EP2127877B1 - Method for measuring the colour register in a printer - Google Patents

Description

The invention relates to a method for color register measurement on a printing press.

For process monitoring during printing, it is customary to provide printed control strips on the sheets or webs to be printed outside the subject with colored test patterns. These control strips, whose longitudinal direction is transverse to the transport direction of the printing material, comprise a longitudinally periodically repeating set of measuring fields, on each of which a specific parameter characterizing the print quality can be measured. Even during the movement of the printed product to be examined in the printing press, an image of at least part of the control strip is detected and evaluated.

In the DE 10 2004 003 613 A1 There are disclosed an apparatus for capturing an image of a predetermined portion of a moving printed matter and a method of operating such apparatus. The device comprises at least one camera with a two-dimensional electronic image sensor and a lighting device, which is directed to the observation area of the camera and suitable for its pulse-like illumination during the local residence of the predetermined section of the printed product. The illumination device consists of a multiplicity of individual light sources, the light of which has a different spectral composition due to different emission characteristics and / or filtering, and which are arranged in a regular pattern such that the light sources of each individual group form a regular arrangement whose illumination area is the observation area fully covers the camera on the printed product. By means of a control device, the individual groups of light sources can be sequentially switched on and off in such a way that the observation area of the camera can be illuminated with a sequence of light pulses of different spectral composition. The latter changes periodically and with each light pulse an image of the section of the printed product currently in the observation area of the camera is recorded.

By means of this device, a printing process can be monitored in the manner described in the introduction, by using as a predetermined section, the image of which is detected by the device, a region of the area printed with a suitable control strip Printed product is provided. The characteristic of the print quality, for the monitoring of which this device is designed and particularly suitable, is the color density.

Compared to the color density measurement, in which the light remission of an area covered by a printing ink is of interest, a significant, i.e. required by at least an order of magnitude higher resolution to gain measurement data of sufficient accuracy for the control of the color register of a printing press. Therefore specially designed measuring devices with a correspondingly high resolution are used for color register measurement and special measuring marks, which allow a simple determination of the color register position, are printed on the printed product. A high resolution color registration meter is a costly item of equipment of a printing press. Further, the color register marks consume a certain, albeit small, portion of the area of the printed product.

The US 2002/178952 A1 discloses a method of color register measurement on a printing machine wherein at least one image is taken of a printed test pattern consisting of a plurality of fields of full and / or halftone tones of the inks and containing a plurality of separate fields of each ink. For each printing ink, a positional dimension of the respective printing ink is determined, the color register position being determined on the basis of the position of the positional dimensions determined for the individual printing colors.

From the US Pat. No. 4,534,288 A and from the DE 197 38 923 A1 Further prior art is known.

The invention has the object of demonstrating a new and cost-effective solution for the measurement of the color register position.

This object is achieved by a method having the features of claim 1. Advantageous embodiments are specified in the subclaims.

The inventive method is characterized in that at least one image of a printed on a printed product test structure, which consists of a plurality of fields with full and / or halftone tones of the inks and contains a plurality of separate fields of each ink, is recorded that for every printing ink a position measurement is determined by averaging over positional dimensions of a plurality of fields of the respective printing ink, and that the color register position is determined relative to one another on the basis of the position of the positional dimensions determined for the individual printing colors.

This means that the function of the color register measurement is taken over by an already existing device for color density measurement and the color density test structure is used as a test structure. Neither a costly own measuring device for the Farbregisterlage is required, nor must be provided on the printed product's own Farbregistermarken. All that is required is an extension of the evaluation software running in the color density measuring device. The invention results in comparison to the conventional color register measurement a considerable simplification and a corresponding cost savings.

In order to achieve the accuracy required for a color registration measurement, despite the much lower resolution of a color density measurement device, averaging is performed over a plurality of fields of each ink. As a result, resolution-related local errors that would occur in an isolated measurement and evaluation on a single field of an ink, balanced on average, so that no special measuring device with higher resolution is required.

Since the remission of the different printing inks depends on the spectral composition of the illumination, it is expedient, with the availability of illumination whose spectral composition can be varied, to take several images of spectrally different illuminations from the test structure and to determine position dimensions for fields of the test structure use. In particular, the difference in the remission intensity of directly adjacent fields can also depend on the illumination, so that for the detection of a certain field boundary a particular illumination may be better suited than other illumination.

Since the color register setting is a relative quantity, the measured absolute positional dimensions of the individual printing colors are advantageously used to determine relative positional dimensions with respect to an absolute positional dimension of an ink selected as the reference color and the dimensions of the position of the color register are the deviations of these relative positional dimensions from the respective nominal values calculated.

In order to minimize the manual operating effort, it is particularly advantageous if the setpoint values of position dimensions and / or dimensions of the fields of the individual inks from the prepress are transmitted as a data set to the measuring apparatus used for carrying out the method, since the required data are available there anyway. In principle, however, it is also possible to manually input position data of the test structure, in particular on the basis of a test printed copy of the test structure using a digital measuring device.

One way to calculate coordinates of the positional dimension of an ink is to first identify those pixels that belong to the fields of the respective ink, and then to calculate a positional dimension of the ink by averaging over positions of these pixels. This is then represented by the center of gravity of all the pixels that enter into averaging.

As an alternative to the center of gravity calculation via the coordinates of individual pixels, coordinates of a positional dimension of an ink can also be calculated from the position of lines which were previously determined as field boundaries of fields of this printing ink. For example, a center of gravity coordinate of an ink can be calculated by calculating the corresponding centroid coordinate from respective boundary lines for each individual field of this ink and then averaging over the centroid coordinates of all of these fields.

For the determination of field boundaries, the location dependence of the remission intensity offers itself, wherein the existing extensive instrumentation of image processing for edge detection can be used.

Since a color density test structure is usually in the form of a contiguous strip extending transversely to the direction of movement of the printed product from a plurality of directly adjacent fields of different colors, gaps and overlaps between adjacent fields in their desired positions will occur in color registration errors in the longitudinal direction of such a strip. In determining field boundaries between such fields adjacent to each other in their nominal positions, gaps between adjacent fields can be identified from maxima of the remission intensity and overlapping areas of adjacent fields by minima of the remission intensity.

A location-dependent statement about the color register position can be obtained by evaluating a plurality of zones of the test structure separately across the width of the printed product transversely to its direction of movement and determining a plurality of respectively color-specific color register positions. This means that to determine a zone-specific color register location, not all the fields of each individual color present over the entire width of the printed product are used, but only those fields which are located in the respectively considered zone. In order for the gain in accuracy by the averaging is still given, this must be in each zone several fields of each color.

Fig. 1

eine streifenförmige Farbdichte-Teststruktur mit einem Farbregisterfehler in Bewegungsrichtung,

Fig. 2

eine streifenförmige Farbdichte-Teststruktur mit einem Farbregisterfehler quer zur Bewegungsrichtung und

Fig. 3

vergrößerte Ausschnitte aus dem Testmuster von Fig. 2 mit zugehörigen Verläufen der Remissionsintensität.

Hereinafter, embodiments of the invention will be described with reference to the drawings. In these shows

Fig. 1

a stripe-shaped color density test structure with a color registration error in the direction of movement,

Fig. 2

a strip-shaped color density test structure with a color register error transverse to the direction of movement and

Fig. 3

enlarged excerpts from the test pattern of Fig. 2 with associated courses of remission intensity.

Fig. 1 shows an example of a section of a test structure 1 for the color density control of a printing press. It has the shape of a stripe with a linear periodic sequence of contiguous fields 2, of which in Fig. 1 only three are marked at the left edge of the section shown with the reference number 2. Fields 2 each print a solid or halftone tone of an ink. The longitudinal direction of the strip-shaped test structure 1 runs in Fig. 1 in the x-direction, ie transversely to the direction of movement y of the printed product in the printing press.

The different colors or tones are in Fig. 1 symbolized by different hatching or dot patterns. For example, four-color printing can be full tones of black, yellow, magenta, and cyan inks. Furthermore, individual fields 2 may also be unprinted in the test structure 1 in order to serve as a white reference.

In Fig. 1 For example, three fields 2D, 2E, and 2F of test structure 1 are slightly offset from the common horizontal centerline of all other fields, reflecting a color registration error of the color printed in boxes 2D, 2E, and 2F in the negative y direction. The following describes how this color register error is determined according to the invention by means of the test structure 1 actually designed for color density measurement using a measuring device likewise designed for color density measurement. For this purpose, expressly again on the above-mentioned DE 10 2004 003 613 A1 referenced, which discloses a device suitable for this purpose and their operation in detail.

First, for color density measurement, two-dimensional images of the test structure are recorded at a plurality of spectrally different illuminations and evaluated as intended to determine the color density distribution over the width of the printed product. The images are determined to determine the color register location According to the present invention quasi-parallel to the color density evaluation of an additional evaluation in the same evaluation unit subjected, which must have appropriate reserves of computing power and storage capacity for the assumption of this additional task.

For this purpose, the evaluation unit must have data about the shape of the test structure 1, the sequence of the printing inks in the individual fields 2 and, in particular, about the desired positions of the individual fields 2 relative to each other. In the case of in Fig. 1 2, the fields 2 are all the same size and lie with perfect color register setting on a straight line running in the x-direction, so that the difference of the y-coordinates of the respective field centers should be zero and the difference of the x-coordinates of the respective field centers between adjacent fields should correspond exactly to the field width in the x direction. The target positions of the fields 2 are present in the prepress as data and can preferably be transmitted to the evaluation unit directly from there.

In order to calculate the color register positions of the individual printing inks, the recorded images are evaluated for the course of the remission intensity in both coordinate directions. In this case, both in the y-direction at the boundary between a printed field and the unprinted background, as well as in the x-direction at the border between two differently colored printed fields 2 each jumps in the measured remission intensity, said jumps between a printed field and the background are higher than between adjacent printed fields 2 and their height in each case depends on the spectral content of the illumination.

Based on the known remission behavior of the individual printing inks, a threshold or a range can be defined for each color with or without or above which the remission intensity must lie, so that a pixel can be clearly identified as belonging to a field of a particular printing ink , If necessary, several images taken at different illuminations can also be used for each to allow a clear distinction from the other inks on the basis of the measured remission intensity.

The image or images are evaluated by assigning each pixel based on its intensity value of a particular ink and the x and / or y-coordinates of all pixels of the same ink added and the respective average is calculated. Here, the averaging extends not only over a single contiguous field 2 of an ink, but over all fields 2 of the same ink. The mean values of the x and / or y coordinates that result in this way represent the coordinates of the common center of gravity of all the fields 2 of this printing ink.

For further evaluation, an ink is assumed to be the reference color and thus the common center of gravity of the fields 2 of this ink as a reference position, ie as the origin of a new coordinate system. Based on this, the coordinates of the centroids of the fields 2 of the remaining inks are calculated relative to this reference position. For example, the reference color in Fig. 1 printed in boxes 2A, 2B and 2C.

Finally, the deviations of the thus obtained relative coordinates of the centroids of the fields 2 of the other printing inks are calculated from respective desired values. If these deviations are all zero, the color register setting is perfect. Otherwise, these deviations indicate the position of the color register in terms of magnitude and direction and can be used to control the color register. In the example of Fig. 1 the center of gravity of the 2D, 2E and 2F fields lies below that of the fields 2A, 2B and 2C in the y-direction, while in the other y-fields it coincides with that of the fields 2A, 2B and 2C. Thus, there is only a color register error in the negative y direction for the color printed in boxes 2D, 2E and 2F.

As an alternative to the previously described method, the position of the fields of a color can also be calculated on the basis of edge detection. In contrast to the assignment of pixels to fields 2 of a certain color On the basis of the value of the remission intensity, the edge detection does not require a definition of characteristic thresholds or areas of the remission intensity, but only the recognition of sudden changes in the remission intensity is important.

This recognition is 1 in the test structure Fig. 1 particularly easy in the y-direction, as here only the transition between the printed area and the white background has to be detected. In the x-direction, the intensity jumps in the transition from a printed field 2 to the adjacent less pronounced, but present and detectable, and here the evaluation of multiple images with spectrally different lighting can be beneficial.

For the detection of edges in images, the state of the art in the field of image processing offers a comprehensive range of instruments, which can be used here to determine the field boundaries. This instrument is known to those skilled in the art and therefore needs no explanation at this point.

If the boundaries of all fields 2 are detected, then from the boundaries of each field its center of gravity coordinates are calculated. If the fields 2 are in the shape of squares like in Fig. 1 or rectangles, this is very simple. Based on the known sequence of the individual printing inks in the fields 2, the respective mean values of the center of gravity coordinates of all fields 2 of this color, ie the coordinates of the common center of gravity of all fields of this color, are then calculated for each individual ink.

Instead of first calculating the center of gravity coordinates of each individual field 2 of an ink and then averaging these center of gravity coordinates, an averaging over the coordinates of the field boundaries of the individual fields of an ink can also be performed first and then the common center of gravity calculation can be made on the basis of these already averaged field boundary coordinates.

If the coordinates of a common center of gravity are determined for the fields 2 of each printing ink, then the calculation of the color register position can proceed as in the previously described method with the calculation of the relative coordinates with respect to the center of gravity of a reference color. It is clear that this is the case in Fig. 1 The calculation of the y-coordinates of the centroids of fields 2 by their upper and lower edges in the x-direction obviously results in the same offset of fields 2D, 2E and 2F compared to fields 2A, 2B and 2C in FIG Negative y-direction results like a center of gravity calculation by adding up pixel coordinates.

It is essential that in both previously explained methods not only a single field of each ink is considered, but an averaging over a multiple fields of each ink comprehensive range is made. It will be understood that the larger the range, the more pronounced is the increase in accuracy due to averaging.

Although the previous statements have general validity, but in the case of a test structure 1 of in Fig. 1 shown with a linear sequence of directly adjacent fields 2 in a color register error in the longitudinal direction of the sequence, ie in the x direction of Fig. 1 , special effects that need to be considered. These effects are explained below Fig. 2 and Fig. 3 explained.

It shows Fig. 2 the same test structure 1 as Fig. 1 but with a color registration error in which fields 2D, 2E and 2F are offset from their nominal position in the positive x-direction, leaving gaps between these fields 2D, 2E and 2F and their respective left-adjacent fields 2A, 2B and 2C overlaps between these fields 2D, 2E and 2F and the fields 2G, 2H and 2J respectively adjacent to each other on the right side. Bottom right is in Fig. 2 an area comprising fields 2C, 2F and 2J is shown enlarged to illustrate gap 3 between fields 2C and 2F and overlap 4 of fields 2F and 2J.

In Fig. 3 2C, 2F and 2J shows the situation of a perfect color register setting (left) of the situation of a color register error in the positive x-direction (right), as already described in FIG Fig. 2 is shown opposite, wherein below the field comprising these fields of the test structure 1, the respective course of the reflectance intensity R for a given illumination over the x-coordinate direction is shown.

If all fields 2C, 2F and 2J have their respective set position, then at the field boundaries, as on the left side of Fig. 3 can be seen, jumps of the reflectance intensity of each characteristic height recognizable, the jump heights depend on the spectral composition of the illumination. On the basis of such jumps, the field boundaries are detected according to the invention, whereby, as stated above, this can be done either simply by means of threshold values or by means of one of the edge detection methods known in the field of image processing.

If there is a color register error in the x-direction, then, as on the right side of FIG Fig. 3 It can be seen that the resulting gap 3 between the fields 2C and 2F, also referred to as a blitzer in the present context, is noticeable by a pronounced narrow maximum of the remission intensity. At the same time, the overlap 4 of the fields 2F and 2J is manifested by a minimum of the remission intensity, which, although less pronounced than the maximum caused by the gap 3, is nevertheless clearly detectable. It is understood that these extremes in reality are not those in Fig. 3 schematically illustrated rectangular shape, but have a rounded shape. The range in which the reflectance intensity has the characteristic value for the field 2F is indicated by the in Fig. 3 on the right assumed color register error compared to in Fig. 3 shown on the left by the double of the color registration error, ie by the width of the gap 3 and the width of the overlap 4 is reduced.

By recognizing the two extremes and determining the coordinates of their flanks, it is possible to determine the x-coordinates of the field boundaries of fields 2C, 2F and 2J needed for the calculation of the x-coordinates of their centroids. The left field boundary of the field 2F is given by the right edge of the maximum, the right field boundary of the field 2F by the right edge of the minimum. The left edge of the minimum indicates the left field boundary of field 2J.

The occurrence of such maxima and minima due to gaps 3 and overlaps 4 must be evaluated in the evaluation of the course of the reflectance intensity in the x-direction, i. in the direction in which fields 2 of different printing inks adjoin one another directly, are taken into account when pixels are assigned to fields 2 of individual colors on the basis of the measured remission intensity, be it by means of threshold values or by means of edge detection. It is understood that this is due to the actually non-rectangular, but rounded course of the reflectance intensity over the x-coordinate with a certain blurring afflicted. By means of the averaging according to the invention over a plurality of periods of the periodic test structure 1, however, the accuracy of the measurement can be increased to a level sufficient for the regulation of the color register.

It is understood that the previously described calculation of the center of gravity of a printing ink as a positional measure is meant purely as an example. In principle, it would also be possible, for example in the case of rectangular fields 2, to determine the position of a specific corner in each case and to calculate the average values of the coordinates of the corresponding corners of all the fields 2 of this printing ink as the positional dimension of an ink. Such and similar variations of the position measurement and / or variations of the mean value calculation by interchanging operations in the context of mathematical transformations are part of the inventive idea and are intended to be encompassed by the protection of the claims.

Claims (8)

1. A method for measuring the color register on a printing press, wherein at least one image of a color density test structure (1), which is printed onto a printed product and which consists of a plurality of fields (2) comprising full and/or half tones of the print colors and which contains a plurality of fields (2) of each print color, which are separated from one another, is recorded by means of a color density measuring device, wherein a position measure is determined for every print color by average determination via position measures of a plurality of fields (2) of the respective print color, and wherein the color register position is determined by means of the position of the position measures, which are determined for the individual print colors relative to one another in such a way that the color density test structure (1) records several images with spectrally different illuminations and are used to determine position measures for fields (2) of the color density test structure (1), wherein the recorded images are evaluated with regard to a course of the remission intensity in order to calculate the color register positions of the individual print colors.
2. The method according to claim 1, characterized in that relative position measures with regard to an absolute position measure of a print color, which is selected as reference color, are determined from absolute position measures of the individual print colors, and that the deviations of these relative position measures from respective setpoint values are calculated as measures for the position of the color register.
3. The method according to claim 1 or 2, characterized in that setpoint values of position measures and/or dimensions of the individual fields (2) of the color density test structure from the pre-printing stage are transmitted as data set to the measuring device, which is used to perform the method.
4. The method according to one of claims 1 to 3, characterized in that at least one coordinate of the position measure of a print color is calculated by an average determination via positions of a plurality of image points, which had previously been identified as belonging to the fields (2) of this print color.
5. The method according to one of claims 1 to 4, characterized in that at least one coordinate of the position measure of a print color is calculated from the position of lines, which had previously been determined as field limits of fields (2) of this print color.
6. The method according to claim 4 or 5, characterized in that the identification of the association of image points to fields (2) of a print color and/or the determination of filed limits occurs by means of the location dependency of the remission intensity.
7. The method according to claim 6, characterized in that gaps (3) between adjoining fields (2C, 2F) are identified by means of maxima of the remission intensity and/or overlap areas (4) between adjoining fields (2F, 2J) by means of minima of the remission intensity in response to the determination of field limits between fields (2C, 2F, 2J) adjoining one another in their setpoint positions.
8. The method according to one of claims 1 to 7, characterized in that a plurality of zones of the color density test structure (1) is evaluated separately via the width of the printed product at right angles to its direction of movement, and several respective zone-specific color register positions are determined.

Images