Cigarette packer detection equipment spot inspection method, equipment, medium and product

Abstract

The invention discloses a point inspection method, equipment, medium and product for detection equipment of a cigarette packaging machine, and the method comprises the following steps: in the process that the cigarette packaging machine carries out cigarette packet production through a plurality of stations, acquiring results through an image acquisition device and / or at least one first encoder; abnormal cigarette packets occurring in the production process are recognized in real time; when it is determined that the abnormal cigarette packet is recognized, the target moving state of the abnormal cigarette packet when the abnormal cigarette packet moves to the removing station of the cigarette packaging machine is tracked in real time; when it is determined that the abnormal cigarette packet is in the target moving state and is separated from the non-rejected station as an unqualified product, a point inspection abnormity prompt matched with the abnormal cigarette packet is generated, the point inspection accuracy is improved, production monitoring and quality control are enhanced, the point inspection efficiency and the product quality are improved, and the product quality is improved. By means of real-time prompts associated with specific abnormal cigarette packets, fault processing time is shortened, production line stagnation is reduced, and production continuity and the intelligent level of the cigarette packaging machine are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cigarette processing equipment, and particularly relates to a method, device, medium and product for point inspection of a cigarette packaging machine detection device. BACKGROUND

[0002] In the field of cigarette production, point inspection of a detection device is a key control means for ensuring product process quality, and the core lies in confirming whether the device can accurately identify a defective sample and correctly remove it at a subsequent removal station.

[0003] At present, in the process of cigarette production, a traditional manual point inspection method is usually used. This method needs to manufacture a defective sample manually under a shutdown state, track the detection and removal of the defective sample in the production process at a low speed, and verify the functionality of the cigarette packaging machine.

[0004] However, in actual operation, since there are many detection devices on the packaging machine, each time of point inspection can only be performed on a single item, and the path length from the detection station to the removal station is not the same. Especially when the path is long, the entire process involves multiple time-consuming links, which is difficult to meet the efficient production demand and reduces the point inspection efficiency. In addition, this method relies on manual experience and lacks a unified point inspection standard, which reduces the accuracy of point inspection. SUMMARY

[0005] The present application provides a method, device, medium and product for point inspection of a cigarette packaging machine detection device to solve the problems of low point inspection efficiency and low accuracy of the cigarette packaging machine detection device.

[0006] According to an aspect of an embodiment of the present application, a method for point inspection of a cigarette packaging machine detection device is provided, comprising:

[0007] In the process of cigarette production by the cigarette packaging machine through multiple stations, an abnormal cigarette packet appearing in the production process is identified in real time through the collection results of an image collection device and / or at least one first encoder;

[0008] When it is determined that the abnormal cigarette packet is identified, a target moving state of the abnormal cigarette packet when moving to a removal station of the cigarette packaging machine is tracked in real time;

[0009] When it is determined that the abnormal cigarette packet is in the target moving state and is not separated as an unqualified product by the removal station, a point inspection abnormality prompt matched with the abnormal cigarette packet is generated.

[0010] According to another aspect of an embodiment of the present application, a device for point inspection of a cigarette packaging machine detection device is provided, comprising:

[0011] The collection module is configured to identify an abnormal cigarette packet in real time by using the image collection device and / or the collection result of the at least one first encoder during the production of the cigarette packets by the cigarette packaging machine.

[0012] The tracking module is configured to track a target moving state of the abnormal cigarette packet when the abnormal cigarette packet moves to the rejection station of the cigarette packaging machine in real time.

[0013] The prompting module is configured to generate an inspection abnormality prompt matched with the abnormal cigarette packet when the abnormal cigarette packet is in the target moving state and is not separated from the cigarette packaging machine as a substandard product.

[0014] According to another aspect of the embodiments of the present application, an electronic device is provided, and the electronic device comprises:

[0015] at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the method for inspection of the cigarette packaging machine detection device according to any one of the embodiments of the present application.

[0016] According to another aspect of the embodiments of the present application, a computer readable storage medium is provided, and the computer readable storage medium stores computer instructions for enabling a processor to implement the method for inspection of the cigarette packaging machine detection device according to any one of the embodiments of the present application when the computer instructions are executed by the processor.

[0017] According to another aspect of the embodiments of the present application, a computer program product is also provided, and the computer program product comprises a computer program for enabling a processor to implement the steps of the method according to any one of the embodiments of the present application when the computer program is executed by the processor.

[0018] The technical scheme of the embodiment of the present application, in the process of cigarette package production by a plurality of workstations of a cigarette packaging machine, identifies the abnormal cigarette package in real time through the collection results of an image collection device and / or at least one first encoder; when it is determined that an abnormal cigarette package is identified, the target moving state of the abnormal cigarette package when moving to the rejection workstation of the cigarette packaging machine is tracked in real time; when it is determined that the abnormal cigarette package is in the target moving state and is not separated by the rejection workstation as unqualified products, a point inspection abnormality prompt matched with the abnormal cigarette package is generated. Compared with the traditional manual point inspection, the stop, defect making, and low-speed tracking of the defect sample through the detection port and the rejection port and other operations are reduced, the human perception of fuzzy defects is avoided, and the point inspection accuracy is improved. Through the trajectory tracking and rejection verification mechanism, the production monitoring and quality control are strengthened, the point inspection efficiency and product quality are improved, the instant prompt associated with the specific abnormal cigarette package helps the operator quickly locate the problem node, shortens the fault handling time, reduces the production line stagnation, and improves the production continuity and the intelligent level of the cigarette packaging machine.

[0019] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0021] Figure 1 is a flowchart of a point inspection method of a cigarette packaging machine detection device according to the first embodiment of the present application;

[0022] Figure 2 is a flowchart of another point inspection method of a cigarette packaging machine detection device according to the second embodiment of the present application;

[0023] Figure 3 is a flowchart of still another point inspection method of a cigarette packaging machine detection device according to the third embodiment of the present application;

[0024] Figure 4 is a structural schematic diagram of a point inspection device of a cigarette packaging machine detection device according to the fourth embodiment of the present application;

[0025] Figure 5 is a structural schematic diagram of an electronic device for implementing the point inspection method of the cigarette packaging machine detection device according to the present application. DETAILED DESCRIPTION

[0026] In order to better understand the technical scheme of the present application, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts should fall within the protection scope of the present application.

[0027] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] Embodiment one

[0029] Figure 1 A flowchart of a method for point inspection of a cigarette packaging machine detection device is provided for the first embodiment of the present application. The present embodiment can be applicable to the point inspection of a cigarette packaging machine detection device. The method can be executed by a cigarette packaging machine detection device point inspection device, which can be realized in the form of hardware and / or software and can generally be configured in an electronic device. As shown in the figure, the method comprises: Figure 1

[0030] S110, during the production of cigarette packets by the cigarette packaging machine through a plurality of workstations, the abnormal cigarette packets appearing in the production process are identified in real time through the acquisition results of the image acquisition device and / or at least one first encoder.

[0031] In the embodiment of the present application, the image acquisition device can be specifically understood as a high-speed camera or a visual sensor and other image acquisition equipment installed on the production line. The image acquisition device can generally be installed at key detection workstations, such as a small packet oil sealing machine workstation, a carton packaging machine workstation, and a carton oil sealing machine workstation, etc.

[0032] ​The small pack oil sealing machine is used for transparent paper packaging of single cigarette packs, and completes the packaging process of cigarette small pack oil sealing. At this station, the cigarette pack is wrapped with transparent paper, and operations such as hot bonding are completed through related components, so that the cigarette pack is protected and preliminarily packaged. Meanwhile, the station can also have cigarette pack detection and replenishment functions, such as detecting unqualified cigarette packs and rejecting them, replenishing cigarette packs in the empty positions after rejection, and qualified cigarette packs being sent to the outlet and entering the next process flow.

[0033] The carton packaging machine station is a downstream station of the small pack oil sealing machine station, and is used for packaging the cigarette packs from the small pack oil sealing machine station into cartons according to a certain number and arrangement (such as five packs in horizontal arrangement in two layers). During carton conveying and forming, the station performs operations such as gluing and folding on the carton, so that the carton is completed and formed. In addition, the carton packaging machine station can also be provided with a rejection device for rejecting unqualified products with defects or incomplete cigarette pack stacking in the carton, and qualified cigarette cartons are sent to the outlet and enter the next process flow.

[0034] The carton oil sealing machine station is a downstream station of the carton packaging machine station, and is used for re-packaging the carton cigarettes packaged by the carton packaging machine station, usually by wrapping a layer of transparent paper, to further protect the carton cigarettes and improve the appearance and moisture-proof performance of the products. The wrapping operation is mainly driven by a cam swing lever mechanism, and the power is generally input by the carton packaging machine through a transmission shaft. The station can ensure mechanical synchronization between the carton packaging machine station and the carton oil sealing machine station through the coupling of the carton packaging machine and the carton oil sealing machine, to ensure the stability and accuracy of the packaging process. In addition, the carton oil sealing machine station can also be provided with a rejection device for rejecting unqualified products with defects in carton packaging.

[0035] The first encoder can be specifically understood as a rotary encoder installed on a mechanical component directly related to the movement of the cigarette pack, such as a drive shaft of a conveyor belt or a main shaft of a packaging machine. It can be understood that the number of pulses output by the encoder per revolution is fixed, and it is assumed that the encoder outputs P pulse signals per revolution. When the encoder rotates N pulses, the corresponding actual rotation angle θ is θ = 360° × (N / P). It can be understood that the rotation of the encoder drives the movement of the conveyor belt. Assuming that the circumference of the conveyor belt drive wheel is L, the movement distance S of the conveyor belt (and the cigarette pack thereon) can be calculated by the following formula: S = L × θ / 360°, and the movement distance S = L × N / P is obtained by substituting the actual rotation angle calculation formula, which represents the relationship between the movement distance of the cigarette pack and the number of pulses output by the encoder.

[0036] Assuming that the encoder outputs N pulses in time t, the rotation speed of the encoder (i.e. the number of pulses per unit time) can be expressed as N / t. The angular speed ω corresponding to the rotation speed of the encoder can thus be calculated as ω = (360° / P) x (N / t), i.e. the proportion of the number of pulses P output per second multiplied by the angle of the circumference 360°, to obtain the angle of rotation per second. The moving speed V of the cigarette packet can be obtained by converting the angular speed into linear speed, i.e. V = L x ω / 360°, and substituting ω to obtain V = L x N / (P x t). Thus, the motion state information such as the moving position, moving distance and moving speed of the cigarette packet can be acquired by the first encoder.

[0037] Specifically, in the process of cigarette packet production by a cigarette packaging machine through multiple workstations (such as a small packet oil sealing machine workstation, a carton packaging machine workstation and a carton oil sealing machine workstation), the image of the cigarette packet at the workstation in the production process is acquired by an image acquisition device. The target region and the background region in the image can be identified by an image gray threshold segmentation algorithm. The edge of the target region is identified by an edge detection algorithm and matched with a pre-prepared standard cigarette packet template to calculate the similarity. If the similarity is lower than a set threshold, the current cigarette packet is determined as an abnormal cigarette packet.

[0038] In addition, in the process of cigarette packet production by a cigarette packaging machine through multiple workstations (such as a small packet oil sealing machine workstation, a carton packaging machine workstation and a carton oil sealing machine workstation), the moving position of the cigarette packet on the production line can be monitored by at least one first encoder. The position and moving distance of the cigarette packet are calculated according to the output pulse signal of the encoder, and the correspondence between the position and the packaging process is established. The moving speed of the cigarette packet can be calculated by monitoring the output pulse signal of the encoder to ensure that the cigarette packet moves at a predetermined speed. When the speed deviation (such as fluctuation caused by accumulation or sliding of the cigarette packet) or the position deviation exceeds the set range (such as the cigarette packet not moving at the predetermined position), it is determined that there is an abnormality, and the current cigarette packet is determined as an abnormal cigarette packet.

[0039] It can be understood that the method of identifying an abnormal cigarette packet by the first encoder is mainly applicable to packaging problems caused by mechanical motion abnormalities. In order to more accurately identify an abnormal cigarette packet, the acquisition result of the encoder can also be synchronously analyzed with other data of the production line. For example, the image information of the image acquisition device is combined, i.e. if the moving speed of the cigarette packet displayed by the encoder suddenly changes or the moving track is abnormal, or the image shows signs of deformation or damage of the cigarette packet, these information can be comprehensively analyzed to accurately identify an abnormal cigarette packet. At the same time, it can also avoid the missed detection caused by poor image acquisition angle when only using the image acquisition device to identify an abnormal cigarette packet.

[0040] Optionally, based on the above embodiments, during the production of cigarette packets by the cigarette packaging machine through multiple workstations, the image acquisition device and / or the acquisition result of the at least one first encoder are used to identify abnormal cigarette packets in real time during the production process, the point inspection process can be triggered regularly or manually, the point inspection frequency can be flexibly configured, the point inspection can be automatically started, the point inspection process and the identification of abnormalities can be recorded, the missed inspection problem can be eliminated, the labor cost of point inspection can be reduced, and the process quality of the product can be ensured.

[0041] Optionally, based on the above embodiments, the image acquisition device is used to identify abnormal cigarette packets in real time during the production process, which can include:

[0042] The image acquisition device arranged at different detection positions in the cigarette packaging machine acquires the cigarette packet images of the cigarette packets in the actual production process in real time, and inputs the cigarette packet images into the pre-trained image detection model;

[0043] According to the output result of the image detection model, the abnormal cigarette packets are identified.

[0044] Specifically, during the production of cigarette packets by the cigarette packaging machine through multiple workstations (such as a small packet oil sealing machine workstation, a carton packaging machine workstation, and a carton oil sealing machine workstation), the image acquisition device acquires the images of the cigarette packets at the workstations during the production process and transmits them to the pre-trained image detection model. The model uses a pre-trained image recognition algorithm (such as a deep learning algorithm such as a convolutional neural network) to analyze the appearance of the cigarette packets. For example, it can detect whether the trademark of the cigarette packet is complete, whether the color meets the standard, and whether the packaging is tight, etc. When the model detects that the difference between the current cigarette packet and the normal cigarette packet characteristics exceeds the preset threshold, the current cigarette packet is determined to be an abnormal cigarette packet, and the detection result is output as abnormal.

[0045] By arranging the image acquisition device at multiple detection positions of the cigarette packaging machine, the cigarette packet images are acquired in real time and input into the pre-trained image detection model. The model analyzes the input images based on the learned normal cigarette packet characteristics. When the difference between the characteristics of the current cigarette packet exceeds the preset threshold, it is determined to be an abnormal cigarette packet. Real-time monitoring of the quality of the cigarette packets during the production process and automatic identification of abnormalities are achieved. The error of manual detection is reduced, the accuracy and efficiency of detection are improved, and the stability of product quality is ensured.

[0046] Optionally, based on the above embodiments, the acquisition result of the at least one first encoder is used to identify abnormal cigarette packets in real time during the production process, which can include:

[0047] At least one first encoder arranged on the target rotating component is configured to collect at least one level sampling signal of the passing cigarette packet in each detection period, thereby forming a sampling signal set corresponding to each cigarette packet;

[0048] The first encoder is configured to rotate by at least one preset angle in the same detection period to collect at least one level sampling signal of the same cigarette packet.

[0049] The sampling signal set corresponding to each cigarette packet is input into a pre-trained decision rule model.

[0050] According to the output result of the decision rule model, the abnormal cigarette packet is identified.

[0051] In the embodiment of the present application, the detection period can be understood as the time interval for detecting each cigarette packet during the production process of the packaging machine. For example, the packaging machine performs a detection period once for every cigarette packet produced or every preset number of cigarette packets conveyed. The level sampling signal can be understood as a level signal, i.e., a pulse signal, which is converted from mechanical position information by the rotation of the encoder. These level signals reflect the motion state of the rotating component, such as the motion position or motion speed of the cigarette packet.

[0052] The preset angle can be understood as an angle value set according to the mechanical structure and production requirements of the packaging machine. For example, if the encoder is installed on the drive shaft of the conveying belt and the drive shaft needs to rotate by a certain angle (e.g., 10 degrees) when each cigarette packet passes through a certain detection station on the conveying belt, the preset angle can be set to 10 degrees. Correspondingly, the encoder will collect a level sampling signal each time it rotates with the rotating component to reach the preset angle. It can be understood that for the same cigarette packet, multiple level sampling signals thereof need to be collected by the first encoder arranged at different positions in the detection period to comprehensively reflect its motion state.

[0053] The decision rule model can be understood as a machine learning or deep learning model for analyzing the sampling signal set and identifying whether it contains abnormal features. The sampling signal set can be understood as a data set containing multiple level sampling signals of the same cigarette packet at different positions or different time points within one detection period.

[0054] Specifically, at least one first encoder is installed on a target rotating component (such as a driving shaft of a conveyor belt or a rotating component related to a detection station of a cigarette packaging machine) of the cigarette packaging machine, and each encoder installed at different positions collects a level sampling signal of a passing cigarette packet in each detection period to form a sampling signal set corresponding to each cigarette packet. The sampling signal set corresponding to each cigarette packet is input into a pre-trained decision rule model, and the model analyzes the input sampling signal set. If the model determines that the characteristics of the sampling signal set differ from the signal characteristics of a normal cigarette packet by more than a set threshold, the cigarette packet is determined to be an abnormal cigarette packet.

[0055] In a specific example, assuming that the encoder is installed on the driving shaft of the conveyor belt, one cigarette packet is conveyed in each detection period, and the preset angle is 10 degrees. Then, the encoder collects a level signal every 10 degrees of rotation in the detection period, and assumes that a total of 3 times (30 degrees of rotation) are collected to form a set containing 3 level sampling signals. The set is input into the decision rule model, and the model outputs a result after analysis. If the result is determined to be abnormal, the current cigarette packet corresponding to the set is identified as an abnormal cigarette packet.

[0056] The first encoder is installed on the target rotating component of the cigarette packaging machine, and a level sampling signal is collected by the rotation of the first encoder by a preset angle in a detection period to form a sampling signal set corresponding to each cigarette packet. The sampling signal set is input into a pre-trained decision rule model to identify an abnormal cigarette packet. Real-time monitoring and automatic identification of abnormalities of the cigarette packet are achieved, the labor cost is reduced, the missed detection and misjudgment are reduced, the influence of the abnormal cigarette packet on the subsequent production process is reduced, the detection accuracy is improved, and the production efficiency is improved.

[0057] Further, on the basis of each of the above embodiments, before the sampling signal set corresponding to each cigarette packet is input into the pre-trained decision rule model, the following steps can also be included:

[0058] Unsupervised clustering is performed on all historical detection period level sampling signals to obtain at least one clustered signal cluster;

[0059] Potential defect clusters are screened and marked in the clustered signal cluster, the clustered signal cluster and the marking information are input into a supervised model for training, the supervised model is iteratively optimized through a semi-supervised learning mechanism until a preset convergence condition is reached, and a pre-trained decision rule model is obtained.

[0060] Specifically, the sampling signal set corresponding to each cigarette pack can be first converted into a feature vector of uniform dimension (such as by a data processing method such as standardization or normalization), to ensure that all feature vectors have the same scale and format, facilitating the clustering algorithm to process the feature vectors of all cigarette packs. An unsupervised learning algorithm (such as a K-means clustering algorithm or a density-based spatial clustering of applications with noise algorithm) is used to cluster the level sampling signals collected in all detection periods. After processing by the unsupervised clustering algorithm, at least one cluster of signal is obtained. In the obtained cluster of signal, an outlying cluster or a small cluster can be manually screened according to business knowledge and the characteristics of the clustering results and marked as a potential defect cluster, or an anomaly score is calculated for each cluster, such as a distance based on a cluster center or a consistency based on samples within a cluster, and clusters with an anomaly score exceeding a preset anomaly threshold are screened and marked as potential defect clusters. The marked cluster of signal and its marked information are input to a supervised learning model (such as a support vector machine, a random forest, an extreme gradient boosting, or a neural network) for training.

[0061] In the training process, a semi-supervised learning mechanism (such as label propagation and confident learning) can be used to iteratively optimize the supervised model. The label propagation algorithm initializes a small amount of labeled data and iteratively updates the labels of unlabeled data points by calculating the similarity between data points and simulating the natural diffusion of labels between data points until convergence. The confident learning algorithm estimates the confidence of unlabeled data points and selects high-confidence samples for labeling, and then iteratively trains the model, so that the supervised model can more fully utilize all available data. The model is iteratively optimized by cross-validation and adjustment of hyperparameters, until the model meets the preset convergence conditions, such as the accuracy reaching a preset threshold or reaching a maximum number of iterations, to obtain a pre-trained decision rule model.

[0062] By using an unsupervised clustering method to automatically cluster the level sampling signals collected in all detection periods, potential defect clusters are screened and marked in the cluster of signal, and these information is input to a supervised model for training, and the model is iteratively optimized in combination with a semi-supervised learning mechanism until the preset convergence conditions are met, to obtain a pre-trained decision rule model, which enhances the generalization ability of the model, makes it better adapt to different production conditions and abnormal types, and improves the accuracy, point inspection efficiency, and point inspection automation level of detecting abnormal cigarette packs.

[0063] Further, on the basis of the above embodiments, before unsupervised clustering of the level sampling signals of all detection cycles collected historically, preprocessing of the level sampling signals of all detection cycles collected historically can also be included, which can specifically include processing missing values and outliers. Among them, missing values can be processed by deleting missing values, filling in mean values, medians, modes, or using interpolation methods, etc. Outliers can be identified by statistical analysis (such as standard deviation or interquartile range, etc.) or machine learning algorithms (such as Isolation Forest), and replaced by deleting, filling in mean values, medians or modes. Through preprocessing, unsupervised clustering analysis can more accurately identify the feature patterns of normal and abnormal cigarette packets from the data and perform clustering, providing a data basis for defect detection and quality control, and helping to improve the automation level of the production process and product quality.

[0064] S120, when it is determined that the abnormal cigarette packet is identified, tracking a target moving state of the abnormal cigarette packet when moving to a rejection station of the cigarette packaging machine in real time.

[0065] In the embodiments of the present application, the target moving state can be specifically understood as the motion state of the abnormal cigarette packet when it reaches the rejection station, such as the position, speed and motion direction of the cigarette packet, etc.

[0066] Specifically, at each detection point of the cigarette packaging machine, the image acquisition device captures the cigarette packet image in real time, and transmits the image to the image detection model. At the same time, the sensor (such as a photoelectric sensor or a proximity sensor, etc.) installed at each detection point or the encoder corresponding to each monitoring point monitors the arrival and passing of the cigarette packet. The pre-trained image detection model analyzes the input cigarette packet image, and when the model detects that the cigarette packet is abnormal, a defect signal can be generated, indicating that the current cigarette packet is identified as an abnormal cigarette packet. After generating the defect signal, the system can mark the abnormal cigarette packet according to the detection time point from the database or the log file and other data recording modules. The marking information can include the specific time of detecting the abnormality, the position of the detection station, the abnormal type code or the identifier of the cigarette packet, etc.

[0067] Correspondingly, as the production process proceeds, when the cigarette packet passes through the image acquisition device of the subsequent station, the system can continuously track the generated defect signal and the corresponding marking to identify the motion trajectory of the cigarette packet, and determine whether the abnormal cigarette packet moves to the rejection station of the cigarette packaging machine and whether it is in the target moving state according to the motion trajectory of the cigarette packet.

[0068] The image acquisition device captures the cigarette package image in real time at each detection station of the cigarette packaging machine and transmits the image to the image detection model, and cooperates with the sensor or encoder to monitor the arrival and passing of the cigarette package. When the model detects an abnormal cigarette package, a defect signal is generated, the abnormal cigarette package is marked, and the defect signal is continuously tracked to identify the movement track of the cigarette package in the subsequent production process. Whether the cigarette package is in a target moving state is determined according to the track of the cigarette package, thereby improving the accuracy and efficiency of the point inspection and enhancing the quality control and point inspection abnormality tracing capability.

[0069] S130, when it is determined that the abnormal cigarette package is in the target moving state and is not separated by the rejection station as unqualified products, a point inspection abnormality prompt matched with the abnormal cigarette package is generated.

[0070] Specifically, when the abnormal cigarette package reaches the rejection station, the system performs a rejection operation. If the sensor of the rejection station does not detect that the cigarette package is successfully rejected within a preset rejection time, that is, the abnormal cigarette package is still on the production line, the point inspection abnormality is confirmed, and a point inspection abnormality prompt matched with the abnormal cigarette package is generated. Specifically, the point inspection abnormality prompt can include identification information of the abnormal cigarette package, time or position of the abnormality, etc.

[0071] Further, based on the above embodiments, the method for point inspection of the cigarette packaging machine detection device can further include:

[0072] Obtaining test samples from the historical defect collection results and qualified collection results, and combining the obtained test samples to construct a test sample set;

[0073] Verifying the performance of the image detection model based on the test sample set, and generating a point inspection abnormality prompt when the performance of the image detection model does not meet a preset performance condition.

[0074] Specifically, the defect collection results (i.e., samples marked as abnormal) and the qualified collection results (i.e., samples marked as normal) are extracted from the historical data as test samples according to a preset proportion or quantity, and the extracted test samples are combined to construct a test sample set. The test sample set data is input into the image detection model to evaluate the identification accuracy and recall rate of the model for defects and qualified cigarette packages. If the performance indicators of the model do not meet the preset indicator threshold, it indicates that the model has problems or needs to be further optimized, and the system generates a point inspection abnormality prompt to notify relevant personnel to adjust or retrain the model.

[0075] By obtaining test samples from historically constructed defects and qualified acquisition results, and combining them to construct a test sample set, the performance of the image detection model is verified. When the model performance does not meet the preset standards, the system will automatically generate an inspection anomaly prompt, which can promptly discover and correct image detection model performance deviations, ensuring the reliability and stability of the image detection model and strengthening quality control in the production process.

[0076] The technical solution of the embodiments of the present invention uses image acquisition devices and / or at least one first encoder to identify abnormal cigarette packets in real time during the production process of cigarette packs at multiple stations. Upon identifying an abnormal cigarette packet, the system tracks the target movement state of the abnormal cigarette packet as it moves to the cigarette packer's rejection station in real time. If the abnormal cigarette packet is determined to be in the target movement state but not separated as a defective product at the rejection station, a spot inspection abnormality prompt matching the abnormal cigarette packet is generated. Compared with traditional manual spot inspection, this system reduces machine downtime, production defects, and low-speed tracking of defective samples through inspection and rejection ports, avoiding defects that can be blurred by human perception and improving spot inspection accuracy. Through trajectory tracking and rejection verification mechanisms, production monitoring and quality control are strengthened, improving spot inspection efficiency and product quality. Instant prompts associated with specific abnormal cigarette packets help operators quickly locate problem nodes, shorten troubleshooting time, reduce production line downtime, and enhance production continuity and the intelligence level of cigarette packers.

[0077] Example 2

[0078] Figure 2 A flow chart of another method for spot inspection of detection equipment of a cigarette packaging machine provided in Example 2 of the present invention. This embodiment is a refinement of the "real-time tracking of the target movement state of the abnormal cigarette package when it moves to the rejection station of the cigarette packaging machine when it is determined that an abnormal cigarette package is identified" in the above embodiment. Specifically, it may include: when it is determined that the abnormal cigarette package is identified, obtaining the abnormal detection position and detection time point of the abnormal cigarette package; tracking the real-time movement trajectory of the abnormal cigarette package from the abnormal detection position in real time through the real-time detection results of the detection sensors set at the cigarette package entry position of each station, the abnormal detection position and detection time point of the abnormal cigarette package; when it is determined that the detection sensor at the cigarette package entry position of the rejection station generates an entry signal for the abnormal cigarette package, determining that the abnormal cigarette package is in the target movement state.

[0079] Correspondingly, such as Figure 2 As shown, the method includes:

[0080] S210, in the process of cigarette pack production by the cigarette packaging machine through multiple workstations, through the acquisition results of the image acquisition device and / or at least one first encoder, the abnormal cigarette packs appearing in the production process are identified in real time.

[0081] S220, when it is determined that the abnormal cigarette pack is identified, the abnormal detection position and detection time point of the abnormal cigarette pack are acquired.

[0082] Specifically, when it is determined that the abnormal cigarette pack is identified, the specific position where the abnormal cigarette pack is detected, i.e., the abnormal detection position, and the corresponding abnormal detection time point are confirmed by the encoder or the sensor.

[0083] S230, through the real-time detection results of the detection sensor arranged at the cigarette pack entering position of each workstation, the abnormal detection position and detection time point of the abnormal cigarette pack, the real-time moving track of the abnormal cigarette pack from the abnormal detection position is tracked in real time.

[0084] Specifically, through the detection sensor arranged at each workstation of the production line, the system can acquire the position information of the cigarette pack in real time, and through the real-time detection results and the abnormal detection position and detection time point of the abnormal cigarette pack, the real-time moving track of the abnormal cigarette pack from the abnormal detection position in the production process can be constructed.

[0085] S240, when it is determined that the detection sensor at the cigarette pack entering position of the rejection workstation generates an entering signal for the abnormal cigarette pack, it is determined that the abnormal cigarette pack is in the target moving state.

[0086] Specifically, when the abnormal cigarette pack reaches the cigarette pack entering position of the rejection workstation, the detection sensor at this position will generate an entering signal, indicating that the abnormal cigarette pack has reached the predetermined position and is in a state that can be rejected, i.e., the target moving state.

[0087] S250, when it is determined that the abnormal cigarette pack is in the target moving state and is not separated by the rejection workstation as unqualified products, a point inspection abnormality prompt matched with the abnormal cigarette pack is generated.

[0088] The technical scheme of the embodiment of the present application, in the process of cigarette packaging machine producing cigarette packets through multiple stations, the image acquisition device and / or the acquisition result of at least one first encoder are used to identify the abnormal cigarette packet in real time in the production process; when it is determined that the abnormal cigarette packet is identified, the abnormal detection position and detection time point of the abnormal cigarette packet are obtained; through the real-time detection result of the detection sensor arranged at the cigarette packet entering position of each station, the abnormal detection position and detection time point of the abnormal cigarette packet, the real-time moving track of the abnormal cigarette packet from the abnormal detection position is tracked in real time; when it is determined that the detection sensor at the cigarette packet entering position of the rejection station generates an entering signal for the abnormal cigarette packet, it is determined that the abnormal cigarette packet is in the target moving state, the automation level and accuracy of abnormal cigarette packet identification and tracking are improved, manual errors are reduced, and the point inspection reliability is enhanced; when it is determined that the abnormal cigarette packet is in the target moving state and is not separated by the rejection station as unqualified products, a point inspection abnormality prompt matched with the abnormal cigarette packet is generated. Compared with the traditional manual point inspection, the stop, defect making, and low-speed tracking defect sample passing through the detection port and the rejection port and other operations are reduced, human perception of fuzzy defects is avoided, and the point inspection accuracy is improved. Through the track tracking and rejection verification mechanism, the production monitoring and quality control are strengthened, the point inspection efficiency and product quality are improved, the instant prompt associated with the specific abnormal cigarette packet helps the operator quickly locate the problem node, shortens the fault handling time, reduces the production line stagnation, and improves the production continuity and the intelligent level of the cigarette packaging machine.

[0089] Embodiment three

[0090] Figure 3 The flowchart of the method for point inspection of the cigarette packaging machine detection equipment provided in Embodiment Three of the present application, this embodiment is a refinement of the above-mentioned embodiment "when it is determined that the abnormal cigarette packet is identified, the real-time moving track of the abnormal cigarette packet moving to the rejection station of the cigarette packaging machine is tracked", specifically can include: when it is determined that the abnormal cigarette packet is identified, the moving sampling signal of the abnormal cigarette packet in the process of moving to the rejection station of the cigarette packaging machine is acquired in real time through at least one second encoder, and the moving sampling array is updated according to the moving sampling signal; when it is confirmed that the abnormal cigarette packet moves to the rejection station based on the element value corresponding to the rejection station in the moving sampling array, it is determined that the abnormal cigarette packet is in the target moving state.

[0091] Correspondingly, as shown in Figure 3 , the method comprises:

[0092] S310, in the process of cigarette packaging machine producing cigarette packets through multiple stations, the image acquisition device and / or the acquisition result of at least one first encoder are used to identify the abnormal cigarette packet in real time in the production process.

[0093] S320, when determining that the abnormal cigarette packet is identified, acquiring a movement sampling signal in a movement process of the abnormal cigarette packet to a rejection station of a cigarette packaging machine in real time through at least one second encoder, and updating a movement sampling array according to the movement sampling signal.

[0094] In the embodiments of the present application, the second encoder can be specifically understood as an encoder for acquiring a sampling signal in a movement process of a cigarette packet to a rejection station in real time, and updating a movement sampling array. The movement sampling array can be specifically understood as a data structure for storing signals collected by the encoder at different positions. Each element in the array corresponds to a specific detection position in the movement process of the cigarette packet. Each movement sampling array corresponds to each cigarette packet.

[0095] Specifically, when the abnormal cigarette packet is identified, the second encoder acquires a sampling signal in a movement process of the abnormal cigarette packet from the position where the abnormal cigarette packet is detected in real time, and updates a corresponding element value in the movement sampling array in real time according to the movement sampling signal collected by the encoder. Optionally, in the movement sampling array, an element value of 1 indicates that the state of the cigarette packet at the position is normal, and 0 indicates that the state is abnormal.

[0096] Optionally, on the basis of each of the above embodiments, the second encoder is arranged on a target rotating component of a small packet oil sealing machine, a subsequent conveying mechanism of the small packet oil sealing machine, a carton packaging machine, and a carton oil sealing machine.

[0097] The second encoder arranged on the target rotating component of the small packet oil sealing machine generates a movement sampling signal every first preset angle of rotation.

[0098] The second encoder arranged on the target rotating component of the subsequent conveying mechanism of the small packet oil sealing machine generates a movement sampling signal every second preset angle of rotation.

[0099] The second encoder arranged on the target rotating component of the carton packaging machine or the carton oil sealing machine generates a movement sampling signal every third preset angle of rotation.

[0100] Specifically, the second encoder is arranged on the rotating part of the small package oil sealing machine, used for monitoring the movement of the cigarette package during the oil sealing process, and the encoder generates a movement sampling signal every first preset angle (e.g. one revolution), indicating that the cigarette package has moved one position; arranged on the rotating part of the conveying mechanism after the oil sealing, used for monitoring the movement of the cigarette package during the conveying process from the stacking to the entering of the carton packaging machine, and the encoder generates a movement sampling signal every second preset angle (e.g. two revolutions), indicating that the cigarette package has moved one position; arranged on the rotating part of the carton packaging machine, used for monitoring the packaging process of the carton, and the encoder generates a movement sampling signal every third preset angle (e.g. ten revolutions), indicating that the carton has moved one position; arranged on the rotating part of the carton oil sealing machine, used for monitoring the oil sealing process of the carton, and the encoder generates a movement sampling signal every third preset angle (e.g. ten revolutions), indicating that the carton has moved one position.

[0101] By arranging the second encoder on the target rotating part of the small package oil sealing machine, the conveying mechanism after the small package oil sealing machine, the carton packaging machine and the carton oil sealing machine, and generating the movement sampling signal by the encoder according to the preset angle, the accurate monitoring of the movement of the cigarette package and the carton at each step during the production process can be realized, the accuracy and reliability of the oil sealing process, the conveying process of the cigarette package and the packaging and oil sealing process of the carton are improved, the production efficiency is improved, the quality control is enhanced, and the automation level of the production line is improved.

[0102] S330, when it is confirmed that the abnormal cigarette package moves to the rejection station based on the element value in the movement sampling array corresponding to the rejection station, it is determined that the abnormal cigarette package is in the target movement state.

[0103] Specifically, when the cigarette package approaches the rejection station, the system checks the element value in the movement sampling array corresponding to the rejection station. If the value represents an exception, it is confirmed that the cigarette package has moved to the rejection station and is in the target movement state.

[0104] In a specific example, each element in the moving sampling array corresponds to a specific station or position, and each cigarette packet corresponds to a moving sampling array. For example, the array [1, 1, 1, 1, 1] indicates that the cigarette packet is in a normal state at all stations (1 represents normal and 0 represents abnormal). Assuming that the five stations are a small packet oil sealing machine, a small packet oil sealing machine subsequent conveying mechanism, a carton packaging machine, a carton oil sealing machine, and a rejection station. Assuming that the small packet oil sealing machine station identifies an abnormal cigarette packet, the corresponding element will be updated to 0, that is, [0, 1, 1, 1, 1]. When the cigarette packet moves from one station to the next station, the elements in the array will change accordingly to reflect the new position and state of the cigarette packet. When the abnormal cigarette packet moves to the small packet oil sealing machine subsequent conveying mechanism, the moving sampling array is updated to [1, 0, 1, 1, 1]. By analogy, when the moving sampling array is updated to [1, 1, 1, 1, 0], the last position becomes 0, indicating that the cigarette packet is at the rejection station and is in the target moving state.

[0105] S340, generating an inspection exception prompt matched with the abnormal cigarette packet when it is determined that the abnormal cigarette packet is in the target moving state and is not separated by the rejection station as unqualified products.

[0106] The technical scheme of the embodiment of the present application, in the process of cigarette packaging machine through multiple stations for cigarette packet production, through the collection results of the image acquisition device and / or at least one first encoder, real-time identification of the abnormal cigarette packet appearing in the production process; when it is determined that the abnormal cigarette packet is identified, at least one second encoder is used to acquire the moving sampling signal of the abnormal cigarette packet in the process of moving to the rejection station of the cigarette packaging machine in real time, and the moving sampling array is updated according to the moving sampling signal; when it is confirmed that the abnormal cigarette packet moves to the rejection station based on the element value corresponding to the rejection station in the moving sampling array, it is determined that the abnormal cigarette packet is in the target moving state, the accuracy of moving information capture is ensured by updating the moving sampling array in real time, the accuracy and real-time of the abnormal cigarette packet moving track monitoring are improved, a reliable basis is provided for the rejection operation, the reliability of the inspection is improved, dynamic tracking and data management of the production process are realized, data support is provided for optimizing the production process; when it is determined that the abnormal cigarette packet is in the target moving state and is not separated by the rejection station as unqualified products, an inspection exception prompt matched with the abnormal cigarette packet is generated. Compared with traditional manual inspection, the operation of stopping, making defects, and low-speed tracking defect samples through the detection port and the rejection port is reduced, human perception of fuzzy defects is avoided, and the inspection accuracy is improved. Through the track tracking and rejection verification mechanism, the production monitoring and quality control are strengthened, the inspection efficiency and product quality are improved, and with the instant prompt associated with the specific abnormal cigarette packet, the operator can quickly locate the problem node, shorten the fault handling time, reduce the production line stagnation, and improve the production continuity and the intelligent level of the cigarette packaging machine.

[0107] Embodiment Four

[0108] Figure 4 A structural schematic diagram of a device for point inspection of a cigarette packaging machine detection equipment according to Embodiment Four of the present application is shown in FIG. 4. As shown in FIG. 4, the device includes a collection module 410, a tracking module 420, and a prompt module 430, wherein: Figure 4

[0109] The collection module 410 is configured to identify abnormal cigarette packets in real time during the production process of the cigarette packets by the cigarette packaging machine through a plurality of workstations through the collection results of the image collection device and / or at least one first encoder.

[0110] The tracking module 420 is configured to track the target movement state of the abnormal cigarette packets when moving to the rejection workstation of the cigarette packaging machine in real time when it is determined that the abnormal cigarette packets are identified.

[0111] The prompt module 430 is configured to generate a point inspection abnormality prompt matched with the abnormal cigarette packets when it is determined that the abnormal cigarette packets are in the target movement state and are not separated by the rejection workstation as unqualified products.

[0112] The technical solution of the embodiment of the present application identifies abnormal cigarette packets in real time during the production process of the cigarette packets by the cigarette packaging machine through a plurality of workstations through the collection results of the image collection device and / or at least one first encoder. The target movement state of the abnormal cigarette packets when moving to the rejection workstation of the cigarette packaging machine is tracked in real time when it is determined that the abnormal cigarette packets are identified. A point inspection abnormality prompt matched with the abnormal cigarette packets is generated when it is determined that the abnormal cigarette packets are in the target movement state and are not separated by the rejection workstation as unqualified products. Compared with the traditional manual point inspection, the present application reduces the operations such as shutdown, defect making, and low-speed tracking of the defect sample through the detection port and the rejection port, avoids the fuzzy defects perceived by humans, and improves the point inspection accuracy. Through the trajectory tracking and rejection verification mechanism, the production monitoring and quality control are strengthened, the point inspection efficiency and product quality are improved, the instant prompt associated with the specific abnormal cigarette packets helps the operator quickly locate the problem node, shortens the fault handling time, reduces the production line stagnation, and improves the production continuity and the intelligent level of the cigarette packaging machine.

[0113] On the basis of the above embodiments, the collection module 410 is specifically configured to:

[0114] collect the cigarette packet images of each cigarette packet in the actual production process in real time through each image collection device arranged at different detection positions in the cigarette packaging machine, and input each cigarette packet image into a pre-trained image detection model;

[0115] ​According to an output result of the image detection model, the abnormal cigarette packet is identified.

[0116] On the basis of the above embodiments, the collection module 410 is further used for:

[0117] At least one level sampling signal of the passing cigarette packet is collected in each detection period through at least one first encoder arranged on the target rotating component, to form a sampling signal set corresponding to each cigarette packet respectively;

[0118] The first encoder rotates at least one preset angle in the same detection period to sample at least one level sampling signal of the same cigarette packet;

[0119] The sampling signal set corresponding to each cigarette packet respectively is input into a pre-trained decision rule model;

[0120] According to an output result of the decision rule model, the abnormal cigarette packet is identified.

[0121] On the basis of the above embodiments, the tracking module 420 is specifically used for:

[0122] When it is determined that the abnormal cigarette packet is identified, an abnormal detection position and a detection time point of the abnormal cigarette packet are acquired;

[0123] The real-time movement track of the abnormal cigarette packet from the abnormal detection position is tracked in real time through the real-time detection result of the detection sensor arranged at the cigarette entering position of each station, the abnormal detection position and the detection time point of the abnormal cigarette packet;

[0124] When it is determined that the detection sensor at the cigarette entering position of the rejection station generates an entering signal for the abnormal cigarette packet, it is determined that the abnormal cigarette packet is in the target movement state.

[0125] On the basis of the above embodiments, the tracking module 420 is further used for:

[0126] When it is determined that the abnormal cigarette packet is identified, the movement sampling signal of the abnormal cigarette packet in the process of moving to the rejection station of the cigarette packaging machine is acquired in real time through at least one second encoder, and the movement sampling array is updated according to the movement sampling signal;

[0127] When it is determined that the abnormal cigarette packet moves to the rejection station based on the element value in the movement sampling array corresponding to the rejection station, it is determined that the abnormal cigarette packet is in the target movement state.

[0128] Further, on the basis of the above embodiments, the device for point inspection of the cigarette packaging machine detection equipment can further include a set construction module and a model detection module, wherein:

[0129] The construction set module is configured to acquire test samples from defect collection results and qualified collection results of historical constructions, and combine the acquired test samples to construct a test sample set.

[0130] The model detection module is configured to verify the performance of the image detection model based on the test sample set, and generate a point inspection abnormality prompt when the performance of the image detection model does not satisfy a preset performance condition.

[0131] Optionally, based on the above embodiments, the collection module 410 can include a clustering unit and a training unit, wherein:

[0132] The clustering unit is configured to perform unsupervised clustering on all level sampling signals of historical collection detection periods to obtain at least one clustering signal cluster before inputting the sampling signal set corresponding to each cigarette packet into the pre-trained decision rule model.

[0133] The training unit is configured to filter and mark a potential defect cluster in the clustering signal cluster, input the clustering signal cluster and its marking information into a supervised model for training, and iteratively optimize the supervised model through a semi-supervised learning mechanism until a preset convergence condition is reached to obtain the pre-trained decision rule model.

[0134] The device for point inspection of the cigarette packaging machine detection equipment provided in the embodiments of the present application can perform the method for point inspection of the cigarette packaging machine detection equipment provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0135] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information comply with relevant laws and regulations and do not violate public order and good customs.

[0136] Embodiment five

[0137] Figure 5 A structural schematic diagram of an electronic device 10 that can be used to implement embodiments of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smart phones, wearable devices (e.g., headsets, glasses, watches, etc.), and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present application described and / or claimed in this document.

[0138] As Figure 5As shown, the electronic device 10 includes at least one processor 11, and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11, wherein the memory stores a computer program that can be executed by the at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0139] A plurality of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0140] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs various methods and processes described above, such as the method of detecting the cigarette packaging machine, that is:

[0141] In the process of the cigarette packaging machine producing cigarette packets through a plurality of stations, the abnormal cigarette packets appearing in the production process are identified in real time through the acquisition results of the image acquisition device and / or at least one first encoder;

[0142] When it is determined that the abnormal cigarette packet is identified, the target movement state of the abnormal cigarette packet when moving to the rejection station of the cigarette packaging machine is tracked in real time;

[0143] When it is determined that the abnormal cigarette packet is in the target movement state and is not separated by the rejection station as unqualified products, a point inspection abnormality prompt matched with the abnormal cigarette packet is generated.

[0144] In some embodiments, the method of cigarette packaging machine inspection equipment point inspection can be implemented as a computer program tangibly embodied in a computer readable storage medium, e.g., storage unit 18. In some embodiments, parts or all of the computer program can be loaded and / or installed onto electronic device 10 via, e.g., ROM 12 and / or communication unit 19. When the computer program is loaded onto RAM 13 and executed by processor 11, one or more steps of the method of cigarette packaging machine inspection equipment point inspection described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the method of cigarette packaging machine inspection equipment point inspection by way of other any suitable means (e.g., by way of firmware).

[0145] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0146] Computer programs used to implement the methods of the present application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program running on the processor implements the functions / operations specified in the flowcharts and / or the block diagrams. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine as part of a standalone software package, or entirely on a remote machine or server.

[0147] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer-readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium will include one or more lines of a program of instructions in a transitory signal, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0148] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0149] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0150] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

[0151] It should be understood that the various forms of flow shown above can be used to reorder, add or delete steps. For example, each step described in the present application can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solutions of the present application can be achieved, which is not limited herein.

[0152] The above detailed description does not constitute a limitation on the scope of protection of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for spot inspection of cigarette packaging machine detection equipment, characterized in that: include: During the cigarette packaging machine's production of cigarette packets at multiple stations, abnormal cigarette packets occurring during the production process are identified in real time using the image acquisition device and / or the acquisition results of at least one first encoder; When an abnormal cigarette pack is identified, real-time tracking of the target movement state of the abnormal cigarette pack when it moves to the rejection station of the cigarette packaging machine; When it is determined that the abnormal cigarette pack is in the target moving state but has not been separated as a non-conforming product by the rejection station, an inspection abnormality prompt matching the abnormal cigarette pack is generated.

2. The method according to claim 1, characterized in that Through the collection results of the image acquisition device, abnormal cigarette packs appearing in the production process are identified in real time, including: The image acquisition devices arranged at different detection positions in the cigarette packaging machine are used to collect images of cigarette packs in real time during the actual production process, and the images of the cigarette packs are input into a pre-trained image detection model; Abnormal cigarette packets are identified based on the output results of the image detection model.

3. The method according to claim 1, characterized in that Real-time identification of abnormal cigarette packets occurring during the production process is performed based on the collection results of at least one first encoder, including: At least one first encoder provided on the target rotating component collects at least one level sampling signal from a passing cigarette packet in each detection cycle to form a sampling signal set corresponding to each cigarette packet; The first encoder samples at least one level sampling signal for the same cigarette pack by rotating at least one preset angle within the same detection period; Inputting the sampled signal set corresponding to each cigarette pack into the pre-trained decision rule model; Abnormal cigarette packets are identified according to the output results of the determination rule model.

4. The method according to any one of claims 1 to 3, characterized in that When an abnormal cigarette pack is identified, the target movement state of the abnormal cigarette pack when it moves to the rejection station of the cigarette packaging machine is tracked in real time, including: When it is determined that the abnormal cigarette packet is identified, obtaining the abnormal detection position and detection time point of the abnormal cigarette packet; By using the real-time detection results of the detection sensors set at the cigarette pack entry position of each station, the abnormal detection position and detection time point of the abnormal cigarette pack, and the real-time movement trajectory of the abnormal cigarette pack from the abnormal detection position; When it is determined that the detection sensor at the cigarette pack entry position of the rejecting station generates an entry signal for the abnormal cigarette pack, it is determined that the abnormal cigarette pack is in the target movement state.

5. The method according to any one of claims 1 to 3, characterized in that When an abnormal cigarette pack is identified, the target movement state of the abnormal cigarette pack when it moves to the rejection station of the cigarette packaging machine is tracked in real time, including: When it is determined that the abnormal cigarette pack is identified, a movement sampling signal of the abnormal cigarette pack during its movement to the rejection station of the cigarette packaging machine is acquired in real time by at least one second encoder, and a movement sampling array is updated according to the movement sampling signal; When it is confirmed that the abnormal cigarette packet has moved to the rejection station based on the element value corresponding to the rejection station in the movement sampling array, it is determined that the abnormal cigarette packet is in the target movement state.

6. The method according to claim 2, characterized in that The method for spot checking of cigarette packaging machine detection equipment also includes: Acquire test samples from defect collection results and qualified collection results of historical constructions, and combine the acquired test samples to construct a test sample set; The performance of the image detection model is verified based on the test sample set. When the performance of the image detection model does not meet the preset performance conditions, an inspection anomaly prompt is generated.

7. The method according to claim 3, characterized in that Before inputting the sampled signal set corresponding to each cigarette pack into the pre-trained decision rule model, the method further includes: Performing unsupervised clustering on the level sampling signals of all detection cycles collected historically to obtain at least one clustered signal cluster; Potential defect clusters are screened and marked in the clustered signal clusters, and the clustered signal clusters and their marking information are input into the supervised model for training. The supervised model is iteratively optimized through a semi-supervised learning mechanism until the preset convergence conditions are reached, thereby obtaining a pre-trained judgment rule model.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for spot inspection of cigarette packaging machine detection equipment according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for spot inspection of cigarette packaging machine detection equipment according to any one of claims 1 to 7 when executed.

10. A computer program product, characterized in that The computer program product comprises a computer program, which, when executed by a processor, implements the method for spot inspection of cigarette packaging machine detection equipment according to any one of claims 1 to 7.

Images