An optical flow detection device for cigarette packaging glue marks
By using an optical process-oriented detection device, combined with an image acquisition module and machine learning, the problem of identifying adhesive residue on cigarette packaging has been solved, enabling efficient and accurate identification of genuine and counterfeit cigarettes, reducing the difficulty of counterfeiting, and improving detection efficiency.
Patent Information
- Application Number
- CN202310353047.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-04-04
- Publication Date
- 2025-12-05
- Estimated Expiration
- 2043-04-04
AI Technical Summary
Existing technologies make it difficult to quickly and effectively identify adhesive residue on cigarette packaging, leading to difficulties in distinguishing genuine from counterfeit cigarettes, especially in the face of improved counterfeiting techniques.
Design an optical process-oriented inspection device that employs an image acquisition module, a lighting module, and an image acquisition device. Combined with machine learning, it acquires adhesive trace images through combinations of different angles and light sources. The rotating smoke shell fixing module and sensors inside the black box of the image acquisition module enable process-oriented operation, thereby enhancing the accuracy and efficiency of adhesive trace image acquisition.
It enables efficient and accurate identification of adhesive residue on cigarette packaging, better distinguishing genuine from counterfeit cigarette packs, reducing the difficulty of counterfeiting, and improving detection efficiency and accuracy.
Smart Images

Figure CN116858828B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of cigarette box inspection technology, specifically relating to an optical process-oriented inspection device for adhesive residue on cigarette packaging. Background Technology
[0002] In the past, the methods for detecting the authenticity of cigarettes were relatively traditional: the main means of identification still relied on manual operation and subjective sensory judgment by inspectors, without the use of specialized equipment. However, in recent years, with the rapid development of information detection and artificial intelligence information processing technologies, machine vision has been widely applied in many fields such as defect detection, process monitoring, traffic navigation, and target recognition, and its market size is expanding rapidly. Currently, cigarette identification mainly focuses on four aspects: first, the internal chemical composition of the cigarette; second, the cigarette itself; third, the outer tobacco pack; and finally, the outer cigarette casing. However, due to improvements in manufacturing processes in recent years, the counterfeiting technology of cigarettes and their packaging has become increasingly sophisticated, making it impossible to effectively, quickly, and accurately identify the authenticity of cigarettes and their casings using machine learning methods.
[0003] Research has revealed that the glue residue inside cigarette packs is produced by specific machinery. These large-scale machines are difficult and costly to counterfeit. Distinguishing whether the industrial equipment that produces the glue residue is registered by identifying the edge shape of the glue residue is a new and effective way to distinguish genuine from counterfeit cigarettes.
[0004] Therefore, an optical process-oriented detection device for adhesive residue on cigarette packaging is proposed. Summary of the Invention
[0005] The purpose of this invention is to propose an optical process-oriented detection device for adhesive residue in cigarette packaging, which can efficiently and systematically solve the problem of adhesive residue identification and detection in tobacco packaging.
[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:
[0007] An optical process-oriented detection device for adhesive residue on cigarette packaging includes an image acquisition module black box. A light module and image acquisition equipment are installed at the top inside the black box. The light module and image acquisition equipment are connected to a cigarette pack adhesive residue comparison processor.
[0008] The image acquisition equipment uses existing technology; any device capable of image acquisition, such as a mobile phone or camera, can be used.
[0009] The cigarette pack adhesive residue comparison processor uses existing technology; the processing method can be found in the following:
[0010] The data processing and comparison are performed at https: / / blog.csdn.net / weixin_63983775 / article / details / 123902956.
[0011] The angle between the lighting module and image acquisition equipment and the object being inspected is 30-80°.
[0012] The image acquisition module has a spindle inside its black box. A base is located at the bottom of the spindle, and a rotatable cigarette shell fixing module is mounted on the spindle via bearings. A fixing frame is located at the bottom of the rotatable cigarette shell fixing module. A bearing is located between the fixing frame and the spindle. A first gear is located on the side wall of the fixing frame. The first gear meshes with a second gear, and the second gear is connected to a motor. The motor is mounted on a fixing plate connected to the spindle.
[0013] Furthermore, the rotatable cigarette shell fixing module includes four cigarette shell placement slots, a first position sensor is set on the base, and a second position sensor is set at the bottom of the cigarette shell placement slot. The first position sensor, the second position sensor and the motor are all connected to an external controller.
[0014] Furthermore, the four cigarette packing slots are symmetrically arranged.
[0015] Furthermore, the cigarette casing placement slot is equipped with a positioning groove and a rotatable pressure strip.
[0016] Furthermore, the lighting module includes a white LED array, a blue LED array, and a purple LED array, and the arrangement on the top of the image acquisition module black box is a blue LED array, a white LED array, a purple LED array, a white LED array, and a blue LED array.
[0017] Furthermore, the inside of the image acquisition module's black box is coated with a light-absorbing paint layer.
[0018] Compared with the prior art, the present invention has at least one of the following beneficial effects:
[0019] 1. Regarding the glue marks inside cigarette packaging, different machines produce glue mark images with different geometric shapes. By using machine learning to distinguish the glue mark images, the authenticity of cigarette packs can be more accurately verified, thus identifying the authenticity of cigarettes. Currently, the market and industry only have chemical and visual inspection technologies for cigarettes themselves and visual inspection technologies for packaging themselves. There is no inspection technology for glue marks. Moreover, glue marks are more difficult to counterfeit, which makes the verification of authenticity by this technology more effective.
[0020] 2. Four cigarette shell placement slots are provided. A first position sensor is installed on the base, and a second position sensor is installed at the bottom of the cigarette shell placement slot. The first position sensor, the second position sensor, and the motor are all connected to an external controller, and the first gear meshes with the second gear. When collecting and photographing the adhesive residue on the cigarette shell, the next cigarette shell to be collected can be placed in another part of the rotatable cigarette shell placement slot, saving time. In use, the cigarette shell to be detected is first placed in the four cigarette shell placement slots. After the first photo is taken, the motor switch is pressed, causing the rotatable cigarette shell fixing module to rotate. When the next cigarette shell placement slot rotates to the bottom of the image acquisition module black box, the first position sensor and the second position sensor are perpendicular to each other, and a signal is transmitted to the controller. The controller sends a signal to stop the motor, and then the rotatable cigarette shell fixing module stops rotating, and the next cigarette shell is photographed, thus performing a streamlined operation.
[0021] 3. It is equipped with positioning grooves and pressure strips, which can increase the stability of the cigarette shell that needs to be photographed.
[0022] The lighting module includes a white LED array, a blue LED array, and a purple LED array. The surrounding blue light source provides overall illumination of the cigarette pack in the black box, while the single-point blue light provides focused illumination of key glue marks on the cigarette pack in the black box. The surrounding white light provides supplementary lighting for key areas of the cigarette pack in the black box, enhancing contrast. When collecting and photographing the glue marks on the cigarette pack, all three light sources are lit simultaneously and remain constantly lit for 5 seconds until image acquisition is complete. The image acquisition device takes three photos within 5 seconds.
[0023] The inside of the image acquisition module black box is coated with a light-absorbing paint layer. This coating can prevent the light from the three light sources from reflecting in the limited black box space and affecting the acquisition effect of the three sample images. Attached Figure Description
[0024] Figure 1 This is a schematic diagram of the structure of the present invention.
[0025] Figure 2 Top view of the rotatable smoke shell fixing module.
[0026] Figure 3 Schematic diagram of the cigarette shell placement slot structure Figure 1 .
[0027] Figure 4 Schematic diagram of the cigarette shell placement slot structure Figure 2 .
[0028] Figure 5 This is a schematic diagram of the image acquisition equipment.
[0029] Figure 6 This is a diagram showing the arrangement of the lighting modules.
[0030] Figure 7 This is a schematic diagram showing the feasible angles for the lighting module and image acquisition equipment.
[0031] Figure 8 This is a schematic diagram of the motor connection structure.
[0032] Figure 9 This is a process flow diagram of the work.
[0033] In the diagram, 1-image acquisition module black box; 2-lighting module; 3-image acquisition device; 4-cigarette shell glue mark comparison processor; 5-spindle; 6-base; 7-rotatable cigarette shell fixing module; 8-fixing frame; 9-first gear; 10-second gear; 11-motor; 12-fixing plate; 13-cigarette shell placement slot; 14-positioning slot; 15-pressure strip. Detailed Implementation
[0034] like Figure 1-9 As shown, to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.
[0035] Example
[0036] An optical process-oriented detection device for adhesive residue on cigarette packaging, such as... Figure 1-8 As shown, the device includes an image acquisition module black box 1. Inside the image acquisition module black box 1, a light module 2 and an image acquisition device 3 are installed at the top. The light module 2 and the image acquisition device 3 are connected to the cigarette shell glue mark comparison processor 4. The angle between the light module and the image acquisition device and the object to be detected is 30-80°. In this embodiment, 45° is selected.
[0037] The image acquisition module black box 1 contains a main shaft 5, with a base 6 at its bottom. A rotatable cigarette shell fixing module 7 is mounted on the main shaft 5 via bearings, and a fixing frame 8 is located at the bottom of the rotatable cigarette shell fixing module 7. A bearing is installed between the fixing frame 8 and the main shaft 5. A first gear 9 is mounted on the side wall of the fixing frame 8, meshing with a second gear 10. The second gear 10 is connected to a motor 11, which is mounted on a fixing plate 12 connected to the main shaft 5. For the glue marks inside the cigarette packaging box, different glue marks will produce glue mark images with different geometric shapes. The method of distinguishing glue mark images by machine learning can more accurately verify the authenticity of the cigarette shell and thus identify the authenticity of the cigarettes. Currently, the market and industry only have chemical and visual inspection technologies for cigarettes themselves and visual inspection technologies for packaging themselves. There is no inspection technology for glue marks, and glue marks are more difficult to counterfeit, which makes the authenticity identification of this technology more effective.
[0038] Preferably, the rotatable cigarette shell fixing module 7 includes four cigarette shell placement slots 13. A first position sensor is set on the base, and a second position sensor is set at the bottom of the cigarette shell placement slot 13. The first position sensor, the second position sensor, and the motor 11 are all connected to an external controller. The four cigarette shell placement slots 13 are symmetrically arranged, allowing the next cigarette shell to be sampled for adhesive residue to be placed in another part of the circular rotatable cigarette shell placement slot, saving time. In use, the cigarette shell to be detected is first placed in the four cigarette shell placement slots. After the first image is taken, the motor switch is pressed, causing the rotatable cigarette shell fixing module to rotate. When the next cigarette shell placement slot rotates to the bottom of the image acquisition module's black box, the first and second position sensors are perpendicular to each other, transmitting a signal to the controller. The controller sends a signal to stop the motor, thereby stopping the rotatable cigarette shell fixing module from rotating, allowing the next cigarette shell to be photographed, thus performing a streamlined operation.
[0039] Preferably, the cigarette shell placement groove 13 is provided with a positioning groove 14 and a rotatable pressure strip 15, which can increase the stability of the cigarette shell that needs to be photographed.
[0040] Preferably, the lighting module 2 includes a white LED array, a blue LED array, and a purple LED array. The arrangement of these three LED arrays on the top of the black box 1 of the image acquisition module is as follows: blue LED array, white LED array, purple LED array, white LED array, and blue LED array. The blue light source surrounds the cigarette pack in the black box to provide overall illumination, while the single-point blue light provides focused illumination on key glue marks on the cigarette pack. The surrounding white light provides supplementary lighting for key areas of the cigarette pack in the black box, enhancing contrast. When capturing images of the cigarette pack glue marks, the three light sources are lit simultaneously and remain constantly lit for 5 seconds until image acquisition is complete. The image acquisition device takes three photos within 5 seconds.
[0041] Preferably, the inside of the black box 1 of the image acquisition module is coated with a light-absorbing paint layer. This coating can prevent the light from the three light sources from reflecting in the limited black box space and affecting the acquisition effect of the three sample images.
[0042] like Figure 9 As shown, the working process of this detection device is as follows:
[0043] Step 1: Turn on the power to the device and open the pressure strip 15 of the cigarette shell fixing module;
[0044] Step 2: Place the cigarette shell to be tested into the space in the fixing module, align the upper left corner of the cigarette shell with the limiting frame, and close the pressure strip 15 to fix the cigarette shell.
[0045] Step 3: Rotate the smoke shell fixing module 7 to rotate the smoke shell module to be acquired into the image acquisition module black box 1;
[0046] Step 4: Click the image acquisition button on the connected cable;
[0047] Step 5: Activate the top light group of the image acquisition module black box 1 to provide illumination;
[0048] The top light assembly includes a white LED array, a blue LED array, and a purple LED array.
[0049] Among them, white LEDs are used for fill light, blue LEDs are used for shadow enhancement, and purple LEDs are used to highlight the edge area of the glue mark outline;
[0050] The wave function of white light is:
[0051] The wave function of blue light is
[0052] The wave function of violet light is
[0053] Superimposed light field of three colors for
[0054]
[0055] A w A b A p The amplitudes of the three colors of light, k w k b k p These are the wavenumbers of the three colors of light, ω w ω b ω p ω represents the angular frequency of the three colors of light, z represents the direction of light wave propagation, and t represents the propagation time of the light wave.
[0056] Step 6: Using the light field of the top light group of the black box in the image acquisition module described in Step 5, three photos of glue marks in different directions are acquired. There is a 1-second interval between each acquisition. The long-cycle image acquisition can ensure the stability of the image acquisition quality. Then, the cigarette shell glue mark comparison processor selects the photo with the most obvious edge contour as the detection sample.
[0057] Step 7: Repeat steps 2-6. The cigarette shell placement slot 13 of the equipment can fix four cigarette shell packages at the same time for streamlined operation.
[0058] Step 8: The cigarette shell glue mark comparison processor identifies the current best sample based on the built-in machine learning algorithm to distinguish the cigarette shell type and glue mark type, and to determine whether the equipment that produced the glue mark is registered.
[0059] Step 9: The cigarette pack adhesive residue comparison processor completes, stores, and outputs the cigarette pack authenticity detection results based on the adhesive residue authenticity detection data;
[0060] Step 10: End.
[0061] Although the invention has been described herein with reference to several illustrative embodiments, it should be understood that many other modifications and implementations can be devised by those skilled in the art, which will fall within the scope and spirit of the principles disclosed herein. More specifically, various variations and modifications can be made to the components and / or layout of the subject matter combination within the scope of the disclosure, drawings, and claims. Besides variations and modifications to the components and / or layout, other uses will be apparent to those skilled in the art.
Claims
1. An optical flow detection device for cigarette wrapping gum marks, characterized by: The image acquisition module black box (1) is provided with a light module (2) and an image acquisition device (3) on the top, and the light module (2) and the image acquisition device (3) are connected with a cigarette shell glue mark comparison processor (4); the image acquisition module black box (1) is provided with a main shaft (5), the bottom of the main shaft (5) is provided with a base (6), and the main shaft (5) is provided with a rotatable cigarette shell fixing module (7) through a bearing, and the bottom of the rotatable cigarette shell fixing module (7) is provided with a fixing frame (8), and the fixing frame (8) and the main shaft (5) are provided with a bearing, the side wall of the fixing frame (8) is provided with a first gear (9), the first gear (9) is engaged with a second gear (10), and the second gear (10) is connected with a motor (11), and the motor (11) is arranged on a fixed plate (12) connected with the main shaft (5); The optical flow detection device for the cigarette package glue mark, the detection steps are: Step 1: turn on the power supply of the device, and open the pressing strip 15 of the cigarette shell fixing module; Step 2: place the cigarette shell to be detected into the space in the fixing module, align the left upper corner of the cigarette shell with the limiting frame, and close the pressing strip 15 to fix the cigarette shell; Step 3: rotate the cigarette shell fixing module 7 to rotate the cigarette shell module to be collected into the image acquisition module black box 1; Step 4: click the image acquisition button connected by a wire; Step 5: activate the light of the lamp group on the top of the image acquisition module black box 1; Step 6: use the light field of the lamp group on the top of the image acquisition module black box 1 to collect three glue mark photos in different directions, and the interval between each collection is 1s, and the long-period image acquisition can ensure the stability of the image acquisition quality, then the cigarette shell glue mark comparison processor selects the photo with the most obvious edge contour as the detection sample Step 7: cycle steps 2-6, the cigarette shell placing groove 13 of the device can fix four cigarette shell packages at the same time, and the flow operation is performed; Step 8: the cigarette shell glue mark comparison processor identifies the current optimal sample according to the built-in machine learning algorithm, distinguishes the cigarette shell category and the glue mark category, and distinguishes whether the equipment for manufacturing the glue mark is registered; Step 9: the cigarette shell glue mark comparison processor completes, stores and outputs the cigarette shell authenticity detection result according to the glue mark authenticity detection data; Step 10: end.
2. The optical flow detection device for cigarette packaging glue marks according to claim 1, characterized in that: The rotatable cigarette shell fixing module (7) includes four cigarette shell placing grooves (13), a first position sensor is arranged on the base, a second position sensor is arranged at the bottom of the cigarette shell placing groove (13), and the first position sensor, the second position sensor and the motor (11) are connected with an external controller.
3. The apparatus for optical flow detection of cigarette wrapping gum marks according to claim 2, characterized in that: The four cigarette shell placing grooves (13) are symmetrically arranged.
4. The optical flow detection device for cigarette package glue marks according to claim 2 or 3, characterized in that: The cigarette shell placing groove (13) is provided with a positioning groove (14) and a rotatable pressing strip (15).
5. The apparatus for optical flow detection of cigarette wrapping gum marks according to claim 1, wherein: The light module (2) includes a white LED lamp array, a blue LED lamp array and a purple LED lamp array, and the arrangement mode on the top of the image acquisition module black box (1) is a blue LED lamp array, a white LED lamp array, a purple LED lamp array, a white LED lamp array and a blue LED lamp array.
6. The apparatus for optical flow detection of cigarette wrapping gum marks according to claim 1, wherein: The inside of the image acquisition module black box (1) is coated with a light absorption paint layer.
Citation Information
Patent Citations
Optical process detection device for cigarette packaging glue marks
CN219496176U