JPH1199634A - Inter-color automatic register correcting method and apparatus therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce effect on a register errorr caused by the vibration of a web by a method in which a register mark is read by a CCD image sensor, an RGB signal is converted into a CMY signal and stored, a stored pattern is compared with a standard pattern, normalized, and used as a register regulating correcting value. SOLUTION: An inter-color automatic register apparatus 6 for outputting a control signal to register regulating motors 5a-5h on the basis of a detection signal from an image sensor 4 consisting of a color CCD camera arranged in the web outlet 3 of a multiple color printing machine 2 converts an RGB signal into a CMY signal by a signal conversion part 8, and accumulates it in a frame memory part 9, and the image information is compared with a registered register mark, normalized to correct an apparent register error, and an actual register error is obtained. An operation processing part 10 which equalizes registered marks taken up, registers them as a new register pattern, and do other things is provided. A correct register error is obtained, and registering is done well.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic color registration adjustment method for a multicolor printing press using process colors and an automatic color registration adjustment apparatus for executing the method.

[0002]

2. Description of the Related Art In a conventional automatic color registration correcting device, an image sensor is installed at a position where web flapping is small, for example, at a position where a guide roller is provided at an exit of a printing machine, in a state perpendicular to the paper surface. I was This is to prevent the detection failure or undetection of the register mark due to the flapping of the web, and to reduce the size of the image of the detected register mark formed on the image sensor light receiving portion due to the change in the distance between the image sensor and the web surface. This is necessary to prevent the calculation amount of the register shift from including an error due to the change.

[0003]

As described above, conventionally, the image sensor has to be installed perpendicular to the web stable surface, and in the existing printing press, the flutter of the web such as the guide roller is suppressed. Some devices do not have a stabilizing device, and even if a stable web surface is obtained, there is no space for mounting an image sensor perpendicular to the web surface. In many cases, it was not even possible to attach the device. On the other hand, when the image sensor is mounted obliquely with respect to the stable web surface due to space restrictions, it takes time to correct the registration error according to the inclination of the image sensor. In addition, even when the image sensor is mounted perpendicularly to the web stable surface, a great deal of time is required for adjusting the installation of the image sensor, such as adjusting the distance from the web surface and adjusting the angle with respect to the web surface. did.

Further, the conventional register device sometimes uses a special register mark (register mark) for each manufacturer, and a register mark printing device must be newly provided.
In some cases, equipment costs increased.

The present invention has been made in view of such a problem, and an object of the present invention is to obtain an accurate register error regardless of the inclination and distance of an image sensor with respect to a web surface. It is an object of the present invention to provide an automatic color registration correction method and apparatus capable of reducing the influence of web flapping on registration errors and easily coping with special register marks (register marks).

[0006]

In order to achieve the above-mentioned object, in the automatic color registration correction method according to the present invention, a register mark is read by an image sensor comprising a color CCD camera, and the register mark is read from the image sensor. R
The GB signal is converted into a YMC signal, the YMC signal is stored, the stored register mark pattern is compared with a reference pattern by a method such as pattern matching and normalized, and the data obtained thereby is used for register adjustment correction. It is used as a value.

Further, in the automatic color register correction method according to the present invention, the register mark pattern is compared with a reference pattern by a pattern matching technique.

Further, in the automatic color registration correcting apparatus according to the present invention, (a) an image sensor comprising a color CCD camera for reading register marks; and (b)
A signal conversion unit for converting the RGB signals from the image sensor into a YMC signal; (c) a frame memory unit for storing the YMC signal; and (d) a frame memory separated into yellow, magenta, and cyan. An arithmetic processing unit that calculates a register adjustment amount in order to use the data normalized by comparing the register mark pattern by a method such as pattern matching as a correction value for register adjustment,
And (e) a control unit that controls a register adjusting motor of the multicolor printing press based on the calculation result of the calculation processing unit.

Further, the automatic register correcting device for colors according to the present invention has a function of averaging read register marks, and registering and holding an averaged pattern obtained by the averaging process as pattern matching data. I have it.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to FIGS.

FIG. 1 shows an automatic color register correcting apparatus 1 for implementing the automatic color register correcting method according to the present invention. The apparatus 1 is arranged at a web exit 3 of a multi-color printing press 2. Image sensor 4 consisting of a color CCD camera
An automatic color registration device 6 for outputting a control signal (register correction correction value) for registration to the register adjustment motors 5a to 5h of the multicolor printing press 2 based on the detection signal from the image sensor 4. And an input device 7 for inputting a new register pattern (register mark pattern) or the like to the automatic color registration device 6 as a comparison reference.

As shown in FIG. 1, the above-mentioned automatic color registration apparatus 6 converts a RGB signal into a CMY signal, and stores the CMY signal converted by the signal conversion section 8 as image information. The frame memory unit 9 compares the image information with the registered register marks, and performs a normalization process described later to obtain an actual register error in which an apparent register error has been corrected. An arithmetic processing unit 10 for averaging the register marks (averaging the luminance of each color of the read register marks) and registering it as a new register pattern, and a register error output from the arithmetic processing unit 10 The motors 5a to 5h for register adjustment of the multi-color printing press 2 are operated based on the parameters regarding the blanket cylinders 11a to 11h. Control unit 1 that functions to match the register by changing the phase of the plate cylinder 12a~12h to
And 3. Here, the RGB signals are R (red), G (green), and B (blue) signals, and are CMY.
The signals are C (cyan), M (magenta), and Y (yellow) signals.

Next, the operation of the automatic color register correcting apparatus 1 having such a configuration will be described as follows.
First, at the web exit 3 of the multi-color printing press 2, the web 1
The register mark on the image sensor 4 is detected and read by the image sensor 4, and the register data taken in by the image sensor 4 is converted from the RGB signal into the CMY signal by the signal converter 8 of the automatic color registration device 6, and the frame is read. It is held (stored) in the memory unit 9. Here, a case will be described in which a register mark line of a cross as shown in FIG. 2 (one register mark line is printed on each plate for printing each color) is used as the register mark. FIG. 2 shows the misregistration of C (cyan) and M (magenta). As shown in FIG. 2, the deviation of the register mark lines is defined as a registration error in the vertical direction and the driving direction (left and right width directions with respect to the flow on the page). In FIG. 2, the operation side refers to the side that operates the offset printing press, and the drive side refers to the opposite side. For example, the register error of each color with respect to M (magenta) is obtained by calculation in the calculation processing unit 10, and the calculation result is input to the control unit 13 as a parameter. The registration control motors 5a to 5h are operated by the output control signal to perform registration. This is the same as the conventional method.

In this specification, the term "normalization" is used.
Will be described with reference to FIG. One record
Reference pattern when looking at the register mark (dragonfly line)
FIG. On the other hand, FIG.
The registered mark line is opposite to the registered mark line shown in FIG.
Then L_bx<L_ax, L_by= L_ayIt has become. this is,
Indicates that the registration mark line has become shorter in the driving direction
ing. That is, the image sensor 4 is tilted in the driving direction.
Will be. Also, in FIG. 3C, L _cx=
L_ax, L_cy<L_ayThis is an image sensor
4 indicates that it is tilted in the vertical direction.

Next, the change in the size of the register mark detected with the change in the distance between the register mark (web surface W) and the image sensor 4 will be described with reference to FIGS.
This will be described with reference to FIG. First, as shown in FIG. 5, light exiting the vertex of the arrow of the register mark O is a lens (convex lens) L.
Go straight through the center P of Further, the light emitted in parallel from the apex of the arrow of the register mark O intersects with the light passing through the center P of the lens L on the CCD image forming surface M to generate an image R.
The size of the register mark O in this case the S _1, when the size of the imaging R and S _2, of S ₂ relative to S ₁ ratio X
Is expressed by the following equation (1) by the distance a from the register mark O to the lens L and the distance b from the lens l to the CCD imaging plane M.

[0016]

(Equation 1)

The following equation (2) holds between the distances a and b and the focal length f of the lens L.

[0018]

(Equation 2)

When the magnification X is obtained by eliminating the distance b from the above equations (1) and (2), the following equation (3) is obtained.

[0020]

(Equation 3)

In the above equation (3), since the focal length f is a fixed value unique to the lens, the magnification X is inversely proportional to the distance a between the register mark O and the lens L. That is, when the register mark O moves away from the lens L, the distance a increases, and the magnification X decreases. Conversely, when the register mark O and the lens L approach each other, the distance a decreases and the magnification X increases. FIG.
Shows this relationship more clearly. Specifically, when the register mark O _F moves away from the lens L CCD
Imaging R _F, which is generated imaging plane M is reduced, imaging R _N is increased to register marks O _N are generated by the CCD image plane M approaches the lens L.

The register mark input from the image sensor 4 and stored in the frame memory unit 9 is subjected to a process of projecting onto a projection plane V where L _by = L _bxR as shown in FIG. Thereby, the apparent register error L _EA is obtained from the projection plane V as the actual register error L _ER . In addition, an actual register error in the vertical direction can be obtained by the same procedure.

On the other hand, as for the distance between the web surface W (see FIG. 1) and the image sensor 4, the above-described method is the same as the reference pattern lengths _Lax and _Lay (see FIG. 3A). What is necessary is just to project on a projection surface.

An example of the judgment required for the above-described processing is shown by a flowchart.
This is shown in FIG. In this case,
Length of X and Y of mark line (register mark) (length of vertical and horizontal lines)
Is compared with the reference pattern. As a result of the comparison, L_bx
= L_axAnd L_by<L_ayWhen the image sensor 4
Is tilted in the up-down direction, and the web surface W and image
It is determined that the distance to the sensor 4 is appropriate, and L_bx<L
_axAnd L_by= L_ayThe image sensor 4 operates.
It is inclined in the driving direction, and the web surface W and the image sensor
It is determined that the distance to the device 4 is appropriate. Also, L_bx
<L_ax, L_by<L _ayAnd L_bx= L_by, L_ax=
L_ayIn the case of, the image sensor 4 is separated from the web surface W.
Is determined to be too much_bx> L_ax, L_by> L_ayIs
One, L_bx= L _by, L_ax= L_ayFor the web surface W
It is determined that the image sensor 4 is too close. Also, L
_bx<L_axOr L_bx= L_axAnd L_by> L_ayNoto
The image sensor 4 is too close to the web surface
It is determined that the image sensor 4 is tilted in the driving direction, and L_bx
> L_axAnd L_by<L_ayOr L_by= L_ayWhen
Means that the image sensor 4 is too close to the web surface
It is determined that the image sensor 4 is tilted in the ground direction. Ma
L_bx<L_axAnd L_by<L_ayOr L _by= L
_ayThe image sensor 4 is separated from the web surface
Judgment that the image sensor 4 is tilted in the vertical driving direction
Then L_bx<L_axOr L_bx= L_axAnd L_by<L
_ayThe image sensor 4 is separated from the web surface
Judge that the image sensor 4 is tilted in the driving direction
You.

According to the automatic color registration correction method having the above-described configuration and the automatic color registration correction apparatus 1 for implementing the method, the register mark taken in by the image sensor 4
Registered register mark pattern (reference pattern)
By performing the normalization as compared with the above, even if the attachment of the image sensor 4 is inclined with respect to the web surface (that is, even if the correspondence between the web surface and the image sensor 4 is slightly shifted). Also, even if the distance between the web surface and the image sensor 4 is slightly deviated from the regular distance (appropriate distance), an accurate registration error can be obtained.

Further, by averaging the above-described normalization processing and a plurality of patterns obtained thereby, it is possible to reduce the influence on the registration error due to the flapping of the web. Then, the averaged pattern can be registered as a reference pattern, or a reference pattern can be input from the input device 7 and registered.

With these functions, the restriction on the mounting location of the image sensor 4 is greatly eased, and the labor required for the installation adjustment of the image sensor 4 can be reduced. Also, even when a special register mark is used, registration as a new pattern is easy,
No renewal of the register mark printing device is required, so that significant savings are also possible in terms of cost.

Although the embodiment of the present invention has been described above, the present invention is not limited to this embodiment.
Various modifications and changes are possible based on the technical concept of the present invention. For example, register mark pattern comparison / normalization is a pattern matching method (two-dimensional normalized correlation)
Other methods, such as taking a shadow on the vertical and horizontal axes and calculating the center point, may be used, and a register mark other than a register mark line (register mark) can also be used. It is.

[0029]

As described above, according to the automatic color registration correction method according to the present invention, the register mark is read by an image sensor comprising a color CCD camera, and the RGB signal from the image sensor is converted into a YMC signal. This is because the YMC signal is stored, the stored register mark pattern is compared with a reference pattern by a method such as pattern matching and normalized, and data obtained thereby is used as a correction value for register adjustment. According to the present invention, according to the present invention, by performing the above-described comparison and normalization, even if there is some deviation in the correspondence between the web surface and the image sensor, Even if the image sensor is tilted or the distance between the web surface and the image sensor is misaligned, accurate registration errors can be determined. Kill.

An automatic color registration correcting apparatus according to the present invention comprises: an image sensor comprising a color CCD camera for reading register marks; a signal conversion unit for converting RGB signals from the image sensor into YMC signals; The frame memory unit for storing the YMC signal and the register mark patterns on the frame memory decomposed into yellow, magenta, and cyan are compared by a technique such as pattern matching, and normalized data is registered. And a control unit that controls a register adjustment motor of the multicolor printing press based on the calculation result of the calculation processing unit. Therefore, according to the present invention, the above-described automatic color registration correction method according to the present invention can be implemented, and accurate The registration can be satisfactorily performed can ask for those errors.

The automatic color register correcting device according to the present invention has a function of averaging read register marks, and registering and holding an averaged pattern obtained by the averaging process as pattern matching data. Therefore, according to the present invention, it is possible to reduce the influence on the register error due to the flapping of the web by averaging the above-described normalization processing and a plurality of patterns obtained thereby. The averaged pattern can be registered as a reference pattern, or a reference pattern can be input and registered from an input device.

Further, according to the method and the apparatus for automatically registering colors between colors according to the present invention, it is possible to greatly reduce the restrictions on the mounting location of the image sensor and also reduce the labor required for adjusting the installation of the image sensor. it can. Furthermore, even when a special register mark is used, it is easy to register as a new pattern, so that it is not necessary to update the register mark printing device, and thus it is possible to save a great deal of money. .

[Brief description of the drawings]

FIG. 1 is a configuration diagram of an automatic color registration correction device according to the present invention.

FIG. 2 is an explanatory diagram showing a relationship between a register mark (register mark line) and a register error.

3A and 3B are explanatory diagrams showing the relationship between the inclination and the perspective (distance) of an image sensor with respect to a reference pattern and a register mark, wherein FIG. 3A is an explanatory diagram of a reference pattern, and FIG. Explanatory diagram showing the case of tilting,
(C) is an explanatory diagram showing a case where the image sensor is tilted in the vertical direction, (d) is an explanatory diagram showing a case where the image sensor is too close to the web surface, and (e) is an image sensor which is too far from the web surface. It is explanatory drawing which shows the case.

FIG. 4 is a projection diagram for describing normalization.

FIG. 5 is an explanatory diagram showing the principle of a CCD imaging system.

FIG. 6 is an explanatory diagram showing a difference in size of a register mark according to a distance from a color CCD camera.

FIG. 7 is a flowchart showing determination conditions for normalization.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Automatic registration correction device between colors 2 Multicolor printing press 3 Web exit 4 Image sensor 5a-5h Motor for registration adjustment 6 Automatic registration device between colors 7 Input device 8 Signal processing unit 9 Frame memory unit 10 Arithmetic processing unit 11a-11h Blanket cylinder 12a-12b Plate cylinder 13 Control unit 14 Web W Web surface

Claims (4)

[Claims] An image sensor comprising a color CCD camera reads a register mark, converts an RGB signal from the image sensor into a YMC signal, and converts the YMC signal into a YMC signal.
A method of automatically registering and correcting colors between colors, comprising storing a signal, normalizing the stored register mark pattern by comparing it with a reference pattern, and using data obtained thereby as a correction value for register adjustment. 2. A method according to claim 1, wherein said register mark pattern is compared with a reference pattern by a pattern matching technique. 3. An image sensor comprising a color CCD camera for reading register marks, and
A signal conversion unit for converting the RGB signals from the image sensor into a YMC signal; (c) a frame memory unit for storing the YMC signal; and (d) a frame memory separated into yellow, magenta, and cyan. An arithmetic processing unit that calculates a register adjustment amount in order to use the data normalized by comparing the register mark pattern by a method such as pattern matching as a correction value for register adjustment,
And (e) a control unit for controlling a register adjusting motor of the multi-color printing press based on the calculation result of the calculation processing unit. 4. The read register mark is averaged,
4. The automatic color register correcting device according to claim 3, further comprising a function of registering and holding an averaged pattern obtained by the averaging process as pattern matching data.