Method and apparatus for detecting and / or testing portions of adhesive arranged on a substrate

Abstract

The invention relates to a method and an apparatus in which portions of adhesive (15) arranged on a substrate (12) are detected and / or tested during a production process, in which the portions of adhesive (15) are applied to the substrate (12), by means of a sensor (18, 29 A, 29 B) which generates measured values and is directed at the substrate (12), wherein the actual measured values (25) generated by the sensor (18, 29 A, 29 B) during the course of the method are evaluated, in particular by means of an analysis device, as to whether they have a measured-value group (21-24, 32) characteristic of a portion of adhesive (15). The method is characterized in that, within the scope of the evaluation, in the event of an actual measured-value group (21-24, 32) which could be characteristic of a portion of adhesive (15) being established, a correlation method is used to determine a measure of possible correlation - correlation measure - between on the one hand said actual measured-value group (21-24), or a value group derived from said actual measured-value group (21-24, 32), and on the other hand a target value group (28, 34) stored in a memory.

Description

[0001] Method and device for detecting and / or testing portions of adhesive arranged on a substrate

[0002] Description

[0003] The present invention relates to a method and a device for detecting and / or checking adhesive portions arranged on a substrate during a production process in which the adhesive portions are applied to the substrate, by means of a sensor directed towards the substrate that generates measured values, wherein the actual measured values ​​generated by the sensor during the method are evaluated, in particular by means of an analysis device, to determine whether they exhibit a group of measured values ​​characteristic of an adhesive portion.

[0004] In production machines, such as packaging machines for the manufacture or packaging of smoking, hygiene, food, or pharmaceutical products (e.g., cigarettes), it is often necessary to apply portions of adhesive to substrates used in the production process, such as packaging materials. Packages are frequently made from flat packaging material blanks that are folded during production, and the folded flaps are glued together. During the packaging manufacturing process, it must be checked whether the adhesive portions have actually been applied to the expected positions and / or whether they meet specific requirements to ensure the necessary quality of the adhesive bonds.

[0005] Several methods are known for this purpose. One method involves illuminating the surface area of ​​the packaging material to be inspected from the side and indirectly detecting the presence of adhesive particles via their shadows. This method has the disadvantage that the contour of each adhesive particle changes with increasing speed at which the packaging material is conveyed during the manufacturing process, thus reducing the accuracy of shadow detection. Furthermore, there is usually insufficient space in the packaging machine for the necessary illumination.

[0006] Especially when using hot melt adhesive portions, it is known to measure their thermal radiation using suitable sensors and compare it to limit values. One disadvantage of this method is that variations in the quantity or shape of the adhesive portions lead to different intensities of thermal radiation, so that the corresponding adhesive portions may then be classified as defective, even though they could still ensure a reliable bond.

[0007] It is also known to use distance sensors to record height profiles on the respective substrate and then analyze whether they contain elevations representing adhesive portions. However, such elevations can also be caused by contamination of the substrate, which in such a case might be incorrectly interpreted as adhesive portions.

[0008] Based on this, the object of the present invention is to further develop the aforementioned method and the aforementioned device.

[0009] This problem is solved by a method having the features of claim 1 and a device having the features of claim 12.

[0010] Accordingly, a method according to the invention is characterized in that, in the context of the evaluation of the actual measured values, if an actual measured value group is determined that could be characteristic of an adhesive portion, a measure of possible correlation - correlation measure - is determined by means of a correlation method between, on the one hand, this actual measured value group or a value group derived from this actual measured value group (for example, derived by normalization of the actual measured value group) and, on the other hand, a target value group stored in a memory.

[0011] The use of such a correlation method according to the invention makes it possible, among other things, to identify actual measurement groups that deviate unexpectedly from standard or target adhesive portions and would not have been recognized as adhesive portions or might have been classified as defective using other methods, as proper adhesive portions. Advantageously, the type of measurement acquisition or the sensor used is not particularly important in this regard. The method according to the invention can be applied, among other things, to the aforementioned infrared sensor-based method as well as, for example, to distance measurements using distance sensors.

[0012] As already indicated above, the production process during which the detection and / or testing method according to the invention is applied may preferably be a production process for the manufacture or packaging of smoking, hygiene, food or pharmaceutical products.

[0013] Preferably, the substrate can also be, for example, a blank for packaging, such as made of paper, cardboard, or another packaging material, which is processed as part of such a production process. Alternatively, the substrate can also be web-shaped packaging material.

[0014] Preferably, as part of the evaluation, the aforementioned correlation measure can be compared with a correlation measure comparison value stored in a memory, in particular with a limit value for this correlation measure, and depending on this comparison, the actual measurement group can be evaluated as being caused or not caused by a portion of adhesive or by a portion of adhesive corresponding to at least one predetermined criterion.

[0015] If the actual measurement group is then evaluated as being caused or not caused by a portion of adhesive or by a portion of adhesive corresponding to at least one predetermined criterion, it may be provided that a suitable signal is preferably generated by the analysis device, which is representative of this evaluation.

[0016] If, for example, the actual measurement group is assessed as not being caused by a portion of adhesive or as not being caused by a portion of adhesive that meets at least one predetermined criterion, such a signal can be an error signal. This error signal could then, for example, trigger the removal of the respective substrate from the production process or the stopping of the production process, such as by stopping a production machine involved in the process.Within the scope of the invention, it can further be provided that several different correlation measure comparison values ​​are stored in the memory, in particular in a table, and that one of the stored correlation measure comparison values ​​is selected manually (by an operator) or automatically for comparison, in particular according to one or more parameters and / or the shape and / or size and / or temperature and / or the application speed and / or the type of adhesive of the adhesive portions to be applied to the substrate during the production process. In this way, it would be easily possible to adapt the method according to the invention should it become necessary, within the production process or even after a machine stoppage in a later production process, to use different adhesive portion parameters for applying the adhesive portions than originally intended.Unlike certain state-of-the-art detection or testing methods, it would then not be necessary to recalibrate the sensor used or to adjust it to the new parameters of the adhesive portions.

[0017] The aforementioned selection of one of the stored target value groups could, for example, be made via a user interface of the analysis device, such as a touchscreen or the like, through which the stored target value groups can be selected.

[0018] Regarding the actual measurement group or the derived value group, each could be assigned its own characteristic curve (for example, through curve fitting) and the target value group could form such a curve, with the evaluation then based on these characteristic curves. The use of such characteristic curves facilitates the evaluation of the actual measurement values ​​of the actual measurement group, which are usually (but not always) discrete and generated by the sensor at a predefined sampling rate.

[0019] The correlation method can then involve comparing the geometric shape of the characteristic curve of the actual measurement group, or the characteristic curve of the derived value group, with the geometric shape of the characteristic curve of the target value group. From this comparison, the correlation measure can be determined, which is then compared with the stored correlation measure reference value.

[0020] Regarding the target or reference value group, several different target value groups, preferably derived from empirical data, can be stored in the memory, similar to the correlation measure comparison values. This is typically done in a table. For the correlation procedure, one of the stored target value groups can then be selected manually or automatically (again, if necessary, in a similar or analogous manner to that described above for the correlation measure comparison values, by a user via a user interface). This selection can also be based on one or more parameters and / or the shape and / or size and / or temperature and / or application speed and / or type of adhesive of the adhesive portions to be applied to the substrate during the production process.

[0021] In further specifying the concepts of the invention, the substrate can be moved past the sensor during the production process, particularly by a conveyor, to record the actual measured values. In particular, a control device queries the current actual measured value of the sensor at several different times, as well as the current actual rotation angle of a drive shaft of the conveyor and / or the actual machine angle of a production machine encompassing the conveyor. Each actual measured value of the sensor queried at a given time is linked to the actual rotation angle of the drive shaft or the actual machine angle queried at the same time, forming a pair of values. This procedure, as those skilled in the art will recognize, allows the position of the respective adhesive portions on the substrate to be determined and, if necessary,to verify, for example by comparing it with a respective target position.

[0022] Furthermore, it can be additionally or alternatively provided that the control unit queries both the current actual measurement value of the sensor and the actual position of the substrate, particularly relative to a reference point, at several different times. Each actual measurement value of the sensor queried at a given time is then linked with the actual position queried at the same time, forming a pair of values. This procedure allows, in particular, the determination or verification of the number of adhesive portions on the respective substrate, especially by comparison with a target number.

[0023] Further features of the present invention will become apparent from the attached claims, the following description of preferred embodiments, and the attached drawings. These show:

[0024] Fig. 1 shows an oblique view of a section of a production machine, in this case a cigarette packaging machine, with a testing device with infrared sensors, on which adhesive portions, in this case made of hot melt adhesive, which have previously been applied to partially completed packages, are detected and / or tested using the method according to the invention.

[0025] Fig. 2 shows an oblique view of an alternative testing device.

[0026] Production machine in which distance sensors are used to detect portions of adhesive,

[0027] Fig. 3 shows a diagram illustrating one of the sensors of the

[0028] The test device from Fig. 1 shows actual measured values ​​(Y-axis) with several groups of actual measured values, in this case intensity measured values ​​of the infrared sensors, at their respective assigned time, position, or machine angle values ​​(X-values).

[0029] Fig. 4 shows a diagram after the actual measured values ​​of the actual measured value groups were taken from

[0030] Fig. 3 were nominated and a characteristic curve was laid through the actual measured values ​​of the actual measured value groups,

[0031] Fig. 5 shows a diagram of a target value group stored in curve form in a memory of the test device for the actual measurement value groups from Fig. 3.

[0032] Fig. 6 shows the diagram from Fig. 4, supplemented by the respective characteristic curve of the target-

[0033] Figure 5 shows a value group for various actual measurement groups, as well as an associated diagram in which the values ​​r of a correlation coefficient r, which represents the degree of possible correlation between the characteristic curve of the respective actual measurement group and the stored target value group, are assigned to the X-values ​​of the diagram from Figure 4. Figure 7 shows a diagram in which, as an example for a pair of sensors on the left in Figure 2, consisting of an upper and a lower sensor of the test device, actual measurement groups are shown, which this pair has generated, namely height profiles or distance measurements (Y-axis) as characteristic curves for each assigned time, position, or machine angle values ​​(X-values), and in which the characteristic curve of a difference height profile formed from the difference of these actual measurement values ​​is shown.

[0034] Figs. 8, 9 Diagrams of two selected actual measurement groups from Fig. 4, each relative to the stored target value group from Fig. 4.

[0035] Figs. 10, 11 Diagrams of two selected actual measurement groups from Fig. 7, each relative to a corresponding stored target value group from Fig. 11.

[0036] Figures 1 and 2 show sections of a production machine 10, a packaging machine for manufacturing packages for products of the cigarette industry, such as classic tobacco products like cigarettes, cigarillos, and the like, but also novel products such as heat-not-burn products or liquid carriers for e-cigarettes. In the present embodiment, which is not to be understood as limiting, the production machine 10 serves to manufacture packages 11, each containing at least one group of cigarettes. The packages 11 are manufactured in a manner known per se from packaging material, such as blanks or webs of cardboard or paper, blanks or webs of film, etc.

[0037] In the machine section shown in Fig. 1, folded packages 11 made from a blank 12 are conveyed by a conveyor 13 along a (straight) conveyor path or conveyor section in an upstream machine section (not shown). In the area of ​​the conveyor path is a device 14 for detecting and / or testing adhesive portions 15 of hot melt adhesive applied to the blank 12 (which in this respect forms a substrate). These portions are applied in the production process directly upstream of the testing device 14 by means of an adhesive application device 16, also located in the area of ​​the conveyor 13, to opposite narrow side walls 17 of the respective package 11. The device 14 is hereinafter collectively referred to as the "testing device".

[0038] Each narrow side wall 17 is assigned a sensor 18 of the test device 14. The viewing direction of each sensor 18 is directed towards the respective narrow side wall.

[0039] 17 is directed so that it can detect the adhesive portions 15. The respective sensor 18 is a sensor that can detect heat radiation emanating from the adhesive portions 15, for example an infrared sensor.

[0040] As explained with reference to Figures 3-6, 8, and 9, the measured values ​​recorded or generated by the test device 14 or the respective sensor 18 are evaluated in a special manner, namely, in this case, with an analysis device (not shown), which can also be part of the control system of the production machine. As those skilled in the art will recognize, the analysis device can, for example, comprise an (industrial) PC or other computing device, which is configured with suitable software to perform the evaluations described below.

[0041] The diagram in Fig. 3 shows individual actual measured values ​​25 of the intensity of the thermal radiation (Y-axis) that the respective sensor 18 measures when the conditions in Fig.

[0042] 1. The package 11 or the blank 12 shown is moved past the respective sensor 18 by the conveyor 13.

[0043] The actual measured values ​​25 are shown here as a function of the respective position of the package 11 or blank 12 relative to the sensor 18 (X-axis). The corresponding assignment is achieved by the control unit of the production machine querying the current actual measured value of the sensor at several different times, as well as, for example, the current actual rotation angle of a drive shaft 19 of the conveyor 13, which can be detected by means of an encoder or rotary encoder 20.

[0044] Each actual measured value queried at a given time 25 of the respective sensor

[0045] The control unit or the analysis unit then links the actual rotation angle of the drive shaft 19, which is queried at the same time, to form a pair of values. These pairs of values ​​can then be displayed in a diagram as shown in Fig. 3. It is understood that, instead of the actual rotation angle of the drive shaft 19, the actual machine angle of the production machine 10 could also be used to form such pairs of values, or any other actual value that allows for a suitable assignment, such as simply the respective time of measurement.

[0046] As can be seen, the sensor 18 has acquired or generated four actual measurement groups 21, 22, 23, 24 as raw data, each consisting of individual actual measurement values ​​25, which, as the evaluation device can recognize using known analysis methods, could each be characteristic of the presence of an adhesive portion 15. The actual measurement group 24 also includes an outlier measurement value 26. Furthermore, the sensor 18 has generated disturbance values ​​27, which are arranged between the actual measurement groups 22 and 23.

[0047] Crucially, according to the invention, in the further evaluation, a measure of possible correlation is determined for each of the actual measurement groups 21-24 using a correlation method. This measure indicates whether, and to what extent, the respective actual measurement group 21-24 correlates with a predetermined target value group 28 (of length n) stored in a memory assigned to the analysis device (see Fig. 5). Depending on the magnitude of the determined measure, the analysis device then decides whether the respective actual measurement group 21-24 was actually caused by a (proper) portion of adhesive 15 or not.

[0048] Specifically, in this case, the aforementioned raw data are first normalized using software in the analysis unit. During this normalization process, the disturbance variables 27 are filtered out and the outlier measurement value 26 is eliminated. The analysis unit then assigns a separate actual characteristic curve 21A, 22A, 23A, and 24A to each of these normalized raw data, or in particular to the actual measurement groups 21-24, for further evaluation (for example, using a fitting procedure), although this is not mandatory.

[0049] For the purpose of determining the measure of the correlation, the target value group 25 is also provided as a function or characteristic curve 28 A in the present case.

[0050] Such a reference or target value group 28, or its characteristic curve 28A, can be empirically determined in advance, i.e., before the respective production process, by recording the respective actual measurement groups or actual measurement signals of adhesive portions on target blanks in experiments and deriving an average value signal or average value group from them. In this way, the corresponding target value groups or their characteristic curves can be determined and stored for different adhesive portions or adhesive applications on the respective blank. These can then be selected by an operator, preferably via a suitable user interface, depending on the type of adhesive portions to be applied in the production process, particularly before the production process.

[0051] The actual, well-known correlation method can then be described, for example, by the following relationship:

[0052] Here, r represents a measure of possible correlation, also called correlation coefficient, n represents the length of the respective (normalized) target value group or its characteristic curve, S cr?7l for the respective standardized actual value group or its standardized characteristic curve and Ä °™ for the target value group or its characteristic curve.

[0053] During the correlation process, the respective normalized characteristic curve 21 A - 24 A of the respective actual value group 21 - 24 is compared with the characteristic curve 28 A of the target value group 28. The result is a comparison of the geometric shape of the characteristic curve 21 A - 24 A of the actual measurement group 21 - 24 with the geometric shape of the characteristic curve 28 A of the target value group 28. To determine the degree of correlation, the analysis device software-wise shifts the characteristic curve 28 A stepwise over the respective characteristic curve 21 A - 24 A of the respective actual value group 21-24, see the upper diagram in Fig. 6.

[0054] For each actual value group 21-24, an assigned amount of the determined correlation coefficient r or the correlation measure results, cf. the respective correlation peak 33 in the lower diagram in Fig. 6.

[0055] Depending on the magnitude of the correlation peak 33, the analysis device then assesses the respective actual value group 21-24 as either triggered by a proper adhesive portion 15 or not. In the case of actual measurement group 22, for example (see also Fig. 8), the correlation is so high that the assigned adhesive portion 15 is classified as proper; in the case of actual measurement group 23 (see also Fig. 9), this is not the case.

[0056] For the aforementioned evaluation, as already indicated above, it may be provided that the analysis device compares the respective correlation peak 33 with a correlation measure comparison value stored in a memory or the memory, in particular with a limit value for this correlation measure, and depending on this comparison evaluates the respective actual measurement group 21 - 24 as being caused or not caused by an adhesive portion 15 or by a proper adhesive portion 15 that meets at least one predetermined criterion.

[0057] In other words, the corresponding (potential) adhesive portion 15, assigned to the respective actual value group 21-24, can then be classified as either proper and / or present, or as improper and / or absent. Subsequently, or depending on this, the analysis unit can then generate a representative signal and forward it to the control unit of the production machine 10, which may then execute an action, such as stopping the machine or triggering the rejection of the corresponding blank 12 or package 11, for example, if the adhesive portion 15 was not assessed as such or at least not as proper and does not meet the specified quality criteria.

[0058] In this context, it may also be provided that several different correlation measure comparison values ​​are stored in the memory assigned to the analysis device, particularly in a table, and that one of the stored correlation measure comparison values ​​is selected manually or automatically for comparison. This is done in particular according to one or more parameters and / or the shape and / or size and / or temperature and / or the speed of application and / or the type of adhesive of the adhesive portions that are to be applied to the blank 12 during the production process.

[0059] Fig. 2 shows a test device 14 that is slightly modified compared to Fig. 1, in which distance sensors 29 A and 29 B are used instead of infrared sensors 18.

[0060] The distance sensors 29B are arranged below the conveying path of the conveyor 13. They allow for the measurement of the distance between the packages 11 and the surface area of ​​the respective folding flaps 30A and 30B located on the underside of the respective folding flaps 30A and 30B during the movement of the packages 11.

[0061] 15 adhesive portions are expected, a first height profile, cf. h un ten in the diagram in Fig. 7, the surface of the underside of the respective folding flaps 30 A, 30 B are measured.

[0062] Furthermore, distance sensors 29A are arranged above the conveyor section. These sensors measure a second height profile on the surface of the upper side of each folded flap 30A, 30B, in a surface area directly opposite the upper side of the respective folded flap 30A, 30B. At the time of measurement, no adhesive portions 15 are expected in this area, as no adhesive portions 15 are provided or applied there. In Fig. 7, the X-axis diagram shows the travel path of the packages 10 along the conveyor section, and the Y-axis shows the distance h of each folded flap 30A, 30B to its respective distance sensor 29A or 29B.

[0063] The diagram also shows the edge profiles of the folding flaps 30A, 30B, which indicate that they are not exactly horizontal or straight, but curved. Furthermore, a height offset 31 is visible between the respective folding flaps 30A, 30B of the package lid of package 11 and the respective folding flaps 30A, 30B of the cup-shaped box part of package 11.

[0064] The measuring principle for detecting the adhesive portions 15 using the distance sensors 29 A, 29 B is known and is based on the principle that at the points on the respective folded flap 30 A, 30 B where the distance between an adhesive portion 15 and the distance sensor 29 A, 29 B is measured as actual values, this distance is smaller than at the points where no adhesive portion 15 is present, and consequently the full distance between the surface of the respective folded flap 30 A, 30 B and the distance sensor 29 A, 29 B is measured directly. For this reason, the adhesive portions 15 appear in the height profile h un ten as elevations 32. Without the aforementioned curvatures of the fold flaps 30 A, 30 B and without the aforementioned height offset 31 (or generally without unevenness) it would possibly be sufficient to use only the height profile h unten then to examine whether the surveys 32 that are representative for the adhesive portions 15 are present therein.

[0065] The curvatures or the height offset 31, however, make such an evaluation of the height profile h more difficult. un ten. For example, these can also form elevations which are then erroneously perceived as portions of adhesive 15.

[0066] It is therefore known that the second elevation profile h is to be used first for the analysis of the measurements. un ten to be consulted and thereby the difference h obgn of the two height profiles h un ten, to let it flow in above. In the

[0067] The diagram in Fig. 7 is a correspondingly derived difference height profile h. s also registered.

[0068] As shown in the difference height profile h sAs can be seen, the influence of the height offset 31 and the curvature of the fold flaps 30 A, 30 B is eliminated or no longer visible due to the difference calculation. All (larger) elevations 32 in the difference height profile are caused by the adhesive portions 15.

[0069] According to the invention, however, in extension and improvement of the known procedure mentioned above, the individual actual characteristic curves 32 A of the surveys 32 forming the respective actual measurement groups are generated and treated by the analysis device similarly to how described above for the test device 14 from Fig. 1, namely, among other things, they are checked to see whether they exhibit a certain degree of correlation to a target characteristic curve 34 A of a stored target value group 34, cf. Figs. 10 and 11.

[0070] In the example of Fig. 10, a degree of correlation is found that leads to the analysis device evaluating the corresponding portion of adhesive 15 as proper, whereas in the case of Fig. 11 it does not.

[0071] ***** Reference symbol list

[0072] 10 Production machine 34 Target value group

[0073] 11 packs 34 A characteristic curve

[0074] 12 Cutting / Substrate

[0075] 13 sponsors

[0076] 14 Test device

[0077] 15 portions of adhesive

[0078] 16 Adhesive application device

[0079] 17 narrow side walls

[0080] 18 Sensor

[0081] 19 Drive shaft

[0082] 20 rotary encoders

[0083] 21 Actual measurement group

[0084] 21 A characteristic curve

[0085] 22 Actual measurement group

[0086] 22 A characteristic curve

[0087] 23 Actual measurement group

[0088] 23 A characteristic curve

[0089] 24 Actual measurement group

[0090] 24 A characteristic curve

[0091] 25 actual measured values

[0092] 26 outlier measurements

[0093] 27 disturbance values

[0094] 28 Target value group

[0095] 28 A Characteristic curve Target value group

[0096] 29 A upper distance sensors

[0097] 29 B lower distance sensors

[0098] 30 A Folding flaps

[0099] 30 B Folding flaps

[0100] 31 Height offset

[0101] 32 surveys

[0102] 32 A characteristic curves

[0103] 33 Correlation peak

Claims

Patent claims 1. A method for detecting and / or checking adhesive portions (15) arranged on a substrate (12) during a production process in which the adhesive portions (15) are applied to the substrate (12), by means of a sensor (18, 29 A, 29 B) directed towards the substrate (12) that generates measured values, wherein the actual measured values ​​(25) generated by the sensor (18, 29 A, 29 B) during the method are evaluated, in particular by means of an analysis device, to determine whether they exhibit a group of measured values ​​(21-24, 32) characteristic of an adhesive portion (15), characterized in that, within the scope of the evaluation, if an actual measured value group (21-24, 32) is determined that could be characteristic of an adhesive portion (15), a measure of possible correlation – correlation measure – between, on the one hand, this actual measured value group (21-24) or a group derived from it is determined by means of a correlation method. Actual measurement group (21-24,32) derived value group and, on the other hand, a target value group (28, 34) stored in a memory is determined.

2. Method according to claim 1, characterized in that, within the framework of the evaluation, this correlation measure is compared with a correlation measure comparison value stored in a memory or the memory, in particular with a limit value for this correlation measure, and depending on this comparison, the actual measurement group (21-24, 32) is evaluated as being caused or not caused by an adhesive portion (15) or by an adhesive portion (15) corresponding to at least one predetermined criterion, in particular by generating a signal representative of this, preferably by the analysis device.

3. Method according to claim 2, characterized in that several different correlation measure comparison values ​​are stored in the memory, in particular in a table, and that one of the stored correlation measure comparison values ​​is selected manually or automatically for the comparison, in particular according to one or more parameters and / or the shape and / or size and / or temperature and / or the speed of application and / or the type of adhesive of the adhesive portions (15) to be applied to the substrate (12) during the production process.

4. Method according to claim 1, 2 or 3, characterized in that the actual measurement group (21-24, 32) or the value group derived therefrom on the one hand and the target value group (28, 34) on the other hand are each assigned or are assigned to their own characteristic curve or form such a curve, and that the evaluation is carried out on the basis of these characteristic curves.

5. Method according to claim 4, characterized in that the correlation method comprises a comparison of the geometric shape of the characteristic curve of the actual measurement group (21-24, 32) or of the characteristic curve of the derived value group with the geometric shape of the characteristic curve of the target value group (28, 34).

6. Method according to claim 5, characterized in that the correlation measure is determined from this comparison and compared with the stored correlation measure comparison value.

7. Method according to one or more of the preceding claims, characterized in that several different target value groups (28, 34), preferably derived from empirical data, are selectably stored in the memory, in particular in a table, and that for the correlation method one of the stored target value groups (28, 34) is selected manually or automatically, in particular according to one or more parameters and / or the shape and / or size and / or temperature and / or the speed of application and / or the type of adhesive of the adhesive portions (15) to be applied to the substrate (12) during the production process.

8. Method according to one or more of the preceding claims, characterized in that the sensor (18, 29 A, 29 B) is a sensor for detecting thermal radiation (18), in particular an infrared sensor, or a sensor for detecting height profiles in surface areas of the substrate (29 A, 29 B), in particular a distance sensor.

9. Method according to one or more of the preceding claims, characterized in that the substrate (12) during the production process to record the actual measured values ​​(25), in particular with a conveyor (13) moving past the sensor (18, 29 A, 29 B), in particular that a control device queries at several different times both the current actual measured value (25) of the sensor (18, 29 A, 29 B) and simultaneously the current actual rotation angle of a drive shaft (19) of the conveyor (13) and / or the actual machine angle of a production machine (10) encompassing the conveyor (13), and that each actual measured value of the sensor (18, 29 A, 29 B) queried at a respective time is linked with the actual rotation angle of the drive shaft (19) or the actual machine angle queried at the same time, forming a pair of values.

10. Method according to one or more of the preceding claims, characterized in that the substrate (12) is moved past the sensor (18, 29 A, 29 B) during the production process to record the actual measured values ​​(25), in particular by means of a conveyor (13), that in particular a control device queries both the current actual measured value (25) of the sensor (18, 29 A, 29 B) and simultaneously the actual position of the substrate (12), in particular relative to a reference point, at several different times, and that each actual measured value of the sensor (18, 29 A, 29 B) queried at a given time is linked with the actual position queried at the same time to form a pair of values.

11. Method according to one or more of the preceding claims, characterized in that the production process is a production process for the manufacture or packaging of smoking, hygiene, food or pharmaceutical products.

12. Device for detecting and / or checking adhesive portions (15) arranged on a substrate during a production process in which the adhesive portions (15) are applied to the substrate (12), in particular for carrying out the method according to one or more of the preceding claims, comprising a sensor (18, 29 A, 29 B) generating measured values ​​and directed towards the substrate, and comprising an analysis device with which the actual measured values ​​generated by the sensor (18, 29 A, 29 B) can be evaluated to determine whether they exhibit a group of measured values ​​characteristic of an adhesive portion (15), thereby characterized in that the analysis device is designed and set up in such a way that, in the context of the evaluation, if an actual measurement group (21-24, 32) is determined which could be characteristic of an adhesive portion (15), it can determine a measure of possible correlation - correlation measure - by means of a correlation procedure between, on the one hand, this actual measurement group (21-24, 32) or a group of values ​​derived from this actual measurement group (21-24, 32) and, on the other hand, a target value group (28, 34) stored in a memory.

13. Device according to claim 12, characterized by one or more features of claims 1 - 11. *****

Images