Title of Invention

A METHOD FOR EVALUATING THE QUALITY OF PRINTED MATTER PRODUCED BY A PRINING PRESS OR PRINTING MACHINE

Abstract This invention relates to a method for evaluating the quality of printed matter produced by a printing press or printing machine, with the printing machine producing several copies of the same printed matter, comprising the method steps of; providing an inspection system having an image sensor adapted to create an image data signal in response to sensing the printed copies; selecting a quantity of copies from the produced copies of the printed matter; evaluating said image data for the copies in said selected quantity with regard to at least one error type selected from a group of different error types comprising a color error, an intensity error, a contour error or a positioning error, where, within the selected quantity of copies, an error of a particular error type detected on at least one of the copies is evaluated in relation to at least one error of a different error type detected on the same copy or on a different copy. Classifying the printed matter as being of good or poor quality based on the evaluation, wherein said selected copies define a common capture of image data and wherein all errors are evaluated relative to one another as detected from the image data of the same capture.
Full Text The invention relates to a method for evaluating the quality of printed matter
produced by printing press or a printing machine with the printing machine
producing several copies of the same printed matter.
Print images generates by a printing machine are checked for print since a long
time by the operating personnel of the printing machine during a running
production, either visually or with the help of optical auxiliary agents. According
to the assessment of the operating person, a classification takes, i.e. a
categorization of the assessed print products in a group of print products with
prior laid-down features, i.e. certain error features that were the object of the
previous inspection. The total quality of the checked print products is classified in
a part quantity or grade with a good quality and a part quantity or grade with a
bad quality, i.e. not usable or marketable quality, whereby the print image of the
print products to be assessed is judged as either good or as bad, i.e. as
defective. Assessment of the quality of a print product by a printing machine by
the operating personnel is subject to extensive fluctuations, as the assessment
depends on the judging capacity and particularly the knowledge and experience
of the assessing person, and as a result varies from person to person.
Nowadays, camera systems are increasingly being used in the printing industry
for different applications, e.g. in inspection systems, web observation systems or
registration measuring systems, whereby these systems are arranged either in a
printing machine or in a machine processing print material. These systems carry


out their functions, for example, 'inline', i.e. during the running production of the
print product to be used, which poses a significant challenge on account of the
large data quality supplied by a camera system and the fast process sequence in
production of the print product for the respective camera system and an image
processing system evaluating its image data. The problem becomes more acute
if the print product has spectral-photometric features or identification features
that are difficult to identify, and in a quality control a reliable assessment is
required even for these identification features in spite of the usually high


transportation speed of the print product in the short time that is available. Furthermore,
valuable print products, i.e. in the production of currency notes, security bonds or
certificates, each individual identification feature of the print product has to be subjected
to a test. At the same time, due to economic reasons, there is the requirement that
particularly in high-valued print products, e.g. in the production of currency notes or
security bonds, precisely because of their high material costs and production costs, the
quantity of wastage has to be kept as less as possible, as long as it is justifiable on the
basis of a prior laid-down quality standard.
In the camera systems mentioned above, often electronic image sensors are used for
taking pictures, particularly colour cameras with at least one image sensor consisting of a
CCD-chip, whose light-sensitive pixel gives out an output signal corresponding to the
colour scanned in the observation zone, e.g. in three separated signal channels, i.e. the
colour channels at least for the colours red, green and blue.
There is the need to evaluate the output signal of a photographing unit, i.e. image data of
the picture taken by the photographing unit, with an image processing system connected
to the photographing unit in such a way, that a need-based, balanced assessment of the
quality of the print product produced by a printing machine can take place. For assessing
the quality, the print product is preferably tested for different criteria.
From the post-published document DE 103 35 147 A1 we know of a method for
determining the condition of bank notes in which data of at least two different properties
of the bank notes are evaluated, in which the data of at least two different properties of
each bank notes are linked with one another and the condition of the bank note is derived
from the linked data of different properties. It can also be foreseen that an average value
for each of the different properties is determined for a quantity of bank notes in order to
determine the condition of the quantity of bank notes for the respectively different
properties, or that an average value for the linked properties for a quantity of bank notes
is determined in order to determine the total condition of the quantity of bank notes.



The similarly post-published document DE 103 14 071 B3 relates to a methods for
qualitative assessment of a material with at least one identification feature, whereby with
an electronic image sensor at least one colour image is scanned of the identification
feature, whereby the image sensor directly or indirectly provides at least one first
electrical signal correlated to the colour image, whereby an evaluation device connected
to the image sensor evaluates the first electrical signal, whereby from at least one
reference image a second electrical signal is obtained and stored in a data storage,
whereby a second electrical signal has for at least two different properties of the reference
image respectively a rated value for the first electrical signal, whereby the first signal is
compared with at least two rated values contained in the second electrical signal, whereby
in the comparison at least the colour image of the identification feature is tested for a
colour variation from the reference image and the identification feature is tested for
association to a certain grade of identification features or tested for a definite geometric
contour or for a relative arrangement with respect to at least another identification feature
of the material. The material could be a bank currency note or a security bond. In any
case, it has to do with testing of the material, i.e. testing of a single piece with which at
least one identification feature of the concerned material is checked with respect to
certain criteria in different but independently conducted test sequences parallel to one
another.
From the document DE 102 34 086 A1 we know of a method for signal evaluation of an
electronic image sensor in the sample identification of image contents of a test body, in
which the categorisation of the test body to a particular grade of a test bodies is decided.
In this method, the content of an image scanned by the test body is evaluated on the basis
of an associative function formed on the basis of methods of fuzzy logic, whereby even
several associative functions can be linked with one another to a super-ordained
associative function.
From the document DE 102 34 085 Al we know of a method for analysing colour
variations of images scanned by an image sensor, whereby the image signal received by
the image sensor is analysed pixel-wise.


From the document DE 101 32 589 A1 we know of a method for qualitative assessment
of printed material with at least one identification feature, in which an image of the
material to be assessed is scanned by an image sensor, and for this image the geometric
contour and/or the relative arrangement of several identification features among one
another is evaluated in an evaluating unit.
It is the task of this invention to create a method for assessing a quality of a print product
produced by a printing machine, in which errors occurring in the production of several
copies of these print products are assessed in a balanced manner.
This task is fulfilled by the invention with the help of the features mentioned in claim 1.
The advantages that can be achieved with the help of the invention are mainly, that the
assessment of the quality of print product produced by a printing machine takes place in a
very balanced manner on a broad basis, because each detected error is not assessed
singularly but in the context of other identified errors, because a holistic assessment of all
errors that have occurred in a selective quantity of copies takes place in such a way, that
the errors are assessed in relation to one another, through which in the final result the
yield of copies of the print product classified as marketable or at least worthy of further
processing gets increased. The method thus increases productivity and cost-effectiveness
in the production process of the print products. A required quality standard is ensured
and unnecessary wastage is avoided.
As samples from the selected quantity of produced copies of the print product is
subjected to scanning by an image sensor generating common image data, there cannot be
any position differences between the detected errors, i.e. errors in their respective
location data, because the obtained image data is evaluated pixel-precise, so that in the
calculations for assessing the quality of samples of the print product produced in the
printing machine, corrections in the determined positions of the detected errors can be
dispensed with, which could however be necessary if errors are determined by multiple


sensors that are positioned differently with respect to the samples of the print product to
be tested. Therefore, it is a special advantage of the suggested solution, that all the errors
to be assessed in relation to one another can be detected from the image data of the same
scan.
Errors identified in a produced print product are weighed in relation to their topology,
e.g. combined in a super-ordained, multi-dimensional associational function and assessed
in a combined viewing of all detected errors on the basis of a preferably need-based-
parameter-able classification threshold. The method can also be utilized to evaluate the
assessment obtained from the image data with respect to a gradual variation in the quality
of produced print products. A gradual variation in quality of produced print products can
be identified before it grows into defects causing large wastage.
The method is particularly suited for assessing the quality of a high-valued print product
whose production is expensive, e.g. in valuable print product, e.g. a bank currency note or
a security bond.
A design example of the invention is depicted in the accompanying drawings and
described in details
below.
The following are shown:
Fig.1. A schematic depiction of an inspection system;
Fig.2. A part of the method in a signal flow diagram;
Fig.3. A depiction of a first aggregated associative function;
Fig.4. A depiction of second aggregated associative function;
Fig.5. A depiction of a parametric second aggregated associative function.
An inspection system used as a model for assessing the quality of a print product
produced by a printing machine has one or more colour-lines cameras 01 coupled with
one another, as shown in fig. 1, or an image scanning unit 01 designed as colour surface


camera 01, that scans a print image 03 lit up by a lighting unit 02, whereby a print image
03 has been generated with a printing machine on a printing material (not shown)
consisting of paper. The image data of the individual colour channels determined by the
image scanning unit 01 from the picture of print image 03 are evaluated in an image
processing system 04. The output of the result takes place, for example, on a monitor 06
connected to the image processing system 04. Inputs, e.g. parameters to be essentially
conveyed to the image processing system 04 for its calculations, are fed through a
keyboard 07 connected to the image processing system 04. The image scanning unit 01
is arranged in the printing machine in such a way that with each scan the respective print
image of several samples of the print product produced in this printing machine is
determined.
The printing machine is preferably designed as a rotary printing press, especially as a
printing machine printing in offset method, in a steel plate relief method, silk screen
method or in a hot-press method. If the printing machine is designed as a sheet-fed offset
machine, it should be ensured that the sheet can be inspected even at a machine speed of
say 18000 sheets per second. If the printing substance to be printed on is a material track,
then the inspection system should be in a position to subject the quality of samples of the
print product that are guided through the printing machine at a machine velocity of 15
m/sec. to inspection of single pieces.
Errors occurring in the production of the print product, e.g. a bank currency note, can be
categorised for certain types of errors, e.g.
a) colour error, if at a certain position of the print material a wrong colour has
been printed;
b) error of intensity, if the correct colour tone has been printed at a particular
position of the printing substance, but however not in the desired correct
colour intensity;
c) contour errors, if the print image or an identification feature of the print image
is at least partly defective in its outline, i.e. particularly incomplete; or


d) layout errors, if a mask thread or any other identification feature of the print
image is missing or appearing in a wrong location.
The error types can once again the categorised with respect to certain properties, namely
whether the defect of a particular defect type occurs in a series of several samples of the
print product produced in a printing machine, e.g. as single error or as multiple error.
Also colour defect and intensity defect can be classified and evaluated with respect to the
respective error magnitude, i.e. with respect to the area-wise extent of the defect. Thus
the quality of the print product at least with respect to certain types of defects and/or a
defect quantity and/or a defect magnitude can be assessed.
On account of the production method applied in a printing machine it can be assumed
that in several sample of the print product printed one after the other, the occurring
printing defects occur column-wise relative to the printing cylinders of the printing
mechanism of the printing press, i.e. the errors get repeated on the print substance on a
line in its movement direction through the printing mechanism, by which a further
property of an error can be defined. If required, the mentioned error types and/or
properties of the errors can be supplemented with further error features. Particularly for
conducting the method for assessment of quality of a print product in a machine
processing the print material and connected to the printing press, it should also be
considered as a side-condition that an inspection of the print material can take place in a
so-called half-sheet evaluation or even alternatively only half-wise.
A starting situation for the method for assessing the quality of a print product could
consist of the following: that several inspection channels i with i = 1 to imax, here e.g.
with imax = 4 correspond to the four error types - colour error, intensity error, contour
error and layout error- are foreseen, that for each individual sample of print product the
error quantity M should be assessed M = 1 to Mmax or the error magnitude N with N = 1
to Nmax pixel of the image sensor, and that for several printed samples of the print
product following one another, occurring print errors the number K of the errors m
content in a column s should be considered as K = 1 to Kmax and the application or non-


application of the half-sheet evaluation should be considered as yes/no decision. The
method foresees that the four inspection channels i, the error quantity M or the error
magnitude N and the number K of the errors m contained in the same column s should be
fuzzified. A defuzzification for assessing the quality of a print product can consist of a
simple evaluation of an integer value L obtained from the method, in that it is checked
whether the integer value L obtained from the method is greater than a set threshold value
Lmax, i.e. L > Lmax by setting the threshold value Lmax follows the fixing of a degree,
from which the produced print product can fuzzified as good or as bad, i.e. defective, i.e.
with the threshold value Lmax the quality standard required for the print product is fixed.
All errors and/or properties to be assessed in relation to one another are detected from the
image data of the same scan of the image sensor, which is why a coordination of the
image data with respect to the location of a detected error and/or a detected peculiarity is
not required.
A part of a sequence of this method has been depicted as an example in fig. 2 in a signal
flow diagram. The signal shows a hierarchic structure of the method.
The fuzzification can foresee that the inspection channels i are arranged linearly in a first
associative function uc, e.g The error quantity M is similarly allocated linearly in a
second associative function uf, whereby the error quantity M is preferably limited to a
maximum number Mmax of the error M, in that the detected errors m with the maximum
number Mmax of the errors m can be weighed as a weighing factor. As a second
associative function µx one then obtains
The method for assessing the quality of a print product seeks to conjunctively aggregate
the first associative function uc and the second associative function ux, i.e. to link both
the associative functions µc; µf multiplicatively with one another. The multiplication of
both associative functions µc: µf gives a new first aggregated associative function µg1,
which can be depicted according to the example described here as follows:


Fig. 3 shows a graphic depiction of a first aggregated associative function µgl, whereby
as example 4 inspection channels i and for the error quantity the value Mmax = 20 have
been selected. The first aggregated associative function µgl has a value range between 0
and 1.
Even the number K of the errors m contained in a column s can be fuzzified, once again
preferably in a linear allocation, so that under the knowledge that in the column a number
Ns of successive printed samples of the print product are evaluated, a third associative
function us can be set up as us = with s The third associative function us
can similarly be conjunctively aggregated with the first associative function us and/or
second associative function uf. For example, by means of a conjunctive aggregation of
all three associative functions uc; uf; us one obtains a second aggregated associative
function µg2 that can be depicted as follows:
For the sake of simplicity, in the three associative functions µc; µf; µs linear allocations
were made for their respective elements. Of course, depending on the requirement, for
one or more of the associative functions µc; µf; µs also non-linear allocations are
possible.
Fig. 4 shows a graphic depiction of this second aggregated associative function µg2,
whereby as an example 4 inspection channels i, for the error quantity M the value Mmax
= 20 and for the number Ns of the successive printed samples the value Ns = 6 were
selected. Like the first aggregated associative function µgl, the second aggregated
associative function µg2 has a value range between 0 and 1 as given on the Y-axis of the
diagram. The linear allocations selected here as examples are clearly identifiable. The
second aggregated associative function µg2 is a multi-dimensional, here four-
dimensional function, in which for its depiction the Y-axis of the diagram has been used



twice, mainly for depicting the number Ns of successive printed samples of print product
per column s and for depicting the value range of this second aggregated associative
function ug2. The double utilization is made possible by the superimposition of the
individual samples of the print product per column s with a respective inspection channel
i, whereby a block size shown in the diagram gets enlarged with each edition of a further
inspection channel i.
According to the depiction of the second aggregated associative function µg2 in fig. 4,
for its value of µg2 = 0.3 a threshold value Lmax is defined through a horizontal surface
parallel to the basic surface of the diagram, whereby the surface forms a classification
threshold Lmax. Depending on the respective application, i.e. the respective required
quality of the print product to be produced, the classification threshold Lmax is
preferably set for ug2 in the range of 0.2 to 0.4. From the example shown in fig. 4 it is
clear that in the parameters selected here as model for error detection, with only one
single inspection channel i, i.e. i = 1, even for an error quantity M of 15 errors m, a
sample of the print product to be tested for its quality is still assessed as good. Only
during error detection with two inspection channels i, i.e. i = 2, and an error quantity M
of 10 errors m per sample of the print product to be tested, a printed sheet, assuming that
Ns = 6 samples of the print product in a certain column s are arranged on the print sheet,
is assessed to be of bad quality and preferably removed from the production flow.
The second aggregated associative function µg2 is parameter-able in an extension to the
extent that a weight-age g with respect to the inspection channel i can be controlled. In
this case, for the second aggregated associative function µg2 one obtains the following
depiction:
µg2 =
with i
Fig. 5 shows a graphic depiction of a parameterized second aggregated associative
function ug2, in which as an example 5 inspection channels i, for Mmax the value Mmax


= 20, for Ns the value Ns = 6 and for the inspection channels i the weightage g of g = 0.3
have been selected. The classification threshold Lmax was again fixed at µg2 = 0.3.
Similarly, in the method for assessing the quality of a print product, the error quantity M
can be substituted by the error magnitude N, or the error magnitude N can be taken in as
a further criterion.
The method described for assessing the quality of a print product means in the application
that not each individual error detected on a printed sheet should lead to the fact that this
printed sheet should be removed as wastage. Rather, each individual detected error is
assessed in its context, whereby with the help of mathematical means, particularly by
applying method of fuzzy logic, the gravity of each error is weighed and/or assessed
particularly in alternating relationship in other detected errors and/or in proportion to
other detected errors. Accordingly, a holistic assessment of all errors takes place which
have been defined within the quantity of samples of the print product printed on a
particular sheet, whereby the errors detected within the selected quantity are assessed in
their respective relation to one another. The assessment of errors in their respective
relation to one another is favourable, in that all errors to be assessed are almost
simultaneously determined by the same image scanning unit 01 and all information
required for assessing the quality of the print product can be taken from the image data
corresponding to the scan.
For example, the risk in errors bank currency notes produced in the steel-plate relief
offset method is comparatively high; however even the material costs and the total
production costs of this print product are relatively high. With the help of the described
method, in the actual printing process a preliminary selection with respect to the printed
sheet can be made. Sheets that do not exceed the number of errors laid down in the
classification threshold Lmax are fed to a machine that will further process the sheet,
whereby each sample of the print product printed on the respective sheet can be subjected
once again to an individual test. Such a machine connected to the printing press could be
a cutting unit, particularly a cutting unit for individually separating the samples of the


print product printed on each sheet, which have already formed a number-wise restricted
quantity of samples for assessment of their quality. Such a number-wise restricted
quantity of samples could be in a few tens or even a few hundreds or more samples of the
print product. In the production of the print product, due to the successive sequence of
the produced samples of the print product, several such number-wise equal batches can
be selected one after the other in the production flow for quality test. Thus a fixed
number of associatively produced samples can be combined to a batch of samples,
whereby successively several batches of samples are formed. All produced samples are
preferably allocated to one of these batches. From each of these batches also an image of
their respective sample is made in order to subject the produced samples of the print
product to a flawless assessment of their qualities.
Each sheet having several copies of the print product, e.g. several bank currency notes
can be subjected to a further quality tests, in that during the post-processing those copies
of the print product which had been earlier classified as good quantity are removed, that
either reveal very big errors or a particularly high number of errors. As in the
preliminary test the entire sheet is not being classified as wastage, the yield of the
produced print product increases. At the same time, the post-processing does not have to
bear the load of sheets in which strong errors were not detected during pre-sorting.
The additional evaluation of the assessment obtained from the image data with respect to
a gradual variation in the quality of produced print product takes place by means of a link
with a data of at least one machine sensor. Such a machine sensor could be a vibration
recorder on a machine frame of the printing press. In a printing press printing in the
(wet-) offset method, the machine sensor can also be designed as sensor controlling the
feed of the wet agent. In a printing press printing in the (wet) offset method or in a steel-
plate relief method, it may be appropriate to measure the temperature of a tempering
agent tempering the forming cylinder of a printing press, especially a cooling agent
cooling the cylinder, with the help of a sensor to take the measured data of this sensor
into account additionally in the quality test of the print product produced in the printing
press. In a printing press printing in the steel-plate relief method it could also be



meaningful to additionally monitor the power consumption a wiping unit removing the
ink overflow from the steel-plate with a machine sensor and to consider the information
derived from the measured signal of the machine sensor about a too high or too low
wiping in the quality test of the print products produced by the printing press.
In the result, the errors detected from the common picture in the print image of the print
products to be produced are assessed in their relation to one another, whereby the
assessment thus obtained can be additionally linked to the information of at least one
further machine sensor in a control unit carrying out the evaluation, in order to especially
identify at an early stage any variation, especially gradually setting in variation in the
quality of the produced print product. From the measured signals of the at least one
additional machine sensor, the control unit can obtain the information that the printing
machine is for example in a print-technically critical operating condition, so that it is
probable that errors causing wastage in a short time manifest themselves on the sample of
the produced print products. Already now the control unit can intervene in the printing
process, in that at least one unit of the printing press influencing the printing process is
automatically post-guided by the control unit in order to bring the printing press back
from its print-technically critical operating condition to its proper operating condition.
Thus a control process or a regulating process evaluating measured signals of the
machine sensors serves the purpose of early detection of negative influences relevant for
the printing process, whereas the assessment obtained from the print image of the
produced print products especially confirms the adherence to quality specifications and, if
required, documents it in the sense of a quality proof.
On the other hand, depending on the assessment of the quality obtained from the print
image of the produced print products, those aggregates of the printing machine can be
post-regulated that are in a critical operating condition, whereby each of these units
influencing the printing process has a machine sensor monitoring these units, whereby
the control unit determines at least one unit negatively influencing the printing process on
the basis of the detected errors and/or the relevant measured signals of the respective
machine sensors and alters the setting of the at least one determined unit till such time as


the assessment of the quality of the print products obtained from the print image of the
produced print products again reaches the classification level of good. In this case, in
conjunction with the assessment obtained from the print image of the produced print
products, settings of units of the printing machine with the allied machine sensors are
checked regarding their relevance with respect to their quality of the print product to be
produced and, if required, automatically altered by the control unit for fulfilling the
quality specifications.
List of reference signs
1 Image scanning unit, colour lines camera, colour surface camera
2 Lighting unit
3 Print image
4 Image processing system
05

6 Monitor
7 Keyboard
g weightage
i, imax inspection channel
m errors
M, Mmax error quantity
N, Nmax error magnitude
Ns number of copies per column
K, Kmax number of errors per column
L numerical value
Lmax threshold value, classification threshold
s, smax column


µc associative function, first
µf associative function, second
µs associative function, third
µg1 aggregated associative function, first
µg2 aggregated associative function, second.


WE CLAIM
1. A method for evaluating the quality of printed matter produced by a
printing press or printing machine, with the printing machine producing
several copies of the same printed matter, comprising the method steps
of:
- providing an inspection system having an image sensor adapted to
create an image data signal in response to sensing the printed
copies;
- selecting a quantity of copies from the produced copies of the
printed matter;
- evaluating said image data for the copies in said selected quantity
with regard to at least one error type selected from a group of
different error types comprising a color error, an intensity error, a
contour error or a positioning error, where, within the selected
quantity of copies, an error of a particular error type detected on at
least one of the copies is evaluated in relation to at least one error
of a different error type detected on the same copy or on a
different copy; characterized by comprising:
- classifying the printed matter as being of good or poor quality
based on the evaluation, wherein said selected copies define a
common capture of image data and wherein all errors are


evaluated relative to one another as detected from the image data
of the same capture.
2. The method as claimed in claim 1, wherein the image data are captured
with an image sensor that detects colors.
3. The method as claimed in claim 1, wherein the evaluation derived from
the image data is analyzed with regard to a slowly building deviation in
the quality of the produced printed matter.
4. The method as claimed in claim 1, wherein the copies of the printed
matter are produced in a sequence.
5. The method as claimed in claim 4, wherein copies of the sequentially
printed matter are evaluated with regard to their quality by columns.
6. The method as claimed in claim 1, wherein a numerically limited
quantity of copies is selected out of the copies of printed matter that are
produced.
7. The method as claimed in claim 1, wherein the quality of the copies
belonging to the selected quantity of copies of the printed matter to be
evaluated is classified by an overall evaluation of all of the errors
detected within the selected quantity of copies.


8. The method as claimed in claim 1, wherein the severity of each error is
evaluated relative to the other detected errors using methods of fuzzy
logic.
9. The method as claimed in claim 1, wherein the types of errors are each
fuzzified in a classification function (µc; µf; µs).
10.The method as claimed in claim 9, wherein an aggregated classification
function (µg1; µg2) is formed by an aggregation of classification
functions (µc; µf; µs) that fuzzify at least two different error types.
11. The method as claimed in claim 10, wherein at least two classification
functions (µc; µf; µs) are conjunctively aggregated to the aggregated
classification functions (µg1; µg2).
12. The method as claimed in claim 10, wherein the aggregated
classification function (µg2) is formed at least four dimensionally.
13. The method as claimed in claim 9, wherein at least one of the
classification functions (µc; µf; µs) relates to a linear classification.
14. The method as claimed in claim 9, wherein at least one element in at
least one of the classification functions (µc; µf; µs) is weighted with one
parameter (g).


15. The method as claimed in claim 9, wherein the aggregated classification
function (µg1; µg2) is evaluated with regard to a classification threshold
(Lmax).
16. The method as claimed in claim 15, wherein the classification threshold
(Lmax) for the evaluation of the printed matter as belonging to a
quantity of copies that has been classified as good or poor is set at a
value.
17. The method as claimed in claim 16, wherein the classification threshold
(Lmax) is set at a value between 0.2 and 0.4.
18. The method as claimed in claim 1, wherein the method is performed in
the printing press or in a machine processing the printed copies of the
printed matter.
19. The method as claimed in claim 1, wherein an established number of
sequentially produced copies are respectively combined into a quantity
of copies.
20. The method as claimed in claim 19, wherein several quantities of copies
are formed one after the other.
21. The method as claimed in claim 20, wherein all produced copies are
assigned to one of these quantities of copies.


22. The method as claimed in claim 21, wherein, for each of these
quantities, a capture is produced of its respective copies.
23. The method as claimed in claim 1, wherein the evaluation of the quality
is made by a comparison of the image captured by the inspection
system with at least one reference image.
24. The method as claimed in claim 1, wherein the evaluation of the quality
occurs during the ongoing production process of the printing press.
25. The method as claimed in claim 1, wherein the evaluation of the quality
occurs in the ongoing production process of a machine processing the
printed copies of the printed matter.
26. The method as claimed in claim 24, wherein a quantity of copies of the
printed matter that has been classified as poor is removed from the
production process.
27. The method as claimed in claim 1, wherein the printed matter is
printed
in an offset printed method, in a steel engraving printing method, in a
serigraphy printing method, or in a hot embossing method.
28. The method as claimed in claim 1, wherein copies of the printed matter
are printed on several printed sheets.


29. The method as claimed in claim 28, wherein the copies of the printed
matter printed on the printed sheet are evaluated with regard to the
quality of their copies at a machine speed of up to 18,000 sheets per
hour.
30. The method as claimed in claim 1, wherein the copies of the printed
matter are printed on a material web.
31. The method as claimed in claim 30, wherein the copies of the printed
matter printed on the material web are evaluated with regard to their
quality at a machine speed of up to 15 m/s.
32. The method as claimed in claim 1, wherein the selected quantity of
copies of the printed matter is presorted by its classification with regard
to a subsequent process step.
33. The method as claimed in claim 32, wherein individual copies of the
classified quantity of copies of the printed matter are subjected to an
individual examination in the subsequent processing step.
34. The method as claimed in claim 3, wherein the analysis of the evaluation
derived from the image data with regard to a slowly building deviation in
the quality of produced printed matter occurs by correlating it with a
measurement signal of at least one machine sensor.


35. The method as claimed in claim 34, wherein a control device keeps the
printing press in an operating state that is proper with regard to printing
technology or returns it to such an operating state by evaluating the
measurement signal from the at least one machine sensor.
36. The method as claimed in claim 1, wherein the evaluation derived from
the printed image of the produced printed matter documents adherence
to quality standards.
37. The method as claimed in claim 3, wherein, in connection with the
evaluation derived from the printed image of the produced printed
matter, the settings of aggregates of the printing press are checked for
their relevance with regard to the quality of the printed matter to be
produced and changed by the control device in order to adhere to
quality standards.
38. The method as claimed in claim 4, wherein the characteristic of the error
relates to a number of errors in sequentially printed copies.
39. The method as claimed in claim 1, wherein the characteristic of the error
relates to the presence of an individual error or multiple errors.
40. The method as claimed in claim 1, wherein the characteristic of the error
relates to its respective error magnitude.


41. The method as claimed in claim 1, wherein the printed copies of the
printed matter are inspected in a half-sheet evaluation.

Documents:

01003-kolnp-2006 abstract.pdf

01003-kolnp-2006 claims.pdf

01003-kolnp-2006 correspondence others.pdf

01003-kolnp-2006 description (complete).pdf

01003-kolnp-2006 drawings.pdf

01003-kolnp-2006 form-1.pdf

01003-kolnp-2006 form-2.pdf

01003-kolnp-2006 form-3.pdf

01003-kolnp-2006 form-5.pdf

01003-kolnp-2006 international publication.pdf

01003-kolnp-2006 international search authority report.pdf

01003-kolnp-2006 pct form.pdf

01003-kolnp-2006-correspondence-1.1.pdf

01003-kolnp-2006-form-18.pdf

1003-KOLNP-2006-CORRESPONDENCE.pdf

1003-KOLNP-2006-EXAMINATION REPORT.pdf

1003-KOLNP-2006-FORM 18.pdf

1003-KOLNP-2006-FORM 26.pdf

1003-KOLNP-2006-FORM 3.pdf

1003-KOLNP-2006-FORM 5.pdf

1003-KOLNP-2006-FORM-27-1.1.pdf

1003-KOLNP-2006-FORM-27.pdf

1003-KOLNP-2006-GRANTED-ABSTRACT.pdf

1003-KOLNP-2006-GRANTED-CLAIMS.pdf

1003-KOLNP-2006-GRANTED-DESCRIPTION (COMPLETE).pdf

1003-KOLNP-2006-GRANTED-DRAWINGS.pdf

1003-KOLNP-2006-GRANTED-FORM 1.pdf

1003-KOLNP-2006-GRANTED-FORM 2.pdf

1003-KOLNP-2006-GRANTED-SPECIFICATION.pdf

1003-KOLNP-2006-OTHERS.pdf

1003-KOLNP-2006-REPLY TO EXAMINATION REPORT.pdf

abstract-01003-kolnp-2006.jpg


Patent Number 250573
Indian Patent Application Number 1003/KOLNP/2006
PG Journal Number 02/2012
Publication Date 13-Jan-2012
Grant Date 10-Jan-2012
Date of Filing 20-Apr-2006
Name of Patentee KOENIG & BAUER AKTIENGESELLSCHAFT
Applicant Address FRIEDRICH-KOENIG-STR. 4, 97080 WURZBURG
Inventors:
# Inventor's Name Inventor's Address
1 DIEDERICHS, CARSTEN ROSENGARTEN 4A, 32657 LEMGO
2 TURKE, THOMAS MYRTENWEG 19, 33699 BIELEFELD
3 WILLEKE, HARALD HEINRICH ROBERT-KOCH-STR. 12A, 33102 PADERBORN
4 LOHWEG, VOLKER LINNERSTR. 35, 33699 BIELEFELD
PCT International Classification Number G07D7/00; B41F33/00; B41F33/14
PCT International Application Number PCT/EP2005/051525
PCT International Filing date 2005-04-06
PCT Conventions:
# PCT Application Number Date of Convention Priority Country
1 102004019978.7 2004-04-23 Germany