METHOD FOR DETECTING THE CONTENTS OF A THREAD SPOOL

Description

Technical field

[0001] The invention relates to a method for detecting the contents of a spool of thread that is unwound from the inside out. State of the art

[0002] There are essentially two basic types of thread spools: those where the thread is wound onto an axle from the inside out, and those where the thread is unwound, usually without an axle, from the inside out. With thread spools wound onto an axle, it's easy to recognize when the end is approaching because the spool diameter decreases continuously, and eventually only a small amount of thread remains on the axle. This is different with a thread spool where the thread is unwound from the inside out, as the end of the thread is generally only noticeable when the outermost layer is reached. By then, however, it may be too late for an orderly thread or spool change, or too much thread may remain in the residual thread spool, which is usually enclosed in a cover.

[0003] This case focuses primarily on the detection of the unwinding of thread spools used in the meat processing industry, particularly in sausage-tying machines. The thread's function is to seal the ends of sausages, preventing any sausage meat from escaping. A more recent development even uses the thread to squeeze out any remaining sausage meat from the tying area between two binding points, thus meeting increased hygiene requirements.

[0004] Currently known binding machines lack a method for detecting the end of the thread on a spool. Only portion counters are known, indicating how many portions, for example of sausages, have been bound. Depending on the thread length used, the machine is stopped after a certain number of portions, based on experience, and the remaining thread is visually checked in the spool container. Depending on the result, the machine is either restarted or the spool is changed. However, a significant amount of thread usually remains unused and must be disposed of.

[0005] Only EP 3 169 160 B1 deals, in the broadest sense, with the detection of the thread spool's thickness. The thread spool, or rather its thickness, is scanned by a light beam; that is, the scanning takes place at the bottom of the spool. A thread end cannot be determined this way, at least not with sufficient accuracy, because the threads are relatively thin and are also applied in different layer structures. This means that when the thread runs out, two to three or more layers of thread still remain on the outer diameter of the spool, which in turn equates to a residual thread length of between 15 m and 40 m.

[0006] It therefore remains up to the operator to estimate how many more sausage portions can be produced. However, if they overestimate the number of portions and the sausage meat leaks out uncontrollably, this will cause significant machine problems, as the sausage meat will not be tied off and will leak into the machine, leading to considerable contamination and cleaning effort. Especially with very high-quality sausage products, such as premium salami, this leaked product must be disposed of, as refilling a high-quality salami product is not permitted. Object of the invention

[0007] The object of the present invention is to eliminate the disadvantages described above and to develop a method or device in which the end of a thread on a thread spool is detected before it runs out, but also before too much thread remains unused. Solution to the task

[0008] The solution to the task involves creating an image or a section of an image of the thread spool from the outside, possibly comparing it with a stored image, detecting changes, and drawing conclusions about the contents of the thread spool from the changes.

[0009] In contrast to the prior art, where the bottom of the thread spool is observed, the thread spool is observed from the side, i.e., radially from the outside, looking at the outer surface of the spool. As soon as an artificial intelligence detects a change in the image, it infers the contents of the thread spool. Ideally, this change should be compared with stored images so that a machine, for example, does not react to every slight movement of the thread on the outer surface. This should be understood as an improved embodiment of the fundamental concept of the invention, namely that every change in the thread spool's outer surface is detected and the contents of the spool are inferred.

[0010] In this context, artificial intelligence (AI) refers to the processing of images of the spool's surface by a suitable processor. During the application, the AI ​​can learn that not every slippage of the thread necessarily indicates the approach of the thread end. It is preferred to further support the AI ​​with a large number of stored images of thread in various structured and unstructured arrangements.

[0011] A second possibility of the inventive method involves determining the approaching thread end almost mathematically. For this purpose, the open area of ​​the observed image section is determined and calculated. As soon as the value of the open area changes, particularly due to a decrease in the amount of thread visible through the image section, an approaching thread end is inferred, and appropriate actions are initiated.

[0012] The invention also includes a corresponding device for carrying out the method described above. For the sake of simplicity, this device consists of a means for detecting the surface of the outer sheath of a thread spool or a section thereof. This means will generally be a camera for the sake of simplicity. However, should other means appear more suitable, they may also be chosen.

[0013] This camera is then connected to a corresponding artificial intelligence to detect changes in the images that the camera delivers.

[0014] As mentioned above, the inventive method or device is intended primarily for use in machinery in the meat processing industry, particularly in binding machines. Only two states occur during binding, namely... The rotor with the thread is turning, i.e., a binding takes place and the filling machine does not portion; the rotor with the binding thread is stationary, i.e., no binding takes place and the filling machine portions.

[0015] When the rotor is stationary, it always stops in the same position in which the aforementioned camera can take the picture, with the thread container also always being in the same position. Character description

[0016] Further advantages, features and details of the invention will become apparent from the following description of preferred embodiments and from the drawings; these show in: Figure 1 A perspective view of a section of a device according to the invention for tying sausages on a sausage string; according to the prior art according to EP 3 169 160 B1 Figure 2 a schematic and block diagram representation of a device according to the invention for carrying out the method according to the invention for recognizing the contents of a thread spool; Figure 3 a top view of a full and an empty spool of thread; Figures 4-9 Images of windows of a cage according to the invention with cutouts of an outer shell of the thread spool in various unwinding positions.

[0017] In accordance with the state of the art, sausages not shown here are arranged on a sausage string according to Figure 1The sausage casing is guided through an opening 1 of a dynamic part 2 of a displacement shear and further through a tube-like part 3. The dynamic part 2, in conjunction with the static part 3, clamps and ties off a sausage casing at specific points. A support ring 4 rotates around the sausage casing, supported by a bearing (not shown) in a machine-mounted bearing housing 5 of the device according to the invention, and rotates in this bearing in the direction of rotation 6.

[0018] A counterweight 7 is mounted on this support ring 4, which rotates together with the support ring 4 in the direction of rotation 6. Furthermore, a cable is located on the support ring 4 opposite the counterweight 7. Fig. 8 for holding a spool of thread. An outer layer of the spool is held by a Figure 1 light beam 11, indicated by a dashed line, passes through an opening 12 in a lid 13 of the cheese. figs 8observed through. This light beam 11 can either follow a reduction in the thickness of the thread spool 9, or several light beams 11 are provided as a light curtain, via which a decrease in the thickness of the thread spool can be determined. A light transmitter 14 is provided for emitting the light beam 11, and a light receiver 15 for receiving the light beams.

[0019] According to the present invention Figure 2 In contrast, a window 16 is formed in cage 8.1, through which part of an outer shell 17 of in Figure 3 This is recognizable by the schematically represented thread spools 18. This window 16 is viewed by a camera 19 which is connected to a processor 20 which includes artificial intelligence.

[0020] According to Figures 3a and 3bThe thread spools 18 are those that are unwound from the inside out. They are usually located in a possibly transparent cover 21. The thread winding, in particular of the outer casing 17 of the thread spool 18, is visible through this cover 21. Above all, it is also according to Figure 3b It can be recognized when thread 22 comes to an end, as the outer sheath then assumes a disordered or altered structure.

[0021] The aim is to detect when the thread 22 on the thread spool 18 runs out, so that, for example, this thread spool can be replaced with a new, full thread spool without leaving large remnants of thread. According to the present invention, there are preferably two approaches for this: According to the Figures 4 to 7 The camera observes a changing structure of the outer sheath of the thread spool through window 16. The internal structure remains the same almost to the end of the thread spool, as shown in Figure 4is indicated. In Figure 4 It is evident that both the outermost layer 23 and the subsequent inner layer 24 are intact, so that the artificial intelligence in the processor 20 can conclude that the thread spool is not yet worn out and needs to be changed.

[0022] In Figure 5 In contrast, the artificial intelligence in processor 20 recognizes via camera 19 that while the outermost layer 23 is still intact, only a small amount of thread 22 is visible from the next inner layer 24. The artificial intelligence thus compares the image of Figure 5 with that of Figure 4 and notes that only the outermost layer 23 of the outer casing of the thread spool remains, which means that only a few meters of thread are available and therefore the thread spool should be changed.

[0023] According to the exemplary embodiment Figure 6In window 16, a closed thread structure of the outer layer of the spool's outer sheath appears, as is the case with any other commercially available spool of thread. Here, the artificial intelligence in the processor 20 recognizes this via the camera 19 according to... Figure 7 The number of threads in the outermost layer decreases, and, for example, only one thread from the next innermost layer is visible. From this, the artificial intelligence can infer the length of the remaining thread on the spool. A second method for determining the thread on the outer surface of the thread spool is described in the Figures 8 and 9As shown, the main focus in window 16 is on how much of window 16's area F is covered by thread 22. If the entire area F is covered, there is still sufficient thread on the spool. If the covered area F falls below a predefined threshold stored in the artificial intelligence of processor 20, this artificial intelligence concludes that only a short length of thread remains, and the spool should be changed. Reference symbol list

[0024] 1 opening 2 dynamic part 3 static part 4 Support ring 5 Bearing housing 6 Direction of rotation 7 counterweight 8 cage 9 Thread spool 10 External location 11 beam of light 12 opening 13 Lid 14 Light transmitter 15 Light receiver 16 Window 17 outer shell 18 Thread spool 19 camera 20 processor 21 Covering 22 thread 23 lettering

Claims

1. Method for determining the content of a thread spool (18) which is unwound from the inside to the outside characterized in in that an image or an image section of the thread spool (18) is produced from the outside, if necessary, compared with a stored image, changes are recognized and the content of the thread spool (18) is deduced from the changes.

2. Method for determining the content of a thread bobbin (18) which is unwound from the inside to the outside, characterized in that an image or an image section of the thread spool (18) is produced from the outside, a portion of the image or image section covered by the thread (22) is determined and the content of the thread spool (18) is deduced from changes.

3. Method according to claim 1 or 2, characterized in that the image or the image section is generated radially to an imaginary spool axis from the outside.

4. Method according to at least one of the preceding claims, characterized in that when it is determined that the content of the thread spool (18) has fallen below a predetermined threshold value, it is replaced by a new one.

5. Device for determining the content of a thread spool (18) which can be unwound from the inside to the outside, a thread (22) forming an outer sheath for the thread spool, characterized in that a means for determining a change in the outer sheath (17) is associated with the outer sheath (17).

6. Device according to claim 5, characterized in that the means for determining a change in the outer sheath (17) of the thread spool (18) is a camera (19).

7. Device according to claim 5 or 6, characterized in that the means for determining a change in the outer sheath (17) of the thread spool (18) is assigned an artificial intelligence for determining this change.

8. Device according to at least one of claims 5 to 7, characterized in that the thread spool (18) is located in a cage (8.1) which has a window (16) for viewing the outer sheath (17) of the thread spool (18) by the means for determining a change in the outer sheath (17).

9. Use of a method according to at least one of claims 1 to 4 or of a device according to at least one of claims 5 to 8 in the manufacture of a product in the meat-processing industry, in particular in the manufacture of sausages.

10. Use of a method according to at least one of claims 1 to 4 or a device according to at least one of claims 5 to 8 in the production of sausages in tying machines.