Nothing Special   »   [go: up one dir, main page]

WO2023042112A2 - Method and system for inspecting smoking articles - Google Patents

Method and system for inspecting smoking articles Download PDF

Info

Publication number
WO2023042112A2
WO2023042112A2 PCT/IB2022/058702 IB2022058702W WO2023042112A2 WO 2023042112 A2 WO2023042112 A2 WO 2023042112A2 IB 2022058702 W IB2022058702 W IB 2022058702W WO 2023042112 A2 WO2023042112 A2 WO 2023042112A2
Authority
WO
WIPO (PCT)
Prior art keywords
image
smoking article
machine
imm
learned model
Prior art date
Application number
PCT/IB2022/058702
Other languages
French (fr)
Other versions
WO2023042112A3 (en
Inventor
Dalia COPPI
Riccardo VASUMINI
Giuliano Gamberini
Original Assignee
G.D S.P.A.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by G.D S.P.A. filed Critical G.D S.P.A.
Priority to JP2024516745A priority Critical patent/JP2024534412A/en
Priority to KR1020247009677A priority patent/KR20240054310A/en
Priority to EP22785792.7A priority patent/EP4401582A2/en
Priority to CN202280061878.9A priority patent/CN117956910A/en
Publication of WO2023042112A2 publication Critical patent/WO2023042112A2/en
Publication of WO2023042112A3 publication Critical patent/WO2023042112A3/en

Links

Classifications

    • A—HUMAN NECESSITIES
    • A24—TOBACCO; CIGARS; CIGARETTES; SIMULATED SMOKING DEVICES; SMOKERS' REQUISITES
    • A24C—MACHINES FOR MAKING CIGARS OR CIGARETTES
    • A24C5/00—Making cigarettes; Making tipping materials for, or attaching filters or mouthpieces to, cigars or cigarettes
    • A24C5/32—Separating, ordering, counting or examining cigarettes; Regulating the feeding of tobacco according to rod or cigarette condition
    • A24C5/34—Examining cigarettes or the rod, e.g. for regulating the feeding of tobacco; Removing defective cigarettes
    • A24C5/3412—Examining cigarettes or the rod, e.g. for regulating the feeding of tobacco; Removing defective cigarettes by means of light, radiation or electrostatic fields
    • A—HUMAN NECESSITIES
    • A24—TOBACCO; CIGARS; CIGARETTES; SIMULATED SMOKING DEVICES; SMOKERS' REQUISITES
    • A24C—MACHINES FOR MAKING CIGARS OR CIGARETTES
    • A24C5/00—Making cigarettes; Making tipping materials for, or attaching filters or mouthpieces to, cigars or cigarettes
    • A24C5/32—Separating, ordering, counting or examining cigarettes; Regulating the feeding of tobacco according to rod or cigarette condition
    • A24C5/34—Examining cigarettes or the rod, e.g. for regulating the feeding of tobacco; Removing defective cigarettes
    • A24C5/345—Removing defective cigarettes
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • G01N21/952—Inspecting the exterior surface of cylindrical bodies or wires
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]

Definitions

  • This invention relates to a method and a system for inspecting smoking articles.
  • an automatic machine for making smoking article may be provided with an inspection system, as known in the prior art, for example, from patent document EP3520631 A1 , in the name of the present Applicant.
  • an inspection system as known in the prior art, for example, from patent document EP3520631 A1 , in the name of the present Applicant.
  • Examples of methods for inspecting smoking articles are described in the following prior art documents: WO2018185722A1 , CN1 11972700A, EP3476228A1 , ITB020080755A1 and EP3067823A1.
  • the prior art provides the solutions described in documents DE102014203158A1 and EP0653170A1 .
  • the aim of this invention is to provide a method and a system for inspecting smoking articles to overcome the above mentioned disadvantages of the prior art.
  • each smoking article includes an elongate tubular wrapper extending along a longitudinal axis, a filling material wrapped in the wrapper and an elongate body embedded in the filling material at a predetermined position.
  • the method comprises a step of capturing an image of the smoking article or a part thereof. This image is viewed along an optical path oriented along the longitudinal axis.
  • the method comprises a step of feeding the image to a machine-learned model.
  • the machine-learned model is trained to identify the body in the image. This step is carried out by a processor.
  • the machine-learned model is trained to generate positioning data for each image.
  • the positioning data represents a position of the body relative to the tubular wrapper.
  • the positioning data represents a planar geometrical figure delimiting the object represented in the image. That way, besides checking for the presence of the body inside the filling material, it is possible to identify its position relative to the wrapper.
  • the method comprises a step of processing the positioning data, through the processor, based on the reference data representing the predetermined position of the body; for example, processing comprises a step, performed through the processor, of comparing the positioning data with the reference data representing the predetermined position of the body.
  • the processor can check that the body is effectively positioned at an acceptable position relative to the predetermined position. For example, the processor has access to deviation values identifying maximum deviation values between the position of the body determined from the positioning data and the predetermined position.
  • the machine-learned model includes a deep neural network.
  • the neural network is a convolutional neural network.
  • the neural network might include one or more filtering stages of the max pooling type.
  • the machine-learned model is trained through a step of feeding a plurality of example images. For each image, the machine-learned model is trained through a step of feeding positioning target data representing a planar geometrical figure delimiting the object represented in the image.
  • the examples include a first plurality of images of semi-finished smoking articles with objects positioned correctly.
  • the examples include a second plurality of images of semi-finished smoking articles with objects positioned incorrectly according to a selection of predetermined positioning defects.
  • the method comprises a step, performed through a control unit, of determining a positive or negative identification result of identifying the object in the image.
  • the step of identifying is performed by the machine- learned model.
  • the method comprises a step, performed through the control unit, of generating information for rejecting the smoking article.
  • the method comprises a step of starting the step of comparing the positioning data with reference data representing the predetermined position of the body.
  • the method comprises generating information for rejecting or approving the smoking article, responsive to the comparison. This allows, firstly, rejecting the articles without the body and, secondly, where the body is present, rejecting the articles in which the body is positioned incorrectly.
  • the method comprises a step, for each image captured, of optically inspecting the wrapper.
  • the method comprises a step of generating wrapper data representing a result of inspecting the wrapper.
  • the method comprises a step of comparing the wrapper data with reference wrapper data, representing predetermined specifications for the wrapper.
  • the method comprises a step of generating (i) information for rejecting or (ii) proceeding with the step of feeding, in response to (i) a negative result or (ii) a positive result of the step of comparing, respectively.
  • the method comprises a full inspection of the smoking article, which also includes checking the wrapper.
  • the method is a method for inspecting smoking articles one by one.
  • the method comprises capturing an image of each individual smoking article.
  • the method comprises a step of positioning the individual smoking article in an inspection zone at which the inspection device is aimed in order to capture the image of it.
  • the inspection zone is illuminated with constant lighting parameters to allow the smoking articles to be illuminated uniformly.
  • the lighting parameters of the inspection zone are the same as those used for capturing the plurality of example images.
  • the smoking articles are positioned in the inspection zone with the same orientation relative to the inspection device.
  • the smoking article framed in the plurality of example images has the same orientation as the smoking articles inspected during the production process.
  • the inspection device is synchronized with a conveyor which conveys the smoking articles (or the semifinished products or the components used to make up the smoking articles) in and out of the inspection zone.
  • the smoking articles or the semifinished products or the components used to make up the smoking articles
  • the smoking articles are conveyed in an orderly manner, one by one; preferably, they are conveyed into the inspection station in such a way that their orientation is predetermined and known to the system.
  • This increases inspection precision and facilitates training of a machine-learned model for recognizing defects or for quality control.
  • the control unit knows the result of the inspection. Consequently, the control unit is able to control a rejection device to reject a single defective smoking article.
  • singulating the smoking articles also facilitates rejection operations.
  • it is not only more difficult to identify a single defective smoking article but also more complicated to reject it, because: (i) it must be singulated retrospectively or (ii) multiple smoking articles must be rejected on account of the presence of a single defective article.
  • the smoking article or a part of it is moved individually as a single item along a predetermined path within the apparatus that produces the smoking articles (which, in one example, is an assembling apparatus but which, for the purposes of this and other aspects of the disclosure regarding the inspection system and the image processing mode, could be an apparatus of a different type).
  • At least one inspection stage (for example, the first inspection stage or the first and the second inspection stage) is located along the predetermined path.
  • the inspection stage includes at least one illuminator and one camera.
  • the smoking articles, or the parts thereof are moved along the predetermined path one by one; preferably, also, the smoking articles, or the parts thereof, are moved along the predetermined path in such a way that, in the inspection stage, they have a predetermined orientation (known to the system in relation to the optical path of the camera; for example, they are oriented so they are aligned with the camera).
  • a predetermined orientation known to the system in relation to the optical path of the camera; for example, they are oriented so they are aligned with the camera.
  • each of the smoking articles, or parts thereof, moved along the predetermined path is individually viewed and photographed (that is, made the object of image capture); that way, each image pertains to a single smoking article or part thereof.
  • the machine-learned model it should be noted that it is trained preferably by feeding to it a plurality of example images, each pertaining to a single smoking article or part thereof. Further, the machine- learned model may also be trained by feeding target data to it (in this case, learning is supervised); alternatively, target data are not fed to it (in which case, learning is unsupervised).
  • the machine-readable instructions contained in the non- transitory data storage are such that, for each smoking article, or part thereof, in the flow, the processor:
  • the machine-learned model is trained to identify predetermined categories of defects for the smoking article or part thereof.
  • a category of defects pertains to the absence or position of a body located (at a predetermined position) in the smoking article.
  • the machine-learned model might, however, be trained to recognize other types of defects such as, purely by way of example, the presence of stains or impurities or a shape that is unlike a predetermined (reference) shape for the smoking article or part thereof.
  • this disclosure provides a computer program including instructions for performing the steps of the method described in this disclosure, if performed by the processor.
  • this disclosure provides an inspection system for inspecting a smoking article including an elongate tubular wrapper extending along a longitudinal axis, a filling material wrapped in the wrapper and an elongate body embedded in the filling material at a predetermined position.
  • the system comprises a processor and a non-transitory data storage including machine-readable instructions.
  • the machine-readable instructions tell the processor to capture an image of the smoking article or a part thereof, viewed along an optical path oriented along the longitudinal axis.
  • the machine-readable instructions tell the processor to feed the image to a machine-learned model that is machine-trained to identify the body in the image.
  • the non-transitory data storage includes further instructions telling the processor to process the image to generate, for each image, positioning data representing a planar geometrical figure delimiting the body represented in the image.
  • the non-transitory data storage includes further instructions telling the processor to compare the positioning data with reference data representing the predetermined position of the body.
  • the instructions tell the processor to determine a positive or negative result of identifying the object in the image, performed by the machine-learned model.
  • the instructions tell the processor, in the event of a negative identification result, to generate information for rejecting the smoking article.
  • the instructions tell the processor, in the event of a positive identification result, to compare the positioning data with the reference data representing the predetermined position of the body and to generate information for rejecting or approving the smoking article, in response to the comparison.
  • the machine-learned model includes a deep neural network.
  • the deep neural network may be a convolutional, deep neural network.
  • the deep neural network may include one or more filtering stages of the max pooling type.
  • this disclosure provides an apparatus for continuous-cycle production of smoking articles, where each smoking article includes an elongate tubular wrapper extending along a longitudinal axis, a filling material wrapped in the wrapper and an elongate body embedded in the filling material at a predetermined position.
  • the apparatus comprises an inspection device according to any of the features described in this disclosure.
  • this disclosure provides a method for inspecting smoking articles, where each smoking article includes a body placed at a predetermined position inside the smoking article.
  • the method is applicable in the context of operations performed by an apparatus which produces the smoking articles in a continuous-cycle through a succession of processing stages, from an initial processing stage to a final processing stage.
  • the method comprises a step of capturing an image of a semifinished product generated by the machine at a processing stage intermediate between the initial processing stage and the final processing stage.
  • the method comprises a step, performed through a control unit, of feeding the image to a machine-learned model trained to identify the body in the image.
  • each smoking article includes a plurality of rod sections defining respective central axes.
  • the plurality of rod sections includes a first rod section, provided with a flavouring element, and a second rod section.
  • the assembling apparatus comprises a combining unit.
  • the combining unit is configured to make groups of rod sections, each including at least the first rod section and the second rod section.
  • the first and the second rod section are axially aligned with each other.
  • the first and the second rod section are abutted end to end.
  • the groups of rod sections are advanced perpendicularly to their central axes.
  • the first rod section are obtained from rods, each having a first and a second end which are spaced along a rod axis.
  • the combining unit is configured to couple respective second rod sections to the first and the second end of the rod.
  • the combining unit is configured, in a separating stage, to separate the rod into a pair of portions by cutting it transversely to the rod axis so as to make a corresponding pair of groups of rod sections.
  • Each group of rod sections includes a respective portion of the rod and a respective second rod section.
  • the assembling apparatus comprises a wrapping unit.
  • the wrapping unit is configured to receive from the combining unit a succession of groups of rod section fed in a feed direction, to feed the groups of rod sections perpendicularly to their central axes and to wrap a sheet of wrapping material around each group of rod sections.
  • the assembling apparatus comprises an inspection system.
  • the inspection system includes a first inspection stage.
  • the first inspection stage is positioned upstream of the separating stage in the feed direction.
  • the first inspection stage is configured to see axially the first and the second end of each rod.
  • the inspection system comprises a second inspection stage.
  • the second inspection stage is positioned downstream of the separating stage in the feed direction.
  • the second inspection stage is configured to see axially a free end of a respective first rod section of each group of rod sections of the pair of groups of rod sections.
  • the assembling apparatus comprises a spacing stage, interposed between the separating stage and the second inspection stage relative to the feed direction.
  • the spacing stage is configured to axially space the groups of rod sections of the pair of groups of rod sections.
  • the assembling apparatus is configured to feed the groups of rod sections of the pair of groups of rod sections perpendicularly to the rod axis, between the separating stage and the spacing stage.
  • the second inspection stage includes a central camera, axially interposed between the groups of rod sections of the pair of groups of rod sections.
  • the second inspection stage includes a further central camera, axially interposed between the groups of rod sections of the pair of groups of rod sections.
  • the second inspection stage includes a right-hand inspection unit, including the central camera, facing the right relative to the feed direction, to see the free end of one of the groups of rod sections of the pair of groups of rod sections.
  • the second inspection stage includes a left-hand inspection unit, including the further central camera, facing the left relative to the feed direction, to see the free end of the other of the groups of rod sections of the pair of groups of rod sections.
  • the right-hand inspection unit and the left-hand inspection unit are offset along the feed direction.
  • the first inspection stage is provided with a pair of cameras.
  • the cameras of the pair of cameras have respective optical paths oriented in opposite, converging directions.
  • the second pair of cameras have respective optical paths that are aligned but oriented in opposite, converging directions.
  • the assembling apparatus comprises a control unit.
  • the control unit is connected to the first inspection stage and to the second inspection stage to drive them.
  • the control unit is provided with a processor and a non- transitory data storage that includes machine-readable instructions telling the processor to capture, through the first and the second inspection stage, a plurality of images of the smoking article or a part thereof, viewed along an optical path oriented along the longitudinal axis.
  • the smoking article includes a body placed at a predetermined position inside the smoking article.
  • the control unit is provided with a processor and a non-transitory data storage that includes machine-readable instructions telling the processor to feed the plurality of images to a machine-learned model trained to identify the body in the image.
  • this disclosure provides an apparatus for the production of smoking articles (preferably multicomponent articles), each smoking article including a component that is inspectable by an optical device in at least one inspection station of the apparatus and hidden from the optical device when the smoking article is finished.
  • this disclosure relates to those apparatuses in which the smoking articles are obtained by progressive processes that define inspectable semifinished products which are not inspectable with optical instruments when the smoking article is finished.
  • These apparatuses comprise an inspection system.
  • the inspection system is located in an inspection station of the apparatus, corresponding to an inspectable semifinished product.
  • the apparatus comprises a first processing station, configured to make the inspectable semifinished product from in-process components.
  • the apparatus comprises a second processing station, configured to make the smoking article from the inspectable semifinished product.
  • the inspection station is interposed between the first and the second processing station along a feed path of the smoking article in the apparatus.
  • the apparatus comprises a control unit.
  • the control unit is connected to the inspection system to drive it.
  • the control unit is provided with a processor and a non-transitory data storage that includes machine- readable instructions.
  • the control unit is programmed to receive a plurality of images of the inspectable semifinished product or part thereof captured by the inspection system.
  • the images are, for example, captured along an optical path oriented preferably along a longitudinal axis of the smoking article. Solely by way of example, the inspection system captures images of a body located at a predetermined position in the smoking article.
  • the inspection system might, however, be configured to identify other aspects of the smoking article otherwise not visible when it is finished, such as the presence of dirt or stains on a surface before being wrapped, or the length of a rod section before it is glued to another rod section.
  • the control unit is programmed to feed the plurality of images to a machine- learned model.
  • the machine-learned model is trained to identify the body inside the smoking article in the image. In other embodiments, for example, it might identify the presence of dirt or stains on a surface before being wrapped, or the length of a rod section before it is glued to another rod section.
  • this disclosure provides an assembling method for the production of multicomponent smoking articles, where each smoking article includes a plurality of rod sections defining respective central axes, and where the plurality of rod sections includes a first rod section, provided with a flavouring element, and a second rod section.
  • the method comprises a step, performed by a combining unit, of feeding a flow of rods, each rod having the flavouring element and being elongate along a rod axis from a first to a second end in a feed direction transverse to the rod axis.
  • the method comprises a step of coupling respective second rod sections to the first and second ends of the rod.
  • the method comprises a step, performed through a combining unit, in a separating stage, of separating the rod into a pair of portions by cutting it transversely to the rod axis so as to make a corresponding pair of groups of rod sections.
  • Each group of rod sections includes a respective rod portion, defining a respective first rod section and a respective second rod section which are axially aligned along a central axis and abutted end to end.
  • the method comprises a step of feeding the groups of rod sections perpendicularly to their central axes by means of a wrapping unit.
  • the method comprises a step of feeding the groups of rod sections perpendicularly to their central axes.
  • the method comprises a step of wrapping the groups of rod sections in a sheet of wrapping material by means of a wrapping system.
  • the method comprises a first step of seeing the first and the second end of each rod in a first inspection stage, positioned upstream of the separating stage in the feed direction.
  • the method comprises a second step of seeing a free end of a respective first rod section of each group of rod sections of the pairs of groups of rod sections in a second inspection stage, positioned downstream of the separating stage in the feed direction.
  • the method further comprises a step of spacing the groups of rod sections of the pair of groups of rod sections in an axial direction after the step of separating and before the second step of seeing.
  • the groups of rod sections of the pair of groups of rod sections are also fed transversely to their axes.
  • the step of coupling precedes the step of separating.
  • the second step of seeing includes the following substeps, in succession:
  • the smoking article includes a body placed at a predetermined position inside the smoking article.
  • the method further comprises a step of feeding images taken in the first and second inspection stages to a machine-learned model trained to identify the body in each image.
  • Figure 1 shows a schematic side view of an apparatus for continuous cycle production of smoking articles
  • Figure 2 shows a schematic top view of the apparatus of Figure 1 ;
  • Figure 3 schematically represents a method for inspecting a smoking article
  • Figure 4 schematically illustrates a transverse cross section of a smoking article including filling material and a body CP.
  • each smoking article includes a plurality of rod sections SP defining respective central axes.
  • the plurality of rod sections includes a first rod section SP1 , provided with a flavouring element, and a second rod section SP2.
  • the smoking article also comprises an outer wrapper IC.
  • the apparatus 1 comprises a combining unit 10.
  • the combining unit is configured to make groups of rod sections GS, each including at least the first rod section SP1 and the second rod section SP2, axially aligned and abutted end to end.
  • the groups of rod sections GS are advanced perpendicularly to their central axes AC.
  • the apparatus 1 comprises a portioning unit 1 1 , configured to make, from corresponding rods, a plurality of first rod sections SP1 and second rod sections SP2 that will subsequently be grouped into pairs to define respective groups of rod sections.
  • the second rod section SP2 comprises a first portion SP2' and a second portion SP2".
  • the portioning unit 1 1 comprises a first separating element 1 1 1 , configured to separate a first rod into corresponding rods B1 that define the first portion SP2' of the second segment SP2 and having a predetermined length along the respective central axis AC.
  • the portioning unit 1 1 comprises a second separating element 1 12, configured to separate a second rod into corresponding rods B2 that define the second portion SP2" of the second rod section SP2 and having a predetermined length along the respective central axis AC.
  • the portioning unit 1 1 comprises a third separating element 1 13, configured to separate a third rod into corresponding rods B3 that define the first rod section SP1 and having a predetermined length along the respective central axis AC.
  • the third rod B3 is a rod containing flavoured filling material MTA.
  • the first rod section SP1 should ideally include a body CP, embedded in the flavoured filling material.
  • the apparatus 1 comprises an inspection system 12, configured to capture inspection data to be processed to derive information regarding the smoking article.
  • the inspection system comprises a first inspection stage 121 , configured to capture inspection data relating to the rods B3 defining the first rod section SP1.
  • the rods B3 each include a first end B31 and a second end B32.
  • the first inspection stage 121 comprises a first pair of cameras 122, each configured to capture a corresponding image of the first end B31 and of the second end B32 of the rod B3.
  • the first pair of cameras 122 capture image data 122' representing an image of the first end B31 of the rod B3 and an image of the second end B32 of the rod B3.
  • the apparatus 1 comprises a control unit 2, including a processor and configured to control the apparatus 1 and to process the data received from the inspection system 12.
  • the control unit 2 is configured to recognize the presence or absence of the body CP in the rod B3. In addition, in an embodiment, the control unit 2 is also configured to check that the body CP is in the correct position relative to the filling material MTA of the rod B3.
  • the step of recognizing the body CP is carried out by feeding to the machine-learned model MA the image data captured by the first inspection stage 121 through the first pair of cameras 122.
  • the machine-learned model MA is trained to identify the body CP in the image. The step of recognizing is described in more detail below.
  • the apparatus is provided with a first plurality of rods B1 and a second plurality of rods B2, which together will define the second rod section SP2, and a third plurality of rods B3, which will define the first rod section SP1 .
  • the combining unit 10 comes into action to combine the three pluralities of rods B1 , B2, B3.
  • the combining unit is configured to feed the rods B1 , B2 and B3 along a feed direction DA, perpendicular to the central axes AC of the rods B1 , B2, B3.
  • the combining unit includes a first stage 101 .
  • the first stage 101 comprises a respective first cutting station 101 T, configured to cut the rods B1 along a direction perpendicular to the central axis AC, that is to say, along the feed direction DA, to make a first segment B1 ' and a second segment B1".
  • the first stage 101 also comprises a first separator 101 S, configured to space the first and the second segment B1 ', B1" along a spacing direction DD, perpendicular to the feed direction DA and parallel to the central axis of the rod B1 (already separated). Separating the two segments is necessary to allow inserting the rod B2 of the second plurality of rods B2 at an intermediate position along the spacing direction DD between the first segment B1 ' and the second segment B1".
  • the combining unit 10 then comprises a second stage 102, in which a second rod B2 of the second plurality of rods B2 is placed at an intermediate position along the spacing direction DD between the first segment B1 ' and the second segment B1".
  • the second stage 102 comprises a first gluing device 1021, configured to connect a first end of the rod B2 to a corresponding end of the first segment B1 ' of the rod B1 and to connect a second end of the rod B2 to a corresponding end of the second segment B1" of the rod B1 .
  • the second stage comprises a second cutting element 102T, configured to separate the second rod B2 into a corresponding first segment B2', connected to the first segment B1 ' of the first rod B1 to define a first rod sectionSPI of a smoking article, and a second segment B2", connected to the second segment B1" of the first rod B1 to define a further second rod section SP2 of a further smoking article.
  • a second cutting element 102T configured to separate the second rod B2 into a corresponding first segment B2', connected to the first segment B1 ' of the first rod B1 to define a first rod sectionSPI of a smoking article, and a second segment B2", connected to the second segment B1" of the first rod B1 to define a further second rod section SP2 of a further smoking article.
  • the second stage 102 comprises a second separator 102S, configured to separate the second rod section SP2 from the further second rod section SP2.
  • the combining unit 10 comprises a third stage 103, in which a third rod B3 of the third plurality of rods B3 is placed at an intermediate position along the spacing direction DD between the second rod section (formed of the first segment B1 ' of the first rod B1 and the first segment B2' of the second rod B") and the second further rod section (formed of the second segment B1" of the first rod B1 and the second segment B2").
  • the third stage 103 comprises a second gluing device 1031, configured to connect a first end of the rod B3 to a corresponding end of the second rod section SP2, that is to say, to an end of the first segment B2' of the second rod B2.
  • the second gluing device 1031 is configured to connect a second end of the rod B3 to a corresponding end of the further second rod section SP2, that is to say, to an end of the second segment B2" of the second rod B2.
  • the third stage 103 comprises a third cutting element 103T, configured to separate the third rod B3 into a corresponding first segment B3', connected to the first segment B2' of the second rod B2 to define a first group of rod sections GS1 (formed of the first rod section SP1 , defined by the first segment B3' of the third rod B3 and by the second rod section SP2) and a second segment B3", connected to the second segment B2" of the second rod B2 to define a second group of rod sections GS2 (formed of the further second rod section SP2 and by the first rod section SP1 , defined by the second segment B3" of the third rod B3).
  • a third cutting element 103T configured to separate the third rod B3 into a corresponding first segment B3', connected to the first segment B2' of the second rod B2 to define a first group of rod sections GS1 (formed of the first rod section SP1 , defined by the first segment B3' of the third rod B3 and by the second rod section SP2) and a second
  • the third stage 103 comprises a third separator 103S, configured to separate the first group of rod sections GS1 and the second group of rod sections GS2 along the spacing direction DD.
  • the apparatus 1 therefore contemplates further inspection to check for the presence of the body CP.
  • the apparatus 1 thus comprises a second inspection stage 13, configured to capture image data representing an image of an outer end of the first rod section SP1 of the first group of rod sections GS1 and of the first rod section SP1 of the second group of rod sections GS2.
  • the second inspection stage 13 is configured to check that the body CP extends up to the outer end of the first rod section SP1 of the first group of rod sections GS1 and up to the outer end of the first rod section SP1 of the second group of rod sections GS2.
  • the second inspection stage 13 comprises a second pair of cameras 131 , including a right-hand camera 131 A and a left-hand camera 131 B.
  • the right-hand camera 131 A is spaced from the left-hand camera 131 B along the feed direction DA.
  • the cameras 131 A, 131 B of the second pair of cameras 131 are each configured to capture image data of the outer end of the first rod section SP1 of the first group of rod sections GS1 and of the outer end of the first rod section SP1 of the second group of rod sections GS2, respectively.
  • the two cameras of the pair of cameras 131 are disposed at the same position along the feed direction DA and look in opposite directions.
  • the cameras of the first pair of cameras face each to capture images of two opposite ends of the third rod B3, while the cameras of the second pair of cameras 131 face opposite directions so as to capture images of the free ends of the first rod sections SP1 of the first and the second group of rod sections GS1 , GS2.
  • the second pair of cameras 131 is positioned above the conveyor and faces a direction oriented from the conveyor to the outside. In other words, the second pair of cameras 131 is positioned, along the spacing direction DD, between the first group of rod sections GS1 and the second group of rod sections GS2. This allows framing zones of the smoking articles that face each other and are thus not visible from one side of the conveyor of the smoking articles.
  • the apparatus 1 comprises a wrapping device 14, configured to wrap the first group of rod sections GS1 and the second group of rod sections GS2.
  • the first and the second group of rod sections GS1 , GS2 are processed by the wrapping device 14 which wraps each of them to define the respective smoking article.
  • the apparatus comprises a third inspection stage 15, configured to control wrapping of the smoking articles.
  • the third inspection stage 15 comprises a camera configured to capture images of the wrapper IC on the smoking article. These images are sent to the control unit 2, which is configured to process them and to compare them with reference images to check that the wrapper IC conforms with predetermined quality standards.
  • the apparatus 1 comprises an unwrapping device 18, configured to unwrap smoking articles or rods B3 that do not meet specified quality requirements.
  • control unit 2 is connected to the unwrapping device 18 to send drive signals representing information on whether or not to unwrap the smoking article or the rod B3.
  • the drive signals are generated by the control unit, based on the information received from the first inspection stage 121 and/or the second inspection stage 13 and/or the third inspection stage 15. Described in detail below is the inspection method implemented by the apparatus 1 and combining unit of this disclosure.
  • the method comprises a step of capturing an image IMM of the smoking article or a part thereof, that is to say, of the rod B3.
  • the image is seen along an optical path oriented along a longitudinal axis L that is parallel to and coincides with the central axis AC.
  • the method comprises a step of feeding the image IMM to a machine- learned model MA.
  • the machine-learned model MA is trained to identify the body CP in the image. This step is carried out by a processor of the control unit 2.
  • the machine-learned model MA is trained to generate, for each image, positioning data 201 .
  • the positioning data 201 represents a position of the body CP relative to the tubular wrapper.
  • the positioning data 201 represents a planar geometrical figure delimiting the object represented in the image IMM. That way, besides checking for the presence of the body CP inside the filling material MTA, it is possible to identify its position relative to the wrapper.
  • the method comprises a step FCN, performed by the processor, of comparing the positioning data 201 with reference data 20T representing the predetermined position of the body CP.
  • the machine-learned model MA includes the reference data 20T, which determines its operation in the step of identifying (that is, the machine-learned model MA defines an identification algorithm based on the reference data 20T).
  • the reference data 20T is derived automatically by the machine-learned model MA, in the step of learning, from a plurality of training examples (which include corresponding positioning data vectors associated with respective target values).
  • the training example include a plurality of example images and corresponding target values.
  • the machine-learned model MA includes a deep neural network.
  • the neural network is a convolutional neural network.
  • the neural network might include one or more filtering stages of the max pooling type.
  • the machine-learned model MA is trained through a step of feeding a plurality of example images. For each image, the machine-learned model is trained through a step of feeding positioning target data representing a planar geometrical figure delimiting the object represented in the image.
  • the examples include a first plurality of images IM1 of semi-finished smoking articles with objects positioned correctly.
  • the examples include a second plurality of images IM2 of semi-finished smoking articles with objects positioned incorrectly according to a selection of predetermined positioning defects.
  • the method comprises a step FDT, performed through the control unit 2, of determining a positive or negative identification result of identifying the object in the image. Identification is performed by the machine-learned model MA.
  • the method comprises a step FGN, performed through the control unit 2, of generating information 203 for rejecting the smoking article.
  • the method comprises a step FCN, performed through the control unit 2, of starting the step of comparing the positioning data with reference data representing the predetermined position of the body CP.
  • the method comprises generating information 203 for rejecting or approving the smoking article, responsive to the step FCN of comparing.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Toxicology (AREA)
  • Pathology (AREA)
  • Immunology (AREA)
  • Biochemistry (AREA)
  • Analytical Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Mathematical Physics (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
  • Image Analysis (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)
  • Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)

Abstract

This disclosure relates to a method for inspecting smoking articles, wherein each smoking article includes an elongate tubular wrapper extending along a longitudinal axis (AC), a filling material (MTA) wrapped in the wrapper (IC) and an elongate body (CP) embedded in the filling material (MTA) at a predetermined position. The method comprises the following steps: capturing an image (IMM) of the smoking article or a part thereof, viewed along an optical path oriented along the longitudinal axis (AC); through a processor, feeding the image (IMM) to a machine-learned model (MA), the machine-learned model (MA) being trained to identify the body (CP) in the image (IMM).

Description

DESCRIPTION
METHOD AND SYSTEM FOR INSPECTING SMOKING ARTICLES
Technical field
This invention relates to a method and a system for inspecting smoking articles.
Background art
In the sector of smoking articles, such as, for example, cigarettes and heat- not-burn (HNB) devices, there is a need to perform a quality check on the individual articles. In effect, the automatic machines that make smoking articles sometimes produce defective smoking articles. In particular, there are smoking articles that include a body placed in a wrapper, where the body must be located at a predetermined position; in this case, the defects may be the result of incorrect positioning of the object. For example, some cigarettes have filters that include an object whose cross section is in the form of a curved object, as described in patent document WO2020128827A1 in the name of the present Applicant, where the object is labelled 2 in the drawings. In another example, an HNB device includes an object having a circular cross section, as described in patent document WO201 2016795A1 , where the object is labelled 6 in the drawings.
In this context, an automatic machine for making smoking article may be provided with an inspection system, as known in the prior art, for example, from patent document EP3520631 A1 , in the name of the present Applicant. Examples of methods for inspecting smoking articles are described in the following prior art documents: WO2018185722A1 , CN1 11972700A, EP3476228A1 , ITB020080755A1 and EP3067823A1. Further, with reference to assembling apparatuses for the production of smoking articles, the prior art, for example, provides the solutions described in documents DE102014203158A1 and EP0653170A1 .
The need remains, however, for systems that are particularly efficient, reliable and fast in performing inspections on the smoking articles. In particular, it would be desirable for such systems to allow a reliable quality inspection to be carried out in process, during production.
Aim of the invention
The aim of this invention is to provide a method and a system for inspecting smoking articles to overcome the above mentioned disadvantages of the prior art.
This aim is fully achieved by the method and system according to the invention as characterized in the appended claims.
According to an aspect of it, this disclosure provides a method for inspecting smoking articles. Each smoking article includes an elongate tubular wrapper extending along a longitudinal axis, a filling material wrapped in the wrapper and an elongate body embedded in the filling material at a predetermined position.
The method comprises a step of capturing an image of the smoking article or a part thereof. This image is viewed along an optical path oriented along the longitudinal axis.
The method comprises a step of feeding the image to a machine-learned model. The machine-learned model is trained to identify the body in the image. This step is carried out by a processor.
This allows the machine to (at least) check for the presence of the body inside the filling material.
Preferably, the machine-learned model is trained to generate positioning data for each image. The positioning data represents a position of the body relative to the tubular wrapper. The positioning data represents a planar geometrical figure delimiting the object represented in the image. That way, besides checking for the presence of the body inside the filling material, it is possible to identify its position relative to the wrapper.
The method comprises a step of processing the positioning data, through the processor, based on the reference data representing the predetermined position of the body; for example, processing comprises a step, performed through the processor, of comparing the positioning data with the reference data representing the predetermined position of the body.
Thus, the processor can check that the body is effectively positioned at an acceptable position relative to the predetermined position. For example, the processor has access to deviation values identifying maximum deviation values between the position of the body determined from the positioning data and the predetermined position.
In an embodiment, the machine-learned model includes a deep neural network. In an embodiment, the neural network is a convolutional neural network.
The neural network might include one or more filtering stages of the max pooling type.
The machine-learned model is trained through a step of feeding a plurality of example images. For each image, the machine-learned model is trained through a step of feeding positioning target data representing a planar geometrical figure delimiting the object represented in the image.
The examples include a first plurality of images of semi-finished smoking articles with objects positioned correctly. In an embodiment, the examples include a second plurality of images of semi-finished smoking articles with objects positioned incorrectly according to a selection of predetermined positioning defects.
The method comprises a step, performed through a control unit, of determining a positive or negative identification result of identifying the object in the image. The step of identifying is performed by the machine- learned model. In the event of a negative identification result, the method comprises a step, performed through the control unit, of generating information for rejecting the smoking article. In the event of a positive identification result, the method comprises a step of starting the step of comparing the positioning data with reference data representing the predetermined position of the body. Next, the method comprises generating information for rejecting or approving the smoking article, responsive to the comparison. This allows, firstly, rejecting the articles without the body and, secondly, where the body is present, rejecting the articles in which the body is positioned incorrectly.
In an embodiment, the method comprises a step, for each image captured, of optically inspecting the wrapper. The method comprises a step of generating wrapper data representing a result of inspecting the wrapper.
The method comprises a step of comparing the wrapper data with reference wrapper data, representing predetermined specifications for the wrapper. The method comprises a step of generating (i) information for rejecting or (ii) proceeding with the step of feeding, in response to (i) a negative result or (ii) a positive result of the step of comparing, respectively.
That way, the method comprises a full inspection of the smoking article, which also includes checking the wrapper.
According to an aspect, the method is a method for inspecting smoking articles one by one. In other words, the method comprises capturing an image of each individual smoking article. Preferably, the method comprises a step of positioning the individual smoking article in an inspection zone at which the inspection device is aimed in order to capture the image of it. In an embodiment of the method, the inspection zone is illuminated with constant lighting parameters to allow the smoking articles to be illuminated uniformly. Also, the lighting parameters of the inspection zone are the same as those used for capturing the plurality of example images. In an embodiment of the method, the smoking articles are positioned in the inspection zone with the same orientation relative to the inspection device. Also, the smoking article framed in the plurality of example images has the same orientation as the smoking articles inspected during the production process.
The uniformity of the lighting and positioning between the images captured and the example images considerably enhances the reliability of inspection with the machine-learned model. This would not be possible if inspection were not individual but in groups, where each smoking article has a position of its own relative to the inspection device.
In an embodiment, the inspection device is synchronized with a conveyor which conveys the smoking articles (or the semifinished products or the components used to make up the smoking articles) in and out of the inspection zone. Preferably, the smoking articles (or the semifinished products or the components used to make up the smoking articles) are conveyed in an orderly manner, one by one; preferably, they are conveyed into the inspection station in such a way that their orientation is predetermined and known to the system. This increases inspection precision and facilitates training of a machine-learned model for recognizing defects or for quality control. Further more, for each smoking article inspected, the control unit knows the result of the inspection. Consequently, the control unit is able to control a rejection device to reject a single defective smoking article.
It is noted that singulating the smoking articles also facilitates rejection operations. In effect, in solutions where the smoking articles are inspected in groups, it is not only more difficult to identify a single defective smoking article but also more complicated to reject it, because: (i) it must be singulated retrospectively or (ii) multiple smoking articles must be rejected on account of the presence of a single defective article.
In at least one example embodiment, the smoking article or a part of it is moved individually as a single item along a predetermined path within the apparatus that produces the smoking articles (which, in one example, is an assembling apparatus but which, for the purposes of this and other aspects of the disclosure regarding the inspection system and the image processing mode, could be an apparatus of a different type). At least one inspection stage (for example, the first inspection stage or the first and the second inspection stage) is located along the predetermined path. The inspection stage includes at least one illuminator and one camera. Preferably, as also explained above), the smoking articles, or the parts thereof, are moved along the predetermined path one by one; preferably, also, the smoking articles, or the parts thereof, are moved along the predetermined path in such a way that, in the inspection stage, they have a predetermined orientation (known to the system in relation to the optical path of the camera; for example, they are oriented so they are aligned with the camera).
In an example, during the step of capturing an image, each of the smoking articles, or parts thereof, moved along the predetermined path is individually viewed and photographed (that is, made the object of image capture); that way, each image pertains to a single smoking article or part thereof.
With regard to the machine-learned model, it should be noted that it is trained preferably by feeding to it a plurality of example images, each pertaining to a single smoking article or part thereof. Further, the machine- learned model may also be trained by feeding target data to it (in this case, learning is supervised); alternatively, target data are not fed to it (in which case, learning is unsupervised).
In an example, the machine-readable instructions contained in the non- transitory data storage are such that, for each smoking article, or part thereof, in the flow, the processor:
- captures, through at least one inspection stage (for example, through the first and the second inspection stage), a plurality of images of the smoking article or part thereof, viewed along a path having a predetermined orientation;
- feeds the plurality of images to the machine-learned model.
The machine-learned model is trained to identify predetermined categories of defects for the smoking article or part thereof. In an example, a category of defects pertains to the absence or position of a body located (at a predetermined position) in the smoking article. The machine-learned model might, however, be trained to recognize other types of defects such as, purely by way of example, the presence of stains or impurities or a shape that is unlike a predetermined (reference) shape for the smoking article or part thereof. According to an aspect of it, this disclosure provides a computer program including instructions for performing the steps of the method described in this disclosure, if performed by the processor.
According to an aspect of it, this disclosure provides an inspection system for inspecting a smoking article including an elongate tubular wrapper extending along a longitudinal axis, a filling material wrapped in the wrapper and an elongate body embedded in the filling material at a predetermined position. The system comprises a processor and a non-transitory data storage including machine-readable instructions. The machine-readable instructions tell the processor to capture an image of the smoking article or a part thereof, viewed along an optical path oriented along the longitudinal axis. The machine-readable instructions tell the processor to feed the image to a machine-learned model that is machine-trained to identify the body in the image.
The non-transitory data storage includes further instructions telling the processor to process the image to generate, for each image, positioning data representing a planar geometrical figure delimiting the body represented in the image.
The non-transitory data storage includes further instructions telling the processor to compare the positioning data with reference data representing the predetermined position of the body.
The instructions tell the processor to determine a positive or negative result of identifying the object in the image, performed by the machine-learned model. The instructions tell the processor, in the event of a negative identification result, to generate information for rejecting the smoking article. The instructions tell the processor, in the event of a positive identification result, to compare the positioning data with the reference data representing the predetermined position of the body and to generate information for rejecting or approving the smoking article, in response to the comparison. In an embodiment, the machine-learned model includes a deep neural network. The deep neural network may be a convolutional, deep neural network. The deep neural network may include one or more filtering stages of the max pooling type.
According to an aspect of it, this disclosure provides an apparatus for continuous-cycle production of smoking articles, where each smoking article includes an elongate tubular wrapper extending along a longitudinal axis, a filling material wrapped in the wrapper and an elongate body embedded in the filling material at a predetermined position. The apparatus comprises an inspection device according to any of the features described in this disclosure.
According to an aspect of it, this disclosure provides a method for inspecting smoking articles, where each smoking article includes a body placed at a predetermined position inside the smoking article. The method is applicable in the context of operations performed by an apparatus which produces the smoking articles in a continuous-cycle through a succession of processing stages, from an initial processing stage to a final processing stage.
The method comprises a step of capturing an image of a semifinished product generated by the machine at a processing stage intermediate between the initial processing stage and the final processing stage.
The method comprises a step, performed through a control unit, of feeding the image to a machine-learned model trained to identify the body in the image.
According to an aspect of it, this disclosure provides an assembling apparatus for producing multicomponent smoking articles. Each smoking article includes a plurality of rod sections defining respective central axes. The plurality of rod sections includes a first rod section, provided with a flavouring element, and a second rod section.
The assembling apparatus comprises a combining unit. The combining unit is configured to make groups of rod sections, each including at least the first rod section and the second rod section. The first and the second rod section are axially aligned with each other. The first and the second rod section are abutted end to end. The groups of rod sections are advanced perpendicularly to their central axes. The first rod section are obtained from rods, each having a first and a second end which are spaced along a rod axis.
The combining unit is configured to couple respective second rod sections to the first and the second end of the rod. The combining unit is configured, in a separating stage, to separate the rod into a pair of portions by cutting it transversely to the rod axis so as to make a corresponding pair of groups of rod sections. Each group of rod sections includes a respective portion of the rod and a respective second rod section.
The assembling apparatus comprises a wrapping unit. The wrapping unit is configured to receive from the combining unit a succession of groups of rod section fed in a feed direction, to feed the groups of rod sections perpendicularly to their central axes and to wrap a sheet of wrapping material around each group of rod sections.
The assembling apparatus comprises an inspection system.
The inspection system includes a first inspection stage. The first inspection stage is positioned upstream of the separating stage in the feed direction. The first inspection stage is configured to see axially the first and the second end of each rod.
The inspection system comprises a second inspection stage. The second inspection stage is positioned downstream of the separating stage in the feed direction. The second inspection stage is configured to see axially a free end of a respective first rod section of each group of rod sections of the pair of groups of rod sections.
In an embodiment, the assembling apparatus comprises a spacing stage, interposed between the separating stage and the second inspection stage relative to the feed direction. The spacing stage is configured to axially space the groups of rod sections of the pair of groups of rod sections.
The assembling apparatus is configured to feed the groups of rod sections of the pair of groups of rod sections perpendicularly to the rod axis, between the separating stage and the spacing stage. The second inspection stage includes a central camera, axially interposed between the groups of rod sections of the pair of groups of rod sections. In an embodiment, the second inspection stage includes a further central camera, axially interposed between the groups of rod sections of the pair of groups of rod sections.
In an embodiment, the second inspection stage includes a right-hand inspection unit, including the central camera, facing the right relative to the feed direction, to see the free end of one of the groups of rod sections of the pair of groups of rod sections.
In an embodiment, the second inspection stage includes a left-hand inspection unit, including the further central camera, facing the left relative to the feed direction, to see the free end of the other of the groups of rod sections of the pair of groups of rod sections.
In a preferred embodiment, the right-hand inspection unit and the left-hand inspection unit are offset along the feed direction.
Preferably, the first inspection stage is provided with a pair of cameras.
In an embodiment, the cameras of the pair of cameras have respective optical paths oriented in opposite, converging directions. Further, according to an aspect of this disclosure, the second pair of cameras have respective optical paths that are aligned but oriented in opposite, converging directions. The assembling apparatus comprises a control unit. The control unit is connected to the first inspection stage and to the second inspection stage to drive them. The control unit is provided with a processor and a non- transitory data storage that includes machine-readable instructions telling the processor to capture, through the first and the second inspection stage, a plurality of images of the smoking article or a part thereof, viewed along an optical path oriented along the longitudinal axis. It should be noted that the smoking article includes a body placed at a predetermined position inside the smoking article.
The control unit is provided with a processor and a non-transitory data storage that includes machine-readable instructions telling the processor to feed the plurality of images to a machine-learned model trained to identify the body in the image.
According to an aspect, this disclosure provides an apparatus for the production of smoking articles (preferably multicomponent articles), each smoking article including a component that is inspectable by an optical device in at least one inspection station of the apparatus and hidden from the optical device when the smoking article is finished. In other words, this disclosure relates to those apparatuses in which the smoking articles are obtained by progressive processes that define inspectable semifinished products which are not inspectable with optical instruments when the smoking article is finished.
These apparatuses comprise an inspection system. The inspection system is located in an inspection station of the apparatus, corresponding to an inspectable semifinished product.
The apparatus comprises a first processing station, configured to make the inspectable semifinished product from in-process components. The apparatus comprises a second processing station, configured to make the smoking article from the inspectable semifinished product.
The inspection station is interposed between the first and the second processing station along a feed path of the smoking article in the apparatus. Advantageously, the apparatus comprises a control unit. The control unit is connected to the inspection system to drive it. The control unit is provided with a processor and a non-transitory data storage that includes machine- readable instructions.
The control unit is programmed to receive a plurality of images of the inspectable semifinished product or part thereof captured by the inspection system. The images are, for example, captured along an optical path oriented preferably along a longitudinal axis of the smoking article. Solely by way of example, the inspection system captures images of a body located at a predetermined position in the smoking article. The inspection system might, however, be configured to identify other aspects of the smoking article otherwise not visible when it is finished, such as the presence of dirt or stains on a surface before being wrapped, or the length of a rod section before it is glued to another rod section.
The control unit is programmed to feed the plurality of images to a machine- learned model.
For example, but not necessarily, the machine-learned model is trained to identify the body inside the smoking article in the image. In other embodiments, for example, it might identify the presence of dirt or stains on a surface before being wrapped, or the length of a rod section before it is glued to another rod section.
According to an aspect, this disclosure provides an assembling method for the production of multicomponent smoking articles, where each smoking article includes a plurality of rod sections defining respective central axes, and where the plurality of rod sections includes a first rod section, provided with a flavouring element, and a second rod section.
The method comprises a step, performed by a combining unit, of feeding a flow of rods, each rod having the flavouring element and being elongate along a rod axis from a first to a second end in a feed direction transverse to the rod axis.
The method comprises a step of coupling respective second rod sections to the first and second ends of the rod.
The method comprises a step, performed through a combining unit, in a separating stage, of separating the rod into a pair of portions by cutting it transversely to the rod axis so as to make a corresponding pair of groups of rod sections.
Each group of rod sections includes a respective rod portion, defining a respective first rod section and a respective second rod section which are axially aligned along a central axis and abutted end to end.
The method comprises a step of feeding the groups of rod sections perpendicularly to their central axes by means of a wrapping unit.
The method comprises a step of feeding the groups of rod sections perpendicularly to their central axes.
The method comprises a step of wrapping the groups of rod sections in a sheet of wrapping material by means of a wrapping system.
The method comprises a first step of seeing the first and the second end of each rod in a first inspection stage, positioned upstream of the separating stage in the feed direction.
The method comprises a second step of seeing a free end of a respective first rod section of each group of rod sections of the pairs of groups of rod sections in a second inspection stage, positioned downstream of the separating stage in the feed direction.
In an embodiment, the method further comprises a step of spacing the groups of rod sections of the pair of groups of rod sections in an axial direction after the step of separating and before the second step of seeing. According to an aspect of this disclosure, during the step of spacing, the groups of rod sections of the pair of groups of rod sections are also fed transversely to their axes.
In an embodiment of the method, the step of coupling precedes the step of separating.
In an embodiment, the second step of seeing includes the following substeps, in succession:
- through a central camera, seeing the free end of one of the groups of rod sections of the pair of groups of rod sections;
- feeding the groups of rod sections in the feed direction;
- through a further central camera, seeing the free end of the other of the groups of rod sections of the pair of groups of rod sections.
In an embodiment, the smoking article includes a body placed at a predetermined position inside the smoking article. The method further comprises a step of feeding images taken in the first and second inspection stages to a machine-learned model trained to identify the body in each image. Brief description of the drawings
These and other features will become more apparent from the following description of a preferred embodiment, illustrated by way of non-limiting example in the accompanying drawings, in which:
Figure 1 shows a schematic side view of an apparatus for continuous cycle production of smoking articles;
Figure 2 shows a schematic top view of the apparatus of Figure 1 ;
Figure 3 schematically represents a method for inspecting a smoking article;
Figure 4 schematically illustrates a transverse cross section of a smoking article including filling material and a body CP.
Detailed description of preferred embodiments of the invention
With reference to the accompanying drawings, the reference numeral 1 denotes an assembling apparatus for producing multicomponent smoking articles. Each smoking article includes a plurality of rod sections SP defining respective central axes. The plurality of rod sections includes a first rod section SP1 , provided with a flavouring element, and a second rod section SP2. The smoking article also comprises an outer wrapper IC.
The apparatus 1 comprises a combining unit 10. The combining unit is configured to make groups of rod sections GS, each including at least the first rod section SP1 and the second rod section SP2, axially aligned and abutted end to end. In the combining unit, the groups of rod sections GS are advanced perpendicularly to their central axes AC.
In particular, the apparatus 1 comprises a portioning unit 1 1 , configured to make, from corresponding rods, a plurality of first rod sections SP1 and second rod sections SP2 that will subsequently be grouped into pairs to define respective groups of rod sections.
In particular, in an embodiment, the second rod section SP2 comprises a first portion SP2' and a second portion SP2".
The portioning unit 1 1 comprises a first separating element 1 1 1 , configured to separate a first rod into corresponding rods B1 that define the first portion SP2' of the second segment SP2 and having a predetermined length along the respective central axis AC.
The portioning unit 1 1 comprises a second separating element 1 12, configured to separate a second rod into corresponding rods B2 that define the second portion SP2" of the second rod section SP2 and having a predetermined length along the respective central axis AC.
The portioning unit 1 1 comprises a third separating element 1 13, configured to separate a third rod into corresponding rods B3 that define the first rod section SP1 and having a predetermined length along the respective central axis AC. It should be noted that the third rod B3 is a rod containing flavoured filling material MTA. In addition, the first rod section SP1 should ideally include a body CP, embedded in the flavoured filling material.
The apparatus 1 comprises an inspection system 12, configured to capture inspection data to be processed to derive information regarding the smoking article.
The inspection system comprises a first inspection stage 121 , configured to capture inspection data relating to the rods B3 defining the first rod section SP1. In particular, the rods B3 each include a first end B31 and a second end B32. The first inspection stage 121 comprises a first pair of cameras 122, each configured to capture a corresponding image of the first end B31 and of the second end B32 of the rod B3.
In the first inspection stage, the first pair of cameras 122 capture image data 122' representing an image of the first end B31 of the rod B3 and an image of the second end B32 of the rod B3.
The apparatus 1 comprises a control unit 2, including a processor and configured to control the apparatus 1 and to process the data received from the inspection system 12.
The control unit 2 is configured to recognize the presence or absence of the body CP in the rod B3. In addition, in an embodiment, the control unit 2 is also configured to check that the body CP is in the correct position relative to the filling material MTA of the rod B3.
The step of recognizing the body CP is carried out by feeding to the machine-learned model MA the image data captured by the first inspection stage 121 through the first pair of cameras 122. The machine-learned model MA is trained to identify the body CP in the image. The step of recognizing is described in more detail below.
We then note that, in this step, the apparatus is provided with a first plurality of rods B1 and a second plurality of rods B2, which together will define the second rod section SP2, and a third plurality of rods B3, which will define the first rod section SP1 .
At this point, the combining unit 10 comes into action to combine the three pluralities of rods B1 , B2, B3. The combining unit is configured to feed the rods B1 , B2 and B3 along a feed direction DA, perpendicular to the central axes AC of the rods B1 , B2, B3.
The combining unit includes a first stage 101 . The first stage 101 comprises a respective first cutting station 101 T, configured to cut the rods B1 along a direction perpendicular to the central axis AC, that is to say, along the feed direction DA, to make a first segment B1 ' and a second segment B1". The first stage 101 also comprises a first separator 101 S, configured to space the first and the second segment B1 ', B1" along a spacing direction DD, perpendicular to the feed direction DA and parallel to the central axis of the rod B1 (already separated). Separating the two segments is necessary to allow inserting the rod B2 of the second plurality of rods B2 at an intermediate position along the spacing direction DD between the first segment B1 ' and the second segment B1".
The combining unit 10 then comprises a second stage 102, in which a second rod B2 of the second plurality of rods B2 is placed at an intermediate position along the spacing direction DD between the first segment B1 ' and the second segment B1".
The second stage 102 comprises a first gluing device 1021, configured to connect a first end of the rod B2 to a corresponding end of the first segment B1 ' of the rod B1 and to connect a second end of the rod B2 to a corresponding end of the second segment B1" of the rod B1 .
The second stage comprises a second cutting element 102T, configured to separate the second rod B2 into a corresponding first segment B2', connected to the first segment B1 ' of the first rod B1 to define a first rod sectionSPI of a smoking article, and a second segment B2", connected to the second segment B1" of the first rod B1 to define a further second rod section SP2 of a further smoking article.
The second stage 102 comprises a second separator 102S, configured to separate the second rod section SP2 from the further second rod section SP2.
The combining unit 10 comprises a third stage 103, in which a third rod B3 of the third plurality of rods B3 is placed at an intermediate position along the spacing direction DD between the second rod section (formed of the first segment B1 ' of the first rod B1 and the first segment B2' of the second rod B") and the second further rod section (formed of the second segment B1" of the first rod B1 and the second segment B2").
The third stage 103 comprises a second gluing device 1031, configured to connect a first end of the rod B3 to a corresponding end of the second rod section SP2, that is to say, to an end of the first segment B2' of the second rod B2. The second gluing device 1031 is configured to connect a second end of the rod B3 to a corresponding end of the further second rod section SP2, that is to say, to an end of the second segment B2" of the second rod B2.
The third stage 103 comprises a third cutting element 103T, configured to separate the third rod B3 into a corresponding first segment B3', connected to the first segment B2' of the second rod B2 to define a first group of rod sections GS1 (formed of the first rod section SP1 , defined by the first segment B3' of the third rod B3 and by the second rod section SP2) and a second segment B3", connected to the second segment B2" of the second rod B2 to define a second group of rod sections GS2 (formed of the further second rod section SP2 and by the first rod section SP1 , defined by the second segment B3" of the third rod B3).
The third stage 103 comprises a third separator 103S, configured to separate the first group of rod sections GS1 and the second group of rod sections GS2 along the spacing direction DD.
Thus, after the third stage, in the combining unit 10, there are two groups of rod sections which are spaced from each other along the spacing direction DD and which will then be wrapped to make two smoking articles. At this stage, the two ends of the first group of rod sections GS1 and of the second group of rod sections GS2 have not been checked for the presence of the body CP in the filling material MTA.
The apparatus 1 therefore contemplates further inspection to check for the presence of the body CP.
The apparatus 1 thus comprises a second inspection stage 13, configured to capture image data representing an image of an outer end of the first rod section SP1 of the first group of rod sections GS1 and of the first rod section SP1 of the second group of rod sections GS2.
In particular, the second inspection stage 13 is configured to check that the body CP extends up to the outer end of the first rod section SP1 of the first group of rod sections GS1 and up to the outer end of the first rod section SP1 of the second group of rod sections GS2.
In an embodiment, the second inspection stage 13 comprises a second pair of cameras 131 , including a right-hand camera 131 A and a left-hand camera 131 B. In an embodiment, the right-hand camera 131 A is spaced from the left-hand camera 131 B along the feed direction DA. Furthermore, the cameras 131 A, 131 B of the second pair of cameras 131 are each configured to capture image data of the outer end of the first rod section SP1 of the first group of rod sections GS1 and of the outer end of the first rod section SP1 of the second group of rod sections GS2, respectively.
In other embodiments, the two cameras of the pair of cameras 131 are disposed at the same position along the feed direction DA and look in opposite directions.
Thus, it is noted that in a preferred embodiment, the cameras of the first pair of cameras face each to capture images of two opposite ends of the third rod B3, while the cameras of the second pair of cameras 131 face opposite directions so as to capture images of the free ends of the first rod sections SP1 of the first and the second group of rod sections GS1 , GS2.
In an embodiment, the second pair of cameras 131 is positioned above the conveyor and faces a direction oriented from the conveyor to the outside. In other words, the second pair of cameras 131 is positioned, along the spacing direction DD, between the first group of rod sections GS1 and the second group of rod sections GS2. This allows framing zones of the smoking articles that face each other and are thus not visible from one side of the conveyor of the smoking articles.
The apparatus 1 comprises a wrapping device 14, configured to wrap the first group of rod sections GS1 and the second group of rod sections GS2.
At the end of the second inspection stage 13, the first and the second group of rod sections GS1 , GS2 are processed by the wrapping device 14 which wraps each of them to define the respective smoking article.
In an embodiment, the apparatus comprises a third inspection stage 15, configured to control wrapping of the smoking articles. In particular, the third inspection stage 15 comprises a camera configured to capture images of the wrapper IC on the smoking article. These images are sent to the control unit 2, which is configured to process them and to compare them with reference images to check that the wrapper IC conforms with predetermined quality standards.
In an embodiment, the apparatus 1 comprises an unwrapping device 18, configured to unwrap smoking articles or rods B3 that do not meet specified quality requirements.
In particular, the control unit 2 is connected to the unwrapping device 18 to send drive signals representing information on whether or not to unwrap the smoking article or the rod B3. The drive signals are generated by the control unit, based on the information received from the first inspection stage 121 and/or the second inspection stage 13 and/or the third inspection stage 15. Described in detail below is the inspection method implemented by the apparatus 1 and combining unit of this disclosure.
The method comprises a step of capturing an image IMM of the smoking article or a part thereof, that is to say, of the rod B3. The image is seen along an optical path oriented along a longitudinal axis L that is parallel to and coincides with the central axis AC.
The method comprises a step of feeding the image IMM to a machine- learned model MA. The machine-learned model MA is trained to identify the body CP in the image. This step is carried out by a processor of the control unit 2.
The machine-learned model MA is trained to generate, for each image, positioning data 201 . The positioning data 201 represents a position of the body CP relative to the tubular wrapper. The positioning data 201 represents a planar geometrical figure delimiting the object represented in the image IMM. That way, besides checking for the presence of the body CP inside the filling material MTA, it is possible to identify its position relative to the wrapper.
The method comprises a step FCN, performed by the processor, of comparing the positioning data 201 with reference data 20T representing the predetermined position of the body CP. The machine-learned model MA includes the reference data 20T, which determines its operation in the step of identifying (that is, the machine-learned model MA defines an identification algorithm based on the reference data 20T). The reference data 20T is derived automatically by the machine-learned model MA, in the step of learning, from a plurality of training examples (which include corresponding positioning data vectors associated with respective target values). Preferably, the training example include a plurality of example images and corresponding target values.
In an embodiment, the machine-learned model MA includes a deep neural network. In an embodiment, the neural network is a convolutional neural network.
The neural network might include one or more filtering stages of the max pooling type.
The machine-learned model MA is trained through a step of feeding a plurality of example images. For each image, the machine-learned model is trained through a step of feeding positioning target data representing a planar geometrical figure delimiting the object represented in the image.
The examples include a first plurality of images IM1 of semi-finished smoking articles with objects positioned correctly. In an embodiment, the examples include a second plurality of images IM2 of semi-finished smoking articles with objects positioned incorrectly according to a selection of predetermined positioning defects.
The method comprises a step FDT, performed through the control unit 2, of determining a positive or negative identification result of identifying the object in the image. Identification is performed by the machine-learned model MA. In the event of a negative identification result 202', the method comprises a step FGN, performed through the control unit 2, of generating information 203 for rejecting the smoking article. In the event of a positive identification result 202", the method comprises a step FCN, performed through the control unit 2, of starting the step of comparing the positioning data with reference data representing the predetermined position of the body CP. Next, the method comprises generating information 203 for rejecting or approving the smoking article, responsive to the step FCN of comparing.

Claims

22
1 ) A method for inspecting smoking articles, wherein each smoking article includes an elongate tubular wrapper (IC) extending along a longitudinal axis (AC), a filling material (MTA) wrapped in the wrapper (IC) and an elongate body (CP) embedded in the filling material (MTA) at a predetermined position, the method comprising the following steps: capturing an image (IMM) of the smoking article or a part thereof, viewed along an optical path oriented along the longitudinal axis (AC); through a processor, feeding the image (IMM) to a machine-learned model (MA), the machine-learned model (MA) being trained to identify the body (CP) in the image (IMM).
2) The method according to claim 1 , wherein the machine-learned model (MA) is trained to generate, for each image (IMM), positioning data (201 ) representing a planar geometrical figure delimiting the object represented in the image (IMM).
3) The method according to claim 2, comprising a step of processing (FCN) the positioning data (201 ), through the processor, based on reference data (201 ’) representing the predetermined position of the body (CP).
4) The method according to any one of the preceding claims, wherein the machine-learned model (MA) includes a deep neural network.
5) The method according to claim 4, wherein the neural network is a convolutional neural network.
6) The method according to claim 4 or 5, wherein the neural network includes one or more filtering stages of the max pooling type.
7) The method according to any one of the preceding claims, wherein the machine-learned model (MA) is trained through the following steps: feeding a plurality of sample images (IM1 , IM2); for each image, feeding positioning target data representing a planar geometrical figure delimiting the object represented in the image.
8) The method according to claim 7, wherein the examples include a first plurality of images (IM1 ) of semi-finished smoking articles with objects positioned correctly and a second plurality of images (IM2) of semi-finished smoking articles with objects positioned incorrectly according to a selection of predetermined positioning defects.
9) The method according to any one of the preceding claims, comprising the following steps: through a control unit (2), determining (FDT) a positive identification result (202”) or a negative identification result (202’) of identifying the body (CP) in the image (IMM), the step of identifying being performed by the machine-learned model (MA); through the control unit (2), in the event of a negative identification result (202’), generating information for rejecting (203) the smoking article; through the control unit (2), in the event of a positive identification result (202”), comparing (FCN) the positioning data (201 ) with the reference data (201 ’) representing the predetermined position of the body and generating (FGN) information for rejecting (203) or information for approving the smoking article, in response to the step of comparing (FCN).
10) The method according to claim 9, further comprising the following steps, for each image captured: optically inspecting the wrapper (IC) and generating wrapper data representing a result of inspecting the wrapper; comparing the wrapper data with reference wrapper data, representing predetermined specifications for the wrapper; in response to (i) a negative result or (ii) a positive result of the step of comparing, generating (i) information for rejecting (203) or (ii) proceeding with the step of feeding the image (IMM) to a machine-learned model (MA), respectively.
11 ) The method according to any one of the preceding claims, wherein, during the step of capturing, the smoking article, or part thereof is moved individually along a predetermined path so that each of the smoking articles or parts thereof moved along the predetermined path is individually viewed and made the object of image capture.
12) The method according to any one of the preceding claims, wherein the machine-learned model (MA) is trained through the following steps: feeding a plurality of sample images (IM1 , IM2); for each image, feeding target data, wherein each example image pertains to a single smoking article or part thereof.
13) A computer program, including instructions for performing the step of the method according to any one of the preceding claims, if performed by the processor.
14) An inspection system (12) for inspecting a smoking article including an elongate tubular wrapper (IC) extending along a longitudinal axis (AC), a filling material (MTA) wrapped in the wrapper (IC) and an elongate body (CP) embedded in the filling material (MTA) at a predetermined position, the system comprising a processor and a non-transitory data storage that includes machine-readable instructions telling the processor: to capture an image (IMM) of the smoking article or a part thereof, viewed along an optical path oriented along the longitudinal axis (AC); to feed the image (IMM) to a machine-learned model (MA), the 25 machine-learned model (MA) being trained to identify the body (CP) in the image (IMM).
15) The inspection system according to claim 14, wherein the non-transitory data storage includes further instructions telling the processor to process the image to generate, for each image, positioning data (201 ) representing a planar geometrical figure delimiting the body (CP) represented in the image (IMM).
16) The inspection system according to claim 15, wherein the non-transitory data storage includes further instructions telling the processor to compare the positioning data (201 ) with reference data (201 ’) representing the predetermined position of the body (CP).
17) The inspection system according to any one of claims 14 to 16, wherein the non-transitory data storage includes further instructions configured to tell the processor: to determine a positive or negative identification result (202”, 202’) of identifying the object in the image (IMM), the step of identifying being performed by the machine-learned model (MA); in the event of a negative identification result (202’), to generate information for rejecting (203) the smoking article; in the event of a positive identification result (202”), to compare the positioning data (201 ) with the reference data (201 ’) representing the predetermined position of the body (CP) and to generate information for rejecting (203) or information for approving the smoking article, in response to the step of comparing (FCN).
18) The inspection system according to any one of claims 14 to 17, wherein the machine-learned model (MA) includes a deep neural network, the deep neural network being a convolutional network and including one or more 26 filtering stages of the max pooling type.
19) The inspection system according to any one of claims 14 to 18, wherein the smoking article or the part thereof is moved individually along a predetermined path by a conveyor which generates a flow of smoking articles or parts thereof moving along the predetermined path, and wherein the processor is programmed to capture the image (IMM) of the single smoking article or part thereof, viewed individually, for each smoking article or part thereof in the flow.
20) An apparatus for continuous-cycle production of smoking articles, wherein each smoking article includes an elongate tubular wrapper (IC) extending along a longitudinal axis (AC), a filling material (MTA) wrapped in the wrapper and an elongate body (CP) embedded in the filling material (MTA) at a predetermined position, comprising an inspection system (12) according to any one of claims 14 to 19.
21 ) A method for inspecting smoking articles, wherein each smoking article includes a body (CP) located at a predetermined position inside the smoking article, the method comprising the following steps: in an apparatus for continuous-cycle production of the smoking articles through a succession of processing stages from an initial processing stage to a final processing stage, capturing an image (IMM) of a semifinished product generated by the machine at a processing stage intermediate between the initial processing stage and the final processing stage; through a control unit (2), feeding the image (IMM) to a machine- learned model (MA), the machine-learned model (MA) being trained to identify the body (CP) in the image (IMM).
22) An assembling apparatus (1 ) for the production of multicomponent smoking articles, each smoking article including a plurality of rod sections 27 defining respective central axes (AC), the plurality of rod sections including a first rod section (SP1 ), provided with a flavouring element, and a second rod section (SP2), the assembling apparatus (1 ) comprising: a combining unit (10), configured to make groups of segments (GS1 , GS2), each of which comprises at least the first segment (SP1 ) and the second segment (SP2) axially aligned and abutted end to end, wherein the groups of segments (GS1 , GS2) are fed perpendicularly to their central axes (AC) and wherein the first segments (SP1 ) are obtained from bars (B3), each having a first and a second end (B31 , B32) which are spaced along a bar axis, the combining unit (10) being configured to couple the first and second ends of the bar (B31 , B32) to respective second segments (SP2) and, during a separating stage (103T), to separate the bar (B3) into a pair of portions (B3’, B3”) by cutting it transversely to the bar axis (AC) so as to make a corresponding pair of groups of segments (GS1 , GS2), each group of segments including a respective bar portion (B3’, B3”) and a respective second segment (SP2); a wrapping unit (14), configured to receive from the combining unit (10) a succession of groups of segments fed in a feed direction, to feed the groups of segments perpendicularly to their central axes and to wrap a sheet of wrapping material around each group of segments; an inspection system (12) according to one or more of claims 14 to 19, wherein the inspection system (12) includes:
-- a first inspection stage (121 ), positioned upstream of the separating stage (103T) in the feed direction and provided with a first pair of cameras (122), configured to see axially the first and the second end (B31 , B32) of each bar (B3);
-- a second inspection stage (13), positioned downstream of the separating stage (103T) in the feed direction and provided with a second pair of cameras (131 ) configured to see axially a free end of a respective first segment (SP1 ) of each group of segments of the pair of groups of segments (GS1 , GS2). 28
23) The assembling apparatus (1 ) according to claim 22, comprising a spacing stage (103S), interposed between the separating stage (103T) and the second inspection stage (13) relative to the feed direction and configured to axially space the groups of rod sections of the pair of groups of rod sections (GS1 , GS2).
24) The assembling apparatus (1 ) according to claim 23, configured to feed the groups of segments of the pair of groups of segments (GS1 , GS2) perpendicularly to the bar axis, between the separating stage (103T) and the spacing stage (103S).
25) The assembling apparatus (1 ) according to any one of claims 22 to 24, wherein the cameras of the first pair of cameras (122) have respective optical paths that are oriented in opposite, converging directions, and wherein the second pair of cameras (131 ) have respective optical paths that are oriented in opposite, diverging directions.
PCT/IB2022/058702 2021-09-17 2022-09-15 Method and system for inspecting smoking articles WO2023042112A2 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
JP2024516745A JP2024534412A (en) 2021-09-17 2022-09-15 Method and system for inspecting smoking articles - Patents.com
KR1020247009677A KR20240054310A (en) 2021-09-17 2022-09-15 Methods and systems for inspecting smoking articles
EP22785792.7A EP4401582A2 (en) 2021-09-17 2022-09-15 Method and system for inspecting smoking articles
CN202280061878.9A CN117956910A (en) 2021-09-17 2022-09-15 Method and system for inspecting smoking articles

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
IT102021000023942 2021-09-17
IT102021000023942A IT202100023942A1 (en) 2021-09-17 2021-09-17 Method and system for inspecting smoking articles

Publications (2)

Publication Number Publication Date
WO2023042112A2 true WO2023042112A2 (en) 2023-03-23
WO2023042112A3 WO2023042112A3 (en) 2023-04-27

Family

ID=78771038

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2022/058702 WO2023042112A2 (en) 2021-09-17 2022-09-15 Method and system for inspecting smoking articles

Country Status (6)

Country Link
EP (1) EP4401582A2 (en)
JP (1) JP2024534412A (en)
KR (1) KR20240054310A (en)
CN (1) CN117956910A (en)
IT (1) IT202100023942A1 (en)
WO (1) WO2023042112A2 (en)

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
IT1393013B1 (en) * 2008-12-18 2012-04-11 Gd Spa METHOD FOR SETTING UP AND MANAGEMENT OF THE CONTROL DEVICE IN A MACHINE FOR THE PRODUCTION OF SMOKE ITEMS.
GB201013072D0 (en) 2010-08-04 2010-09-15 British American Tobacco Co Smoking article
CN104636707B (en) * 2013-11-07 2018-03-23 同方威视技术股份有限公司 The method of automatic detection cigarette
DE102014203158B4 (en) * 2014-02-21 2016-01-14 Hauni Maschinenbau Ag Optical inspection of rod-shaped articles of the tobacco processing industry
WO2018185722A1 (en) * 2017-04-07 2018-10-11 International Tobacco Machinery Poland Sp. Z O.O. Apparatus and method for rejection of defective rod-like articles from a mass flow of rod-like articles of the tobacco industry
PL3476228T3 (en) * 2017-10-25 2020-11-16 International Tobacco Machinery Poland Sp. Z O.O. Method and apparatus for filling transport containers with rod-like articles of tabacco industry
IT201800002250A1 (en) 2018-01-31 2019-07-31 Gd Spa Method and apparatus for the inspection of the ends of rod-shaped pieces of the tobacco industry
IT201800020083A1 (en) 2018-12-18 2020-06-18 Gd Spa MACHINE FOR THE MAKING OF TUBULAR SLICES FOR THE TOBACCO INDUSTRY
CN111972700B (en) * 2019-05-22 2023-04-25 上海烟草集团有限责任公司 Cigarette appearance detection method and device, equipment, system and medium thereof
EP4156983A1 (en) * 2020-05-28 2023-04-05 JT International S.A. Method and system for identifying smoking articles

Also Published As

Publication number Publication date
KR20240054310A (en) 2024-04-25
WO2023042112A3 (en) 2023-04-27
IT202100023942A1 (en) 2023-03-17
CN117956910A (en) 2024-04-30
EP4401582A2 (en) 2024-07-24
JP2024534412A (en) 2024-09-20

Similar Documents

Publication Publication Date Title
RU2407408C2 (en) System and device for cigarette filter identification
EP2768734B1 (en) Cigarette package coding system and associated method
CN106880078B (en) For manufacturing the method and system of bar-type article
EP2295157B1 (en) Method and device for controlling container seals
EP2199883B1 (en) A method of setting up and managing the inspection device in a machine for manufacturing tobacco products
WO2012117499A1 (en) Inspection system of filter rod
KR20170080910A (en) 3d shape inspection system
EP3967414A1 (en) Method for recovering a posteriori information about the operation of a plant for automatic classification and sorting of fruit
WO2023042112A2 (en) Method and system for inspecting smoking articles
WO2023042113A2 (en) Assembling apparatus and assembling method for producing multicomponent smoking articles
JP7248317B2 (en) Tofu production system
Chen et al. Intelligent inspection system based on infrared vision for automated fiber placement
JP2006262900A (en) Method and apparatus for checking product quality of tobacco product in production line
US20060130858A1 (en) Module for a machine and method for perforating rod-shaped articles of the tobacco industry
CN112845117A (en) Automatic detection equipment
JP2014113061A (en) Inspection system and inspection method for cigarette filter
US20240251842A1 (en) Method for determining the conformity of a type of filter being processed in a machine of the tobacco industry
CN211721868U (en) Cigarette rolling and connecting equipment
RU2808967C2 (en) Feeder device for feeding tobacco product segment
CN112754068A (en) Feeding device for feeding tobacco industry segments
CN118980694A (en) Turbine volute processing defect online detection method
KR20240056913A (en) A method for detecting defects in the antistatic needle process
NZ779697A (en) Method for recovering a posteriori information about the operation of a plant for automatic classification and sorting of fruit
CN118613714A (en) Visual inspection auxiliary device
CN116161374A (en) Method for synchronizing indexes during production of gloves or condoms

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22785792

Country of ref document: EP

Kind code of ref document: A2

WWE Wipo information: entry into national phase

Ref document number: 202280061878.9

Country of ref document: CN

ENP Entry into the national phase

Ref document number: 2024516745

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20247009677

Country of ref document: KR

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2022785792

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2022785792

Country of ref document: EP

Effective date: 20240417