. These critical values ​​are the initial thresholds. Finally, the initial thresholds are iteratively optimized based on the misjudgment rate and missed judgment rate of printing anomalies in actual production to ensure that printing mismatch anomalies can be accurately identified while avoiding production interruptions caused by misjudgments.

[0050] Step 2: Identification and correction during the printing process;

[0051] Color correction is performed based on the various time periods within the printing process; abnormal spraying and printing processes are identified, and targeted color correction is performed according to the corresponding time periods. Specific color correction includes spraying abnormality correction and printing abnormality correction. Among them, spraying abnormality correction (adapted to textured fabrics): by acquiring two-dimensional color images and three-dimensional depth images, the raised / depressed areas of the fabric are distinguished, and a correction curve is generated in a targeted manner to solve the sampling deviation caused by uneven texture shadows / reflections.

[0052] Printing anomaly correction: Establish a cross-device color reference library, quantify the density deviation of each printhead, locate the source of color deviation, and adjust the jet pressure and ICC curve respectively.

[0053] Step 3: Printing anomaly correction and detection;

[0054] The printing anomaly correction is detected. During the printing anomaly correction execution phase, the deviation value between the corresponding jet pressure adjustment amount and the actual jet pressure compensation amount of the printing equipment printhead is obtained. At the same time, the deviation value between the drying speed of the ink characteristic parameters equipped with the printing equipment printhead and the drying speed required for printing on the substrate is obtained.

[0055] If the deviation between the adjusted jet pressure of the printhead and the actual compensated jet pressure exceeds the pressure deviation threshold, or if the deviation between the drying speed of the ink characteristic parameters of the printhead and the required drying speed of the substrate exceeds the drying speed deviation threshold, it is inferred that the printing anomaly correction is inefficient, and the current printing anomaly correction process is marked as an inefficient printing correction stage. It should be explained that if there is a deviation when comparing the pressure data and the speed data threshold, it is assumed that the data is not compatible with the substrate setting.

[0056] If the deviation between the spray pressure adjustment amount and the actual compensation amount of the spray pressure of the print head of the printing equipment does not exceed the pressure deviation threshold, and the deviation between the drying speed of the ink characteristic parameters of the print head and the drying speed required for printing on the substrate does not exceed the drying speed deviation threshold, then it is inferred that the printing abnormality correction is performed efficiently, and the current printing abnormality correction process is marked as the printing correction efficient stage.

[0057] The pressure and speed parameters are adjusted during the inefficient printing correction stage. During the adjustment stage, the parameter changes during the efficient printing correction stage are monitored to avoid the efficient printing correction stage of adjacent printheads being affected after the inefficient printing correction stage of the current printhead is completed and adjusted.

[0058] It needs to be explained that the printing anomaly correction and detection step involves two core thresholds:

[0059] First, there is the pressure deviation threshold (the deviation threshold between the jet pressure adjustment amount and the actual compensation amount). This is obtained by conducting printhead pressure adjustment experiments on different models of printing equipment (covering mainstream brands and models in the industry), recording the deviation data between the pressure adjustment amount and the actual compensation amount of each printhead under normal operating conditions, and statistically analyzing the mean and standard deviation of the deviation. The mean plus twice the standard deviation is used as the initial threshold. Subsequently, supplementary experiments are conducted in the early, middle, and late stages of printhead use, taking into account the printhead's working life cycle, to adjust the threshold to adapt to pressure deviation changes caused by printhead wear. Simultaneously, the initial threshold is corrected by referring to the printhead pressure adjustment accuracy standards provided by the equipment manufacturer to ensure that the threshold meets the equipment's operating requirements.

[0060] Second, the drying speed deviation threshold (the deviation threshold between the ink drying speed and the required drying speed of the substrate) is obtained as follows: First, based on the requirements for ink drying speed in GB / T 23979.1-2009 "Digital Printing Inks Part 1: Water-based Inks", combined with experimental data on the water absorption / ink absorption characteristics of different substrates (such as textile fabrics, paper, and plastic films), the standard drying speed range corresponding to each substrate is determined; then, the maximum allowable deviation between the standard drying speed and the actual drying speed (i.e., the drying speed deviation threshold) is calculated. This deviation value is based on the core principle of ensuring that the printed colors do not bleed or fade. Through batch printing experiments, it is verified that when the deviation exceeds this threshold, the probability of bleeding and fading of the printed products exceeds 5%. Based on this, the final drying speed deviation threshold is determined, and it is calibrated separately for different ink types (water-based, oil-based, and reactive inks).

[0061] Step 4: Inspection and correction of spraying anomalies;

[0062] During the abnormal ink absorption correction stage, the maximum deviation of the ink absorption speed of the substrate when receiving ink of different colors is obtained. At the same time, when the ink corresponding to the maximum deviation of the ink absorption speed is completed in succession, the overlap time between the abnormal temperature fluctuation stage of the real-time printing environment and the ink spraying stage is obtained. The abnormal temperature fluctuation indicates that the temperature fluctuation trend affects the ink absorption speed, resulting in it being too fast or too slow.

[0063] If the maximum deviation of the ink absorption speed when the substrate receives ink of different colors exceeds the maximum deviation threshold, or if the overlap time between the abnormal temperature fluctuation stage of real-time printing environment and the ink spraying stage exceeds the overlap time threshold, it is inferred that the spraying abnormality correction is inefficient and the current spraying abnormality correction is marked as the inefficient spraying correction stage.

[0064] If the maximum deviation of the ink absorption speed when the substrate receives ink of different colors does not exceed the maximum deviation threshold, and the overlap time between the abnormal temperature fluctuation stage of real-time printing environment and the ink spraying stage does not exceed the overlap time threshold, it is inferred that the spraying anomaly correction is performed efficiently, and the current spraying anomaly correction is marked as the efficient spraying correction stage.

[0065] The system collects and controls data on ink and substrate during the inefficient spraying correction stage to match the substrate with the appropriate ink. At the same time, it controls the parameters of the spraying environment during the printing process. In this application, temperature is used as the environmental parameter. In actual scenarios, parameters such as humidity also have an impact and are also applicable to this system.

[0066] It needs to be explained that the spraying anomaly correction and detection procedure involves two core thresholds:

[0067] First, the maximum deviation threshold (the maximum deviation threshold of ink absorption speed when the substrate receives different colored inks) is obtained as follows: Eight different colored inks commonly used in the industry (covering the three primary colors and auxiliary colors) are selected, and ink absorption speed experiments are conducted on different substrates. The ink absorption speed data of each colored ink on the same substrate are recorded, and the deviation value of the ink absorption speed of different colored inks is calculated. Then, through the verification of printing visual effects, when the maximum deviation of ink absorption speed exceeds a certain value, problems such as inconsistent color depth and color deviation will occur. This value is the critical deviation value. Combining the recognition accuracy of color differences by machine vision, the critical deviation value is corrected to determine the final maximum deviation threshold. At the same time, for textured fabrics, the influence of their surface texture on ink absorption speed is considered, and the threshold is calibrated separately through experiments.

[0068] Second, the overlap duration threshold (the overlap duration threshold between the abnormal temperature fluctuation stage and the ink spraying stage in real-time printing) is obtained as follows: First, define the standard for abnormal temperature fluctuation (i.e., the critical temperature fluctuation range that affects the ink absorption speed, determined experimentally, for example, a temperature fluctuation of ±3℃ / min is considered abnormal fluctuation); then, conduct printing experiments under different environmental temperature fluctuation scenarios, record the overlap duration between the abnormal temperature fluctuation stage and the ink spraying stage, as well as the corresponding printing color deviation; when the overlap duration exceeds a certain value, the color deviation exceeds the allowable range (refer to the industry standard of color deviation ΔE≤2), and this value is the initial overlap duration threshold; finally, combine the temperature and humidity fluctuation patterns of the actual production environment to optimize the initial threshold to ensure adaptability to environmental changes in different production scenarios.

[0069] In summary, this invention aims to solve the technical problems existing in digital printing color correction, such as poor real-time performance, insufficient correction accuracy, weak targeting, and lack of effective detection and feedback of correction effects, so as to achieve full-process, multi-dimensional, and high-precision online color correction and ensure the color reproduction and consistency of printed materials.

[0070] Thresholds, preset values, preset ranges, etc. are set for result comparison and analysis to determine whether they are good or bad. The value of these thresholds is determined by a combination of large-scale model analysis of sample data and human experience. They can also be adjusted appropriately based on seasonal or common-sense influences.

[0071] Furthermore, the settings for weighting ratios, influence factors, etc., are based on the magnitude of each parameter's influence on the results. The specific values ​​are allocated to ultimately reflect the impact on the results. The settings for input and storage are also determined by a combination of large-scale model analysis of sample data and human experience. Appropriate adjustments can also be made based on seasonal or rational influence conditions.

[0072] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A machine vision-based online color correction method for digital printing, characterized in that, Includes the following steps: Step 1: Printing color recognition and detection. The required printing image is extracted from the substrate using computer image processing technology, and the printing image is divided into sub-regions. Based on the sub-region recognition and detection, the printing process is classified and correction decisions are made. Step 2: Printing process identification and correction; Identify and correct colors according to the different time periods in the printing process, specifically including spraying anomaly correction and printing anomaly correction; Step 3: Printing anomaly correction and detection. During the printing anomaly correction stage, detection and adjustments are performed. Step 4: Spraying Anomaly Correction and Detection. During the spraying anomaly correction stage, detection and control measures are implemented.

2. The online color correction method for digital printing based on machine vision according to claim 1, characterized in that, Color information includes color space, color hierarchy, and color matching degree; Color space is represented by the number and distribution range of colors; color hierarchy is represented by the distribution of color depth and brightness in each area, with brightness used as the quantitative data in actual scenarios. Color matching degree refers to the matching degree between the printhead and ink of a digital printing machine. The specific quantitative data is the deviation between the color setting range and the actual range.

3. The online color correction method for digital printing based on machine vision according to claim 2, characterized in that, The process of print color recognition and detection is as follows: The deviation between the number of colors in the color space of the substrate and the set number is obtained during the printing process. At the same time, the ratio of the number of areas with abnormal color distribution to the number of printed sub-areas is obtained. If the deviation between the number of colors in the color space of the substrate and the set number exceeds the number deviation threshold during the printing process, or if the ratio of the number of areas with abnormal color layer distribution to the number of completed printing sub-areas exceeds the number ratio threshold, it is inferred that the color spraying of the substrate is abnormal during the printing process, and the current printing process is marked as a spraying abnormal process. If the deviation between the number of colors in the color space of the substrate and the set number does not exceed the number deviation threshold during the printing process, and the ratio of the number of areas with abnormal color layer distribution to the number of completed printing sub-areas does not exceed the number ratio threshold, then it is inferred that the color spraying of the substrate is normal during the printing process, and the current printing process is marked as a normal spraying process.

4. The online color correction method for digital printing based on machine vision according to claim 3, characterized in that, Simultaneously, quantitative data of color matching degree during the printing process is acquired and the changes of quantitative data are recorded. If the peak value of the quantitative data of color matching degree exceeds the range area peak value, or the fluctuation frequency of the quantitative data of color matching degree exceeds the fluctuation frequency threshold, it is inferred that the printing process is abnormal and the corresponding printing process is marked as an abnormal printing process. If the peak value of the quantitative data of color matching degree does not exceed the range area peak value, and the fluctuation frequency of the quantitative data of color matching degree does not exceed the fluctuation frequency threshold, it is inferred that the printing process is normal and the corresponding printing process is marked as a normal printing process.

5. The online color correction method for digital printing based on machine vision according to claim 4, characterized in that, Color correction is performed based on the various time periods within the printing process; abnormal spraying and printing processes are identified, and color correction is performed based on the corresponding time periods.

6. The online color correction method for digital printing based on machine vision according to claim 1, characterized in that, The printing anomaly correction and detection process is as follows: The printing anomaly correction is detected. During the printing anomaly correction execution phase, the deviation value between the corresponding jet pressure adjustment amount and the actual jet pressure compensation amount of the printing equipment printhead is obtained. At the same time, the deviation value between the drying speed of the ink characteristic parameters equipped with the printing equipment printhead and the drying speed required for printing on the substrate is obtained. If the deviation between the spray pressure adjustment amount corresponding to the print head of the printing equipment and the actual spray pressure compensation amount exceeds the pressure deviation threshold, or if the deviation between the drying speed in the ink characteristic parameters of the print head of the printing equipment and the drying speed required for printing on the substrate exceeds the drying speed deviation threshold, then the current printing abnormality correction process will be marked as an inefficient printing correction stage. If the deviation between the adjusted spray pressure and the actual compensation of the spray pressure of the print head does not exceed the pressure deviation threshold, and the deviation between the drying speed and the required drying speed of the substrate in the ink characteristic parameters of the print head does not exceed the drying speed deviation threshold, then the current printing anomaly correction process is marked as the high-efficiency printing correction stage.

7. The online color correction method for digital printing based on machine vision according to claim 6, characterized in that, The process for correcting and detecting abnormal coating conditions is as follows: During the abnormal spraying correction stage, the maximum deviation of the ink absorption speed of the substrate when receiving ink of different colors is obtained. At the same time, when the ink corresponding to the maximum deviation of the ink absorption speed is successively sprayed, the overlap time between the abnormal fluctuation stage of the real-time printing environment temperature and the ink spraying stage is obtained. If the maximum deviation of the ink absorption speed when the substrate receives ink of different colors exceeds the maximum deviation threshold, or if the overlap time between the abnormal temperature fluctuation stage of real-time printing environment and the ink spraying stage exceeds the overlap time threshold, then the current spraying abnormal correction will be marked as the spraying correction inefficient stage. If the maximum deviation of the ink absorption speed when the substrate receives ink of different colors does not exceed the maximum deviation threshold, and the overlap time between the abnormal temperature fluctuation stage of real-time printing environment and the ink spraying stage does not exceed the overlap time threshold, then the current spraying abnormal correction is marked as the high-efficiency spraying correction stage.

Images