DE102009051197A1 - Method for correction of register error in e.g. inkjet printing process in ink-jet printing system, during production of book, involves calculating target-bit maps, where portion of target bit maps corresponds to deformation factor - Google Patents

Abstract

The method involves processing data of a raster graphics by one of printing images along an image processing axis (BA) of the raster graphics for avoiding a register error, where the image processing axis corresponds to a shrinking axis (SA). Control data is formed from the raster graphics. Target-bit maps (31) are calculated from adjacent source-bit maps (30) according to a pre-determined calculation rule. A portion of the processed target-and source bit maps along the image processing axis corresponds to a deformation factor. Independent claims are also included for the following: (1) a device for correction of the register error in the printing process (2) a computer program product for correction of the register error in the printing process.

Description

The invention relates to a method, a device and a computer program for correcting a registration error caused by deformation of the recording medium in a printing process.

In printing processes in which a recording medium is subjected to a temperature treatment, the recording medium usually deforms. If, for example, the recording medium is printed several times in the same printing unit or by different printing units during the printing process and a heat treatment occurs between the printing operations, then the problem often arises that during the second printing operation the image to be printed appears larger due to the shrunken recording medium as the picture printed in the first printing. This results in register errors, which are disturbing, for example, when the recording medium is printed on both sides and appears due to the shrinkage effect of the back printing area larger than the front-side printing area. Especially in book production, such a register error is disturbing and incompatible with a high standard of print quality. Even in color printing, in which separations are successively printed in different printing units and a corresponding shrinkage effect occurs between the printing units, the color quality can suffer from the fact that the color separations of an image are not printed in register.

In many printing processes paper is used as a record carrier. In this case, the effect that the moisture naturally contained in the paper by a heat treatment escapes from the paper and shrinkage of the paper takes place both in the direction of the paper fibers and across the paper fibers, more or less strong. The thickness depends, for example, on the direction of the fiber, the type of fiber or paper, the storage conditions of the paper and other conditions in the printing process, such as the ambient humidity, the temperature, etc.

Such a shrinkage effect can occur in any type of printing process, especially in electrographic printing processes and in ink-jet printing processes. In printing processes using water-based inks, the paper's shrinkage effect may be particularly pronounced due to the high moisture absorption of paper as it prints with the ink, and the subsequent drying process for the ink.

Paper is a hygroscopic material that releases water vapor when heated and restores water when it is re-connected. The heat-related shrinkage may be in printing processes of the type mentioned typically 1 to 2%, but also more or less. For example, with a paper width of 20 "and a shrinkage factor of 1%, the shrinkage in the paper width direction is about 5 millimeters. In the course of cooling and storing water, paper can also expand.

In ink-jet printing systems, where a paper web is printed in two printing units and the paper or ink is dried by heat in a drying process after the first printing operation, water is substantially removed from the paper, causing it to shrink. As a result, register deviations between the information printed by the two printing units occur as described above. In such printing devices then it comes in addition to problems if they are designed to print the full paper width, that is, when a so-called edge-to-edge printing is done. The shrinkage of the paper then leads in the region of the second print head to the fact that the information on the outermost edge of the picture can no longer be printed on paper, but the intended ink is sprayed unused past the paper. As a result, areas of the printing unit, such as storage parts, pressure rollers, etc., can also be contaminated, which necessitates cleaning processes. Furthermore, as a result, later processed print material that passes through this printing unit, so heavily polluted that it must be discarded as waste. A significant risk is that important print data of the edge area are not printed on the print material and thus the respective printed document is not complete.

The above-mentioned problems are particularly troublesome when the printing process is carried out by means of a printhead in which a discrete number of writing elements effect the printing process, for example light emitting diodes arranged in a line or matrix for exposing an electrophotographic element or arranged in lines and / or columns inkjet nozzles.

From the DE 10 2007 040 402 A1 a method for register correction in paper shrinkage is known, in which in the second printing a printing element line relative to a printing element line of a first printing operation with respect to a given axis of the paper movement direction is rotated by an angle such that the distance between the lines with the printing elements is reduced in size. As a result, an image shrunk corresponding to the paper shrinkage is written on the paper web in the second printing operation. The disadvantage of such a method is z. B. that mechanical adjustments are necessary on the print head according to the degree of shrinkage, which are relatively expensive.

In the DE 10 2007 040 402 A1 For duplex printing, it is further proposed to shift print frames of the front and back print images against each other so that their central axes coincide. This can cause the front and back sides to lie centered over one another and the register errors to be distributed symmetrically on both sides of the edges of the printed images, ie to be quasi averaged out. In the case of a printing process which is line by line on a lateral edge of the recording medium, this also means that the register error halves at the end of the line. However, complete compensation of the register error can not take place as a result. For large print widths of z. B. 30 '' and a degree of shrinkage of 1% at the edges along the shrinkage axis still leads to a two-sided registration error of about 4 millimeters, which is not acceptable in many printing applications.

From the WO 2005/031470 A1 For example, a method of correcting paper shrinkage is known that starts with the generation of a bitmap. In this case, to produce an image for the first printing operation or for the second printing operation, the number of pixels to be considered for printing is changed in accordance with the expected shrinkage in the direction of an axis. For this purpose, it may be provided to enlarge the screened and dithered image for the first printing operation by inserting pixels in accordance with the degree of shrinkage or to reduce the image for the second printing operation by deleting pixels in accordance with the degree of shrinkage. A disadvantage of the method described there is that it can lead to image distortions or artifacts by the insertion or omission of entire pixels in the bitmap.

2 schematically shows the in the WO 2005/031470 A1 described method. A bitmap 1 in which a binary picture 2 is stored, becomes a shrinkage compensation process 3 subjected. The process takes place along an image processing axis BA in which an image column 6 is deleted, causing the picture 2 to a picture 7 is shrunk. The shrinkage factor achieved in the process or incorporated into the process 4 corresponds to a shrinkage factor of the recording medium 11 in the printing process along its shrinkage axis SA. Deleting remains in the destination image 7 still the neighboring column 6a , The remaining pixels in this area, however, no longer form squares, but only rectangles, that is, there is an artifact in the target image 7 , The blank edge area 9 right of the target image 7 however, is one column larger than the unwritten edge area 8th of the source image showing the image shrinkage.

With the data of the target image 7 then becomes the printhead 10 controlled, the paper web 11 printed. The shrinkage axis SA corresponds to the paper web 11 the image processing axis BA.

From the US 4,721,969 A is an electronically-based method for correcting register errors in electrographic color printers known in which registry errors of the color separations is determined by register marks, which are detected with CCD sensors.

The above publications are hereby incorporated by reference into the present specification.

It is an object of the invention to provide a method, a device and a computer program for correcting a register error in a printing process, which work on the one hand on the basis of electronic image processing and on the other hand cause as little disturbing artifacts of the printed images.

This object is achieved by the invention described in the independent claims. Advantageous embodiments of the invention are specified in the subclaims.

• – aus jeweils n benachbarten Quell-Bildpunkten wird nach einer vorgegebenen Rechenvorschrift ein Ziel-Bildpunkt berechnet, wobei n eine ganze Zahl und größer als 1 ist,
• – der Anteil der verarbeiteten Punkte entlang der Bildverarbeitungsachse entspricht dem Verformungsfaktor.

According to a first aspect of the invention, a registration error is corrected in a printing process in which data for images to be printed are subjected to a raster image process, by which in each case at least one raster graphic structured in rows and columns is generated. Control data for at least one printing unit are formed from the raster graphics. Furthermore, a recording medium is printed in a first printing process with the control data from a first raster graphics with a first printed image and then a treatment, by which he and the first printed image deform at least along a deformation axis. Then, the recording medium is printed a second printing operation with corresponding control data from a second raster graphics with a second printed image. In order to avoid register errors between the two print images, the data of the raster graphics are processed by at least one of the two print images along an image processing axis of the raster graphic corresponding to the deformation axis, where:

• From each n adjacent source pixels, a target pixel is calculated according to a predetermined calculation rule, where n is an integer and greater than 1,
• The proportion of processed points along the image processing axis corresponds to the deformation factor.

According to this aspect of the invention, it has been recognized that by calculating a target pixel from a plurality of adjacent source pixels, enhancement of shrinkage correction can be made because the target pixel is not merely deleted by completely erasing information of the source image it contains information from several neighboring source pixels. It can thereby be achieved that the information of each source pixel is preserved, at least partially or partially, in the target pixel and thus fewer artefacts arise. On the one hand, this type of shrinkage correction can be implemented by software in the generation of a bitmap for controlling a printing unit. In this case, two pixels or two columns of the source image are each combined to form a pixel or a column of the target image. Artifacts, in particular low-frequency repetitions of the same structures, which are disturbing to the eye, can thereby be avoided.

The deformation of the recording medium can also be regarded as a change in length along the deformation axis. The treatment of the recording medium may in particular be a heat treatment. The record carrier can shrink in particular. The deformation axis can then also be referred to as a shrinkage axis. However, the recording medium may also expand, depending on its material, for example metal, and the type of treatment, for example a moistening of paper. Instead of reducing the size of the images in the destination bitmap relative to the source bitmap, a corresponding enlargement can also be provided.

The calculation of a target pixel from a plurality of source pixels of the raster graphics can be particularly effective when the source pixels each comprise a plurality of data bits, for example, each 8 bits. From corresponding gray levels of the source pixels can then calculate a new gray level for the target pixel. From a corresponding source bytemap, a corresponding destination bytemap can then be calculated. This applies in particular to a printing process in which print data are first subjected to a raster process (RIP process) in which rasterized image data are generated from the source data, the rasterized data each comprising several bits, and then a so-called screening, dithering or Halftoning process are subjected, in which the rasterized image data pressure-specific converted into binary image data. The image processing process according to the first aspect of the invention is particularly beneficial when performed after the RIP process and before the screening, dithering or halftoning process.

It is also advantageous that the deformation correction can take place after the RIP process and thus can be decoupled from it.

The calculation rule applied to the raster graphics may comprise an image processing window in which a given weighting for n × p encompasses adjacent and, if appropriate, pixels enclosing the respective source pixel, where n and p are integer numbers greater than zero. In particular, the calculation rule can provide an arithmetic mean of adjacent source pixels.

With the invention it can be provided as an example that images of an original size are used in the first printing process and images compressed in accordance with the shrinkage factor in the second printing process. However, it can also be provided to generate an image of an original size in the second printing process and to produce an image which is enlarged in accordance with the shrinkage factor in the first printing process.

According to a second aspect of the invention, which can also be viewed independently of the first aspect of the invention, a registration error is corrected in a printing process in which data for images to be printed are subjected to a raster image process, by which a raster graphic structured in rows and columns is generated becomes. Control data for at least one printing unit are respectively formed from the raster graphics, wherein a recording medium is printed with the control data of a first image in a first printing operation and the recording medium is subjected to a treatment by which the recording medium and the printed first image at least move along a deformation axis deform a deformation factor. The recording medium is printed in a second printing process with the control data of a second image. To avoid a registration error between the two printed images, the data of the raster graphic are processed by at least one of the two images along an image processing axis of the raster graphic corresponding to the deformation axis. The source pixels to be used in each case are selected stochastically from the source image transversely to the image processing axis.

By image processing according to the second aspect of the invention can be achieved that pixels in the course of the deformation correction for image reduction are not left out in exactly equidistant intervals, but are removed stochastically distributed. In this case, provision may be made in particular for a column and / or row to be selected within the matrix structure of the raster graphics, and for pixels to be stochastically distributed around a given or calculated pixel spacing around the respective row and / or column. A corresponding method with stochastic distribution can also be used to enlarge the raster graphics.

According to a third aspect of the invention, which can also be seen independently of the two aforementioned aspects, an easy-to-use method for determining the deformation factor of a recording medium after a printing operation is proposed in which the recording medium is first printed by a first printing process and then subjected to a treatment by which the recording medium and the printed first image deformed by the deformation factor at least along a deformation axis. It is provided that in the first printing process, a scale is printed and the same scale is printed in the second printing process. The recording medium is guided both in the first printing process 1 and in the second printing process 2 along a fixed reference edge, at which a predetermined point of the scale, preferably the zero point, is printed. After passing through the entire printing process, the two markings of the scale are then determined, which were printed on the opposite side of the recording medium along the deformation direction to be determined, in particular transversely to the reference edge. Due to the deformation of the recording medium, the scale printed in the second printing process will be finished at a small scale value than the scale printed in the first printing process. In the case of shrinkage, the second printed scale will "tear off" at this lateral edge of the recording medium. From the comparison of the two scale value, the deformation factor can be easily derived. This value can then be used for the above image processing processes.

From the third aspect of the invention, the determination of the degree of deformation or deformation factor can be further simplified by a calculation routine and a user interface are provided on the printing device, in which the user read only the value of the first scale and the value of the second scale and at the user interface must enter. The routine then calculates from both numbers the necessary factor for the strain compensation and can pass it directly to the routine performing the compensation on the raster graphics. In a still further automated embodiment, the scale values of sensors, for example CCD cameras, can be automatically detected and read, and the subsequent calculation of the deformation factor can also be carried out automatically.

For the second and the third aspect of the invention, in particular, the statements made to the first aspect apply.

Hereinafter, embodiments of the invention will be described. As far as elements of different figures are the same, the same reference numerals are used.

Show it:

1 A printing system,

2 A shrinkage compensation according to the prior art,

3 A first example of a shrinkage compensation according to the invention,

4 A second example of a shrinkage compensation,

7 a shrinkage compensation with stochastic selection of pixels,

6 Components of a computer system and

7 A method for shrinkage determination of a paper web.

In 1 is a printing system 12 shown at the print data from a data source 13 such as a client computer or a print server to a controller 14 be transmitted in which the print data for output on a printing device 15 be prepared and which is also essentially a computer. In the controller 14 the incoming print data are processed by means of a parser 16 interpreted. You can use any print data language provided by the controller 14 supported. Examples of such print languages are Advanced Function Presentation (AFP), Portable Document Format (PDF), Page Command Language (PCL) or Personalized Page Markup Language (PPML). The analyzed data is then sent to a raster image processor 17 (Raster Image Processor RIP) fed, in the pixel-based raster graphics (bitmaps) are generated. In order to be able to give up grayscale images and color images, each pixel of the bitmap has several bits assigned, for example, 8 bits for 256 gray levels. For color printing, per color separation (C, M, Y, K) such a bitmap (tarpaulin) may be provided. As part of the parsing process and the raster process is in the controller 14 determines which document page of the incoming data stream in the first printhead 20 and which part in the second printhead 21 to be printed. Before the data the screening module 19 are fed, in which bitmaps are generated from the raster graphics with only one bit per pixel and optionally color data are divided into grid-specific angle is in a shrinkage correction module 18 decided which raster graphics are reduced or enlarged. Accordingly, the raster graphics are processed page-specific. The change in the raster graphics can be done in an intermediate step between the screening and the screening (also called dithering or halftoning) (arrow 18a ) or in one go together with the screening (arrow 18b ). The processed data will be sent to the printer 15 fed and there on the two printheads 20 . 21 divided up. For printing, the paper web 11 from an unwinding roll 25 unwound, with the first printhead 20 by pointwise arranged ink jet nozzles with z. As water-based ink printed, the paper web 11 with ink in the first dryer 22 dried, turned and along transport direction TR to the second print head 21 transported, with the appropriate back side image by ink side by side on the paper web 11 is printed. In the second dryer 23 Also the back image is dried and the paper web 11 then on the take-up roll 26 wound.

The two printheads 20 . 21 may be substantially identical and housed in a common housing of a duplex printing device or in different housings of two individually operable, but coupled as so-called. Twin system printing systems. It is also possible that the first and the second printing operation are performed by the same print head on the front or back of the recording medium and the corresponding printing device contains a total of only one print head.

In 3 an example of a processing according to the invention of a raster graphics for shrinkage compensation is shown. On the left is a section of a source bitmap 30 shown in a raster process. Each of the small squares represents one byte in raster memory and represents a pixel of a primary color of the printer. The numerical value given in the squares corresponds to the byte value and indicates the respective tone value of the pixel. The value range goes from 0 to 255 or 0xFF, corresponding to discrete tone values (gray levels) from 0 to 100%. For simplicity, only three different tones were used, namely 0 (0%), 0x80 (50%) and 0xFF (100%). In the example shown is the source image 29 a homogeneous area with 50% halftone value corresponding to a gray tone or the value 0x80 of the corresponding source bitmap 30 , For illustration purposes are in the 3 to 5 For the pixels, only the values "80" for 0x80 and "F" for 0xFF are drawn.

On the right is the dithered matrix 32 represented by the source bitmap 30 the raster graphics after performing the shrinkage compensation 33 under the influence of a shrinkage factor and subsequent dithering with dither matrices 35a and 35b is produced. The dithered bitmap is for the printhead used 10 and may represent one bit of information (as shown) or even multiple bits per pixel if the printhead is capable of representing corresponding gray levels.

In the example shown, the shrinkage correction takes place along the image processing axis BA, which corresponds to the shrinkage axis SA of the paper 11 corresponds and in turn transversely to the transport direction TR of the paper web 11 lies. However, a shrinkage compensation can in principle also take place in accordance with the paper transport direction TR, that is to say perpendicular to the directions SA, BA.

To compress the raster graphic or the source image 29 along the image processing axis BA become the columns 30a and 30b the source bitmap 30 line by line combined in such a way that the arithmetic mean is formed from the two respective pixel values. This will make the two columns 30a . 30b in the destination bitmap 31 only one more column 31a , The quadratic source image formed by the pixels having the values 80 29 thus becomes the target image 36 the destination bitmap 31 Shrunk by exactly one column. The from columns 30a and 30b calculated target column 31a but has in the range of the target image 36 each the same gray values 0x80, because the corresponding gray values of the source pixels of both columns 30a . 30b also have the values 0x80.

Through the dither process 34 (also called screening or halftoning process) with the dither matrices 35a . 35b , which are only representatively represented here as matrices with 8 pixels each, but in fact can be much larger, become the data of the dither bitmap from the target bitmap 32 educated. This is thus in the range of the outermost right marginal columns 37 also compressed accordingly, so that the dithered image 38 exactly to the width b of the shrunken paper web 11 is reduced and to control the second printhead 21 can be used.

wobei ZP_ij der Zielwert der i-ten Zeile der j-ten Spalte ist, Q1_ij der Quellwert der i-ten Zeile in der j-ten Spalte (Spalte 41) und Q2_ij+1 der Quellwert der i-ten Zeile der j + 1-ten Spalte (Spalte 42) ist. At the in 4 The paper shrinkage compensation scheme shown schematically includes the source bitmap 30 Pixels that correspond to a solid (100% black) (value 0xFF) and the image information 2 adjacent capital letters "H" 40a . 40b contain. It is further assumed that a very critical area is selected to calculate the shrinkage, namely the column 41 , which forms the right leg of the left H. The shrinkage factor 4 accordingly, in the shrinkage compensation function 33 this embodiment for a destination in column 44 the arithmetic mean of each adjacent pixels of the selected column 41 and their right neighbor column 42 of the source image calculated line by line and according to the mapping function 43 on the column 44 the destination bitmap. The function for the arithmetic mean is then:

where ZP _{ij is} the target value of the i-th row of the j-th column, Q1 _{ij is} the source value of the i-th row in the j-th column (column 41 ) and Q2 _{ij + 1} the source value of the i-th row of the j + 1-th column (column 42 ).

In the column 44 Therefore, the target pixels of the right leg of the left "H" have the values 0x80. By this way of mapping two source image columns 41 . 42 to a target image column 44 Thus, it is achieved that the flank of the left "H" does not completely disappear, but is only displayed in a lower tone value. The image information is essentially preserved. Submitting the destination bitmap 31 the dither process 34 with the three schematically indicated dither matrices 35a (for 0% tonal value), 35c (for 50% tonal value) and 35b (for 100% tonal value), the result is the dithered bitmap 32 , As in the border area 37 Once again, the image has shrunk in the row direction by one column. The distance of the two letters "H" from each other is one column less than in the image of the source bitmap 30 , Even after the dithering process, the right leg of the left "H" is still largely visible. The partially missing sections are due solely to the dither process. If one prints such dithered letters with high resolution, the human eye can scarcely perceive the missing parts or supplement them automatically, so that the reproduced printed image appears complete.

Instead of in 4 described reduction of the target image for driving the second printing unit can also be provided an enlargement of the target image for driving the first printing unit. From two columns of the source image can then be added, for example, by a corresponding averaging a third column for the target image.

In 5 is shown another example of a shrinkage compensation, in which a source image two hairlines each one column wide 50 . 51 includes. The compensation also takes place here by calculating a new image column from the values of two adjacent columns of the source image. The situation is similar to the one in 4 shown compensation, in which the right leg of the left letter "H" is exactly one image column, that is, represents a hairline. In 5 are two corresponding hairlines in the columns 50 . 51 the source bitmap 30 intended. According to the in 4 provided compensation is in the hairline column 50 in the last line together with the point to the right by an illustration 53 the pixel 52 in the destination bitmap 31 calculated. In this case, the values 0xFF and 0 of the two source pixels again become the value 0x80 for the target pixel 52 calculated. In the shrinkage compensation of this embodiment is additionally provided that based on a random number generator, a random shift 54 is built in such a way that line by line to the left and right of the acting as a base hairline column 50 in each case a corresponding target image point in the destination bitmap is calculated from two adjacent source pixels of a line. For example, in the penultimate image line, the source bitmap becomes 30 from the two pixels, the two or three columns to the left of the base column 50 lie, by means of illustration 55 a new destination in the destination bitmap 31 calculates the only two columns from the hairline pixel with the value 0xFF of the destination bitmap 31 lies.

• 1. Es wird eine Basis-Spalte in der Quell-Bitmap ausgewählt.
• 2. Es wird ein Grenzwert K festgelegt, der bestimmt, um welche Bildpunkt-Entfernung von der Basis-Spalte die Schrumpfungskompensation maximal ansetzen darf.
• 3. Zeile für Zeile wird mit einem Zufallsgenerator ein ganzzahliger Wert bestimmt, der zwischen –K und +K liegt und anhand dessen bestimmt wird, mit welcher gegenüber der Basis-Spalte spaltenweise Verschiebung die beiden Quell-Bildwerte zur Berechnung des Ziel Bildwerts herangezogen werden.
• 4. Aus den beiden Quell-Bildwerten wird der Ziel-Bildwert nach einer vorgegebenen Funktion berechnet.

To calculate a new target pixel, the following criteria and process steps are used:

• 1. A base column is selected in the source bitmap.
• 2. A threshold value K is determined which determines by which pixel distance from the base column the shrinkage compensation may be set to a maximum.
• 3. Line by line, an integer value is determined with a random number generator which lies between -K and + K and is used to determine with which of the base column column-wise displacement the two source image values are used to calculate the target image value.
• 4. From the two source image values, the target image value is calculated according to a predetermined function.

With the method, the combination of source pixels does not always take place in the same column but line by line around the base Column around at stochastic intervals, but at a maximum distance from K columns. This ensures that no regular structures are recognizable on the printout. For example, hatched areas make it quieter.

The resulting target image map gives the hairline of the base column 50 so again, as they are in the field 56 is shown. This area seems to have no hairline and is not continuous. Due to the limited resolution of the eye, however, can be achieved with appropriate resolution of the print image that this seemingly discontinuous area appears as a straight line. This is also promoted in an actual printing process in that the printed pixels are not squares but usually more or less wide or overlapping dots.

This in 5 The method shown in which the pixels used to reduce the image size are determined stochastically line by line can be applied not only as shown for source bitmaps whose pixels are encoded with a plurality of bits, but in particular for source bitmaps, which in turn only one Have bit per pixel. By way of example, bitmaps of this type can already be bit-dithered bitmaps in advance. An after-dithering process 34 as in 5 shown is no longer necessary.

As a recording medium, instead of the paper web described above 11 also other prefabricated paper, such as sheet-by-sheet paper, or other materials such. B. transparent plastic film material can be used, which deform under the influence of heat. If the deformation of the material causes stretching of the recording medium rather than shrinkage, then a corresponding "reverse" image processing process may be used to compensate for the register error caused by the strain. The term "shrinkage" in the context of this description would then be replaced by the term "elongation" or generally "deformation". In the respective processes, a raster graphic or an image would be stretched / enlarged instead of compressed / reduced or vice versa.

In addition to the line- or column-wise deformation is also conceivable to provide a deformation correction in oblique direction. As an algorithm or mapping rule for the calculation of target pixels more than two source pixels can be used. As a mapping function, in addition to the simple arithmetic calculation, it is also possible to use more complex functions, in particular a plurality of pixels of different weighting, for example interpolations, spline functions or other one- or two-dimensional window functions.

In 6 For example, basic components of a computer system that may be used to practice the invention are shown. Essentially they represent a computer 60 representing a random access memory (RAM) 61 in which data can be buffered for processing, a non-volatile memory (eg hard disk) 62 on which a computer program causing the described processes is stored. In the computer program all process steps and variables are programmed. Corresponding parameter values, for example for the calculation functions, can e.g. B. in the computer program or separately stored on the hard disk, in particular as a look-up tables (LUT). The computer 60 also contains a microprocessor 63 and an interface 64 , about which, for example, data with the controller 14 , with camera sensors and / or with the control panel 24 of the printer 15 are interchangeable. Data from source bitmaps and destination bitmaps can be cached in raster graphics stores before, during, and / or after they are used with the microprocessor 63 be processed by the computer program. A store 67 Contains a value for a shrink factor, for example, from the control panel 24 is entered or calculated. In the storage room 68 the limit value for the maximum distance K is stored. On an optional monitor 69 For example, tax and calculation results can be displayed. A random generator 70 can be used to specify the line-by-line stochastic distribution of the source calculation points.

ist der Schrumpfungsfaktor 1% leicht zu bestimmen und die oben gezeigten Verfahren zur Kompensation der Papierschrumpfung anwendbar. Die Breite der Papierbahn kann von einem Bediener anhand des gedruckten Maßstabs manuell abgelesen oder auch mit einem z. B. optischen Sensor bzw. einer Kamera automatisch erfasst werden. Im automatisierten Fall können die Papierbreiten des ungeschrumpften und geschrumpften Papiers auch ohne Drucken eines Maßstabs erfasst werden und damit vollautomatisch während des Druckbetriebs (on-the-fly) der Schrumpfungsgrad bestimmt und die Schrumpfungskorrektur entsprechend durchgeführt werden.In 7a is shown as a shrinkage factor in a given printing system 12 can be easily determined. The paper web 11 is doing by the printing system 12 with a fixed reference edge 73 transported along the transport direction TR and transverse to the transport direction a scale 71 printed. With the full paper width a, the full-width scale, that is to say the scale value "30", can then be printed, for example, by a first printing unit. If you print the same scale in a second printing after the paper web 11 heated and thereby shrunk to a width b, this results in the 7b shown picture in which the printed scale 72 at a scale value of "27" on the right edge of the paper 74 interrupted. Through the equation

For example, the shrinkage factor 1% is easy to determine and the methods shown above for Paper shrinkage compensation applicable. The width of the paper web can be read manually by an operator on the basis of the printed scale or with a z. B. optical sensor or a camera can be detected automatically. In the automated case, the paper widths of the unshrunk and shrunk paper can also be detected without printing a scale and thus fully automatically determined during the printing operation (on-the-fly), the degree of shrinkage and the shrinkage correction can be performed accordingly.

• – aus jeweils n benachbarten Quell-Bildpunkten wird nach einer vorgegebenen Rechenvorschrift ein Ziel-Bildpunkt berechnet, wobei n eine ganze Zahl und größer als 1 ist und
• – der Anteil der verarbeiteten Bildpunkte entlang der Bildverarbeitungsachse entspricht dem Verformungsfaktor.

In summary, it can be explained again:
For digital printing processes in which a recording medium is printed by means of raster graphics at least twice and is subjected to a treatment between the printing processes, by which it deforms along one of the deformation axis, for example by a heat treatment shrinks, and thereby register errors, a method for correcting the Register error applied. The data of the raster graphics of at least one of the two printed images are processed along an image processing axis of the raster graphic corresponding to the deformation axis, where:

• From each n adjacent source pixels, a target pixel is calculated according to a predetermined calculation rule, where n is an integer and greater than 1, and
• - The proportion of processed pixels along the image processing axis corresponds to the deformation factor.

LIST OF REFERENCE NUMBERS

1

bitmap

2

Source image

3

shrinkage compensation

4

shrinkage factor

5

shrinkage column

6

Image column target image

8th

Border area of the source image

9

Border area of the destination image

10

printhead

11

paper web

12

printing system

13

Data Source

14

controller

15

printing device

16

parser

17

Raster image processor

18

correction module

18a

Correction 1

18b

Correction 2

19

Screening Module

20

First printhead

21

Second printhead

22

First dryer

23

Second dryer

24

Control panel

25

unwinding

26

up roll

29

Source image

30

Source bitmap

30a, 30b

Columns of the source bitmap

31

Destination bitmap

31a

Column of the destination bitmap

32

dithered bitmap

33

Shrinkage compensation

34

dithering

35a, 35b

dithermatrices

36

Target image

37

marginal gaps

38

Dixied picture

40a

First letter H

40b

Second letter H

41

Column selected for shrinkage

42

neighboring column

43

Illustration

44

Target column

45

dithered letter

50

First hairline column

51

Second hairline column

52

Target pixel

53

Illustration

54

random shift

55

Illustration

56

Dehydrated area of the hairline

60

computer

61

random access memory

62

hard disk

63

microprocessor

64

interface

65

First raster graphics memory

66

Second raster graphics memory

67

Shrinkage factor

68

Distance memory

69

monitor

70

Random

71

printed scale

72

printed scale

73

Fixed reference edge

74

paper edge

BA

Image processing axis

SA

shrinkage axis

TR

transport direction

a

Web width

b

Width of the shrunken paper web

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102007040402 A1 [0008, 0009]
• WO 2005/031470 A1 [0010, 0011]
• US 4721969 A [0013]

Claims (12)

Method for correcting a registration error in a printing process, in which - data for images to be printed is subjected to a raster image process, by which at least one raster image structured in rows and columns ( 24 ), - from the raster graphic ( 24 ) each control data for, at least one printing unit ( 20 . 21 ), - a record carrier ( 11 ) is printed in a first printing with the control data from a first raster graphics with a first printed image, - the recording medium ( 11 ) is subjected to a treatment by which the record carrier ( 11 ) and the printed first printed image at least along a deformation axis (SA) deform by a deformation factor and - the recording medium ( 11 ) is printed in a second printing operation with the control data from a second raster graphics with a second print image, - wherein the data of the raster graphics of at least one of the two print images along one of the deformation axis (SA) corresponding image processing axis (BA to avoid register error between the two print images ) of the raster graphics, where: - from in each case n adjacent source pixels ( 30 ), a target pixel (in accordance with a prescribed calculation rule) ( 31 . 52 ), where n is an integer and greater than 1, and - the proportion of processed pixels along the image processing axis (BA) corresponds to the deformation factor. Method according to claim 1, wherein the respective source pixels ( 30 ) are stochastically selected. Method according to claim 2, wherein the selection is performed line by line within a given pixel pitch (K) from a given base source image column ( 50 ) he follows. Method according to one of the preceding claims, wherein each source pixel of the raster graphics comprises a plurality of data bits, hereinafter called source byte map, and from the source byte map a corresponding destination byte map for the target pixels ( 31 ) is calculated. Method according to one of the preceding claims, wherein the calculation rule after the raster process on the target pixels ( 31 ) is applied to the raster graphics, and then a dithering process is performed on the target pixels generated by the calculation rule ( 34 ), which generates binary bitmap image data for the printing process. Method according to one of the preceding claims, wherein the calculation rule comprises an image processing window with a predetermined weighting for p × q adjacent pixels to the respective source pixel, and wherein p and q are integers greater than zero. Method according to one of the preceding claims, wherein the calculation rule comprises an arithmetic mean of adjacent source pixels ( 30 ). Method according to one of the preceding claims, wherein the treatment is a heat treatment, the deformation is a shrinkage and the deformation factor is a shrinkage factor. Method according to one of the preceding claims, wherein - for determining the deformation factor along a transverse axis of the recording medium ( 11 ) in the first and in the second printing process in each case a printed image ( 71 . 72 ) of the same scale across the respective full width of the recording medium ( 11 ) is printed such that the record carrier ( 11 ) in both printing processes on the same reference edge ( 73 ) and in each case the same starting point of the scale at this reference edge ( 73 ) and - from the two at the opposite edge of the paper ( 74 ), which has just been printed, and the scale factor of the deformation factor is calculated. Device for correcting a register error in a printing process, wherein in the printing process - data to be printed images are subjected to a raster image process, by the at least one line and column structured grid image ( 24 ), - from the raster graphic ( 24 ) each control data for at least one printing unit ( 20 . 21 ), wherein in the printing process a record carrier ( 11 ) is printed in a first printing with the control data from a first raster graphics with a first printed image, - the recording medium ( 11 ) is subjected to a treatment by which the record carrier ( 11 ) and the printed first printed image at least along a deformation axis (SA) deform by a deformation factor and - the recording medium ( 11 ) is printed in a second printing process with the control data from a second raster graphics with a second printed image, the device comprising: - at least one computer ( 14 . 60 ), on which runs a computer program that, in order to avoid a register error between the two print images, the data of the raster graphics of at least one of the two print images along one of the Processing axis (SA) corresponding image processing axis (BA) of the raster graphics processed, where: - from each n adjacent source pixels ( 30 ), a target pixel (in accordance with a prescribed calculation rule) ( 31 . 52 ), where n is an integer and greater than 1, and - the proportion of processed pixels along the image processing axis (BA) corresponds to the deformation factor. Printing system comprising a device according to claim 10. A computer program product which, when loaded and run on a computer, effects a method according to any one of claims 1 to 9.

Images