Cigarette box inner appearance detection method, device and system, and storage medium

Abstract

The present disclosure relates to a method, device, system, and storage medium for detecting the appearance inside a cigarette box. The method for detecting the appearance inside a cigarette box includes: an intelligent industrial camera acquires the appearance image of the current cigarette pack collected by the image acquisition module in the fifth wheel of the packaging machine; the intelligent industrial camera compares the appearance image of the current cigarette pack with the appearance image of the qualified cigarette pack to determine whether there is a defect in the appearance inside the current cigarette pack; when there is a defect in the appearance inside the current cigarette pack, the intelligent industrial camera sends a rejection signal to the packaging machine controller so that the packaging machine controller controls the rejection solenoid valve to reject the current cigarette pack. The present disclosure can identify the appearance defects inside the cigarette pack, and then reject the small packs with defective appearance inside the box at the rejection port of the sixth wheel of the medium-speed packaging machine. The packaging machine box appearance detection device disclosed in the present disclosure can effectively and accurately identify and reject defective cigarette packs, and ensure the quality of the appearance inside the cigarette box.

Description

Technical Field

[0001] The present invention relates to the field of appearance inspection of packaging machines, and in particular to a method, device, system, and storage medium for inspecting the appearance of cigarette boxes. Background Art

[0002] The appearance inspection device of the related art can only inspect the appearance function of the outside of the box due to the influence of its installation position. The appearance inspection device of the inside of the box of the medium-speed packaging machine of the related art adopts an external industrial computer or a single-chip microcomputer to realize its inspection function. Summary of the invention

[0003] Through research, the inventors found that the related technology's visual detector based on industrial computers occupies a large space, has poor installation processability and reliability, and requires a large amount of maintenance. The existing GDX2 packaging machine has limited space for the five wheels and is difficult to install. The visual detector based on a single-chip microcomputer has a slow image processing speed and poor imaging quality. It can only realize simple internal box appearance defect detection and has no QR code pre-decoding function.

[0004] In view of at least one of the above technical problems, the present disclosure provides a method, device and system, and storage medium for detecting the appearance inside a cigarette box, which can identify appearance defects inside a cigarette box and then remove small packages with appearance defects at the sixth wheel rejection port of a medium-speed packaging machine.

[0005] According to one aspect of the present disclosure, a device for detecting the appearance of a cigarette box is provided, comprising:

[0006] The intelligent industrial camera is used to collect the appearance image of the current cigarette pack box in the fifth wheel of the packaging machine; compare the appearance image of the current cigarette pack box with the appearance image of the qualified cigarette pack box to determine whether there is a defect in the appearance of the current cigarette pack box; if there is a defect in the appearance of the current cigarette pack box, send a rejection signal to the packaging machine controller;

[0007] The packaging machine controller is used to receive the rejection signal sent by the intelligent industrial camera and send the rejection signal to the cigarette package rejection solenoid valve;

[0008] The reject solenoid valve is used to reject the current cigarette pack according to the reject signal sent by the packaging machine controller.

[0009] In some embodiments of the present disclosure, the packaging machine controller includes an encoder;

[0010] The encoder is used to perform phase triggering on the smart industrial camera and trigger the smart industrial camera to take pictures through the rising edge signal.

[0011] In some embodiments of the present disclosure, a packaging machine controller is used to perform data operations, storage, conversion and screen display on the information collected by the intelligent industrial camera, control the corresponding equipment actions, and transmit rejection and alarm information.

[0012] In some embodiments of the present disclosure, the packaging machine controller is used to generate a stop signal to stop the packaging unit when the number of continuously rejected cigarette packs reaches the number of continuously rejected stop packs.

[0013] In some embodiments of the present disclosure, the packaging machine controller includes a human-computer interaction interface, wherein:

[0014] The packaging machine controller is used to set relevant parameters according to the user's input through the human-computer interaction interface, wherein the relevant parameters include at least one of the bypass of the in-box appearance detection device, the in-box appearance detection shooting trigger phase, the in-box appearance result feedback phase, the in-box appearance process step and the number of continuously rejected shutdown packages.

[0015] In some embodiments of the present disclosure, the intelligent industrial camera is used to perform product identification on the current appearance image inside the cigarette pack box, compare the current appearance image inside the cigarette pack box with the threshold of the appearance image inside the box of qualified cigarette packs, determine whether there is a defect in the appearance inside the current cigarette pack box, and determine the type of cigarette pack defect when there is a defect in the appearance inside the current cigarette pack box, wherein the product identification includes at least one of appearance recognition and color recognition of the appearance inside the box, and the cigarette pack defect includes at least one of appearance defect and color recognition of the appearance inside the box.

[0016] In some embodiments of the present disclosure, the intelligent industrial camera is used to perform image acquisition, image processing, data analysis and image judgment on the collected images of the inner appearance of the cigarette package box to determine whether there are any defects in the inner appearance of the current cigarette package box.

[0017] In some embodiments of the present disclosure, the smart industrial camera is installed above the turret of the cigarette pack casing, and the lens of the smart industrial camera photographs the appearance of the cigarette pack box before packaging.

[0018] In some embodiments of the present disclosure, the cigarette box inner appearance detection device further includes:

[0019] Camera stand for adjustment in any direction in space.

[0020] In some embodiments of the present disclosure, the intelligent industrial camera includes:

[0021] The visual light source is integrated into the smart industrial camera.

[0022] In some embodiments of the present disclosure, the visual light source includes a plurality of light emitting diode lamp beads.

[0023] In some embodiments of the present disclosure, the cigarette box inner appearance detection device further includes:

[0024] The photoelectric coupling isolation circuit is arranged between the intelligent industrial camera and the packaging machine controller, and is used to isolate the input and output signals of the intelligent industrial camera from the input and output signals of the packaging machine controller.

[0025] According to another aspect of the present disclosure, a cigarette box inner appearance detection system is provided, comprising:

[0026] A device for inspecting the appearance of cigarette boxes, used to upload on-site data to a cloud device;

[0027] The cloud device is used to perform data annotation and model training according to the field data, and send the training model to the cigarette box inner appearance detection device so that the cigarette box inner appearance detection device can perform scene recognition and defective cigarette package recognition according to the trained model.

[0028] In some embodiments of the present disclosure, the cigarette box interior appearance detection device is the cigarette box interior appearance detection device as described in any of the above embodiments.

[0029] In some embodiments of the present disclosure, the cloud device includes a central cloud device and an edge cloud device, wherein:

[0030] An edge computing platform is deployed on the edge cloud device, which is used to connect with the control of the cigarette box inner appearance detection device; a basic environment of the edge virtual machine is provided, wherein the basic environment includes at least one of debugging tools, access routing, health check and monitoring log collection;

[0031] An edge computing business management platform is deployed on the central cloud device, which is responsible for data analysis and storage, and modeling through high-performance servers and artificial intelligence services; providing business orchestration capabilities, wherein the business orchestration capabilities include at least one of business monitoring statistics, automated operation and maintenance, mirror center, log analysis and image recognition.

[0032] According to another aspect of the present disclosure, a method for detecting the appearance of a cigarette box is provided, comprising:

[0033] In the fifth wheel of the packaging machine, the intelligent industrial camera obtains the appearance image of the current cigarette package box collected by the image acquisition module;

[0034] The intelligent industrial camera compares the appearance image of the current cigarette pack with the appearance image of the qualified cigarette pack to determine whether there are defects in the appearance of the current cigarette pack.

[0035] When there are defects in the appearance of the current cigarette pack, the intelligent industrial camera sends a rejection signal to the packaging machine controller so that the packaging machine controller controls the rejection solenoid valve to reject the current cigarette pack.

[0036] In some embodiments of the present disclosure, the method for detecting the inner appearance of a cigarette box further includes:

[0037] The intelligent industrial camera performs product recognition on the appearance image inside the current cigarette package box, wherein the product recognition includes at least one of appearance recognition and color recognition of the appearance inside the box;

[0038] The intelligent industrial camera compares the current cigarette pack box appearance image with the box appearance image threshold of a qualified cigarette pack to determine whether there are defects in the current cigarette pack box appearance;

[0039] When there are defects in the appearance of the current cigarette package box, the intelligent industrial camera determines the type of cigarette package defect, wherein the cigarette package defect includes at least one of appearance defects and color recognition of the appearance inside the box.

[0040] According to another aspect of the present disclosure, a method for detecting the appearance of a cigarette box is provided, comprising:

[0041] The packaging machine controller receives the rejection signal sent by the intelligent industrial camera, wherein the intelligent industrial camera collects the inner appearance image of the current cigarette pack box in the fifth round of the packaging machine, compares the inner appearance image of the current cigarette pack box with the inner appearance image of the qualified cigarette pack box, and determines whether there is a defect in the inner appearance of the current cigarette pack box; if there is a defect in the inner appearance of the current cigarette pack box, a rejection signal is sent to the packaging machine controller;

[0042] The packaging machine controller sends a rejection signal to the cigarette pack rejection solenoid valve, controlling the rejection solenoid valve to reject the current cigarette pack.

[0043] In some embodiments of the present disclosure, the method for detecting the inner appearance of a cigarette box further includes:

[0044] When the number of continuously rejected cigarette packs reaches the number of continuously rejected stop packs, the packaging machine controller generates a stop signal to stop the packaging unit.

[0045] According to another aspect of the present disclosure, there is provided an intelligent industrial camera, comprising:

[0046] An image acquisition module is used to acquire the appearance image of the current cigarette pack box in the fifth wheel of the packaging machine;

[0047] A camera memory for storing instructions;

[0048] The camera processor is used to execute the instruction so that the intelligent industrial camera performs the operation of implementing the method for detecting the inner appearance of a cigarette box as described in any of the above embodiments.

[0049] According to another aspect of the present disclosure, there is provided a packaging machine controller, comprising:

[0050] A controller memory for storing instructions;

[0051] The controller processor is used to execute the instructions so that the packaging machine controller performs the operations of implementing the method for detecting the inner appearance of a cigarette box as described in any of the above embodiments.

[0052] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein the non-transitory computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the method for detecting the inner appearance of a cigarette box as described in any of the above embodiments is implemented.

[0053] The present invention can identify the appearance defects inside the cigarette pack box, and then remove the small packs with appearance defects inside the box at the sixth wheel rejection port of the medium-speed packaging machine. The present invention can effectively and accurately identify and remove defective cigarette packs, thereby ensuring the quality of the appearance inside the cigarette box. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0055] Figure 1 Schematic diagrams of some embodiments of the device for inspecting the inner appearance of a cigarette box disclosed in the present invention.

[0056] Figure 2 Schematic diagrams of other embodiments of the device for detecting the inner appearance of a cigarette box disclosed in the present invention.

[0057] Figure 3 Schematic diagram of a photoelectric coupling isolation circuit between a camera input terminal and a packaging machine controller output terminal in some embodiments of the present disclosure.

[0058] Figure 4 Schematic diagram of a photoelectric coupling isolation circuit between a camera output terminal and a packaging machine controller input terminal in some embodiments of the present disclosure.

[0059] Figure 5 It is a schematic diagram of the structure of a camera bracket in some embodiments of the present disclosure.

[0060] Figure 6 Schematic diagram of some embodiments of the method for detecting the inner appearance of a cigarette box disclosed in the present invention.

[0061] Figure 7 Schematic diagrams of other embodiments of the method for detecting the inner appearance of a cigarette box disclosed in the present invention.

[0062] Figure 8 Schematic diagrams of some other embodiments of the method for detecting the inner appearance of a cigarette box disclosed in the present invention.

[0063] Fig. 9 Schematic diagrams of some other embodiments of the method for detecting the inner appearance of a cigarette box disclosed in the present invention.

[0064] Fig.10 Schematic diagram of some embodiments of the smart industrial camera disclosed herein.

[0065] Fig.11 Schematic diagrams of some embodiments of the packaging machine controller disclosed in the present invention.

[0066] Fig.12 Schematic diagram of some embodiments of the cigarette box interior appearance inspection system disclosed in the present invention.

[0067] Fig.13 Schematic diagrams of other embodiments of the cigarette box interior appearance detection system disclosed in the present invention.

[0068] Fig.14 Schematic diagram of some embodiments of the on-site terminal architecture of the present disclosure.

[0069] Fig.15 Schematic diagram of some embodiments of edge cloud construction disclosed in the present invention. DETAILED DESCRIPTION

[0070] The technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present disclosure and its application or use. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0071] The relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

[0072] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0073] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.

[0074] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0075] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0076] Figure 1 Schematic diagrams of some embodiments of the device for detecting the appearance of the inside of a cigarette box disclosed in the present invention. Figure 1 As shown, the cigarette box inner appearance inspection device disclosed in the present invention includes an intelligent industrial camera 110, a packaging machine controller 120 and a rejection solenoid valve 130, wherein:

[0077] The intelligent industrial camera 110 is used to collect the inner appearance image of the current cigarette pack at the fifth wheel V station of a packaging machine (such as a medium-speed packaging machine such as X2); compare the inner appearance image of the current cigarette pack with the inner appearance image of a qualified cigarette pack to determine whether there are any defects in the inner appearance of the current cigarette pack; if there are any defects in the inner appearance of the current cigarette pack, send a rejection signal to the packaging machine controller.

[0078] In some embodiments of the present disclosure, the smart industrial camera 110 is disposed at the fifth wheel V station of the packaging machine.

[0079] In some embodiments of the present disclosure, the smart industrial camera 110 is installed above the turret of the cigarette pack opening, and the lens of the smart industrial camera captures the appearance of the cigarette pack before packaging.

[0080] In some embodiments of the present disclosure, the smart industrial camera 110 has integrated a PC (Personal Computer), an I / O (Input / Output) control board, an image acquisition card, a camera, etc. into one.

[0081] In some embodiments of the present disclosure, a smart industrial camera is equipped with a WINDOWS system, and visual recognition is used to detect the appearance quality inside the box and pre-decode and identify the QR code, which is uploaded to the cloud at the same time. It can also meet the requirements of the design of the appearance inspection device inside the box, which reflects a functionally rich and user-friendly information interaction design, such as the integration of some functions such as parameter adjustment, brand switching, and viewing defect pictures.

[0082] In some embodiments of the present disclosure, the smart industrial camera 110 is installed on the fifth wheel of the packaging machine.

[0083] The packaging machine controller 120 is used to receive the rejection signal sent by the intelligent industrial camera and send the rejection signal to the cigarette package rejection solenoid valve.

[0084] The reject solenoid valve 130 is used to reject the current cigarette pack according to the reject signal sent by the packaging machine controller.

[0085] In the related technical industry, the appearance inspection inside the X2 packaging machine box is generally carried out by photoelectric sensors. The above-mentioned embodiment of the present invention is the first to install a smart industrial camera on the fifth wheel of the packaging machine to collect the appearance image of the inside of the cigarette pack box. The collected image is judged by a visual algorithm, and unqualified cigarette packs are output as a rejection signal to the MICRO II controller, and are rejected at the rejection port of the sixth wheel of the packaging machine.

[0086] The above-mentioned embodiments of the present disclosure use a new type of intelligent industrial camera as a carrier, and cleverly implement various visual algorithms on the camera operating system to detect appearance defects inside the box, and perform two-dimensional code recognition and decoding functions while detecting defects, as well as a Micro II control system and peripheral circuit modules to realize signal processing and equipment control.

[0087] In some embodiments of the present disclosure, the intelligent industrial camera 110 is used to perform product identification on the current appearance image inside the cigarette pack box, compare the current appearance image inside the cigarette pack box with the threshold of the appearance image inside the box of qualified cigarette packs, determine whether there are defects in the appearance inside the current cigarette pack box, and determine the type of cigarette pack defect when there are defects in the appearance inside the current cigarette pack box, wherein the product identification includes at least one of appearance recognition and color recognition of the appearance inside the box, and the cigarette pack defect includes at least one of appearance defect and color recognition of the appearance inside the box.

[0088] In some embodiments of the present disclosure, the appearance defects inside the cigarette package may include defects such as poor QR code printing, missing inner frame paper stripes, missing inner frame paper, missing aluminum foil pull tabs, and damaged aluminum foil.

[0089] In some embodiments of the present disclosure, the intelligent industrial camera 110 is used to perform image acquisition, image processing, data analysis and image judgment on the collected images of the inner appearance of the cigarette package box to determine whether there are any defects in the inner appearance of the current cigarette package box.

[0090] In some embodiments of the present disclosure, the cigarette box interior appearance inspection device of the present disclosure may be a five-wheel appearance visual inspection device for a medium-speed packaging machine.

[0091] Figure 2 Schematic diagrams of other embodiments of the device for detecting the appearance of the inside of a cigarette box disclosed in the present invention. Figure 1 Compared with the embodiment, Figure 2 In the embodiment, the cigarette box inner appearance inspection device disclosed in the present invention may further include an image display screen 140, wherein:

[0092] In some embodiments of the present disclosure, the image display screen 140 can also be used as a control interface for an in-box appearance inspection device to realize cigarette QR code recognition, cigarette box in-box appearance defect recognition, defect image self-adding function, defect image storage and history viewing function, intelligent industrial camera configuration function, and parameter saving function design.

[0093] In some embodiments of the present disclosure, the cigarette box inner appearance inspection device of the present disclosure may further include a touch screen display control circuit, wherein the touch screen display control circuit is used to provide power and backlight for the liquid crystal screen, so that the liquid crystal screen can work stably and reliably.

[0094] In some embodiments of the present disclosure, the image display screen 140 can also be used for the recognition of the QR code of the inner frame paper of cigarettes, the automatic generation of various related appearance defects in the box (such as inner frame paper defects), and the position and size adjustment function of the detection area; as well as the adjustment function of the detection point threshold and threshold value, which is convenient for maintenance personnel to maintain and adjust in time. Historical defect pictures can be stored and viewed, and the interface configuration is completed by operating the image recognition points, and the configuration module is automatically generated without writing any code, which reduces the threshold for use. The camera setting interface also provides exposure time and gain related parameter adjustment functions. In addition, detection enable, rejection function, and parameter saving function are provided for debugging and use by operating user personnel. There is also an unlock key when the touch screen enters the system setting to prevent human error.

[0095] In some embodiments of the present disclosure, Figure 1 and Figure 2 As shown, the packaging machine controller 120 may be a host controller of the packaging machine.

[0096] In some embodiments of the present disclosure, the packaging machine controller 120 may be a MICRO II controller of the packaging machine.

[0097] In some embodiments of the present disclosure, the packaging machine controller 120 may include an encoder, wherein the encoder is used to perform phase triggering on the smart industrial camera, and trigger the smart industrial camera to take pictures through a rising edge signal.

[0098] The control method of the above-mentioned embodiment of the present disclosure eliminates the previous method of external PLC controller, directly uses the original machine MICRO II system (MICRO II controller) for control, and performs phase triggering through the encoder of the host, eliminating the installation of an additional encoder.

[0099] In some embodiments of the present disclosure, the packaging machine controller 120 is used to perform data calculation, storage, conversion and screen display on the information collected by the intelligent industrial camera, control the corresponding equipment actions, transmit rejection and alarm information, etc.

[0100] In some embodiments of the present disclosure, the packaging machine controller 120 is used to generate a stop signal to stop the packaging unit when the number of continuously rejected cigarette packs reaches the number of continuously rejected stop packs.

[0101] The above-mentioned embodiments of the present disclosure involve the development of a Micro II control system. When the packaging machine reaches the machine phase set for the corresponding step of the appearance inspection of the packaging machine box, the intelligent industrial camera is triggered to take pictures. The intelligent industrial camera will collect information about the appearance of unqualified boxes and feed it back to the Micro II system, the control brain of the packaging machine, and generate a rejection signal to the rejection solenoid valve to reject unqualified cigarette packs. At the same time, the information of the unqualified cigarette packs at this time is recorded. If the continuous rejection reaches the set number of cigarette packs, a stop signal will be generated to stop the packaging unit.

[0102] In some embodiments of the present disclosure, the packaging machine controller 120 may include a human-computer interaction interface, wherein:

[0103] The packaging machine controller 120 is used to set relevant parameters according to the user's input through the human-computer interaction interface, wherein the relevant parameters include at least one of the bypass of the in-box appearance detection device, the in-box appearance detection shooting trigger phase, the in-box appearance result feedback phase, the in-box appearance process step and the number of continuously rejected shutdown packages.

[0104] The Micro II control system of the above embodiment of the present disclosure has also developed a set of related parameter human-machine interfaces: bypass of the in-box appearance inspection device, trigger phase of in-box appearance inspection shooting, feedback phase of in-box appearance results, in-box appearance steps, and the number of packages that are continuously rejected and stopped, etc., for use by debugging personnel. The bypass parameter of in-box appearance inspection provides exemption from the in-box appearance inspection device for debugging by maintenance personnel; the trigger phase parameter of in-box appearance shooting provides the setting angle (output of the Micro II control system to the appearance photography); the feedback phase parameter of in-box appearance results provides the setting angle (input of the appearance results to the Micro II control system); the in-box appearance steps and the number of packages that are continuously rejected and stopped provide the appearance inspection step position and the number of packages that are rejected and stopped.

[0105] In some embodiments of the present disclosure, Figure 2 As shown, the smart industrial camera 110 disclosed in the present invention may be a Mindvision smart industrial camera, the packaging machine controller 120 may be a Micro II control system, and the rejection solenoid valve 130 may be an output rejection valve.

[0106] In some embodiments of the present disclosure, Figure 2 As shown, the cigarette box inner appearance detection device may further include an IO (input and output) circuit 150, wherein:

[0107] The disclosed intelligent industrial camera is used for appearance and color recognition of the inner appearance of the small package product box of the packaging machine. By comparing the correct product threshold, the intelligent industrial camera interacts with the Micro II control system using their own IO circuits.

[0108] In some embodiments of the present disclosure, Figure 2 As shown, the cigarette pack runs on the box of the packaging machine. When it passes through the relevant process step phase, the Micro II system sends a signal to trigger the LED light of the smart industrial camera to flash, and take pictures to collect the image of the appearance of the small package. The smart industrial camera then performs image acquisition, image processing, data analysis and image judgment to make a result of the appearance of the unqualified cigarette pack.

[0109] In some embodiments of the present disclosure, Figure 2 As shown, if the in-box appearance inspection device identifies unqualified in-box appearance cigarette packs, it will be fed back to the Micro II control system through the IO input control board that interacts with the Micro II, wherein the signal is transmitted through the photoelectric coupling circuit 160, and after the Micro II control system processes the relevant information, the output signal is transmitted to the rejection solenoid valve to reject the unqualified cigarette packs.

[0110] In some embodiments of the present disclosure, the ZB45 electrical control system of the packaging machine controller uses the Micro II system, which centrally stores and processes the information transmitted by various external sensors, and then issues instructions to the actuator through a certain program. The Micro II system monitors the input of the intelligent industrial camera in real time and feeds back the instructions to the solenoid valve in real time.

[0111] The above embodiments of the present disclosure involve the development of a new type of five-wheel appearance inspection system human-machine interface and Micro II control system. The camera uses Mindvision intelligent industrial camera to identify the appearance and color of the inner frame paper of the small package product of the packaging machine. By comparing the correct product template, the intelligent industrial camera and the Micro II control system interact using their own IO. The cigarette package of the above embodiment of the present disclosure runs on the fifth wheel of the packaging machine. When passing through the relevant process step phase, the Micro II system sends a signal to trigger the LED light of the intelligent industrial camera to flash, take pictures and collect images of the appearance of the small package, and the intelligent industrial camera then obtains the appearance result of the unqualified cigarette package after image acquisition, image processing, data analysis and image judgment. In the above embodiment of the present disclosure, if the five-wheel appearance inspection device identifies an unqualified inner frame paper cigarette package, it will be fed back to the Micro II control system through the IO input control board that interacts with Micro II, wherein the signal transmission passes through the photoelectric coupling circuit, and the output signal is transmitted to the rejection solenoid valve after the Micro II control system processes the relevant information, so that the unqualified cigarette package is rejected.

[0112] In some embodiments of the present disclosure, Figure 2 As shown, the intelligent industrial camera 110 may include a visual algorithm module, an image storage system and an image acquisition module, wherein:

[0113] The visual algorithm module is used to realize image processing and analysis of the appearance inside the box through the HALCON visual software, especially various image processing algorithms, and then implement them using the MFC graphical interface in the VS2015 development software.

[0114] Image storage system, the system board hardware is equipped with a USB 3.0 Camera, and uses an SSD hard disk as a storage device. The microcontroller directly stores the appearance image of the defective box and other related information on the SSD (Solid State Disk).

[0115] The image acquisition module is used to trigger the camera to take pictures in each work cycle and transmit the image of the inside and outside of each box back to the image processing system. The camera parameters are set through the image acquisition and processing software, and the camera is controlled to acquire images. The acquired images are transmitted to the computer via USB (Universal Serial Bus) data, and the acquired images are segmented, filtered, smoothed, and other processing methods are used on the computer to extract the parameters of the image required in practice.

[0116] In some embodiments of the present disclosure, Figure 2 As shown, the cigarette box inner appearance detection device may further include a photoelectric coupling isolation circuit 160, wherein:

[0117] The photoelectric coupling isolation circuit (photocoupler circuit) 160 is arranged between the intelligent industrial camera and the packaging machine controller, and is used to isolate the input and output signals of the intelligent industrial camera from the input and output signals of the packaging machine controller.

[0118] The optocoupler circuit 160 of the above-mentioned embodiment of the present disclosure connects the input and output of the intelligent industrial camera with the output and input of the packaging machine controller (such as a MicroII control system) through photoelectric coupling, doubly protecting the working safety of each terminal. When a line overload occurs at either end, it will not interfere with the other end, thereby playing the role of input and output isolation.

[0119] In some embodiments of the present disclosure, Figure 2 As shown, the smart industrial camera may include: a visual light source 170 integrated into the smart industrial camera.

[0120] In some embodiments of the present disclosure, the visual light source 170 may include a plurality of LED (light-emitting diode) flash lamps.

[0121] The above-mentioned embodiment of the present disclosure innovatively integrates 12 LED flash lamps (lamp beads) as a visual light source into an intelligent industrial camera, eliminating the need to install a light source bracket, thereby greatly reducing the installation space.

[0122] Figure 3 Schematic diagram of a photoelectric coupling isolation circuit between a camera input terminal and a packaging machine controller output terminal in some embodiments of the present disclosure. Figure 4 Schematic diagram of the optocoupler isolation circuit between the camera output terminal and the packaging machine controller input terminal in some embodiments of the present disclosure. Figure 3 and Figure 4 As shown, the packaging machine controller disclosed in the present invention is a PLC (Programmable Logic Controller).

[0123] In some embodiments of the present disclosure, Figure 3 and Figure 4 As shown, the input and output of the PLC and the smart industrial camera disclosed in the present invention are all isolated by optical coupling, the input voltage is 5-24V; the output is open drain, and the maximum current shall not exceed 50mA. A high-performance optical coupler is used to isolate the input and output signals of the smart industrial camera from the input and output signals of the Micro II system to avoid mutual interference.

[0124] In some embodiments of the present disclosure, Figure 3 As shown in the figure, the external trigger input terminals of the smart industrial camera are labeled TRIG+ and TRIG-, which change from 0V to 5~24V. Once the voltage changes, the rising edge signal triggers the smart industrial camera to take pictures. The camera uses a photocoupler TLP281 for isolation, which is used to input signals to enable the smart industrial camera to take pictures.

[0125] In some embodiments of the present disclosure, Figure 4 As shown, the external trigger output end of the smart industrial camera is the 4th pin of the photoelectric coupler TLP281, which outputs a rising edge signal to the Micro II control system processor to determine whether the appearance inspection smart industrial camera is a good cigarette package or a bad cigarette package.

[0126] Figure 5 Schematic diagram of the structure of the camera bracket in some embodiments of the present disclosure. Figure 5 As shown, the cigarette box inner appearance detection device disclosed in the present invention may also include a camera bracket, wherein:

[0127] Camera stand for adjustment in any direction in space.

[0128] The camera of the above-mentioned embodiment of the present disclosure is installed in the V-station of the fifth wheel of the packaging machine.

[0129] The camera bracket of the above embodiment of the present disclosure can be adjusted in any direction in space, and has the characteristics of stable installation and convenient adjustment.

[0130] The cigarette box inner appearance inspection device of the above embodiment of the present disclosure can identify the inner appearance defects of cigarette packs, and then remove the small packs with defective inner appearance at the sixth wheel rejection port of the medium-speed packaging machine. The cigarette box inner appearance inspection device of the above embodiment of the present disclosure can effectively and accurately identify and remove defective cigarette packs, ensuring the quality of the inner appearance of the cigarette box. The above embodiment of the present disclosure can prevent such unqualified cigarette packs from entering the next production process, causing serious product quality accidents.

[0131] Figure 6 Schematic diagram of some embodiments of the method for detecting the appearance of the inside of a cigarette box disclosed in the present invention. Preferably, this embodiment can be performed by the intelligent industrial camera disclosed in the present invention, the device for detecting the appearance of the inside of a cigarette box disclosed in the present invention, or the system for detecting the appearance of the inside of a cigarette box disclosed in the present invention. The method includes at least one of the following steps 61 to 63, wherein:

[0132] Step 61, the intelligent industrial camera obtains the appearance image of the current cigarette package box captured by the image acquisition module in the fifth wheel of the packaging machine.

[0133] Step 62, the intelligent industrial camera compares the inner appearance image of the current cigarette pack with the inner appearance image of a qualified cigarette pack to determine whether there are any defects in the inner appearance of the current cigarette pack.

[0134] Step 63, when there are defects in the appearance of the current cigarette pack, the intelligent industrial camera sends a rejection signal to the packaging machine controller so that the packaging machine controller controls the rejection solenoid valve to reject the current cigarette pack.

[0135] In some embodiments of the present disclosure, the method for detecting the inner appearance of a cigarette box may further include: using an intelligent industrial camera to perform product recognition on the inner appearance image of a current cigarette pack, wherein the product recognition includes at least one of appearance recognition and color recognition of the inner appearance; using an intelligent industrial camera to compare the inner appearance image of the current cigarette pack with a threshold value of an inner appearance image of a qualified cigarette pack to determine whether there is a defect in the inner appearance of the current cigarette pack; and using an intelligent industrial camera to determine the type of cigarette pack defect when there is a defect in the inner appearance of the current cigarette pack, wherein the cigarette pack defect includes at least one of appearance defect and color recognition of the inner appearance.

[0136] Figure 7 Schematic diagrams of other embodiments of the method for detecting the appearance of the inside of a cigarette box disclosed in the present invention. Preferably, this embodiment can be performed by a packaging machine controller disclosed in the present invention, a cigarette box inside appearance detection device disclosed in the present invention, or a cigarette box inside appearance detection system disclosed in the present invention. The method includes at least one of the following steps 71 to 73, wherein:

[0137] Step 71, the packaging machine controller receives the rejection signal sent by the smart industrial camera, wherein the smart industrial camera collects the appearance image of the current cigarette pack box in the fifth round of the packaging machine, compares the appearance image of the current cigarette pack box with the appearance image of the qualified cigarette pack box, and determines whether there are defects in the appearance of the current cigarette pack box; if there are defects in the appearance of the current cigarette pack box, a rejection signal is sent to the packaging machine controller.

[0138] In step 72, the packaging machine controller sends a rejection signal to the cigarette packet rejection solenoid valve, controlling the rejection solenoid valve to reject the current cigarette packet.

[0139] In some embodiments of the present disclosure, the method for inspecting the inner appearance of a cigarette box may further include: when the number of continuously rejected cigarette packs reaches a continuous rejection stop number of packs, the packaging machine controller generates a stop signal to stop the packaging unit.

[0140] Figure 8 Schematic diagrams of some other embodiments of the method for detecting the appearance of the inside of a cigarette box disclosed in the present invention. Preferably, this embodiment can be performed by the device for detecting the appearance of the inside of a cigarette box disclosed in the present invention or the system for detecting the appearance of the inside of a cigarette box disclosed in the present invention. Figure 8 As shown, the method may include: the appearance inspection system in the box uses an intelligent industrial camera as an image sensor, after starting the five-round appearance inspection system software, uses the light source LED control, obtains the appearance image of the cigarette package box through image acquisition, and extracts the area to be inspected, analyzes and processes the image module, finds out whether there is a defect, and then performs a threshold judgment on the type of defect; if its image pixel threshold exceeds the set value, it is determined that there is a defect, outputs a signal to the Micro II control system, controls the action of the rejection solenoid valve, and completes the task of rejecting the bad cigarette package; if its image pixel threshold does not exceed the set value, unlocks the interface, performs interface settings, and performs comparison and judgment on the next cigarette package image.

[0141] Fig. 9 Schematic diagrams of some other embodiments of the method for detecting the inner appearance of a cigarette box disclosed in the present invention. Preferably, this embodiment can be performed by the device for detecting the inner appearance of a cigarette box disclosed in the present invention or the system for detecting the inner appearance of a cigarette box disclosed in the present invention. Fig. 9 In the embodiment, the intelligent industrial camera uses edge triggering as the trigger condition. When a valid trigger signal is received on the trigger terminal, the camera starts to collect a frame of image and transmits it to the host. Each valid edge corresponds to a trigger, and only one frame of image is output. Before the acquisition of the previous frame of image is completed, the repeated trigger signal will be ignored.

[0142] Fig. 9 This is a schematic diagram of the operation steps and process of the new intelligent industrial camera for packaging machine box appearance inspection disclosed in this disclosure. Fig. 9 As shown, the specific implementation steps of the method may include:

[0143] Step 1: Load the dynamic link library file MVCAMSDK.DLL of the SDK (Software Development Kit).

[0144] Step 2: Initialize the SDK. After loading the SDK, call the CameraSdkInit (initialization) function to initialize it before using other interfaces.

[0145] Step 3: Call the CameraEnumerateDevice function to enumerate devices and obtain a list of cameras currently connected to the PC, including device name, version number, unique serial number, camera model, etc. If the number of enumerated cameras is less than or equal to 0, the process ends.

[0146] Step 4: If the number of enumerated cameras is greater than 0, initialize the camera and obtain the camera parameters.

[0147] In some embodiments of the present disclosure, the step of initializing the camera and obtaining camera parameters may include: calling the CameraInit function to initialize the specified camera according to the camera device enumeration information obtained in step 3 to obtain the camera parameters.

[0148] In some embodiments of the present disclosure, Fig. 9 As shown, the step of initializing the camera may further include: display initialization.

[0149] In some embodiments of the present disclosure, Fig. 9 As shown, the step of display initialization may also include: initializing the display window, setting the display window size, creating a camera setting page, and establishing a picture taking thread.

[0150] Step 5: Let the SDK enter the image acquisition mode. Call the CameraPlay function to put the camera into working mode, and the SDK starts to receive images from the camera.

[0151] Step 6: Capture image: Actively call CameraGetImageBuffer(buffer) to obtain a frame of image data.

[0152] In some embodiments of the present disclosure, step 6 may include: acquiring camera RAW data.

[0153] Step 7, process the image: call CameraImageProcess to obtain image processing effects, such as color gain adjustment, white balance correction, saturation, LUT (Look Up Table) conversion, noise reduction, etc., and convert the original data in YUV or Bayer format to 24BIT bitmap format.

[0154] Step 8. Overlay the crosshairs, auto-exposure reference window, and white balance reference window: In the CameraImageOverlay (overlay) function, the crosshairs, auto-exposure reference window, and white balance reference window set to a visible state will be overlaid on the input image.

[0155] Step 9: Save or display the image.

[0156] Step 10. Stop thread acquisition; Deinitialize: turn off the camera before exiting the program.

[0157] The cigarette box inner appearance inspection method of the above embodiment of the present disclosure can identify the inner appearance defects of cigarette packs, and then remove the small packs with defective inner appearance at the sixth wheel rejection port of the medium-speed packaging machine. The cigarette box inner appearance inspection method of the above embodiment of the present disclosure can effectively and accurately identify and remove defective cigarette packs, ensuring the quality of the inner appearance of the cigarette box. The above embodiment of the present disclosure can prevent such unqualified cigarette packs from entering the next production process, causing serious product quality accidents.

[0158] Fig.10 Schematic diagram of some embodiments of the intelligent industrial camera disclosed in the present invention. Fig.10 As shown, the smart industrial camera (eg Figure 1-Figure 4 The smart industrial camera 110 of any embodiment may include an image acquisition module 111, a camera memory 112, and a camera processor 113, wherein:

[0159] The image acquisition module 111 is used to acquire the appearance image of the current cigarette package box in the fifth round of the packaging machine.

[0160] The camera memory 112 is used to store instructions.

[0161] The camera processor 113 is used to execute the instruction so that the smart industrial camera can perform any of the above embodiments (for example Figure 5 or Figure 7-Figure 9 Operation of the cigarette box inner appearance detection method described in Example).

[0162] Fig.11 Schematic diagram of some embodiments of the packaging machine controller disclosed in the present invention. Fig.11 As shown, the packaging machine controller (eg Figure 1-Figure 4 The packaging machine controller 120 of any embodiment may include a controller memory 121 and a controller processor 122, wherein:

[0163] The controller memory 121 is used to store instructions. The controller processor 122 is coupled to the controller memory 121. The controller processor 122 is configured to execute and implement any of the above embodiments (for example, Figure 6-Figure 9 The method for detecting the inner appearance of a cigarette box as described in any embodiment).

[0164] like Fig.11 As shown, the computer device also includes a communication interface 123 for information exchange with other devices. At the same time, the computer device also includes a bus 124, through which the controller processor 122, the communication interface 123, and the controller memory 121 communicate with each other.

[0165] The controller memory 121 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory. The controller memory 121 may also be a memory array. The controller memory 121 may also be divided into blocks, and the blocks may be combined into virtual volumes according to certain rules.

[0166] In addition, the controller processor 122 may be a central processing unit (CPU), or may be an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present disclosure.

[0167] Fig.12 Schematic diagram of some embodiments of the cigarette box inner appearance detection system disclosed in the present invention. Fig.12 As shown, the cigarette box inner appearance detection system disclosed in the present invention may include a cigarette box inner appearance detection device 100 and a cloud device 200, wherein:

[0168] The cigarette box inner appearance inspection device 100 is used to upload on-site data to a cloud device.

[0169] In some embodiments of the present disclosure, the cigarette box inner appearance detection device can be any of the above embodiments (for example Figure 1-Figure 4 The device for detecting the inner appearance of a cigarette box as described in any embodiment).

[0170] The cloud device 200 is used to perform data annotation and model training based on the field data, and send the training model to the cigarette box inner appearance inspection device so that the cigarette box inner appearance inspection device can perform scene recognition and defective cigarette package recognition based on the trained model.

[0171] Fig.13 Schematic diagrams of other embodiments of the cigarette box inner appearance detection system disclosed in the present invention. Fig.13 As shown, the cigarette box inner appearance detection system disclosed in the present invention may include a cigarette box inner appearance detection device 100 and a cloud device, wherein the cloud device may include a central cloud device 210 and an edge cloud device 220, wherein:

[0172] The cigarette box inner appearance inspection device 100 is used to upload on-site data to the central cloud device 210 via the edge cloud device 220 .

[0173] The central cloud device 210 is used to perform data annotation and model training based on the field data, and send the training model to the cigarette box inner appearance inspection device 100 through the edge cloud device 220, so that the cigarette box inner appearance inspection device 100 can perform scene recognition and defective cigarette package recognition according to the trained model.

[0174] The edge cloud device 220 is deployed with a MEP (MEC Platform, edge computing platform) for connecting with the control of the appearance inspection device inside the cigarette box; providing an EVM (Edge virtual machine, edge virtual machine) basic environment, wherein the basic environment includes at least one of debugging tools, access routing, health checks, and monitoring log collection.

[0175] In some embodiments of the present disclosure, the edge cloud device 220 is deployed near the edge computer room of the telecommunications service provider, using its self-developed edge cloud platform.

[0176] In some embodiments of the present disclosure, Fig.13 As shown, the edge cloud device 220 may include an AI inference server, an AI management server, a management screen, and a storage server.

[0177] An edge computing business management platform is deployed on the central cloud device 210, which is responsible for data analysis and storage, and modeling through high-performance servers and artificial intelligence services; providing business orchestration capabilities, wherein the business orchestration capabilities include at least one of business monitoring statistics, automated operation and maintenance, mirror center, log analysis, and image recognition.

[0178] In some embodiments of the present disclosure, the central cloud device 210 is deployed in a telecommunications service provider's cloud center node, uses public cloud resources, and is responsible for data analysis and storage.

[0179] In some embodiments of the present disclosure, Fig.13 As shown, the central cloud device 210 is used to realize functions such as image annotation, natural speech processing, text recognition, algorithm research, cloud-edge management, speech recognition, knowledge graph, big data analysis, model training and Internet of Things management.

[0180] In some embodiments of the present disclosure, in the construction of the central cloud, the cloud resource area provides elastic cloud hosts + distributed storage to carry various application services of the center. Distributed storage provides storage resources for database servers and application servers as required. Data backup and disaster recovery functions can also be provided within the Internet cloud area through cloud hard disk backup, cloud host backup and other services. Each business system can be divided into multiple subnets in the VPC (Virtual Private Cloud), and isolated through the subnets. Within the subnet, network security isolation can be performed within the subnet by dividing security groups. If intercommunication between subnets is required, ACLs (Access Control Lists) between subnets can be configured for interconnection, so that AI artificial intelligence (Artificial Intelligence) applications of multiple different modules can be performed. The processed results can be pushed and notified to designated personnel in the form of text messages or emails.

[0181] Fig.14 Schematic diagram of some embodiments of the on-site terminal architecture of the present disclosure. Figure 4 As shown, the field terminal architecture of the edge cloud device of the present disclosure may include a terminal device in a production workshop (i.e., a cigarette box inner appearance inspection device 100), a CPE (Customer Premise Equipment) and an intelligent convergence device in a workshop computer room, and an edge cloud device 220, wherein:

[0182] The terminal device collects photos of cigarette packages, opens the collection port by modifying the program, and transmits the files to the CPE through the on-site industrial wireless WIFI while generating photos or video stream files, and then uploads them to the edge cloud platform. The terminal and the edge cloud are connected through the 5G network. The data collected by the terminal is collected by a dedicated line to the edge cloud node through a public or self-built UPF.

[0183] In some embodiments of the present disclosure, Fig.14 As shown, the edge cloud device 220 may include a private cloud server and a MEC (Multi-Access Edge Computing) server.

[0184] In some embodiments of the present disclosure, Fig.14 As shown, a plurality of cigarette box inner appearance inspection devices 100 are arranged in the production workshop, and each cigarette box inner appearance inspection device 100 may include an edge computing industrial-grade optical terminal (i.e., an intelligent industrial camera), an end-side AI, an I / O control module, and an all-in-one display screen.

[0185] Private cloud servers and MEC (Multi-Access Edge Computing) servers.

[0186] The device for detecting the appearance of the cigarette box in the above-mentioned embodiment of the present disclosure can identify the appearance defects of the cigarette pack, and then remove the small packs with defective appearance at the sixth round rejection port of the medium-speed packaging machine. The device can effectively and accurately identify and remove defective cigarette packs, ensure the quality of the appearance of the cigarette box, and upload it to the cloud in real time using the on-site wireless technology in the workshop. The overall construction adopts the "wireless network + edge computing + industrial Internet platform" architecture model. The wireless network uses 5G clients to support edge computing using edge MEC servers and industrial edge gateways. The industrial Internet platform of the above-mentioned embodiment of the present disclosure adopts the concept of a general platform combined with the network bandwidth support of telecommunications service providers.

[0187] The above-mentioned embodiments of the present disclosure can be based on the needs of the roll packaging workshop and the status of the existing resource pool. The above-mentioned embodiments of the present disclosure adopt a cloud-edge collaborative construction solution of public cloud + edge cloud to build a 5G + industrial Internet application platform for the roll packaging workshop.

[0188] Fig.15 Schematic diagram of some embodiments of edge cloud construction disclosed in the present invention. Fig.15 As shown in the figure, in the edge cloud construction, ECF (Edge Computing Fabric) provides a complete IaaS (Infrastructure as a Service) cloud computing platform solution and the scheduling capability of the public cloud, which can provide a complete off-site / heterogeneous management platform for cloud computing resources for enterprise customers (such as customer cloud applications, customer support systems and customer edge devices).

[0189] The cloud computing platform disclosed herein can provide business support such as service management, resource management, application store, tenant management, application log, and image management.

[0190] The cloud computing platform disclosed in the present invention sets up edge virtual machines, clusters, edge container instances and edge serverless K8S components based on EBM (Edge physics machine), and provides edge multi-cluster container services.

[0191] The cloud computing platform disclosed herein also provides IAM (Identity and Access Management) services.

[0192] The disclosed cloud computing platform provides complete off-site / heterogeneous cloud computing resource services for customer cloud applications, customer support systems and customer edge devices through API (Application Programming Interface) & SDK. The API & SDK modules include ECI (Edge Computing Interconnect) API, VK (Virtual-Key) SDK, K8S API, VM (Virtual Machine) API and EBMAPI.

[0193] The above-mentioned embodiment of the present disclosure mainly focuses on the appearance defects of the X2 box of the tobacco industry packaging machine, such as poor QR code printing, missing inner frame paper stripes, missing inner frame paper, missing aluminum foil pull tabs, and damaged aluminum foil. This patent designs a new five-wheel appearance detection device for packaging machines, which is installed at the V station of the fifth wheel. An industrial camera is used on site to collect product photos and production flow videos, which are uploaded to the edge cloud platform through the on-site industrial wireless WIFI and CPE. By combining vision with cloud technology, functions such as box appearance detection, QR code pre-decoding, product image storage, and product traceability are realized. The new detection device can accurately and efficiently identify cigarette packages with defects in the appearance of the box on the fifth wheel of the packaging machine, and can prevent such unqualified cigarette packages from entering the next production process, causing serious product quality accidents.

[0194] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein the non-transitory computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, any of the above embodiments (for example Figure 5-Figure 9 The method for detecting the inner appearance of a cigarette box as described in any embodiment).

[0195] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, devices, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0196] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0197] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0198] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0199] The packaging machine controller described above can be implemented as a general-purpose processor, a programmable logic controller (PLC), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components or any appropriate combination thereof for performing the functions described in this application.

[0200] So far, the present disclosure has been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

[0201] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or may be accomplished by instructing the relevant hardware through a program, and the program may be stored in a non-transitory computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0202] The description of the present disclosure is given for the purpose of illustration and description, and is not intended to be exhaustive or to limit the present disclosure to the disclosed form. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present disclosure, and to enable those of ordinary skill in the art to understand the present disclosure and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A device for detecting the inner appearance of a cigarette box, characterized in that: include: Intelligent industrial camera, used to collect the appearance image of the current cigarette pack in the fifth wheel of the packaging machine; Realize cigarette package QR code recognition and decoding; Compare the current cigarette pack box inner appearance image with the box inner appearance image of a qualified cigarette pack to determine whether the current cigarette pack box inner appearance has defects; In the case where there are defects in the appearance of the current cigarette pack, a rejection signal is sent to the packaging machine controller, wherein the intelligent industrial camera includes: a visual light source integrated with the intelligent industrial camera, without the need to install a light source bracket, and the visual light source includes a plurality of light-emitting diode lamp beads; the appearance defects of the cigarette pack include poor printing of the two-dimensional code; The packaging machine controller is used to receive the rejection signal sent by the intelligent industrial camera and send the rejection signal to the cigarette package rejection solenoid valve; A rejection solenoid valve is used to reject the current cigarette pack according to the rejection signal sent by the packaging machine controller; The packaging machine controller is also used to generate a stop signal to stop the packaging unit when the number of continuously rejected cigarette packs reaches the number of continuously rejected stop packs; An intelligent industrial camera is used to perform product recognition on the current cigarette package box inner appearance image, compare the current cigarette package box inner appearance image with the box inner appearance image threshold of a qualified cigarette package, determine whether the current cigarette package box inner appearance has defects, and determine the cigarette package defect type when the current cigarette package box inner appearance has defects, wherein the product recognition includes at least one of appearance recognition and color recognition of the box inner appearance, and the cigarette package defect includes at least one of appearance defects and color defects of the box inner appearance; The cigarette box inner appearance inspection device is used to upload field data to a cloud device, wherein the cloud device includes a central cloud device and an edge cloud device, and the cloud device is used to perform data annotation and model training based on the field data, and send the trained model to the cigarette box inner appearance inspection device; receive the model trained by the cloud device, and perform scene recognition and defective cigarette package recognition based on the trained model; An edge computing platform is deployed on the edge cloud device, which is used to connect with the control of the appearance inspection device inside the cigarette box, and provide a basic environment for the edge virtual machine. The basic environment includes at least one of debugging tools, access routing, health checks and monitoring log collection; an edge computing business management platform is deployed on the central cloud device, which is responsible for data analysis and storage, modeling through high-performance servers and artificial intelligence services, and providing business orchestration capabilities. The business orchestration capabilities include at least one of business monitoring statistics, automated operation and maintenance, mirror center, log analysis and image recognition.

2. The cigarette box inner appearance detection device according to claim 1, characterized in that: The packaging machine controller includes an encoder; The encoder is used to perform phase triggering on the smart industrial camera and trigger the smart industrial camera to take pictures through the rising edge signal.

3. The cigarette box inner appearance detection device according to claim 1, characterized in that: The packaging machine controller is used to perform data calculation, storage, conversion and screen display on the information collected by the intelligent industrial camera, control the corresponding equipment actions, and transmit rejection and alarm information.

4. The cigarette box inner appearance inspection device according to any one of claims 1 to 3, characterized in that: The packaging machine controller includes a human-machine interaction interface, wherein: The packaging machine controller is used to set relevant parameters according to the user's input through the human-computer interaction interface, wherein the relevant parameters include at least one of the bypass of the in-box appearance detection device, the in-box appearance detection shooting trigger phase, the in-box appearance result feedback phase, the in-box appearance process step and the number of continuously rejected shutdown packages.

5. The cigarette box inner appearance inspection device according to any one of claims 1 to 3, characterized in that: The intelligent industrial camera is used to perform image acquisition, image processing, data analysis and image judgment on the collected images of the inner appearance of the cigarette package box to determine whether there are defects in the inner appearance of the current cigarette package box.

6. The cigarette box inner appearance inspection device according to any one of claims 1 to 3, characterized in that: The intelligent industrial camera is installed above the turret of the cigarette pack sleeve, and the lens of the intelligent industrial camera takes pictures of the cigarette pack's internal and external appearance.

7. The device for detecting the appearance of the inside of a cigarette box according to claim 6, characterized in that: Also includes: Camera stand for adjustment in any direction in space.

8. The cigarette box inner appearance inspection device according to any one of claims 1 to 3, characterized in that: Also includes: The photoelectric coupling isolation circuit is arranged between the intelligent industrial camera and the packaging machine controller, and is used to isolate the input and output signals of the intelligent industrial camera from the input and output signals of the packaging machine controller.

9. A cigarette box inner appearance inspection system, characterized in that: It comprises a cloud device and a cigarette box inner appearance detection device as described in any one of claims 1 to 8.

10. A method for detecting the inner appearance of a cigarette box, characterized in that: include: The intelligent industrial camera acquires the appearance image of the current cigarette pack box collected by the image acquisition module in the fifth wheel of the packaging machine, wherein the intelligent industrial camera comprises: a visual light source integratedly installed in the intelligent industrial camera, without the need to install a light source bracket, and the visual light source comprises a plurality of light emitting diode lamp beads; Intelligent industrial cameras can recognize and decode the QR codes on cigarette packages; The intelligent industrial camera compares the appearance image of the current cigarette pack with the appearance image of the qualified cigarette pack to determine whether there are defects in the appearance of the current cigarette pack. The appearance defects of the cigarette pack include poor printing of the QR code. When there are defects in the appearance of the current cigarette pack, the intelligent industrial camera sends a rejection signal to the packaging machine controller, so that the packaging machine controller controls the rejection solenoid valve to reject the current cigarette pack. The packaging machine controller is used to generate a stop signal to stop the packaging unit when the number of continuously rejected cigarette packs reaches the number of continuously rejected stop packs. Wherein, the cigarette box inner appearance detection method further comprises: The intelligent industrial camera performs product recognition on the appearance image inside the current cigarette package box, wherein the product recognition includes at least one of appearance recognition and color recognition of the appearance inside the box; The intelligent industrial camera compares the current cigarette pack box appearance image with the box appearance image threshold of a qualified cigarette pack to determine whether there are defects in the current cigarette pack box appearance; The intelligent industrial camera determines the type of cigarette package defect when there is a defect in the appearance of the current cigarette package box, wherein the cigarette package defect includes at least one of an appearance defect and a color defect of the appearance of the box; The cigarette box inner appearance detection device uploads the on-site data to the cloud device, wherein the cigarette box inner appearance detection device includes the intelligent industrial camera; The cloud device performs data annotation and model training based on the field data, and sends the training model to the cigarette box inner appearance inspection device; The cigarette box inner appearance inspection device receives the model trained by the cloud device, and performs scene recognition and defective cigarette package recognition according to the trained model; The edge cloud device is connected to the control of the cigarette box inner appearance inspection device to provide a basic environment for the edge virtual machine, wherein an edge computing platform is deployed on the edge cloud device, the basic environment includes at least one of debugging tools, access routing, health checks and monitoring log collection, and the cloud device includes an edge cloud device and a central cloud device; The central cloud device is responsible for data analysis and storage, and is modeled through high-performance servers and artificial intelligence services to provide business orchestration capabilities. An edge computing business management platform is deployed on the central cloud device. The business orchestration capabilities include at least one of business monitoring statistics, automated operation and maintenance, mirror center, log analysis, and image recognition.

11. A method for detecting the inner appearance of a cigarette box, characterized in that: include: The packaging machine controller receives a rejection signal sent by the intelligent industrial camera, wherein the intelligent industrial camera collects the appearance image of the current cigarette pack box in the fifth round of the packaging machine, realizes the recognition and decoding of the cigarette pack QR code, compares the appearance image of the current cigarette pack box with the appearance image of the qualified cigarette pack box, determines whether there is a defect in the appearance of the current cigarette pack box, and sends a rejection signal to the packaging machine controller when there is a defect in the appearance of the current cigarette pack box. The intelligent industrial camera includes: a visual light source integratedly installed in the intelligent industrial camera, without the need to install a light source bracket, the visual light source includes a plurality of light emitting diode lamp beads, and the appearance defects of the cigarette pack box include poor printing of the QR code; The packaging machine controller sends a rejection signal to the cigarette pack rejection solenoid valve, controlling the rejection solenoid valve to reject the current cigarette pack; Wherein, the cigarette box inner appearance detection method further comprises: When the number of cigarette packs continuously rejected reaches the number of packs to be rejected and stopped, the packaging machine controller generates a stop signal to stop the packaging unit; The cigarette box inner appearance detection device uploads the on-site data to the cloud device, wherein the cigarette box inner appearance detection device includes the packaging machine controller; The cloud device performs data annotation and model training based on the field data, and sends the training model to the cigarette box inner appearance inspection device; The cigarette box inner appearance inspection device receives the model trained by the cloud device, and performs scene recognition and defective cigarette package recognition according to the trained model; The edge cloud device is connected to the control of the cigarette box inner appearance inspection device to provide a basic environment for the edge virtual machine, wherein an edge computing platform is deployed on the edge cloud device, the basic environment includes at least one of debugging tools, access routing, health checks and monitoring log collection, and the cloud device includes an edge cloud device and a central cloud device; The central cloud device is responsible for data analysis and storage, and is modeled through high-performance servers and artificial intelligence services to provide business orchestration capabilities. An edge computing business management platform is deployed on the central cloud device. The business orchestration capabilities include at least one of business monitoring statistics, automated operation and maintenance, mirror center, log analysis, and image recognition.

12. An intelligent industrial camera, characterized in that: include: An image acquisition module is used to acquire the appearance image of the current cigarette pack box in the fifth wheel of the packaging machine; A camera memory for storing instructions; The camera processor is used to execute the instruction so that the intelligent industrial camera performs the operation of implementing the method for detecting the inner appearance of a cigarette box as claimed in claim 10.

13. A packaging machine controller, characterized in that: include: A controller memory for storing instructions; The controller processor is used to execute the instruction so that the packaging machine controller performs the operation of implementing the method for detecting the inner appearance of a cigarette box as claimed in claim 11.

14. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and when the instructions are executed by the processor, the method for detecting the inner appearance of a cigarette box according to claim 10 or 11 is implemented.

Images