DE102017213262B4 - Image acquisition with area-by-area image resolution - Google Patents

Abstract

Method for image acquisition in a machine (14) processing printing materials using a computer (12, 15), image data (16) of a print job being analyzed by the computer (12, 15) and divided into image areas (2, 3, 4) with different image content are assigned, different image resolutions are assigned to the different areas (2, 3, 4), the printed image data (5) are recorded and analyzed by means of an image sensor (9) of an image acquisition system (10), consisting of individual parts, characterized in that the individual parts of the image sensor (9) are able to use different image resolutions and they are controlled by the computer (15) in such a way that the different image areas are recorded with the respectively different, assigned image resolutions (19) and forwarded to the image acquisition system (10). will.

Description

The invention relates to a method and a system for image acquisition in a printing machine for quality assurance of printed products.

The invention lies in the technical field of automatic quality control.

Many modern printing machines have automated image acquisition systems with which the printed products can be examined for the print quality achieved. In the case of multi-color printing, both the target color values achieved and the register, i. H. the superimposition of the individual color separations. The registers, i. H. the correct distance to the edge of the print substrate and the general print quality. The latter is based on the correct representation of certain image objects, where it is important that these z. B. be displayed correctly in terms of edge transitions. An example of this are barcodes or QR codes, where even the smallest deviation in the printed image can lead to a change in the transported data content and thus to a printing error. Another example are the package inserts for medications, where misprints can also have a significant impact.

In larger sheet-fed and web-fed printing presses in particular, the image acquisition systems are installed internally in the press, usually behind the last printing or coating unit, in order to be able to carry out inline measurements during production. For this purpose, the image acquisition systems have one or more cameras or scanner strips, which are used for the corresponding inline image acquisition. The data for color measurement, for general image inspection and for register accuracy can then be obtained from the image data recorded in this way.

The different goals of image acquisition require different quality requirements with regard to the necessary image resolution, which the cameras or scanner bars used must provide. For example, a high image resolution, e.g. of at least 300 dpi up to 600 dpi and more, is required to check the register and register accuracy, while lower image resolutions of 50 to 150 dpi are sufficient for color measurement. For general image inspection, image resolutions between 100 and 300 dpi are usually sufficient. In addition to the image resolution, another factor that places a requirement on the cameras used is the transport speed of the print substrate, which depends on the type of printing machine used, e.g. B. whether it is a sheet-fed or web-fed press is dependent.

The current state of the art primarily uses CCD-based RGB cameras and scanner bars, but these are used at the usual high transport speeds of at least 3 m/s for sheet-fed offset printing machines and at least 0.6 m/s for digital or inkjet printing machines are not able to capture the entire print image with the currently used highest resolution of 600 dpi. The bottleneck mainly affects the data processing in the evaluation electronics of the cameras, or the interface between the camera and the subsequent data processing in the image processing computer.

This therefore results in the need to use several cameras, with one camera only monitoring the color bars on the print substrate, for example, a second carrying out the general image inspection and a third only scanning the register marks in order to be able to carry out all image acquisition objectives. However, this has the disadvantage that the installation effort and thus the cost factor for the image acquisition system increases massively.

In order to solve this problem, it would therefore be advantageous to have an image sensor, i.e. a camera or scanner bar, which can monitor the entire print substrate in accordance with the requirements mentioned. Since the simplest solution, namely the use of a camera or scanner bar, which can always record and process the entire print substrate in the highest resolution, is not possible, since such camera systems would either be far too expensive or overwhelm the subsequent data processing of the image processing computer, alternative camera systems must be sought here.

The object of the present invention is therefore to find a method for image acquisition in a printing press with a camera system as part of the printing press, with which the disadvantages known from the prior art can be overcome.

The object is achieved by a method according to the invention for image acquisition in a machine processing printing materials using a computer, image data of a print job being analyzed by the computer and divided into image areas with different image content, the different areas being assigned different image resolutions which, the printed image data are recorded and analyzed by means of an image sensor of an image acquisition system consisting of individual parts and which is characterized in that the individual parts of the image sensor are able to use different image resolutions and are controlled by the computer in such a way that the different image areas are recorded with the respective different, assigned image resolutions and forwarded to the image acquisition system. The decisive factor here is that the image sensor used for the camera system can display different image resolutions for different areas of the printed image. In this way, within the scope of image capture for quality testing of the printed products produced, the optimal image resolution can be captured for the individual areas. The image acquisition method must be adapted accordingly to these new possibilities. This means that a computer, for example the computer of the image acquisition system, the control computer of the printing press or the prepress computer, must use the prepress data to analyze which image data is currently to be printed. This image data must then be placed in the corresponding image areas with the different image content are subdivided and the corresponding optimum image resolutions must be assigned to these different image areas. Ideally, this can already be done directly in prepress if the prepress system used is able to do this. Otherwise, it is taken over by a downstream computer, such as the image processing computer, which is responsible for monitoring the image acquisition system. This division is then made available to the image acquisition system accordingly, so that the corresponding divided image areas can be independently acquired by the image sensor with the respectively associated optimal image resolution and analyzed in the image processing computer.

Advantageous and therefore preferred developments of the method according to the invention result from the associated subclaims and from the description with the associated drawings.

A preferred development of the method according to the invention is that the areas with different image content include image areas for register and register measurement and/or for color measurement and/or for general image inspection. The fiducial and register marks can therefore be recorded with the highest image resolution, the color control strips with a correspondingly lower image resolution and the areas for general image inspection with a correspondingly medium image resolution.

A further preferred development of the method according to the invention is that the image areas for register and register measurement have an image resolution of at least 200 dpi, the image areas for color measurement have an image resolution of at least 50 dpi and the image areas for general image inspection for the inspection of machine-readable data codes, such as e.g. barcodes, an image resolution of at least 200 dpi, for the inspection of other areas an image resolution of at least 100 dpi must be assigned. For the register measurement, the highest image resolution of at least 200 dpi up to 600 dpi is required with current image acquisition systems in order to be able to record the register and register as precisely as possible. For general image inspection, medium image resolutions of at least 200 dpi are sufficient for certain data codes, such as barcodes or QR codes, while at least 100 dpi is sufficient for other areas of inspection. The color measurement control by capturing the color control strips requires the lowest image resolution of 50 dpi up to 200 dpi with current systems.

A further preferred development of the method according to the invention is that the computer creates an interface specification from the image data with the image areas recorded with different resolutions, which assigns the image areas to the respective image recording location and transmits this to the image recording system, which controls the individual parts of the image sensor in this way that the individual image areas are evaluated with their associated image resolution. In order to make the correspondingly divided image areas with different image resolutions and their image acquisition system available in a correct manner, so that the image acquisition system can also do something with this data, the computer, which for example is the image acquisition computer, the control computer of a printing press or the computer of the prepress stage, creates an interface specification for the image acquisition system, more precisely, for the camera system, from the image data with the differently divided image areas, which is then transmitted to the image acquisition system. The individual modules of the image sensor can then be controlled in such a way that they evaluate the corresponding different image areas with their respectively assigned optimum image resolution. The control of the individual modules of the image sensor is usually implemented independently by the control electronics of the image sensor using the information as to which areas are to be recorded with which image resolution. This means that the correspondingly controlled parts of the image sensors always capture their respective area optically with their maximum image resolution. As this However, leading to the problems mentioned in data evaluation, the recorded image data is reduced by the so-called "binning", i.e. the combining of individual pixels into a common pixel.

A further preferred development of the method according to the invention is that the individual parts of the image sensor are controlled in such a way that the image information is weighted from a plurality of pixels to one pixel in the image areas which are recorded with a reduced image resolution. Since the optics of the image sensors always capture the corresponding print image or the correspondingly assigned divided areas with the full image resolution, the desired image resolution is implemented by controlling the individual parts of the image sensor via their evaluation electronics. The reason for this is that the critical point of image acquisition and image processing of the image sensor is not so much the optics, but that the downstream evaluation electronics are not able to process the image data supplied by the optics with the necessary speed. And even if the evaluation electronics of the image sensor are perhaps able to process the corresponding image data, the image acquisition system itself is overwhelmed with the transmission of the amount of data. The corresponding control of the individual modules of the image sensor for capturing with different image resolutions therefore takes place via the image capturing system to the evaluation electronics of the image sensor.

A further preferred development of the method according to the invention is that the image areas are read out with different image resolutions with full RGB color information. Since, for example, a color measurement is also carried out using the method according to the invention as part of the image acquisition, it must of course be possible to acquire the corresponding image areas which evaluate the color control strips with full color information. General image inspection areas may also require capture with full RGB color information.

A further preferred development of the method according to the invention is that the image acquisition of the image acquisition system takes place at a transport speed of the print substrate of at least 3 m/s for offset printing machines and at least 0.6 m/s for inkjet printing machines. The transport speed of the print substrate in a printing machine, which is monitored by an image acquisition system, is generally at least 3 m/sec in offset printing machines. With inkjet printing machines, the transport speed is currently much lower - currently at least 0.6 m/sec. The image acquisition of the image acquisition system with the different image areas with correspondingly optimal, individual image resolution must also be able to acquire and process the image data generated with a transport speed of at least 3 m/sec or 0.6 m/sec.

A further preferred development of the method according to the invention is that if the image detection system is used in an inkjet printing machine, a printed print image is divided into several image areas, which are each recorded sequentially with the highest possible image resolution and then used to detect faulty print nozzles. Bad print nozzles cause image noise in the image areas where the bad print nozzle should have printed. For the detection of such image disturbances and thus faulty printing nozzles, a very high image resolution is required when capturing the image of the printed image. The method according to the invention can also be used here, in which individual areas of the printed image are recorded sequentially with the highest available image resolution and then evaluated accordingly by the image recording system or by another component. The individual areas are recorded and evaluated sequentially, i.e. one after the other with the highest available image resolution, until the entire print image and thus the entire print head width have been checked with the highest possible image resolution.

A further preferred development of the method according to the invention is that the color resolution of the image sensor is increased to more than 8 bits when capturing the image areas for color measurement. The usual color resolution of 8 bits enables a color depth of 256 color values (0 to 255) in the color range captured by the image sensor. When capturing colors, however, it is advantageous to use a larger color value depth. If the color resolution is increased from 8 bits to 10 bits, for example, the color value recorded by the image sensor can be divided into 1024 color values (0 to 1023), which enables a more detailed examination of the color values generated and thus better color control. However, if the amount of data generated in this way is too high for further processing, the image resolution can be reduced from at least 50 dpi to 200 dpi, thus saving data. However, it is important not to let the image resolution fall below 50dpi.

A further solution according to the invention to the stated object is an image acquisition system for a sheet-fed printing press with an inline measuring device for carrying out a method according to one of the preceding claims is characterized in that the image sensor is a modular CIS scanner bar based on CMOS or CCD or a CMOS camera, with all parts of the CIS scanner bar or the CMOS camera being able to be controlled and read out in sections for image recording with different image resolutions . The image acquisition system according to the invention for the sheet-fed printing press with inline measuring device provides a modular CIS scanner bar, in which case all parts or modules for image acquisition can be controlled individually with correspondingly different image resolutions. The method according to the invention can thus be carried out with the acquisition of the different image areas with a correspondingly optimal image resolution. Alternatively, a modularly controllable CMOS camera can also be used. Whereby modularly controllable means that the individual CMOS sensors of the CMOS camera can be controlled individually in such a way that an area-by-area image capture of the printed print substrate is possible over its entire width. It doesn't matter whether a CIS scanner bar or a CMOS camera is used: It is important that the modular control enables area-by-area image acquisition with different image resolutions. In the case of the CIS scanner bar, this means that the individual modules must be subdivided in such a finely structured way that it is possible to capture all image areas with the associated optimum individual image resolution. With the CMOS camera, this is not a problem anyway, due to the individual controllability of the individual CMOS sensors. In this case, it is quite possible for individual image areas with the corresponding image resolution to be recorded by several parts of the camera, ie CMOS sensors with the same image resolution. However, it is important that the monitoring area of a CMOS sensor is not so large that several image areas with different image resolutions can only be captured by this one CMOS sensor.

A further preferred development of the image acquisition system according to the invention is that the image acquisition system can accept a changed interface at least within 20 image lines. In order to avoid the problem already described, that several image areas with different image resolutions can only be captured by part of the image sensor, the image capture system must be able to accept a changed interface specification, i. H. in the form of using a different image resolution, within a maximum of 20 image lines. The transfer from one image line to the next is the long-term goal, but this cannot be achieved with the current image sensor technology. With a transfer within a maximum of 20 image lines, however, it is ensured that each area with an individual image resolution is correspondingly captured by an individual part of the image sensor.

A further preferred development of the image acquisition system according to the invention is that the image sensor (9) has sensor lines (8) with three or more color channels, with RGB sensor lines (8) for three color channels in a trilinear arrangement, mixed arrangement of sensor lines or by means of a prism camera, and more than three sensor lines with individual color detection can also be used. The image sensor must be able to capture the color information of the print image for the application of the image capture system for general image inspection and especially for color measurement control by capturing the color control strips. This is implemented by using sensor lines with at least three color channels. When using three color channels, for example, three RGB sensor lines can be implemented. These can, for example, be arranged in a classic trilinear manner; i.e. one sensor line captures the R, another the G and another the B color information. However, alternative, mixed arrangements are also possible, where e.g. Another option for RGB sensor lines is the use of a prism camera, in which the individual color channels for RGB are recorded and processed individually and then combined into a common image only in the downstream image processing of the image sensor. It is also possible to use more than three color channels for more precise color detection. In this case, in addition to the RGB sensor lines, sensor lines with an individual frequency range can also be used. However, alternative divisions of the color channels that do not contain any RGB sensor lines are also possible. In this case, the frequency range to be covered by the image sensor is divided up among the sensor lines used with their color channels.

The method according to the invention and functionally advantageous developments of the method are described in more detail below with reference to the associated drawings using at least one preferred exemplary embodiment. In the drawings, corresponding elements are given the same reference numbers.

• 1 einen Druckbogen mit Bildbereichen unterschiedlicher Bildauflösung
• 2 den funktionalen Aufbau einer CIS-Scannerleiste
• 3 den funktionalen Aufbau eines Bilderfassungssystems
• 4 den Ablauf des erfindungsgemäßen Verfahrens

The drawings show:

• 1 a print sheet with image areas of different image resolution
• 2 the functional structure of a CIS scanner bar
• 3 the functional structure of an image acquisition system
• 4 the course of the method according to the invention

Modern image recording technologies such as scanner bars 9 made of CIS sensors and CMOS cameras 9 have electronics that allow the image recording sensors to be controlled differently in different areas. This property can be used to vary the resolution with which the image data is transferred to the connected image processing system depending on the register measurement, image acquisition or color measurement function. Internally, the sensors always work with the highest resolution, but the data is combined in the electronics close to the sensor to the desired resolution. This can be done both in the row direction and in the column direction by averaging image pixels over several rows. Scanner strips 9 consist of several composite sensor modules (>100) that can be controlled independently. Individual sensor modules can also be operated with high image resolutions (600 dpi) at high speeds (>5 m/sec).

RGB scanner bars usually consist of several RGB sensor modules with different electronic controls. RGB cameras with CCD technology are usually made up of just a few RGB sensor modules that can be read out individually. The CMOS technology offers greater possibilities here, but also requires a locally different activation of sensor areas, so that these functions can be implemented with more or less great effort or have to be developed as special configurations.

CIS scanner bars 9, consisting of several sensor modules, are preferably used for the method according to the invention. An example of such a CIS scanner bar 9 is in 2 shown in section. The actual sensor modules 8 are clearly visible here, as well as the LEDs 6, which enable the illumination of the printed print image 5 to be captured and also the optics in the form of the lens 7. The signal processors 21 are also attached directly inside the scanner bar 9, which Process image data and, together with the signal processors of the other sensor modules, generate the digital printed image 16. This is then forwarded to the image processing computer 15 via the hardware interface 22 . These CIS scanner bars 9 preferably use CMOS sensors 8, but the use of CCD sensors is also possible if they provide the required performance data in terms of image resolution and data acquisition speed. Image data with a reduced resolution 3 of 50 to 200 dpi are recorded for color measurement. Image data with a higher resolution such as 200 or 600dpi is only recorded in selected image areas. This image data with higher image resolution is used for register measurement 2 (600dpi) or for inspection with higher resolution (200dpi), eg for barcodes 4.

Another particularly preferred embodiment is the use of a CMOS camera 9, provided that it can capture the entire printed image 5. In the case of CMOS cameras 9, the individual image sensors 8 can be controlled, which makes it even easier to capture images with a region-wise image resolution than in the case of CIS scanner strips 9. The use of a CMOS camera 9 for the method according to the invention does not differ from the use of a CIS Scanner bar 9. The control of the CMOS camera 9 differs from the described control of the CIS scanner bar 9 only insofar as structural differences require this. The following description of the image acquisition system 10 and the method according to the invention uses a CIS scanner bar 9; however, the use of a CMOS camera 9 is just as possible and required.

The image acquisition system 10 used is shown schematically in 3 described. In addition to the printing machine 14 itself, it contains a CIS scanner bar 9 mounted in the printing machine 14, an image processing computer 15 which is connected directly to the CIS scanner bar 9 and is connected to the control computer 12 of the printing machine 14. The control computer 12 with its display 13 represents the interface to the user 11 . The CIS scanner bar 9 itself is installed inside the printing press 14, usually behind the last printing unit. It records the printed image 5 to be inspected. By capturing the printed, original image 5 from the print job, a digital overall image 16 of the print areas 2, 3, 4 is created, which is then compared with a digital target print image by the image processing computer 15 in the course of further image acquisition.

4 shows schematically the sequence of the method according to the invention in a preferred embodiment. For this purpose, the printed image is divided into different areas by a computer 12, 15, preferably the preliminary computer. The areas correspond to the locations 2, 3, 4 of the printed image 5, 16 to be inspected. Each area 2, 3, 4 is then assigned the optimal image resolution 19 for the intended image acquisition. Which image resolution is optimal depends on the type of image acquisition already presented. An example of one printed print image 5 on a print substrate 1 with different image areas 2, 3, 4 is in 1 shown.

From the various areas of the printed image in its digital form with the assigned image resolutions 17, a software interface specification 18 is then created by the computer 12, which precisely assigns at which point in the printed image 17 the scanner bar 9 with which image resolution the respective area 2, 3, 4 should capture. The software interface specification 18 precisely defines which image regions 2, 3, 4 in the x/y direction on the printed image 17 use which image resolution. It is then transmitted from the computer 12 to the image processing computer 15 of the image acquisition system 10 via a software interface and enables this to precisely evaluate the selected image regions 2, 3, 4 with the respectively selected image resolution. The structure of the transmitted image data 17 thus depends on this software interface specification 18 . After that, the printed image data 5 is captured during production as part of the image capture. The software interface regulation 18 also serves to correctly assign the recorded image data to the respective image recording location when the data is read out. The results of the image acquisition with locally optimized image resolution can then also be communicated to the user 11 via the output devices 13 of the computer 12, 15 and/or the printing machine 14 in a completely normal manner, as before. In addition, output or storage in a log is also possible.

It is also advantageous if the image acquisition system 10 can quickly, i.e. within 20 image lines, accept a changed software interface specification 18 so that the acquisition of a printed image 5 can be optimized as far as possible for the image content 2, 3, 4.

The advantage of the method according to the invention is that the interface 22 for image processing usually limits the transmission of overall images with high resolution (600 dpi) and at high speeds of the print substrate transport 1 (>5 m/sec). This disadvantage is circumvented by reading out and transmitting image areas 2, 4 with high image resolution only in certain areas, ie partially. Only the image areas 2, 4 which require a high resolution are also read out with the high resolution. As a result, the limit speed of the image recording is increased for a predetermined bandwidth of the image data transmission and the amount of data of the recorded printed image 16 is reduced to the absolutely necessary amount.

A disadvantage, on the other hand, can be that the greater hardware and software outlay on the part of the scanner bar 9 and when assembling an overall image 16 in image processing may result in higher costs. However, if you compare the costs with those for installing multiple cameras, as is known from the prior art, where the work of the individual cameras also has to be precisely coordinated with one another at great expense, the invention according to the preferred exemplary embodiment is significantly more efficient.

Reference List

1

print substrate

2

Registration mark

3

color bar

4

image object (barcode)

5

printed image

6

LEDs for lighting

7

lens

8th

sensor module

9

CIS scanner bar

10

image capture system

11

user

12

Control computer of the printing press

13

display of the control computer

14

printing press

15

image processing computer

16

digital print image

17

print image areas to be inspected

18

interface rule

19

Print image areas with associated image resolution

20

Image capture result

21

signal processor

22

hardware interface

Claims (12)

Method for image acquisition in a machine (14) processing printing materials using a computer (12, 15), image data (16) of a print job being analyzed by the computer (12, 15) and divided into image areas (2, 3, 4) with different image content are assigned, different image resolutions are assigned to the different areas (2, 3, 4), the printed image data (5) are recorded and analyzed by means of an image sensor (9) of an image acquisition system (10), consisting of individual parts, characterized in that the individual Parts of the image sensor (9) are able to use different image resolutions and they are controlled by the computer (15) in such a way that the different image areas are recorded with the respectively different, assigned image resolutions (19) and forwarded to the image acquisition system (10). . procedure after claim 1 , characterized in that the areas (2, 3, 4) with different image content include image areas for register measurement (2) and/or for color measurement (3) and/or for general image inspection (4). procedure after claim 2 , characterized in that the image areas for register and register measurement (2) have an image resolution of at least 200 dpi, the image areas for color measurement (3) an image resolution of at least 50 dpi and the image areas for general image inspection (4) for the inspection of machine-readable data codes such as barcodes, an image resolution of at least 200 dpi, for the inspection of other areas an image resolution of at least 100 dpi. Method according to one of the preceding claims, characterized in that the computer (12) creates an interface specification (18) from the image data with the image areas (19) with different resolutions, which assigns the image areas (2, 3, 4) to the respective image recording location , transmits this to the image acquisition system (10), which controls the individual parts of the image sensor (9) in such a way that the individual image areas are evaluated with their respective associated image resolution (19). Method according to one of the preceding claims, characterized in that the individual parts of the image sensor (9) are controlled in such a way that in the image areas (3, 4) which are recorded with a reduced image resolution, the image information is weighted from several pixels to one pixel becomes. Method according to one of the preceding claims, characterized in that the image areas (2, 3, 4) are read out with different image resolutions with full RGB color information. Method according to one of the preceding claims, characterized in that the image acquisition of the image acquisition system (10) takes place at a transport speed of the print substrate (1) of at least 3m/s for offset printing machines and at least 0.6m/s for inkjet printing machines. Method according to one of the preceding claims, characterized in that if the image acquisition system (10) is used in an inkjet printing machine, a printed print image is divided into a plurality of image areas which are each sequentially acquired by the image sensor (9) with the highest possible image resolution and then be used to detect faulty print nozzles. Method according to one of the preceding claims, characterized in that when capturing the image areas for color measurement (3), the color resolution of the image sensor (9) is increased to more than 8 bits. Image acquisition system (10) for a sheet-fed printing press (14) with an inline measuring device (9) for carrying out a method according to one of the preceding claims, characterized in that the image sensor (9) is a modular CIS scanner bar (9) based on CMOS or CCD base or a CMOS camera (9), all parts of the CIS scanner bar (9) or the CMOS camera (9) for image recording with different image resolution sections can be controlled and read. Image acquisition system (10) according to claim 10 , characterized in that the image acquisition system (10) can accept a changed interface rule (18) within a maximum of 20 image lines. Image acquisition system (10) according to one of Claims 10 until 11 , characterized in that the image sensor (9) has sensor rows (8) with three or more color channels, with RGB sensor rows (8) for three color channels in a trilinear arrangement, mixed arrangement of sensor rows or by means of a prism camera, and also more than three sensor rows with individual color detection can be used.

Images