DE10037098A1 - Method and device for checking cigarette heads - Google Patents

Abstract

The invention relates to a method for checking cigarette heads (11), wherein several areas (24) of a head (11) are irradiated with structured light. A detector (20) receives the reflected light and generates an image (25) associated with an area (24). When this figure (25) is evaluated and an incorrect cigarette is recognized, an exclusion signal is generated. However, known methods of this kind are imprecise. The invention improves such methods in that the irradiated areas (24) are each areal and each image (25) is evaluated on the basis of several different criteria. The invention also relates to a corresponding device (10).

Description

The invention relates to a method for testing cigarettes behead, with several areas of a cigarette head with the areas that are illuminated according to structured light light reflected from the cigarette head in such a way from one Detector is received that irradiated and received re reflected light at an angle, especially triangulations angle, run to each other, the irradiated areas on the Mapped detector and one each generated by the detector, one Area assigned mapping is evaluated to order at a reject signal due to an improper cigarette to create.

The invention further relates to a device for testing Cigarette heads with a light source for producing struk light to illuminate multiple areas of a cigarette head, a detector to receive from the Ziga rescue head reflected light, using light source and detector spaced apart and at an angle, especially tri Angulation angles are arranged to each other, and an Auswer Unit for evaluating each generated by the detector images assigned to an area and for generating a  Reject signal when there is an improper room Garette.

A test method or a test device for be contactless testing of cigarette heads, one of which is straight Line is blasted onto a tobacco-side cigarette head. If the cigarette is not properly filled with tobacco, the line appears - if it is from a different angle is considered - no longer as a straight line, but as Serpentine line or as a broken serpentine line. This Image is captured by a sensor. Eventually they will Pixels inside and outside a narrow area lie around an imaginary, theoretical straight line, counted and related to each other. If this ratio exceeds a limit, this is to indicate that a cigarette is not properly filled.

The disadvantage of this type of checking is that it is imprecise and no exact statements about the condition of a cigarette laubt.

The invention is therefore based on the problem of testing To improve cigarettes.

This problem is solved with a method of the beginning named type, which is characterized in that the irradiation th areas are each areal and each Ab education evaluated using several different criteria becomes. Furthermore, this problem is solved by a device of the type mentioned at the outset, which is characterized by that the light source irradiates the areas over a wide area and the evaluation unit is designed such that each image evaluation based on several different criteria.

In contrast to the prior art, they are not line-like Areas selected, rather area-like. On the one hand, this has the advantage that larger areas of a cigarette head can be checked. On the other hand, an areal ab feel the possibility of different types of examination methods or to use criteria that are only due to the area th training of the individual areas are possible, not ever  but with line-like formation of the irradiated areas. For example, the size of the area, its circumference, by knife, area center of gravity, as well as derived therefrom Sizes such as elongation and compactness of those generated by the detector image of an irradiated area was examined. hereby the possibility of different results in an advantageous manner Define criteria that can be used to decide whether a cigarette is proper or improper is trained. The use of several different ones Criteria for testing a cigarette thus offer an essential one much higher reliability, only improper cigarette discard and properly trained cigarettes in the To leave the production process. By the inventive Kom combination of extensive training of irradiated areas and Use of several different criteria in the evaluation the images generated by the detector can be checked by Cigarettes can be improved significantly.

Preferred embodiments of the invention result from the Subclaims and based on those shown in the drawing Embodiments. The drawing shows:

Figure 1 shows a test device according to an embodiment of the invention for testing the heads of a cigarette formation with a light source with a lens aperture system and a detector.

Fig. 1a shows a cross section through light emitted by individual light sources according to a section along the Li never IA-IA in Fig. 1;

Fig. 2 is a ver used in the lens aperture system of FIG 1;

Figure 3 shows the end face of a cigarette formation, each with three irradiated areas on each cigarette head.

Fig. 4 shows a cigarette head in which only one of three overlapping areas is currently being irradiated (highlighted in black);

5 shows a cigarette head with three mutually beabstan Deten irradiated areas.

Fig. 6 is an image of an irradiated area, as generated by detector in an improper cigarette;

Fig. 7 is an illustration of an irradiated area with only a partial area, as generated by the detector during a proper cigarette;

Figs. 8A to H the evaluation of cigarettes 25 each with reference ei nes other criterion; and

Fig. 9 shows the results of the evaluation of various criteria Kri in FIG. 8.

Fig. 1 shows a testing device 10 for testing cigarette heads 11 of a cigarette formation 12th The test device 10 preferably checks the tobacco-side ends of cigarettes, but it can also be used to test filter-side cigarette heads.

The test device 10 has a light source 13 which has a plurality of, preferably three, individual light sources, in particular laser diodes. The individual light sources 14 are each connected to light lines 15 , which guide the light emanating from the individual light sources 14 to a lens-aperture system 16 , which comprises one or more lenses 17 and an aperture 18 , which are in particular housed in a housing 19 .

The light emanating from the ends of the light pipes 15 facing the lens diaphragm system 16 is oriented essentially in parallel and has a circular cross section shown in FIG. 1a. The light emitted by the individual light sources 14 thus enters the lens-aperture system 16 essentially in the form of a column. The light is expanded here by the lenses 17 and formed by the aperture 18 in such a way that it emerges on the opposite side - the exit side - of the lens-aperture system 16 again essentially parallel, columnar structure and evenly shaped , Different areas of a cigarette head 11 can be illuminated in a sharply delimited manner by this structuring. That is, a cigarette head 11 is uniformly irradiated with light within a circumferential line defining an area. A cigarette head 11 is divided into several such areas. The areas cover the largest part of the end face of a cigarette head.

In an alternative exemplary embodiment, the light exits at the end of the lens-blending system 16 facing the cigarette heads 11 not in parallel, but rather in a converging or di-converging manner.

Light reflected by the cigarette heads 11 is received by a detector 20 , in particular a CCD camera. The light source 13 or the lens aperture system 16 and the detector 20 are spaced apart from one another at an angle, in particular triangulation angle 21 , arranged. In this way, the light beams irradiating the cigarette heads 11 and the reflected light beams received by the detector 20 also run at this triangulation angle to one another. The triangulation angle 21 is in the range between 10 ° and 80 °, preferably 20 ° and 60 °, more preferably 20 ° to 30 ° and is preferably essentially 25 °. The detector 20 has an output line 22 which is connected to an evaluation unit (not shown) by means of which images of the irradiated areas generated by the detector are evaluated. The evaluation unit generates a reject signal in the presence of an improper cigarette. Such a reject signal leads to the rejection of the cigarette or cigarette formation identified as being defective.

The individual light sources 14 are connected to a control unit (not shown), by means of which the individual light sources 14 can be switched on and off individually. This control unit can be operated in such a way that only one light source can be switched on at a time, while the rest remain switched off. In this way, the areas selected on the cigarette heads can be irradiated one after the other, so that interference between the areas is excluded. This is particularly advantageous if the areas are selected so that they partially overlap. However, if the areas are selected at a distance from one another or tangentially, the control unit can also be operated in such a way that the individual light sources are switched on and off simultaneously. The test takes place in such a way that each area of a cigarette head of a cigarette formation is irradiated only once. This can take place during a current standstill phase of the cigarette formation. Due to the fast reaction time of the individual light sources, however, it is also possible to carry out the test while the cigarette formation is moving past the test device 10 .

Fig. 2 shows the aperture 18 in an enlarged view. The aperture 18 each has a cigarette of the cigarette formation 12 associated opening 23 through which light can pass unhindered. The arrangement of the openings 23 corresponds to the arrangement of the cigarettes within the cigarette formation 12th

Fig. 3 shows the front side of the cigarette formation 12 of 20 cigarettes heads 11. Each cigarette head 11 is assigned several, namely three, uniformly, in particular circularly shaped regions 24 which are irradiated with light in succession or simultaneously. These areas 24 are created in the same way on each cigarette head 11 of the cigarette formation 12 thanks to the plurality of individual light sources 14 and the lens aperture system 16 . The light emanating from an individual light source 14 is expanded by the lenses 17 and structured by the diaphragm 18 in such a way that a circularly illuminated area 24 is generated on each cigarette head of the cigarette formation. The widening leads to an overlap of the regions 24 of different individual light sources 14 according to the manner shown in FIG. 3. With a smaller degree of expansion, an overlap of the regions can be avoided, so that the regions 24 are arranged at a distance from one another or tangentially to one another.

Fig. 4 shows a single cigarette head 11 in an enlarged view. Of the three areas to be irradiated, only the upper area 24 , shown in black, is currently illuminated, while the other two areas are not illuminated and are shown only schematically. All three areas 24 are partially overlapping. Only one of these areas 24 is irradiated with light.

Fig. 5 is a spaced arrangement of the regions 24 shows on a cigarette head 11. With this arrangement, all three areas can be irradiated with light simultaneously. Alternatively, the areas can be arranged such that they only touch at the edges.

FIG. 6 shows an image 25 generated by the detector 20 consisting of several, namely three partial surfaces 26 . Due to the angular arrangement of the light source 13 and detector 20 , in particular in the triangulation angle 21 , and the unevenness of the surface of a cigarette head 11 , an image of the region 24 irradiated with several partial surfaces 26 , for example as shown in FIG. 4, is formed in the detector 20 . lie between de seemingly unlit areas 27 . These apparently unlit areas 27 arise due to shadowing caused by the unevenness of the tobacco structure or filter structure. The more these bumps are formed, the larger the apparently unilluminated areas or the more unlit areas are created. This increases the length of the circumferential line in FIG. 25 or the total length of all circumferential lines of all partial surfaces 26 . The invention has recognized that the length of the circumferential line is a measure of the quality of the surface of a cigarette head 11 . The length of the connecting lines between the centroids of the partial areas 26 is preferably added to the circumferential line, preferably only the lengths of the connecting lines of the largest partial area 26 to the other partial areas being taken into account.

FIG. 7 shows a cigarette head 11 with a more uniform surface than in FIG. 6, ie a surface with less roughness. In this case, Figure 25 consists of only one surface, but the circular structure of the illuminated area is given an irregular border. Although 6 and 7 are shown the cigarette heads 11 circular in Fig.. In fact, however, with a complete illumination of the cigarette heads 11, an elliptical border of the cigarette heads would be imaged by the detector 20 .

If there is no tobacco on a cigarette head 11 , a figure 25 , for example shown in FIG. 7, tears into several partial areas 26 , for example shown in FIG. 6, parts of figure 25 are missing or the outline of figure 25 is less smooth. In order to evaluate these phenomena, the image signal received by the detector is binarized. The binarized image in this way is then evaluated on the basis of criteria described in more detail below.

FIGS. 8A-H show eight different criteria of the analyzed images and their results garetten for 25 ro. The 25 cigarettes are in each case plotted on the abscissa, while the respectively determined values of a criterion are plotted on the ordinate. Cigarettes 1 through 10 are properly formed cigarettes, while cigarettes 11 through 25 are improperly trained cigarettes.

FIG. 5B illustrates the criterion of a total area of an image 25, possibly taking into account their partial surfaces 26, represents. When the total area above a threshold GR _B, a cigarette is found to be good. Accordingly, a cigarette is found to be bad if the value of the total area is below this limit value GR _B.

FIG. 8F shows the sum of the distances of the centroids of the partial areas 26 from the center of gravity of the largest partial area 26 , weighted with the partial areas, without normalization to the total area according to FIG . If this weighted distance sum exceeds a limit value GR _F , the respective cigarette is found to be good. Accordingly, a cigarette is found to be bad if the limit GR _{F is} exceeded.

FIG. 8A shows a representation derived from FIG. 8F of the weighted distance sum of the centroids of the partial areas according to the above explanations, but normalized to the total area according to the representation in FIG. 8B. If a limit value GR _A for the weighted distance sum with standardization for the respective total area is undershot, a cigarette is found to be good. Accordingly, a cigarette is rated as bad if this limit value GR _{A is} exceeded.

Fig. 8D illustrates-making the maximum diameter of a Abbil, where applicable, is measured over a plurality of sub surfaces 26. The maximum diameter can also be determined as the diameter of the convex circumference. The convex circumference denotes the curve with the shortest length that encloses all partial areas of an image. A cigarette is judged to be correct if it falls below a limit value GR _D for the maximum diameter. Accordingly, a cigarette is found to be improper if this limit GR _{D is} exceeded.

Fig. 8H illustrates in a corresponding manner the minimum diameter of an image or by the partial areas of an image and the minimum diameter of the space enclosed by the convex peripheral line structure. If the minimum diameter determined exceeds a limit value GR _H , a cigarette is found to be correct. Accordingly, a cigarette is found to be improper if this limit GR _{H is} undershot.

FIG. 8C illustrates-making the average diameter of a Abbil, which can be determined as the average of the minimum and maximum diam ser. Alternatively, this diameter can also be determined as the mean value of all the diameters resulting with respect to the convex circumferential line. If the average diameter exceeds a limit value GR _C , a cigarette is found to be correct. Accordingly, a cigarette is found to be improper if this limit is exceeded.

FIG. 8E illustrates the so-called elongation, which is determined as the ratio of the maximum diameter to the minimum diameter. If the elongation exceeds a limit value GR _E , a cigarette is found to be correct; accordingly, a cigarette is found to be incorrect if this limit value is exceeded.

Fig. 8G illustrates the so-called. Compactness of an image. The compactness is a measure that is calculated from the squared length of the circumference, which is divided by the product "4 π total area". The length of the circumferential line of an image is calculated taking into account the lengths of the circumferential lines of the individual images belonging to the partial areas, in particular taking into account the lengths of the connecting lines of the centroids from the largest partial area to the other partial areas. If the value of the compactness exceeds a limit value GR _G , a cigarette is found to be correct; accordingly, a cigarette is found to be incorrect if this limit value is exceeded.

In addition to the criteria shown in FIGS. 8A to H, the following additional criteria can be used for the investigation: The length of the peripheral line of an image, possibly taking into account the lengths of the peripheral lines of partial areas of the image. The length of the connecting lines of the centroids of the largest subareas to the other subareas is also preferably taken into account.

Another criterion is the so-called fracture of an image dung. The fissure is determined from the length of the circumference line in the sense of one of the above definitions in the ratio nis to the length of the convex circumference.

Another criterion is the optically measured cigarette length. In particular, two methods are available for this. On the one hand, the cigarette length can be obtained from a triangulation measurement. This is advisable if the light directed by the lens aperture system onto the cigarette heads is formed in parallel. Here, a possible deviation, in particular a distance, the position of the image of an irradiated area from an expected position of an image of a corresponding area of an ideal cigarette head is determined and the length of the tested cigarette is calculated from the deviation. In another method, the length of the cigarette is calculated from the size or area of an image. A prerequisite for this method is that the light emitted by the lens aperture system 16 onto the cigarette heads is not parallel, but rather diverging or converging.

All of the criteria mentioned can be one for each figure irradiated area can be applied. With three irradiated Be Each of the criteria mentioned would therefore suffice per cigarette head  Use three times. The results can then be ge be averaged.

FIG. 9 shows a table of the evaluation of the examinations shown in FIGS. 8A to H. A dot in this table indicates a cigarette that was found to be improper by the criterion assigned to the corresponding column. A reject signal is generated by the evaluation unit if at least one such point is assigned to a cigarette, ie if at least one criterion has been found to be improper. The limits of the criteria GR _A to GR _H are chosen generously, ie a cigarette tends to be found to be good rather than bad.

At least two of the criteria explained are combined which combines to generate a reject signal if necessary. It can also be three, four, five, six, seven, eight or more criteria for forming a reject signal with each other be combined. The greater the number of combinations criteria, the higher the degree of reliability upon detection of an improper cigarette.

Illuminated cigarette head by evaluating images areas, especially within the range of cigarette papers closed face of a cigarette head, based on several different criteria can result in a very high detection rate improper cigarettes. By means of the Invention is therefore ver verifying cigarettes considerably repaired.  

LIST OF REFERENCE NUMBERS

10

Tester

11

cigarettes head

12

cigarette formation

13

light source

14

Single light source

15

light line

16

Lens-stop system

17

lens

18

cover

19

casing

20

detector

21

triangulation

22

output line

23

opening

24

Area

25

Illustration

26

subarea

27

apparently unlit area
GR _A

Limit value for weighted distance sum with normalization to total area
GR _B

Limit for total area
GR _C

Average diameter limit
GR _D

Maximum diameter limit
GR _E

Elongation limit
GR _F

Limit for weighted distance sum
GR _G

Compactness limit
GR _H

Minimum diameter limit

Claims (11)

1. A method for checking cigarette heads ( 11 ), wherein a plurality of areas ( 24 ) of a cigarette head ( 11 ) with these loading areas ( 24 ) are irradiated in accordance with structured light and light received by the cigarette head ( 11 ) is received in this way by a detector ( 20 ) the irradiating and received reflected light run at an angle, in particular a triangulation angle ( 21 ), to each other, the irradiated areas ( 24 ) are imaged on the detector ( 20 ) and one area ( 24 ) generated by the detector ( 20 ) ) associated image ( 25 ) is evaluated in order to generate a reject signal in the presence of a non-proper cigarette, characterized in that the irradiated areas ( 24 ) are each areal and each image ( 25 ) is evaluated on the basis of several different criteria , 2. The method according to claim 1, characterized in that the following are used as criteria:

• - The total area of a figure ( 25 ), possibly taking into account several partial areas ( 26 ) of the figure ( 25 );
• - The sum of the weighted with the partial areas ( 26 ) from the individual center of gravity of the partial areas from the center of gravity of one of the partial areas ( 26 ), in particular the largest partial area ( 26 ) with or without normalization to the total area;
• - minimum, maximum and / or average diameter of an image ( 25 );
• - The elongation of an image ( 25 ), in particular the ratio of the maximum to the minimum diameter;
• - The length of the circumferential line of an image, possibly taking into account the lengths of the circumferential lines of the partial areas of the image ( 25 );
• - The compactness, in particular the ratio of the square of the circumferential line to the total area, and / or
• - The fissure, in particular the ratio of the length of the circumferential line to the length of the convex circumference or the length of the shortest Li never enclosing all partial surfaces.

3. The method according to claim 1 or 2, characterized in that the regions ( 24 ) are arranged partially overlapping, adjacent to one another or spaced apart. 4. The method according to any one of the preceding claims, characterized in that the areas ( 24 ) are irradiated one after the other, in particular in the event of partial overlap, or at the same time, in particular in the case of an adjacent or spaced arrangement. 5. The method according to any one of the preceding claims, characterized in that the regions ( 24 ) each have a uniform, in particular circular, contour. 6. The method according to any one of the preceding claims, characterized in that the reject signal is generated when at least one of the criteria is not met, in particular exceeds or falls below a limit value (GR _AH ). 7. The method according to any one of the preceding claims, carried in that any deviation, in particular a distance of the position of the image (25) of be irradiated region (24) formation (25) from an expected position of a corresponding area from a ( 24 ) of an ideal cigarette head ( 11 ) and the length of the tested cigarette is calculated from the deviation. 8. The method according to any one of the preceding claims, characterized in that the areas ( 24 ) are irradiated with non-parallel light, in particular divergent or converging light, and the length, in particular a diameter, of an image is used to calculate the length of the corresponding cigarette becomes. 9. Apparatus for testing cigarette heads ( 11 ), in particular for carrying out a method according to one of claims 1 to 8, with a light source ( 13 ) for generating structured light, so that several areas ( 24 ) of a cigarette head ( 11 ) to irradiate, a detector ( 20 ) for receiving light reflected from the cigarette head ( 11 ), the light source ( 13 ) and detector ( 20 ) being spaced apart and arranged at an angle, in particular a triangulation angle ( 21 ), and an evaluation unit for evaluating images ( 25 ) generated by the detector ( 20 ), each assigned to an area and for generating a reject signal upon detection of an improper cigarette, characterized in that the light source irradiates the areas ( 24 ) in each case over a large area and the evaluation unit is designed such that each image ( 25 ) can be evaluated on the basis of several different criteria. 10. The device according to claim 9, characterized in that the light source ( 13 ) has several, in particular three, individual light sources ( 14 ), each of which is connected to a light guide ( 15 ) in order to light to a lens aperture system ( 16 ) which is designed to irradiate the same areas ( 24 ) on each cigarette head ( 11 ) of a cigarette formation ( 12 ) located in the area of the test device ( 10 ). 11. The device according to claim 9 or 10, characterized by a control unit for cyclical switching on or off of the individual light sources ( 14 ), in particular for alternating the switching on time of only one individual light source ( 14 ).