EP1958775A2 - Colour decomposition correction procedure - Google Patents

Abstract

The method involves photoelectrically scanning a printed sheet in a point by point manner, and forming color and/or density measured values from scanning signals from scanned points such as image data. The scanned points having brightness, which exceeds a threshold value, are identified. The measured values are corrected based on the identified scanned points. The scanning signals of a part of the scanned points lying within a measuring field are averaged during the formation of the color and/or density measured values for the measuring field on the printed sheet.

Description

The invention relates to a method for correcting the measurement error caused by color-splitting-related surface effects in the color measurement of a printed sheet on a running printing device according to the preamble of the independent claim.

In the document WO2005108083A1 a method and a measuring device for determining color and / or density values for the monitoring and / or control of the printing process in a printing device, especially a sheetfed offset printing machine described, wherein measuring fields of a sheet during the printing process directly in or on the current printing device photoelectrically measured and from the measured values obtained, the color and / or density values for the relevant measuring fields are formed.

In this method, the measured values are detected directly during the printing process with a measuring arrangement which is located inside the printing device - e.g. a sheetfed offset press or generally a printer - is installed. This type of measured value acquisition or measurement is referred to below as "inline". In contrast, "externally" refers to a measured value acquisition outside the printing device in a stable state of the printed product.

Inline measuring technology is significantly more complex than conventional external color measurement technology. The inline measurement must be carried out shortly after the paint application. At this time, the color layer is not yet stable. It is influenced by various printing process parameters and color properties, which decay with different time constants. Depending on the situation, this can result in large differences between the inline measured values and corresponding external measured values on stable dry samples. In addition, the process dependence complicates the interpretation of the measurement data. It is not clear if one measured variation was caused by a change in the color order or by a change in the process parameters. Inline and external measured values are therefore not directly comparable.

The differences between measured values determined inline and externally must be corrected for the practical utilization of the inline measured values. This correction is commonly referred to as color break or color-split correction.

The document WO2005108083A1 describes a solution for this problem. It consists of a special design of measurement technology and measurement geometry in conjunction with computational correction methods for the inline measured values, which enable a conversion into standard-compliant color and density measured values for corresponding stable external samples (printed products). The known approach thus includes a metrological component and a computational component.

The aim of the metrological component is to maximally reduce the influence of the process-dependent disruptive effects and to provide as unambiguous as possible measured values. The remaining measured value deviations from externally determined, standard-compliant measured values are then compensated by means of numerical correction measures or models.

A preferred realization of the mentioned metrological correction component is according to W02005108083A1 the use of polarizing filters in the illumination and receiver channels of the measuring head. The polarization filters consist of linear polarizers and are installed with mutually perpendicular polarization axes in the illumination and receiver channel.

Such an arrangement of polarization filters eliminates or at least greatly attenuates the noise component that results from the surface effect. The likewise in W02005108083A1 described further measurement errors (especially errors due to surface modulation of the disturbed ink layer) are not attenuated or eliminated by such a measure. The mathematical correction components are used for this.

The almost complete elimination of the measurement errors as a result of surface effect by an arrangement with polarizing filters from a metrological point of view is an inherently good solution. In combination with effective computational correction components in particular the negative influence of color splitting on the inline measured values can be turned off.

Although the use of polarizing filters as part of a metrological correction component is very effective, the use of polarizing filters is also associated with various disadvantages. A particular disadvantage is the strong attenuation of the light level. Also space problems, reduced optical imaging quality and costs are other negative factors.

In a non-imaging measurement technique, as well as in WO2005108083A1 is mainly described, a computational correction of the surface effect is most not practical because the disturbance of the ink layer immediately after the pressure at the location of measurement is typically still so strong that the surface effect that the measuring system sees is very pronounced. Especially with dark surfaces, the disturbing signal component, which comes from the surface effect, is large in comparison to the actual useful signal. Since the surface effect is an additive disturbance, a computational correction must ultimately be subtractive on the overall signal. Since, in particular, the printing process shows fluctuations and thus also the measurements fluctuate, such a subtractive correction is problematic or not practicable.

The present invention is intended to provide an alternative method for measuring value correction in inline measurements which does not have the disadvantages of known methods of this type described above. The purpose of the measured value correction is to minimize the differences between inline measurements and corresponding external measurements.

This problem underlying the invention is achieved by the measures listed in the characterizing part of the independent claim. further developments and particularly advantageous embodiments of the invention are the subject of the dependent claims.

The basic idea of the invention is not to correct the surface effect as in W02005108083A1 using a polarizer, but purely computationally based on image data of a line or area camera, which is mounted in the printing press and watching the just printed paper.

The method according to the invention is based on the finding that the surface effect caused by the color breakage or the color splitting, which adds a noise component to the measurement, occurs very locally and with an overall very low area coverage. If the print image is scanned with a camera, the disturbing effects in the camera image are recognizable as individual fine but at the same time extremely bright spots. The basic idea of the invention consists in identifying these extremely bright spots in the image and eliminating their influence on the measurement result. This can be done by not taking into account these bright points for the formation of the measured values of interest or that the area ratio of these points on the total image and their average brightness is calculated, whereby the surface effect can be determined quantitatively and used to correct the measurement result. The method according to the invention thus represents an image-based correction of the surface effect.

The method of the present invention requires an imaging sensor (e.g., a line or area camera) which is mounted so that it can record the print image in a suitable location at a high quality. In a sheet-fed offset printing press, for example, here comes almost exclusively the last printing unit in question, where the measuring location is then on the printing cylinder. However, the method can also be implemented with a point-by-point scanning sensor if it has a sufficiently high spatial resolution.

Particularly useful is the image-based correction according to the invention in measuring systems which work exclusively with imaging sensors. Such a measuring system includes, for example, a line scan camera, which just printed sheet only a few centimeters after the nip still observed on the printing cylinder. For the needs of print control, such a camera is equipped with suitable light sources and / or filters which define a meaningful for the pressure control spectral sensitivity of the camera. A suitable for the needs of the pressure control camera is also equipped with at least three separate spectral sub-areas, such as cyan, magenta, yellow density. Especially with such imaging measuring systems, the equipment with polarizing filters is not easy and therefore a computational correction makes particular sense.

If such a camera system is used for printing control (i.e., in particular for the semi or fully automated control of the actuators in the color boxes of the printing units), this printer will continuously create images of the so-called colorimeter in the printing operation and transmit them to a measuring computer. The measuring computer extracts therefrom, if not already done in an earlier processing stage, from the overall picture the image of the colorimeter and then finally the images of the individual measuring fields of the stiffness. The measurement variables of interest for pressure control can then be determined from these partial images. This is advantageously done by suitable averaging of the measured values of the individual pixels within a measuring field.

The measured values thus determined in the current printing machine are, however, now subject to the errors mentioned at the beginning, which are caused by the color breakage, without the correction of which precise printing process control is not possible.

The method according to the invention now provides that image or sampling points with a particularly large surface effect proportion and / or the strength of the surface effect as a whole are determined on the basis of the image data of the sub-images which are to be used for printing control. This is done by identifying in the image data those image or sampling points whose brightness exceeds a certain threshold.

The correction can now be carried out in a particularly simple embodiment of the invention so that the identified bright pixels (and possibly also adjacent Pixels) can be excluded from further processing. This means that the above-described averaging of the measured values of pixels for the formation of measured values for pressure control then takes place without the image or sampling points falsified by the strong surface effect.

In a further embodiment of the invention, following the identification of the bright pixels, their area fraction in the image of the measurement field is determined (number of particularly bright pixels based on the total number of pixels in the measurement field). In addition, its average brightness is preferably calculated. This allows the surface effect to be determined quantitatively. The quantitative knowledge of the surface effect is of particular interest because it enables the error-prone measurements of a second non-imaging system to be corrected. It is conceivable in particular a combination of an imaging subsystem for image inspection (ie, for example, for image defect detection, but not for color control of the printing press) with a non-imaging subsystem (eg punctiform spectrophotometer) for color control of the printing press (see. Fig. 4 ).

The described corrections correct only the measurement error component caused by the surface effect caused by the color cleavage (color break). The corrections described thus represent a kind of "virtual" polarization filter. In order to optimally correct the total error caused by the color splitting, in practice even more computational corrections, as described in the document WO2005108083A1 are described performed. However, these are not the subject of the present invention and therefore require no further explanation.

But it is not just the system just described with a line or line scan conceivable. The inventive method is not limited to the use of imaging measuring systems (line or area camera). For the scanning of the sample and a non-imaging measuring system can be used, as for example in the document WO2005108083A1 is described. Particularly useful for the inventive method is the combination of non-imaging color measurement sensors (eg, point measuring devices with Spetralphotometern) with imaging Sensors (eg a camera system for image inspection). The image information of the imaging sensors (the camera) can then be used to correct the measurement error (caused by the surface effect) of the non-imaging sensors in the manner described above.

When using imaging sensors, the image information can also be used to detect faulty measurement fields. In particular, so lubrication (especially when starting the printing press) and defects (slugs) can be detected in the print image and so the corresponding measurement fields are excluded from the Ducksteuerung. This is particularly useful when using a combination of non-imaging and imaging sensors.

Fig. 1

eine schematische Darstellung des erfindungsgemässen Verfahrens,

Fig. 2

ein Blockschema einer Ausführungsform des Verfahrens,

Fig. 3

eine Skizze zur Verdeutlichung des Verfahrens und

Fig. 4

ein Blockschema einer weiteren Ausführungsform des Verfahrens.

In the following the invention will be explained in more detail with reference to the drawing. Show it:

Fig. 1

a schematic representation of the inventive method,

Fig. 2

a block diagram of an embodiment of the method,

Fig. 3

a sketch to illustrate the method and

Fig. 4

a block diagram of another embodiment of the method.

According to Fig. 1 a printed sheet B located on the last printing cylinder Z of a printing press is scanned pointwise photoelectrically by means of an imaging scanning device in the form of a line camera K. By scanning the printed sheet line by line, a two-dimensional image is generated. The image data obtained in the form of the scanning signals of all sampled image or sampling points are represented by the box 1. Suitable imaging scanners are known and not the subject of the present invention.

From the scanning signals color and / or density measured values are formed in a conventional manner, which can be used for example for the control of the printing press. In particular, the scanning signals of pixels from the measuring fields are thereby a color control strip usually present on the sheet evaluated. The totality of the color and / or density measurements is represented by box 3.

According to the invention, in the formation of the color and / or density measured values, a correction of the surface effects caused by the color splitting (the color break) is undertaken. This is symbolized by the box 2.

Subsequently, further, eg from the W02005108083A1 made known numerical corrections, which correct the other, not on surface effects attributable to color-fading-related measurement errors. This is represented by box 4.

Finally, as a result of all the correction steps, color-split-corrected color and / or density measured values are available (box 5).

In Fig. 2 is a simplest embodiment of the inventive correction process shown in more detail.

From the image data (box 1) those sampling points are identified, the brightness of which reaches a certain limit I _G ( Fig. 3 ) (box 2a). These identified sampling points are then filtered out (box 2b), ie the sampling signals of these identified sampling points are not taken into account for the subsequent formation of the color and / or density measured values (box 3).

The Fig. 3 shows a section of a typical measured brightness distribution over the sampling points of a sheet. The abscissa indicates the individual sampling points i, the ordinates their brightnesses I. The brightness of most sampling points is below the limit value I _G. Where the surface of the ink layer on the printing sheet is disturbed by the color break, extremely bright sampling points are measured. However, the number or area ratio of these extremely bright sampling points is relatively low. Sample points lying above the limit I _G brightness are, as already said, filtered out or not for the formation of the color and / or Density measured values taken into account. The determination of the limit value I _G is performed empirically on the basis of test measurements in such a way that inline measurement results and external measurement results are as comparable as possible.

In Fig. 4 a variant of the method according to the invention is shown, in which a combination of a line camera K and a point measuring device S, in particular a spectrophotometer, is used to scan the printed sheet B.

From the scanning signals of the line camera K (image data 1), color and / or density measured values (box 3) are formed and used for the image inspection (box 6). Previously, as based on the Fig. 1 described a correction of the surface effects performed (box 2).

The measuring signals of the point measuring device S are first preprocessed in the usual way (box 11) and then converted into color and / or density measured values (box 13). In this case, a correction of the surface effect is performed (box 12), wherein the evaluation of the image data of the line scan camera is used. Subsequently, further numerical corrections for the remaining measurement errors (box 14) and finally the output of the corrected color and / or density measured values (box 15) take place, for example. for the purpose of printing machine control.

Claims (5)

Method for correcting the measurement error caused by color-splitting surface effects in the color measurement of a printed sheet on a running printing device, wherein the sheet is scanned pointwise photoelectrically and color and / or density measured values (3) are formed from the scanning signals (1) of the sampling points, characterized in that those sampling points are identified which have a brightness which exceeds a limit value (I _G ), and in that the measured values are corrected on the basis of the identified sampling points. A method according to claim 1, characterized in that during the formation of a color and / or density measurement value (3) for a measurement field located on the print sheet, the sampling signals of at least a portion of the sample points lying within the measurement field are averaged. Method according to one of the preceding claims, characterized in that the correction of the color and / or density measured values takes place in that the scanning signals of the identified sampling points are not taken into account in the formation of the color and / or density measured values. Method according to one of the preceding claims, characterized in that the surface portion of the sampling points determined by the limit of exceeding brightness and from the size of the surface effect is determined. Method according to one of the preceding claims, characterized in that an imaging scanning system, preferably a line or area camera (K) is used for the identification of the sampling points with the limit of excess brightness.

Images