Cross-machine color preset configuration

Abstract

The invention relates to a method for color adjustment of a printing press (3) by means of a computer (1, 2), wherein measurement values of printed products produced in the printing press (3) are detected by means of a color measuring instrument, and the computer (1, 2) analyzes the measurement values and, on the basis of the analysis, adjusts the color of the printing press (3) for processing a print job, characterized in that the computer (1, 2) analyzes the ink-up preset configuration of the printing press (3) and of additional printing presses by means of a data model (7), whereby for each print job of the printing press (3) a color preset configuration is calculated for all color separations, and by means of which the color of the printing press (3) is adjusted.

Description

TECHNICAL FIELD

[0001] The present invention relates to a method for automated color adjustment of a printing press by means of an artificial intelligence assisted data analysis approach.

[0002] The present invention belongs to the technical field of color adjustment of a printing press. BACKGROUND

[0003] In order to produce high-quality printed products, the ink-up pre-settings of the individual Farbauszüge of a print image must match the color acceptance of the respective Farbauszüge. Also, the amount of applied wetting agent has an influence on the print quality.

[0004] In order to match the ink-up, computer-supported printing press controllers in the prior art employ image data and so-called Farbvoreinstellungskennlinien. The ink-up is essentially composed of the opening of the Farbezone, the Farbestreifenbreite (i.e. the duration of the contact of the Farbwalze with the Farbwalze) and the Farbwalzhebertakt. The distribution in width is generated from the image data and these Farbvoreinstellungskennlinien provide the relationship between the desired Flächendeckung and the Stellung of the respective Farbezone and the Farbestreifenbreite.

[0005] Starting from the initial Farbvoreinstellungskennlinie, the machine controller continuously further optimizes this Farbvoreinstellungskennlinie based on stored rules; this is the so-called Farbvoreinstellungskennlinie training. This optimized Farbvoreinstellungskennlinie is however only applicable to one color in a specific printing group on a specific machine. This means that all influencing factors not contained in the image data are mapped holistically by this Farbvoreinstellungskennlinie.

[0006] In order to set the humidity, the machine currently does not provide a recommendation. In most cases, the setting of the humidity is from a previous job.

[0007] There is a series of influencing factors on the ink-up, for example: the customer ID, the machine ID, the position of the printing group in the machine, the history of the printing group (in particular in the form of cleaning status, wear status, etc.), the properties of the substrate, the properties of the color and others.

[0008] In other words, existing methods have various drawbacks. Firstly, new colors must be trained by performing several printing tasks on the machine. Secondly, besides image data, the inking characteristic curve only maps all other influencing parameters holistically for a specific printing press. Even within the machine, the differences in these mapped influencing parameters can only be considered very indirectly through so-called training (Lernen), and the internal differences within these parameters cannot be mapped at all because the characteristic curve represents an average of all these differences. Furthermore, different humidity requirements are not considered.

[0009] To address these shortcomings, color presets are often attempted by simply copying the characteristic curves, ignoring machine dependencies, or by using machine-dependent default characteristic curves, without considering color dependencies.

[0010] Furthermore, prior art provides a method for color adjustment in a printing press using a computer, wherein, within the scope of the current printing job, the measurement values ​​of the printed matter produced in the printing press are detected by a colorimeter, and the computer uses these measurement values ​​to create a characteristic curve. The computer then uses this characteristic curve for color adjustment in the next printing job, if appropriate. The characteristic curve is characterized by first being trained by the computer at the end of the current printing job, wherein the computer statistically analyzes the measurement values ​​detected up to that point in the printing press, along with additional parameters, and this analysis is considered in the training of the characteristic curve. The term "training" here is similar to the previously mentioned prior art and should be understood as: continuously adjusting these characteristic curves during the printing job setup phase and during the job execution to achieve a pre-set color configuration, and then storing these characteristic curves in their matched state at the end of the printing job. Summary of the Invention

[0011] The object of the present invention is therefore to provide an improved method for color presetting configuration of a printing press, which achieves less waste pages and less setup time compared to methods known in the prior art.

[0012] This objective is achieved through a method for color adjustment of a printing press using a computer. The method involves detecting measurements of printed matter produced in the printing press using a colorimeter, analyzing these measurements using a computer, and adjusting the printing press's color based on this analysis for processing printing tasks. The method is characterized by the computer analyzing the inking pre-setting configuration of the printing press and additional printing presses using a data model. For each printing task, a color pre-setting configuration (Farbvoreinstellung) is calculated for all color separations, and the printing press is color-adjusted using this color pre-setting configuration. The main difference between the method of this invention and previous technologies is that, in addition to analyzing existing inking pre-setting configurations using a self-learning algorithm (or data model), this method considers not only the current inking pre-setting configurations of the printing press to be applied (e.g., in the case of training and learning the color pre-setting configuration characteristic curves using the aforementioned prior art) but also the historical inking pre-setting configurations of other printing presses with the same or similar structures. Therefore, it is possible to automatically calculate the color preset configuration for each printing press using a trained data model, which is significantly closer to the rated color values ​​for each printing task than known methods to date.

[0013] Advantageous and therefore preferred improvements to this method arise from the preferred embodiments of the invention and the description with the accompanying drawings.

[0014] A preferred improvement of the method according to the invention involves a computer creating a color presetting configuration characteristic curve for color presetting configuration of all color separations and ink zones for each printing press, and performing color presetting configuration using this color presetting configuration characteristic curve. Several possible schemes exist for performing color presetting configuration on the printing press. The most known, however, is to use a color presetting configuration characteristic curve for all color separations of the printed image to be produced, which assigns an aperture (Öffnung) or adjustment angle (Stellwinkel) to each ink zone for each target color value to be achieved.

[0015] Another preferred improvement of the method according to the invention is that the computer performs color presetting configuration for all color separations by directly manipulating the ink zone positions in conjunction with the calculated rated coverage. Alternatively, color presetting configuration can be performed using color presetting configuration characteristic curves for all color separations and all (if any) ink zones, wherein the ink zone positions (i.e., adjustment angles) in each printing unit are adjusted for the desired rated coverage (or color value). This results in the ink zone positions having the same dependence on the desired rated color value (or rated coverage) in the final effect, but instead of applying the color presetting configuration characteristic curves for all color separations and ink zones stored in the computer, these values ​​are directly converted.

[0016] Another preferred improvement of the method according to the invention lies in that the computer calculates the position (Stellung) of each individual ink zone by means of a data model, taking into account all influencing factors, and thereby additionally considers the interaction (Interaktion) of adjacent ink zones. If direct manipulation of the position (Stellung) of the ink zones is achieved, then additional influencing factors (especially the influence between adjacent ink zones) can be considered. This is not possible when manipulating the position (Stellung) by means of a color preset configuration characteristic curve, because such characteristic curves are globally applicable to all ink zones of each color separation and printing task.

[0017] Another preferred improvement of the method according to the invention is that, as input parameters to the data model, these inking preset configurations include: historical color preset configuration characteristic curves, rated surface coverage of each ink zone, ink zone position, color band width, inking roller cycle time, moisture content, and the substrate (or printing ink) to be applied. These are the most common input parameters required for the data model based on historical color preset configuration data from all available printing presses. However, the list is not limited to these parameters, but may also include preset values ​​in other parameter forms not mentioned herein.

[0018] Another preferred improvement of the method according to the invention is that, as output parameters of the data model, for each printing task on the printing press, the data model recalculates: the color presetting configuration characteristic curve or ink zone position, rated coverage, color stripe width, inking roller cycle time, and moisture content. These output parameters thus correspond to values ​​that the trained data model outputs for each printing press's color presetting configuration. The data model outputs corresponding output parameters depending on whether the color presetting configuration should occur using the characteristic curve of the color presetting configuration or by direct control of the color presetting configuration. That is, in its trained state, the data model clearly knows, based on historical color presetting configuration data on the applied machine and on other additional machines, which printing press with which color presetting configuration, in the form of ink zone position and other aforementioned input parameters, has achieved which rated color value (or coverage).

[0019] Another preferred improvement to the method according to the invention lies in that the computer continuously adapts the data model to new sets of data by analyzing and processing log data from all available printing presses. This data model should be provided in a trained and learned state to the printing plant where the corresponding printing presses to be configured exist, thereby enabling immediate application of the data model. Advantageously, however, the data model is continuously improved in terms of color preset configuration using real-world data from the operation of each printing press, as well as additional printing press operation data available to the computer (on which the data model runs).

[0020] Another preferred improvement to the method according to the invention is that the data model responds to changes occurring on the color-adjusted printing press, as determined by a computer (particularly changes in machine characteristics due to service or wear). These changes in machine characteristics logically also affect the correlation between ink zone opening (or ink zone position) and the desired rated color value (or area coverage). Therefore, if these machine characteristics change, the data model must also adapt accordingly to these changing conditions in order to continue reliably configuring color presets.

[0021] Another preferred improvement of the method according to the invention is that the computer supplements the missing input parameters by means of a data model, taking into account empirical values ​​and current input parameters. Another advantage of using such a data model is that the trained data model can also handle incomplete data. For example, if a new color is to be applied on a known machine, the data model (even if data for this new printing color does not yet exist) can already provide very good preset configurations for each machine and the new printing color.

[0022] Another preferred improvement to the method according to the invention lies in that the data model analyzes the input data (especially the aforementioned input parameters) in a categorized manner regarding defined characteristics (such as customer characteristics, printing press characteristics, color characteristics, substrate characteristics, etc.). This involves the internal processes of the data model, namely, how to structure the input data so that it can be subsequently analyzed and processed. Preferably, the data model considers all of these categories. However, it is also possible to omit individual parameter categories (or types). However, the more categories omitted, the less specific the color preset configuration generated by the trained data model becomes. Attached Figure Description

[0023] The invention itself, as well as advantageous improvements in construction and / or function, will now be further described with reference to the accompanying drawings and at least one preferred embodiment. In the drawings, corresponding elements are given identical reference numerals. The drawings show:

[0024] Figure 1 The structural design of the printing press system used;

[0025] Figure 2 : A schematic flow diagram of the method according to the present invention. Detailed Implementation

[0026] The method according to the invention is used in a printing press 3, which is located in a defined workflow system 4. This workflow system 4 is exemplarily... Figure 1 As shown in the diagram, the workflow system 4 runs on one or more computers 1, which process the corresponding printing tasks. The printing task to be printed on the printing press 3 is analyzed by a computer regarding the necessary color pre-setting configuration for that task, and its color pre-setting configuration is matched to the printing press 3 to be applied. The color pre-setting configuration characteristic curve 5 thus created is then forwarded to the control computer 2 of the printing press 3 for corresponding color adjustments. Here, the computer can be the printing press 3's own control computer 2, which eliminates the need for dedicated forwarding, or the computer can be equivalent to the workflow computer 1 or another prepress stage computer.

[0027] Taking into account past data and additional influencing parameters, a data model 7 is now developed that provides the optimal inking pre-setting configuration for each color separation for each printing task. In a preferred implementation variant, model 7 is based on data 9 continuously collected across a large number of printing presses.

[0028] The inking preset settings are described here using the following data:

[0029] - Ink region position;

[0030] - Color strip width;

[0031] - Ink transfer roller cycle time;

[0032] - Moisturizing dosage;

[0033] - and others.

[0034] Furthermore, to simplify integration into existing infrastructure, the ink zone position is also described by the color preset configuration characteristic curve 5 and the rated surface coverage of each ink zone. The specific ink zone position is a point on the color preset configuration characteristic curve 5, which corresponds to the desired rated surface coverage extracted from the image data. A better characteristic curve 5 is generated by considering multiple influencing parameters, which contributes to reducing waste pages.

[0035] To date, the "trained" (i.e., stored at a defined time) color preset configuration characteristic curve 5 can only conditionally transfer between different printing presses 3 and, ideally, between structural series. However, this is inherent to the use of data model 7, which calculates new preset values ​​for each printing task after considering a large number of influencing factors. The color preset configuration characteristic curve 5, calculated for each color separation of each task, thus depicts the overall difference / overall variance of these influencing parameters. Figure 2 This document illustrates a method flow for color adjustment using a color preset configuration when applying the proposed data model 7 according to the present invention.

[0036] For each color separation in each task, Model 7 will provide the following output parameters:

[0037] - Color preset configuration characteristic curve 5 calculated separately based on all influencing parameters 6;

[0038] - Rated surface coverage;

[0039] - Color strip width;

[0040] - Ink transfer roller cycle time;

[0041] - Moisturizing dosage.

[0042] Model 7 can also handle incomplete data (6). An example of this is:

[0043] a) Calculations are performed for machine 3, which is unknown to date but has known settings (or color preset configuration characteristic curve 5). Since the customer is known and has operated multiple machines, and these machines are always running at high color density, this preference is automatically taken into account by model 7.

[0044] b) A new color is adopted on the known machine 3, which is described by its Lab value and, if necessary, by the rated layer thickness. Here, model 7 thus provides a very good color preset configuration taking into account all the characteristics of the specific machine 3 and the customer.

[0045] In another preferred implementation variant, model 7 is configured such that it always matches new data sets. For this purpose, all machines 3 with suitable logging configurations provide data through existing machine logs 8. Furthermore, model 7 will be able to react to changes, such as changes in machine characteristics due to service or wear.

[0046] In another implementation variant, data model 7 can directly provide the ink zone positions instead of the color preset configuration characteristic curve 5, based on which the color preset configuration is considered using the rated area coverage. This solution can also depict the interaction between adjacent ink zones because model 7 can output the position of each individual ink zone while taking all influencing factors into account.

[0047] This is not feasible in the preferred variant using color pre-setting to configure characteristic curve 5. The interaction between individual ink zones cannot be considered here because characteristic curve 5 applies globally to all ink zones for each color separation and task. Therefore, integrating the preferred variant using color pre-setting to configure characteristic curve 5 into the existing infrastructure is significantly simpler, and thus this becomes the preferred option.

[0048] In another preferred embodiment, model 7 inherently partitions its input data according to specific hierarchical relationships, resulting in parameters 6 that describe customer characteristics, machine characteristics, color characteristics, substrate characteristics, etc. All these input parameters 6 are considered in the preferred embodiment. However, alternatively, individual parameter types (e.g., machine characteristics) can be omitted. Thus, color preset configuration becomes increasingly globalized, approaching methods known in the prior art today, and therefore accuracy is reduced.

[0049] The preferred implementation variant with color preset configuration characteristic curve 5 can therefore achieve better integration into existing infrastructure and can more accurately depict the differences in these influencing parameters, thereby enabling better color preset configuration.

[0050] Generally, all the implementation variations mentioned improve color preset configuration by considering multiple influencing parameters, which means reducing waste pages. Until now, the "trained" color preset configuration characteristic curve 5 has only been conditionally transferable between different printing presses 3 and also between structural series. However, this is inherently achieved by a data model 7 that calculates new preset values ​​for each task while considering a large number of influencing factors.

[0051] In summary, the advantages of the method according to the present invention are as follows:

[0052] - Even less skilled operators can obtain the best color / humidity preset configuration.

[0053] - No manual intervention required.

[0054] - It can also be well preset to configure new colors.

[0055] - Cost savings due to fewer waste pages (or shorter installation and setup time).

[0056] - Improved customer satisfaction.

[0057] List of reference numerals

[0058] 1. Workflow Computer

[0059] 2. Control computer for the printing press

[0060] 3. Printing press

[0061] 4. Workflow System

[0062] 5. Color Preset Configuration Characteristic Curve

[0063] 6. Input parameters of the data model

[0064] 7 Data Model

[0065] 8. Printing press data log

[0066] 9. Other printing press data

Claims

1. A method for adjusting the color of a printing press (3) using a computer (1, 2), in, The measurement values ​​of the printed matter produced in the printing press (3) are detected using a color measuring instrument, and Computers (1, 2) analyze and process the measured values, and based on the analysis and processing, adjust the color of the printing press (3) for processing printing tasks. Its features are, Computers (1, 2) analyze and process the inking preset configuration of the printing press (3) and additional printing presses using data model (7), thereby calculating the color preset configuration for all color separations for each printing task of the printing press (3) and thereby adjusting the color of the printing press (3).

2. The method according to claim 1, Its features are, For the color preset configuration, the computer (1, 2) creates a color preset configuration characteristic curve (5) for all color separations and ink zones of each printing press, and performs color preset configuration using the color preset configuration characteristic curve.

3. The method according to claim 1, Its features are, Computers (1, 2) achieve color preset configuration for all color separations by means of direct manipulation of the ink zone position and the calculated rated surface coverage.

4. The method according to claim 3, Its features are, Computers (1, 2) use data models (7) to calculate the position of each individual ink zone while taking into account all influencing factors, and thereby additionally consider the interaction between adjacent ink zones.

5. The method according to any one of claims 1 to 4, Its features are, As input parameters (6) of the data model (7), the inking pre-setting configuration includes: historical color pre-setting configuration characteristic curve (5), rated surface coverage of each ink zone, ink zone position, color strip width, ink transfer roller cycle, moisture content, and the printing substrate or printing ink to be applied respectively.

6. The method according to any one of claims 1 to 4, Its features are, As output parameters of the data model (7), for the corresponding printing task on the printing press (3), the computer (1, 2) recalculates the color pre-adjustment configuration characteristic curve (5) or ink zone position, rated surface coverage, color strip width, ink transfer roller cycle and moisture content by means of the data model (7).

7. The method according to any one of claims 1 to 4, Its features are, The computer (1, 2) continuously adapts the data model (7) to new data sets by analyzing and processing data (9) from the logs (8) of all available printing presses (3).

8. The method according to any one of claims 1 to 4, Its features are, The data model (7) responds to changes obtained by the computer (1, 2) on the color-adjusted printing press (3).

9. The method according to any one of claims 1 to 4, Its features are, The data model (7) responds to changes in machine characteristics due to service or wear, obtained by the computers (1, 2).

10. The method according to any one of claims 1 to 4, Its features are, The computer (1, 2) supplements the missing input parameters (6) by taking into account empirical values ​​and existing input parameters (6) with the help of the data model (7).

11. The method according to claim 1, Its features are, The data model (7) analyzes and processes the input data in a classification manner regarding certain characteristics.

12. The method according to claim 10, Its features are, The data model (7) analyzes the input parameters (6) in a categorical manner regarding the defined characteristics.

13. The method according to claim 11 or 12, Its features are, The determined characteristics are customer characteristics, printing press characteristics, ink characteristics, and / or substrate characteristics.

Images