Method for determining fit deviations using recursion analysis

Abstract

Method for determining registration and / or misregistration on substrates (3) during production in multicolor printing presses (7) using a computer (4), wherein the influence of duplication effects is taken into account when measuring the misregistration on the substrate (3) by the computer (4), wherein the duplication effects of at least one preceding and / or one subsequent printed image are recursively taken into account in the computer (4) when measuring on the substrate (3), characterized in that the line thickness of measuring marks (9) on the substrate (7) for detecting registration and / or misregistration is determined by the computer (4) under the assumption of identical print transfer from printing form to substrate or intermediate printing medium.

Description

[0001] The present invention relates to a method for determining registration and / or misregistration on printing materials during production in multicolor printing presses using a computer, wherein the influence of duplication effects is taken into account when measuring the misregistrations on the printing material by the computer.

[0002] In offset printing and other additive printing processes, multiple color separations of a print image must be printed on top of each other to produce a multi-colored print. The precise superimposition of these multiple color separations is called registered printing. Furthermore, the superimposed color separations must maintain a specified distance from the edges of the substrate, a process known as indirect registration. To detect registration and / or misregistration, registration measuring devices are used. These devices measure the registration marks on the substrate and thus detect any deviations. However, malfunctions during the operation of an offset printing press can lead to so-called duplication effects, which also cause shifts between the individual color separations.This means that when using a conventional register measuring device, not only register and crossover deviations are detected, but also superimposed deviations caused by duplication effects. The challenge, therefore, lies in separating deviations caused by duplication effects from the register and / or crossover deviations, as otherwise the actuators for correcting register and / or crossover deviations cannot be controlled correctly, and deviations will persist.

[0003] The problem of duplication effects is known, for example, from German patent application DE 38 00 877 A1. This document describes a method for measuring duplication shifts in printing presses, in which a test halftone pattern is printed onto a substrate and detected by two sensors. These sensors transmit the gray values ​​of the test pattern to a duplication detection device. Based on the measurement results and the geometric relationship of the test strips to each other, duplication effects can then be detected. This method is also suitable for use in printing presses, allowing duplication effects to be detected during the printing process. The sensors essentially detect duplication shifts characterized by the presence of a duplication shadow next to a fresh, saturated halftone dot. This shadow is matte and often smaller in diameter.

[0004] European patent application EP 0 000 328 A1 discloses a measuring method and a device for determining defects on printing substrates, in which remission values ​​on the substrate are recorded and evaluated computationally. To calculate the correct halftone value on the substrate, the effects of duplication are determined and taken into account. If desired, the duplication effects are also displayed on a screen to the printing press operator.

[0005] European patent EP 1 620 266 B1 discloses a method for using a device to control at least one register in a printing press, wherein a computer in the printing press uses a control loop to adjust the supply of ink or dampening solution and the register or registration. The printing press has four printing units in series and uses the offset rotary printing process. The computer for controlling the printing press is integrated into the control room and acquires measured values ​​for dot gain and register and registration deviations, which it detects via sensors and adjusts accordingly. Duplicating effects on changes in the measured values ​​of register and registration are not taken into account, so that the measured values ​​for register and registration are superimposed and distorted by these duplicating effects.

[0006] A method for accounting for geometric changes during printing is disclosed in German patent application JP 2000-82135A. In this method, deviations on the substrate are detected via a sensor during the printing process. Appropriate templates are used to facilitate the detection of these deviations, even at high printing speeds. Compensation for these geometric deviations can then be applied to small areas of the printed image. Duplication effects are not taken into account when recording or correcting the geometric deviations.

[0007] The creation of duplication effects in offset printing presses in the form of double or multiple contours is known from the handbook of print media by Helmut Kipphan ISBN 3-540-66941-8.

[0008] The current state of the art has the disadvantage that, while existing measuring devices can detect duplication effects, they only take them into account with regard to the correct evaluation of halftone values. With register and / or cross-reference measuring devices, the effects of duplication cannot yet be considered, so that correspondingly distorted register and / or cross-reference deviations are determined.

[0009] It is therefore an object of the present invention to provide a method for determining registration and / or misregistration on printing materials, in which the influence of duplication effects is taken into account when measuring registration and / or misregistration on printing materials and one measurement process is sufficient to determine the misregistration.

[0010] This problem is solved according to the invention by claim 1. Advantageous embodiments of the invention can be found in the dependent claims and the drawings. The method according to the invention provides that the registration and / or register deviations, including any duplication effects contained therein, which are detected on the substrate by a registration measuring device in a single measurement process, are fed to a computer. This computer then determines the proportion of duplication effects to the measured deviations by recursively considering at least the duplication effects of a preceding and / or a subsequent printed image. This means that not only are duplication effects detected in a single printed image on a substrate, as in the prior art, but also duplication effects of several consecutive printed images on multiple printed sheets or a printing web.This approach is based on a model in which the calculation of duplication effects is performed recursively on the resulting register fluctuations across multiple printed images. In this way, the duplication effects caused by vibrations can be distinguished from the register and / or registration deviations, and the superimposed effects can be separated. This allows the contribution of the register and / or registration deviations to be factored out, and only these deviations are then correctly corrected by control commands to the register actuators in the printing press. If the duplication effects were not factored out, incorrect control values ​​would be sent to the register actuators in the printing press, and the control system would not function correctly.Nevertheless, the inventive method and device do not require a second measurement process to determine duplication effects; instead, it is sufficient to measure the printed images with a register measuring device, without the need for separate measuring devices to detect duplication effects. The present invention is particularly suitable for use with sheet-fed rotary printing presses.

[0011] In a particularly advantageous embodiment of the invention, the duplication effects of up to five preceding and / or subsequent printed images are taken into account. This number has proven sufficient in practice. It is especially important that, with sheet-shaped substrates, the sheets are measured in the correct sequence as they are produced successively in the printing press. This is particularly relevant if the printed sheets are measured with a separate measuring device only after being removed from the printing press. However, if the measuring device is installed in the printing press after the last printing unit, the measurement of the substrates in the correct sequence is ensured. To separate the effects of duplication and changes in registration and alignment, the present invention employs recursive analysis.Recursion analysis specifically considers color shifts of halftone tones in the Lab color space. From these color shifts, the pattern of duplication effects can be calculated, thus separating the influence of these effects from changes in registration and pass / fade.

[0012] According to the invention, the determination of the line thickness of registration marks on the substrate for detecting registration and / or cross-registration deviations is carried out under the assumption of identical print transfers from the printing form to the substrate or intermediate printing medium. In offset printing, a rubber blanket is typically used as the intermediate printing medium. The line thickness of the registration marks is not initially known as an absolute value, but can be calculated from recursion analysis under the aforementioned assumption. The method according to the invention can, in principle, be carried out during continuous printing as well as before the actual printing operation during the start-up phase of the printing press.

[0013] A conventional handheld registration measuring device can be used, which either has its own computer for performing recursive analysis or is connected to a suitable computer. Furthermore, such a registration and register measuring device can also be integrated into a colorimeter. This has the significant advantage that only one measurement is required to acquire both color and registration values. Particularly when this combined measuring device is located in the last printing unit of the printing press, color and registration values ​​can be acquired for each printed image.

[0014] The present invention is described and explained in more detail below with reference to several figures. These show: Fig. 1 a combined registration and color measuring device with an attached computer and a connected printing press, Fig. 2. A simple example of color shifts in the Lab color space. Fig. 3. Formation of duplicate shadows around raster points, Fig. 4 a printing process with three printing colors with color build-up on the rubber blanket of an offset printing press, Fig. 5. The printing operation during the start-up phase of a printing press without the influence of disturbances, Fig. 6. The printing operation during the start-up phase in case of disruptions in the sheet feed and Fig. 7 the stationary printing process in continuous printing operation in the event of a disruption in the sheet feed.

[0015] In Fig. Figure 1 shows a combined registration and colorimeter 10, which has a measuring bar 1 with a combined measuring head 8. The measuring bar 1 is electrically driven and moved across the entire printed image of a sheet-shaped substrate 3. The measuring bar 1 moves in the X direction, while the measuring head 8 moves within the measuring bar 1 in the Y direction, allowing the measuring head 8 to detect any pixel on the sheet 3. For the measurement process, the sheet 3 is placed on the measuring table 2 beneath the measuring bar 1. A measuring mark 9 is printed on the side of the sheet 3. This measuring mark 9 can contain color measurement fields as well as registration marks, which the colorimeter 10 detects. The measuring device 10 sends the detected registration, register, and color measurement values ​​to a computer 4.

[0016] However, the problem with the registration and register measurements is that they also include duplication effects. These duplication effects are then filtered out in computer 4 using the recursive analysis described below, thus determining the actual registration and register deviations. Computer 4 can calculate corresponding correction values ​​for the connected offset printing press 7 based on these actual registration and register deviations. In addition to fully automatic control, it is also possible for these correction values ​​to first be displayed to the operator on a screen 5. Using input via a keyboard 6 connected to computer 4 or a computer mouse 11, the operator can then approve the corresponding correction values ​​and transfer them to the printing press 7.There, the correction values ​​are converted into corresponding control commands for the register actuators in printing press 7, thus counteracting the determined register and registration deviations.

[0017] Fig. Figure 2 shows an example of the chromatic aberration of a halftone dot on the substrate 3 in the Lab color space. It can be seen that the chromatic aberration F of a halftone dot is shifted to a chromatic aberration F' due to duplication effects. This shift from the actual chromatic aberration F to the chromatic aberration F' can be expressed by a chromatic aberration vector dF.

[0018] This color shift dF is caused by the so-called duplication shadow as seen in Fig. Figure 3 illustrates this. Duplicate shadows are caused by the backprinting of undried pixels from sheet 3 onto the rubber blanket in subsequent printing units of a printing press 7, under the influence of disturbances in the sheet path, e.g., due to vibrations in the machine. Because of these disturbances, the halftone dots backprinted in the subsequent printing units, which are significantly lighter in color, are not printed exactly on top of each other, but are shifted relative to each other, which becomes visible as a duplicate shadow that enlarges the pixel. This is particularly noticeable in the upper area. Fig. Figure 3 shows a halftone dot D1 with a diameter of 100 µm, which has a duplicate shadow DS1 of 80 µm. In the lower half of the image, a halftone dot D2 with a diameter of 50 µm is shown, which has a duplicate shadow DS2 with a diameter of 40 µm. It is realistically assumed that the duplicate shadow is approximately 80% the size of the original halftone dot D1, D2. In the first column, the duplicate shadow disappears behind the halftone dot because there is a displacement of 0 µm between the halftone dot and the duplicate shadow. In the second column, the halftone dot and duplicate shadow are shifted by 10 µm relative to each other. This displacement increases to 30 µm in the third column, and it can be seen that this affects print quality and consequently leads to distorted measurement results when measuring halftone dots to determine registration and misregistration deviations.

[0019] In Fig. Figure 4 illustrates the build-up of ink at a halftone dot during the printing process on the blanket in three printing units of the printing press 7. The first printing unit, DW1, contains black ink, the second printing unit, DW2, contains cyan ink, and the third printing unit, DW3, contains magenta ink. In the first printing unit, DW1, only black dots build up on the blanket because only fresh sheets 3 from the feeder enter this unit. In the second printing unit, DW2, as the cyan ink is transferred to the substrate, the blanket rolls across the substrate 3, printing black components back onto the blanket and then onto the sheet 3. In this way, in the third printing unit, DW3, in addition to the magenta component, black and cyan components from the preceding printing units, DW1 and DW2, are also printed back onto the blanket in the third printing unit, DW3, and are again printed onto the sheet 3.

[0020] Fig. Figure 5 shows the printing process in the first three printing units of the printing press 7 during the printing start-up phase, without any disturbance in the sheet feed.

[0021] In Fig. Figure 6 also shows the printing process in the start-up phase, whereby after the second sheet a disturbance occurs in the sheet travel between printing units DW1 and DW2, causing the printed halftone dots to be shifted in the subsequent printing units DW2 and DW3 and resulting in corresponding duplication effects.

[0022] In Fig. Figure 7 depicts the stationary printing process in continuous operation, where a disruption in the sheet path also occurs. It can be seen that, in this case as well, corresponding duplication effects occur due to the displacement of halftone dots printed on the blanket. These displacements cause the halftone dots of, for example, black in the following sheet (3) to be overprinted by the halftone dots printed on the blankets in printing units DW2 and DW3. Since these dots are not precisely aligned, they are marked with a duplication shadow. The duplication shadow thus arises because ink from the preceding printing units is present on the blanket of the subsequent printing units, and due to disruptions in the sheet path, these superimposed halftone dots are not precisely aligned but are shifted relative to each other.These duplication shadows are calculated out by the recursion analysis described below according to the invention, so that even when duplication effects occur, correct measurement of register and / or pass deviations is possible.

[0023] When several color separations A_j, j=1...m (m = number of printing units) are printed together, the desired color impression F=f(...,A_j, ...), j=1...m is created with the paper white and through the autotypical color mixing of the superimposed color separations. This can be uniquely identified using a metric as an ordered triple (L, a, b) in the Lab color space; see also Fig. 2. Due to influencing factors and disturbances (for example, the color density of the individual separations, the topology and ink absorption behavior of the substrate, etc.), a shift towards the color impression F'=(L', a', b') occurs. The difference between these two states is described by the vector dF=(L'-L, a'-a, b'-b). Using, for example, an IT 8.7 / 3 color chart, a large number (number I) of different halftone patches with the color impressions F_k, k=1...I are printed. If a disturbance occurs, e.g., during sheet transfer in the printing units j and j+1, then, depending on the printing sequence, a number o of the I halftone patches will undergo a characteristic color change dF_j, j=1...o.

[0024] As an alternative to the IT 8.7 / 3 color plate, other elements are also conceivable. A duplicating halftone dot experiences an increase in area coverage, which depends on its absolute size, since this is an "edge effect." A certain minimum registration shift is necessary so that the duplicating shadow emerges from the original image; see [reference]. Fig. 3. With a large halftone dot area, the increase in area coverage is less than with a small area. Therefore, the expected color changes can be influenced by choosing the appropriate halftone dot size.

[0025] The influence of a single color separation A_j or a transfer Ü_j / j+1 on the color impression F of the I halftone patches can be described by a model: F_k=f(...,A_j, Ü_j / j+1), k=1...I. If all I halftone patches are now measured and the actual color impressions F' are determined, a characteristic shift field of the hues in the Lab color space results from dF=(L'-L, a'-a, b'-b). As described above, the values ​​for the affected halftone patches are different from zero.

[0026] By working backwards, the magnitude of the causative disturbance can then be calculated. This calculation can also be performed using an empirically derived model with the aid of recursion analysis.

[0027] Additionally, it is possible, analogous to the current procedure for registration measuring devices, to determine a global shift / duplication value for each color separation and to incorporate it into the above model as further information, but without specifying a direction.

[0028] The underlying model specifically considers the point in time at which a color dot is printed. Therefore, it is essential to measure sheet 3 in a defined sequence.

[0029] The aforementioned method provides both qualitative and quantitative information about disruptions in image or sheet transport at a specific point in the printing process. Due to the nature of the process, the printing process possesses a certain memory. A disruption in the front section of machine 7 also affects subsequent printing processes, both temporally and spatially; see [reference]. Fig.5 to 7. Therefore, the color impression and thus also the desired measured values ​​of a sheet j are influenced by the disturbances in the printing of sheets j-1, j-2. In turn, sheet j naturally also influences the values ​​of its successors j+1, j+2... In general, the influence is limited to approximately + / -5 neighboring sheets 3.

[0030] Taking this into account, a recursive improvement of the measurement results can be performed. In the first step, as is done today, a measurement for the position of the color separations is determined as a starting value for the recursion. Already with the second sheet 3, or measurement, information about any change from sheet j to sheet j+1 is available. Using the model described above, an estimate can now be made for the duplication influence interfering with the measurement. This is then used to improve the duplication shadow correction of sheet j+1. If the value of the duplication shadow correction is now output separately alongside the registration measurement, one measurement per pixel on sheet 3 yields further important information, namely the information about the duplication size, which correlates directly with the visual impression.

[0031] Another parameter to consider is the line thickness of measuring mark 9. This is not initially known as an absolute value. However, assuming that the original image is always printed identically from the printing plate onto the rubber blanket, its thickness can also be derived recursively. Reference symbol list 1 measuring bar 2 Measuring table 3. Printing material 4 computers 5 screen 6 keyboard 7 printing press 8 measuring head 9 measuring mark 10 Colorimeter 11 Computer mouse L, a, b color space axes F Color location F' shifted color point dF chromaticity shift vector D1, D2 Grid dot diameter DS1, DS2 duplicating shadow diameter

Claims

[1] Method for determining registration and / or misregistration on substrates (3) during production in multicolor printing presses (7) using a computer (4), wherein the influence of duplication effects is taken into account when measuring the misregistration on the substrate (3) by the computer (4), wherein the duplication effects of at least one preceding and / or one subsequent printed image are taken into account recursively in the computer (4) when measuring on the substrate (3), characterized by , that the line thickness of measuring marks (9) on the substrate (7) is determined by the computer (4) to detect registration and / or register deviations, assuming identical print transfer from printing form to substrate or intermediate printing carrier. [2] Method according to claim 1, characterized by , that the duplication effects of up to 5 preceding and subsequent printed images are taken into account. [3] Method according to any one of the preceding claims, characterized by , that the computer (4) separates passer and register changes and the influence of duplication effects by recursion analysis. [4] Method according to claim 3, characterized by , that color shifts of halftone tones in the Lab color space are taken into account in the recursion analysis. [5] Method according to claim 3 or 4, characterized by , that the recursion analysis takes into account the time at which a halftone dot is printed in the printing press (7). [6] Method according to any one of the preceding claims, characterized by , that the procedure is carried out before the actual printing operation in the start-up phase of the printing press (7).

Images