Cigarette box transparent paper packaging quality detection method, device, equipment and medium

Abstract

The embodiment of the application discloses a cigarette box transparent paper packaging quality detection method, device, equipment and medium, the method comprises the following steps: obtaining the waveform signal corresponding to the cigarette box through the photoelectric dynamic scanning module, and utilizing the waveform signal to carry out defect detection, obtaining the photoelectric characteristic and the first detection result corresponding to the cigarette box; obtaining the appearance image corresponding to the cigarette box through the machine vision detection module, and utilizing the appearance image to carry out defect identification, obtaining the defect description and the second detection result corresponding to the cigarette box; when the first detection result and the second detection result satisfy the preset accurate detection condition, cross verification and comprehensive analysis are carried out based on the photoelectric characteristic and the defect description, and the third detection result corresponding to the cigarette box is determined; based on the third detection result and the historical packaging data corresponding to the cigarette box, the target detection result corresponding to the cigarette box is determined. Through the technical scheme of the embodiment of the application, the accuracy of the packaging quality detection is improved, and the product quality is further improved.

Description

Technical Field

[0001] This invention relates to the field of quality control technology for cigarette packaging equipment, and in particular to a method, apparatus, equipment, and medium for quality testing of transparent paper packaging for cigarette boxes. Background Technology

[0002] The quality of the cellophane wrapping in cigarette boxes directly affects the product's appearance and anti-counterfeiting performance. Therefore, quality inspection is necessary after the cellophane wrapping is completed, and any defects should be promptly removed. FXS series packaging machines often experience defects such as loose cellophane wrapping, skewing, misaligned stitching, and exposed edges during high-speed operation.

[0003] Currently, the traditional methods for detecting the quality of transparent paper packaging mainly include the following two types: (1) Mechanical contact detection: The transparent paper is sensed by mechanical structures such as levers and springs. However, this method cannot detect non-contact defects such as offset or missing parts, and it is prone to wear and tear and has a high false alarm rate. (2) Static visual detection: Static images are captured by industrial cameras for analysis. However, this method is greatly affected by lighting and the posture of the cigarette pack, and it is difficult to adapt to the dynamic detection requirements of high-speed production lines, and cannot guarantee the accuracy of detection. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, and medium for quality inspection of transparent paper packaging for cigarette boxes, which can accurately and conveniently inspect the quality of transparent paper packaging for cigarette boxes, improve the accuracy of packaging quality inspection, and further improve product quality.

[0005] In a first aspect, embodiments of the present invention provide a method for quality inspection of transparent paper packaging for cigarette boxes, comprising:

[0006] The photoelectric dynamic scanning module performs photoelectric scanning on the cigarette box after it is wrapped in transparent paper to obtain the waveform signal corresponding to the cigarette box. The waveform signal is then used to perform defect detection to obtain the photoelectric features and the first detection result corresponding to the cigarette box.

[0007] The cigarette box is image acquired by the machine vision inspection module to obtain the appearance image of the cigarette box, and the appearance image is used to identify defects to obtain the defect description and the second inspection result of the cigarette box.

[0008] When the first detection result and the second detection result meet the preset fine inspection conditions, cross-validation and comprehensive analysis are performed based on the photoelectric features and the defect description to determine the third detection result corresponding to the cigarette box.

[0009] Based on the third detection result and the historical packaging data corresponding to the cigarette box, the target detection result corresponding to the cigarette box is determined; wherein, the historical packaging data is the historical packaging data corresponding to the equipment that performs transparent paper packaging operation on the cigarette box.

[0010] Optionally, the method further includes: when the first detection result and the second detection result meet the first rejection condition, performing different types of rejection operations on the cigarette box based on the defect type and defect level corresponding to the cigarette box; and releasing the cigarette box when the first detection result and the second detection result meet the first release condition.

[0011] Optionally, the method further includes: when the target detection result meets the second rejection condition, performing different types of rejection operations on the cigarette box based on the defect type and defect level corresponding to the cigarette box; when the target detection result meets the second release condition, releasing the cigarette box; when the target detection result meets the preset equipment failure condition, generating an early warning instruction corresponding to the equipment performing transparent paper packaging operation on the cigarette box, so as to maintain the equipment.

[0012] Optionally, the method further includes: the photoelectric dynamic scanning module, comprising: multiple sets of reflective photoelectric sensor arrays arranged along the cigarette box conveying direction; the photoelectric dynamic scanning module is used to continuously and dynamically scan the edge and surface of the transparent paper; the reflective photoelectric sensor array comprises: four sets of photoelectric sensors; each set of photoelectric sensors comprises: a transmitter and a receiver.

[0013] Optionally, the method further includes: the machine vision inspection module, comprising: at least two industrial cameras, an adjustable light source, and an image processing unit; the machine vision inspection module is used to acquire multi-angle appearance images of cigarette boxes and perform defect identification on the appearance images; the at least two industrial cameras are arranged at an angle; the adjustable light source is an LED diffuse reflection light source with adjustable brightness and angle.

[0014] Optionally, the method further includes: periodically and adaptively adjusting the photoelectric signal threshold in the photoelectric dynamic scanning module and the detection model parameters in the machine vision detection module without stopping the machine.

[0015] Secondly, embodiments of the present invention provide a quality inspection device for transparent paper packaging of cigarette boxes, comprising:

[0016] The first detection result determination module is used to perform photoelectric scanning on the cigarette box packaged in transparent paper through the photoelectric dynamic scanning module, obtain the waveform signal corresponding to the cigarette box, and use the waveform signal to perform defect detection, thereby obtaining the photoelectric features and the first detection result corresponding to the cigarette box.

[0017] The second detection result determination module is used to acquire images of the cigarette box through the machine vision detection module, obtain the appearance image of the cigarette box, and use the appearance image to identify defects, thereby obtaining a defect description and a second detection result for the cigarette box.

[0018] The third detection result determination module is used to determine the third detection result corresponding to the cigarette box by performing cross-validation and comprehensive analysis based on the photoelectric features and the defect description when the first detection result and the second detection result meet the preset fine inspection conditions.

[0019] The target detection result determination module is used to determine the target detection result corresponding to the cigarette box based on the third detection result and the historical packaging data corresponding to the cigarette box; wherein, the historical packaging data is the historical packaging data corresponding to the equipment that performs transparent paper packaging operation on the cigarette box.

[0020] Thirdly, embodiments of the present invention provide an electronic device, the electronic device comprising:

[0021] One or more processors;

[0022] Memory, used to store one or more programs;

[0023] When the one or more programs are executed by the one or more processors, the one or more processors implement the cigarette box transparent paper packaging quality inspection method provided in any embodiment of the present invention.

[0024] Fourthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the cigarette box transparent paper packaging quality inspection method provided in any embodiment of the present invention.

[0025] Fifthly, embodiments of the present invention provide a computer program product, including a computer program that, when executed by a processor, implements the cigarette box transparent paper packaging quality detection method provided in any embodiment of the present invention.

[0026] The technical solution of this invention involves using a photoelectric dynamic scanning module to perform photoelectric scanning on a cigarette box packaged in transparent paper, obtaining a waveform signal corresponding to the cigarette box, and using the waveform signal for defect detection to obtain the photoelectric features and a first detection result corresponding to the cigarette box. A machine vision inspection module then performs image acquisition on the cigarette box to obtain an appearance image corresponding to the cigarette box, and uses the appearance image for defect identification to obtain a defect description and a second detection result corresponding to the cigarette box. By combining photoelectric dynamic scanning with machine vision recognition, simultaneous and comprehensive detection of physical morphology (such as missing parts or wrinkles) and visual appearance (such as skewness or misalignment) defects in the transparent paper is achieved, fundamentally improving the comprehensiveness of the detection and overcoming the blind spots inherent in single detection technologies. When the first and second detection results meet the preset precision inspection conditions, cross-validation and comprehensive analysis are performed based on the photoelectric features and the defect description to determine the third detection result corresponding to the cigarette box. By comparing and synthesizing the underlying features, the recognition accuracy and classification precision of complex and ambiguous defects are significantly improved, and false alarms caused by interference from a single sensor are effectively filtered out, thereby reducing the false rejection rate during the detection process while ensuring a high detection rate. Based on the third detection result and the historical packaging data corresponding to the cigarette box, the target detection result corresponding to the cigarette box is determined, improving the accuracy of packaging quality inspection and further improving product quality; wherein, the historical packaging data refers to the historical packaging data corresponding to the equipment that performs transparent paper packaging operations on the cigarette box.

[0027] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart of a method for quality inspection of transparent paper packaging for cigarette boxes provided in Embodiment 1 of the present invention;

[0030] Figure 2 This is a flowchart of a method for quality inspection of transparent paper packaging for cigarette boxes provided in Embodiment 2 of the present invention;

[0031] Figure 3This is a schematic diagram of the structure of a quality inspection device for transparent paper packaging of cigarette boxes provided in Embodiment 3 of the present invention;

[0032] Figure 4 This is a schematic diagram of the structure of an electronic device for implementing the cigarette box transparent paper packaging quality inspection method of the present invention. Detailed Implementation

[0033] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0034] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0035] Example 1

[0036] Figure 1 This invention provides a flowchart of a method for quality inspection of transparent paper packaging for cigarette boxes, according to Embodiment 1. This embodiment is applicable to the quality inspection of transparent paper packaging for cigarette boxes. The method can be executed by a quality inspection device for transparent paper packaging, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0037] S110. The photoelectric dynamic scanning module performs photoelectric scanning on the cigarette box after the transparent paper packaging is completed, obtains the waveform signal corresponding to the cigarette box, and uses the waveform signal to perform defect detection, and obtains the photoelectric features and the first detection result corresponding to the cigarette box.

[0038] In this embodiment, the photoelectric dynamic scanning module refers to a hardware unit integrating a light source, a photoelectric sensor, and a signal conditioning circuit. The photoelectric dynamic scanning module can be used to non-contactly measure changes in reflected light intensity on an object's surface. In this solution, the photoelectric dynamic scanning module is specifically used for high-speed, continuous scanning and detection of the transparent paper of a moving cigarette box. The transparent paper can refer to the biaxially oriented polypropylene (BOPP) film wrapped around the outermost layer of a cigarette pack (i.e., a hard-pack cigarette pack). The transparent paper can be used for moisture protection, aroma preservation, and improving appearance. The cigarette box can refer to a cigarette box that has completed the inner lining and label packaging and has entered the transparent paper packaging station, such as a hard-pack cigarette pack.

[0039] In this embodiment, the waveform signal can refer to an analog or digital voltage signal output by a photoelectric sensor that varies over time (corresponding to the moving position of the cigarette box). The amplitude change of the waveform signal directly reflects the change in the reflectivity of the transparent paper surface or the distance from the sensor. Photoelectric features can refer to a set of quantitative indicators extracted from the waveform signal to characterize the physical state of the transparent paper. Photoelectric features may include, but are not limited to, the location coordinates of signal anomalies, peak / valley intensity, abnormal pulse width, specific change patterns of the waveform profile (such as step, ramp), and correlation coefficient with a standard waveform. The first detection result can refer to a preliminary quality judgment based on the immediate analysis of the photoelectric features. The first detection result typically includes a binary or multi-dimensional state identifier (such as normal, suspected abnormal, severe abnormal) and an associated confidence score.

[0040] Specifically, the system's main control unit receives encoder signals from the packaging machine's main shaft to accurately determine whether the cigarette box has entered the preset scanning station. When the cigarette box arrives, the main control unit triggers the photoelectric dynamic scanning module to operate. The sensor within the module emits modulated light onto the surface of the transparent paper, and the receiver synchronously collects the reflected light and converts it into a continuous waveform electrical signal. The collected raw waveform signal is filtered (e.g., to remove high-frequency noise), amplified, and converted from analog to digital. The pre-processed digital signal is compared in real-time with a pre-stored standard qualified product waveform template, and abnormal features are extracted using algorithms (e.g., sliding window differential, peak detection, waveform matching), and their feature values ​​are calculated. The extracted photoelectric features are input into a preset logic judge or simple classifier. For example, if the signal strength at a certain point is below a threshold A, it is marked as a suspected missing element; if the signal exhibits fluctuations at a specific frequency, it is marked as a suspected wrinkle. Combining the judgments of multiple features, a first detection result with confidence level is generated.

[0041] As an optional embodiment of this disclosure, the photoelectric dynamic scanning module includes: multiple sets of reflective photoelectric sensor arrays arranged along the cigarette box conveying direction; the photoelectric dynamic scanning module is used to continuously and dynamically scan the edge and surface of the transparent paper; the reflective photoelectric sensor array includes: four sets of photoelectric sensors; each set of photoelectric sensors includes: a transmitter and a receiver.

[0042] In this embodiment, a reflective photoelectric sensor array can refer to a detection assembly composed of multiple photoelectric sensors arranged according to a specific spatial geometric relationship. The reflective photoelectric sensor array can be used to achieve multi-point synchronous coverage measurement of a target area. The photoelectric sensor can refer to a U-shaped (groove-type) photoelectric switch or a through-beam photoelectric sensor. The transmitter and receiver of the photoelectric sensor are arranged opposite each other, forming a detection groove or light curtain. The transmitter can refer to a device that emits a modulated beam of a specific wavelength (such as infrared light). The receiver can refer to a device that receives the beam emitted by the transmitter and reflected by the surface of transparent paper (for diffuse reflection) or directly received (for through-beam), and converts it into an electrical signal.

[0043] Specifically, four U-shaped photoelectric sensor groups are installed on both sides of the cigarette box's transparent paper output channel of the FXS unit. Two groups are aligned with the top and inner edges of the cigarette box, respectively, while the other two groups are aligned with the bottom and outer edges. The beam of each sensor group spans the channel, ensuring that the transparent paper at the corresponding edge of the cigarette box passes through the beam as it passes. When the transparent paper is flat and adhered, the reflected (or transmitted) light intensity is stable; if there are gaps, wrinkles, or flipped edges, it will cause a characteristic abrupt change in light intensity, which will be captured by the sensor and form a unique waveform signal.

[0044] S120. The machine vision inspection module acquires images of the cigarette box to obtain the appearance image of the cigarette box, and uses the appearance image to identify defects, thereby obtaining the defect description and the second inspection result of the cigarette box.

[0045] In this embodiment, the machine vision inspection module can refer to a hardware and software system composed of an industrial camera, an optical lens, an illumination source, and an image processing computing unit. The machine vision inspection module can simulate human visual functions to capture, analyze, and understand images of objects. The appearance image can refer to a high-resolution digital image captured by an industrial camera, showing the overall appearance of the cigarette box's transparent paper packaging. The appearance image typically covers the front, side, or specific angle views of the cigarette box. The defect description can refer to a structured and semantic representation of the identified defects. The defect description can be a data structure containing information such as defect type, defect location (e.g., image coordinates or actual location mapped to the cigarette box), defect size (e.g., area, length, offset), and visual feature vectors. The second inspection result can refer to a quality judgment generated based on intelligent analysis of the appearance image. The second inspection result can include status indicators (e.g., qualified, misaligned stitching, skewed transparent paper) and corresponding confidence levels.

[0046] Specifically, the machine vision inspection module is controlled by a signal from the same encoder or the completion signal of the photoelectric dynamic scanning module. An adjustable light source illuminates the cigarette box at optimal brightness and angle, and two or more industrial cameras simultaneously capture high-definition images from different perspectives. The acquired images undergo denoising, contrast enhancement, and color space conversion to optimize image quality. A trained deep learning model (such as a convolutional neural network CNN) or traditional vision algorithms is used to analyze the preprocessed images. The model locates potential defect areas and classifies them into their specific types (such as skew, folding, misaligned stitching, breakage, contamination, etc.). For identified defects, the algorithm calculates their bounding box, center point coordinates, pixel area, and other quantitative information, which, together with the defect type, constitute a defect description. Finally, all identified information is integrated to output a second detection result.

[0047] As an optional implementation of this disclosure, the machine vision inspection module includes: at least two industrial cameras, an adjustable light source, and an image processing unit; the machine vision inspection module is used to acquire multi-angle appearance images of cigarette boxes and perform defect identification on the appearance images; the at least two industrial cameras are arranged at an angle; the adjustable light source is an LED diffuse reflection light source with adjustable brightness and angle.

[0048] In this embodiment, the industrial camera refers to a digital camera with features such as high frame rate, high resolution, and global shutter, suitable for industrial environments. Industrial cameras typically use interfaces such as GigE Vision or USB 3.0. The adjustable light source refers to an illumination device whose brightness, color temperature, and emission angle can be adjusted via a program or circuit. Adjustable light sources can be used to ensure optimal imaging results under different environments or materials. The image processing unit refers to a computer or embedded device equipped with a high-performance processor (CPU / GPU). The image processing unit can be used to run image processing algorithms and defect recognition models. The LED diffuse reflection light source refers to a light source that uses LEDs as light-emitting elements and employs a diffuse reflector to uniformly and softly illuminate the surface of the object being measured, thereby reducing specular reflections and high-brightness interference.

[0049] Specifically, two 5-megapixel global shutter color cameras are installed above the inspection station. One camera faces the front of the cigarette box at approximately a 30-degree angle, and the other faces the side of the cigarette box at approximately a 60-degree angle, creating a stereoscopic vision. The light source uses a surround-style strip LED diffuse reflection light source, illuminating from both sides. The image processing unit uses an edge computing device equipped with a GPU. When the cigarette box arrives, the two cameras simultaneously capture images, which are transmitted via network to the edge computing device. Deep learning models such as YOLO or U-Net deployed on the device perform real-time inference, outputting a defect description and a second inspection result containing specific information such as a 2mm rightward offset of the pull line and a confidence level of 0.96.

[0050] It should be noted that the modular design of the functions supports remote diagnostics and online upgrades, reducing downtime.

[0051] S130. When the first and second test results meet the preset fine inspection conditions, cross-validation and comprehensive analysis are performed based on photoelectric features and defect descriptions to determine the third test result corresponding to the cigarette box.

[0052] In this embodiment, the preset fine inspection conditions can refer to rules that trigger a more refined and complex analysis process. Preset fine inspection conditions typically include: inconsistencies between the first and second detection results (e.g., one normal and one abnormal); both reporting low-confidence minor anomalies or suspected defects; and both reporting different defect types but indicating potential packaging problems. Cross-validation can refer to spatiotemporally aligning and logically associating the physical state changes implied by photoelectric features (e.g., a sudden drop in signal at position X) with the visual morphological anomalies described in the defect description (e.g., wrinkles identified in region Y) to check whether they point to the same problem at the same location on the cigarette box. Comprehensive analysis can refer to using data fusion algorithms (e.g., DS evidence theory, fuzzy logic, weighted voting, etc.) to comprehensively calculate factors such as the confidence level of photoelectric features, the confidence level of the visual defect description, and the degree of consistency between the two, to arrive at a more reliable and accurate unified judgment. The third detection result can refer to the intermediate-level decision result generated after cross-validation and comprehensive analysis. The third detection result is more reliable than the first and second detection results. The third test result includes a more specific defect classification, grade, and revised overall confidence level.

[0053] Specifically, the system determines in real time whether the first and second detection results meet the preset fine inspection conditions. For cigarette boxes that meet the preset fine inspection conditions, the system maps the abnormal location information in their photoelectric features and the defect location coordinates in the visual defect description to the same physical space (cigarette box coordinate system) through coordinate transformation. A fusion decision algorithm is then invoked. For example, using DS evidence theory, the evidence provided by photoelectric detection (supporting the reliability of propositions such as missing parts and wrinkles) is synthesized with the evidence provided by visual detection. If the two pieces of evidence are highly consistent (e.g., both strongly support a missing top-left corner), the reliability of the proposition is significantly improved after synthesis; if they contradict each other (e.g., the photoelectric display is normal but the visual report is skewed), the synthesized result may produce a highly reliable uncertain proposition, or it may be biased towards one side based on historical weights. Based on the reliability distribution of the fused evidence, the final defect type, level (e.g., minor, moderate, severe), and overall confidence level are determined, forming the third detection result.

[0054] As an optional implementation of this disclosure, the method further includes: when the first detection result and the second detection result meet the first rejection condition, performing different types of rejection operations on the cigarette box based on the defect type and defect level corresponding to the cigarette box; and releasing the cigarette box when the first detection result and the second detection result meet the first release condition.

[0055] In this embodiment of the disclosure, the first rejection criterion can refer to a rapid rejection rule based on the initial detection results (i.e., the first detection result and the second detection result). The first rejection criterion is usually set very strictly, for example, requiring that both the first and second detection results be reported as serious defects and that the confidence levels are both higher than the threshold Th1. Defect types can include missing transparent paper, large-area wrinkles, severe skew, complete detachment of the thread, etc. The defect level can refer to a level divided according to the severity of the defect (such as area size, offset), such as minor, moderate, and severe.

[0056] In this embodiment, the rejection operation can refer to the action of removing defective cigarette boxes from the main production line. Specific methods of rejection operations may include high-pressure air blowing, mechanical push rods, flipping rods, etc. Rejection operations can also correspond to different rejection areas. For specific types of defects with high defect levels, they can be rejected to the repackaging area for subsequent removal and repackaging of the transparent paper. For specific types of defects with low defect levels, they can be rejected to the substandard product area for subsequent graded release. The first release condition can refer to a rapid release rule based on the initial inspection results. The first release condition is also extremely stringent, for example, requiring both the first and second inspection results to be reported as normal with confidence levels higher than the threshold Th2.

[0057] Specifically, upon receiving the first and second inspection results, the system first checks whether the first rejection condition or the first release condition is met. If the first rejection condition is met, a rejection instruction is immediately generated, and the appropriate rejector (such as the main air valve) is selected based on the defect type and defect level to execute the rejection, without waiting for subsequent fine inspection. If the first release condition is met, the cigarette box is immediately allowed to pass. This improves the processing efficiency of clearly defined good and bad products. Cigarette boxes that do not meet either of these conditions flow into the fine inspection process of S130.

[0058] S140. Based on the third detection result and the historical packaging data corresponding to the cigarette box, determine the target detection result corresponding to the cigarette box; wherein, the historical packaging data is the historical packaging data corresponding to the equipment that performs transparent paper packaging operation on the cigarette box.

[0059] In this embodiment, historical packaging data refers to quality data accumulated over time related to the current production equipment (such as a specific FXS packaging machine). Historical packaging data may include: the frequency and distribution patterns of various defects in the recent period (e.g., the past hour, shift), and their correlation with the action phases of key equipment components (such as folding mechanisms, soldering irons). The target detection result refers to the final, authoritative quality judgment output by the system. The target detection result can be used to guide the final execution actions (such as differential rejection, release, and equipment warnings). The target detection result is generated based on the third detection result, combined with contextualized verification of equipment health background information. The equipment may refer to an FXS series packaging machine.

[0060] Specifically, the system queries the recent historical packaging database of the packaging machine that produced the current cigarette box, based on the production time and machine number. The third inspection result is then compared with the historical packaging data. For example, if the third inspection result indicates minor wrinkles, but historical packaging data shows that the machine's minor wrinkle defect rate is significantly higher than the normal baseline during the current period, the system may determine that the equipment condition is deteriorating (e.g., soldering iron temperature is too low), thereby increasing the confidence level of the current cigarette box's target inspection result or directly linking it to an equipment warning. Similarly, if the third inspection result indicates a rare defect, but similar defects have not appeared in historical packaging data, the system may mark the target inspection result as an isolated anomaly for manual review. Combining the third inspection result and historical packaging data, the final target inspection result is generated. This result includes not only the final judgment on the current cigarette box (pass / fail / defect type and level) but may also include a marker of the equipment condition (e.g., equipment condition is suspicious).

[0061] The technical solution of this invention uses a photoelectric dynamic scanning module to perform photoelectric scanning on a cigarette box packaged in transparent paper, obtaining a waveform signal corresponding to the cigarette box. This waveform signal is then used for defect detection, yielding photoelectric features and a first detection result. A machine vision inspection module acquires an image of the cigarette box, obtaining an appearance image. This appearance image is then used for defect identification, yielding a defect description and a second detection result. By combining photoelectric dynamic scanning with machine vision recognition, simultaneous and comprehensive detection of physical morphology defects (such as missing parts or wrinkles) and visual appearance defects (such as misalignment or crooked teeth) in the transparent paper is achieved. This fundamentally improves the comprehensiveness of the detection and overcomes the blind spots inherent in single detection technologies. When the first and second detection results meet preset precision inspection conditions, cross-validation and comprehensive analysis are performed based on the photoelectric features and defect descriptions to determine a third detection result for the cigarette box. By comparing and synthesizing underlying features, the accuracy of identifying and classifying complex and ambiguous defects is significantly improved, and false alarms caused by interference from a single sensor are effectively filtered out. This ensures a high detection rate while reducing the false rejection rate during the detection process. Based on the third test results and the historical packaging data corresponding to the cigarette boxes, the target test results corresponding to the cigarette boxes are determined, which improves the accuracy of packaging quality testing and further improves product quality; among them, the historical packaging data refers to the historical packaging data corresponding to the equipment that performs transparent paper packaging operations on cigarette boxes.

[0062] Example 2

[0063] Figure 2This is a flowchart of a method for quality inspection of transparent paper packaging for cigarette boxes according to Embodiment 2 of the present invention. Based on the above embodiments, this embodiment describes in detail the process of subsequent processing of the cigarette box according to the target inspection results. Explanations of terms that are the same as or corresponding to those in the above embodiments are not repeated here. Figure 2 As shown, the method includes:

[0064] S210. The photoelectric dynamic scanning module performs photoelectric scanning on the cigarette box after the transparent paper packaging is completed, obtains the waveform signal corresponding to the cigarette box, and uses the waveform signal to perform defect detection, and obtains the photoelectric features and the first detection result corresponding to the cigarette box.

[0065] S220. The machine vision inspection module acquires images of the cigarette box to obtain the appearance image of the cigarette box, and uses the appearance image to identify defects, thereby obtaining the defect description and the second inspection result of the cigarette box.

[0066] S230. When the first and second test results meet the preset fine inspection conditions, cross-validation and comprehensive analysis are performed based on photoelectric features and defect descriptions to determine the third test result corresponding to the cigarette box.

[0067] S240. Based on the third detection result and the historical packaging data corresponding to the cigarette box, determine the target detection result corresponding to the cigarette box; wherein, the historical packaging data is the historical packaging data corresponding to the equipment that performs transparent paper packaging operation on the cigarette box.

[0068] S250. When the target detection result meets the second rejection condition, different types of rejection operations are performed on the cigarette box based on the defect type and defect level corresponding to the cigarette box.

[0069] In this embodiment of the disclosure, the second rejection condition may refer to a rejection rule based on the final target detection result. The setting of the second rejection condition is more refined and optimized. For example, the defect level in the target detection result is moderate or severe, or the overall confidence level is greater than the threshold Th3.

[0070] Specifically, the system reads the target detection results and determines whether the second rejection condition is met. If it is, the optimal rejection strategy is selected based on the precise defect type and level in the target detection results. For example, for severe defects, a high-speed main air valve is triggered for powerful blowing; for slight side wrinkles, a less forceful auxiliary air nozzle or mechanical lever is triggered to gently push the defective product into the defective product channel, avoiding damage to the cigarette box. The timing of the rejection command is strictly synchronized with the encoder phase to ensure accurate positioning.

[0071] S260. When the target detection result meets the second release condition, the cigarette box is released.

[0072] In this embodiment of the disclosure, the second release condition can refer to the situation where the target detection result does not meet any rejection or warning conditions, which is considered to meet the release condition. The core of the second release condition is that the target detection result is determined to be qualified.

[0073] Specifically, if the system determines that the target detection result is qualified or no other conditions are triggered, it will not send an action command to any actuator. The current cigarette box will naturally pass through the detection station on the conveyor line and enter the next process.

[0074] S270. When the target detection result meets the preset equipment fault conditions, generate an early warning instruction corresponding to the equipment that performs transparent paper packaging operation on cigarette boxes, so as to maintain the equipment.

[0075] In this embodiment, the preset equipment failure condition can refer to a rule for inferring the health status of the equipment from the target detection results. For example, the same type of defect (such as creases at the same location) occurs M times in N consecutive cigarette boxes; or the incidence rate of a certain type of defect exceeds a dynamic statistical threshold within a time window T. The warning instruction can refer to structured alarm information generated by the system. The warning instruction typically includes a warning level (such as alert, warning, or severe), associated equipment / workstation number, suspected faulty component or cause (such as possible wear on the upper folding mechanism), relevant defect data samples, and a timestamp.

[0076] Specifically, the system continuously monitors the sequence of target detection results and calculates the statistical process control (SPC) indicators for various defects in real time. Once a preset equipment failure condition is triggered (e.g., the system detects that the occurrence rate of a bottom edge folding defect jumps from 1% to 10% within 5 minutes), an early warning command is immediately generated. This command can pop up an alarm via a human-machine interface (HMI), be sent to a mobile device, or be directly uploaded to the manufacturing execution system (MES), prompting maintenance personnel to check the corresponding packaging machine components (such as the heat-sealing soldering iron or folding plate at the corresponding location). The method and process of this embodiment can be integrated with the factory's information system to achieve quality traceability, equipment health management, and process optimization. For example, the system supports the OPC UA protocol and can seamlessly interface with the factory's SCADA and MES systems. Real-time uploaded defect data includes: defect type, workstation where it occurred, timestamp, and photographic evidence. Through a big data dashboard, managers can monitor the quality trends of each machine in real time and achieve quality traceability and push process parameter optimization suggestions.

[0077] The technical solution of this invention, when the target detection result meets the second rejection condition, performs different types of rejection operations on the cigarette box based on the defect type and defect level corresponding to the cigarette box. This allows different actuators to perform differentiated rejection operations on the cigarette box based on the precise defect type and defect level (e.g., using high-speed air blowing to remove severe defects, and using gentle push rods to separate minor wrinkles). This mechanism achieves precise handling, effectively avoiding the loss of qualified products or secondary packaging damage caused by excessive or improper rejection in traditional methods. While strictly isolating defective products, it significantly reduces material waste and production costs. When the target detection result meets the second release condition, the cigarette box is released, ensuring that qualified products can flow unimpeded to the next process, avoiding production cycle loss that may occur due to the detection system itself, and thus ensuring the overall operating efficiency and smoothness of the high-speed packaging line. When the target detection results meet the preset equipment failure conditions (such as the continuous and concentrated occurrence of specific types of defects in a short period of time), a warning instruction is generated for the equipment that performs transparent paper packaging operations on cigarette boxes, so as to maintain the equipment. This mechanism realizes the deduction of equipment health status from product quality judgment, enabling maintenance personnel to intervene before failure occurs or in the early stage of deterioration based on accurate data warnings. This transforms traditional reactive maintenance into advanced predictive maintenance, thereby significantly reducing unplanned downtime, improving overall equipment efficiency (OEE), and extending the service life of key components.

[0078] As an optional implementation of this disclosure, the method further includes: periodically and adaptively adjusting the photoelectric signal threshold in the photoelectric dynamic scanning module and the detection model parameters in the machine vision detection module without stopping the machine.

[0079] In this embodiment of the disclosure, the photoelectric signal threshold can refer to a critical value used in photoelectric detection to determine whether a signal is abnormal. For example, the photoelectric signal threshold may include a lower intensity threshold for determining "missing" signals and a fluctuation amplitude threshold for determining "wrinkles". Detection model parameters can refer to adjustable parameters in a machine vision recognition model, such as the weights of convolutional kernels and the bias of classification layers in a deep learning model, or the gradient threshold for edge detection and the similarity threshold for template matching in traditional algorithms.

[0080] Specifically, the system establishes a background optimization cycle (e.g., after 10,000 cigarette boxes are produced). Within each cycle, operations such as data collection, threshold recalibration, model fine-tuning, and seamless updates are performed.

[0081] Among them, in the data collection stage, the optoelectronic waveforms and images corresponding to all cigarette boxes finally determined to be qualified within this cycle are collected as the current normal sample set. In the threshold recalibration stage, for optoelectronic signals, the distributions of various eigenvalue of the waveforms of the current normal sample set (such as average reflection intensity, standard deviation of fluctuations) are calculated, and based on these distributions, new decision thresholds are recalculated (such as set as the mean minus 3 times the standard deviation) to adapt to changes in the transparent paper material batch or light source aging. In the model fine-tuning stage, for the visual model, the current normal sample set and the defective samples confirmed within this cycle (such as the rejected products confirmed by manual review) are used as incremental data to perform lightweight transfer learning or fine-tuning (Fine-tuning) on the original deep learning model, so that the model can better adapt to the product characteristics in the current production environment and improve the recognition accuracy. In the seamless update stage, the new threshold and new model parameters are hot updated to the running system, and the entire process does not affect the continuous operation of the production line, realizing the self-learning and adaptive optimization of the system.

[0082] The cigarette box transparent paper packaging quality detection device provided by the embodiment of the present invention can execute the cigarette box transparent paper packaging quality detection method provided by any embodiment of the present invention, and has the corresponding beneficial effects of executing the cigarette box transparent paper packaging quality detection method.

[0083] The following is an embodiment of the cigarette box transparent paper packaging quality detection device provided by the embodiment of the present invention. This device and the cigarette box transparent paper packaging quality detection methods of the above embodiments belong to the same inventive concept. For the details not described in detail in the embodiment of the cigarette box transparent paper packaging quality detection device, reference can be made to the embodiments of the above cigarette box transparent paper packaging quality detection methods.

[0084] Embodiment III

[0085] Figure 3 It is a schematic structural diagram of a cigarette box transparent paper packaging quality detection device provided by Embodiment III of the present invention. As Figure 3 shown, the device includes: a first detection result determination module 310, a second detection result determination module 320, a third detection result determination module 330, and a target detection result determination module 340.

[0086] The first detection result determination module 310 is used to perform photoelectric scanning on the cigarette box after transparent paper packaging using a photoelectric dynamic scanning module to obtain the waveform signal corresponding to the cigarette box, and to use the waveform signal for defect detection to obtain the photoelectric features and the first detection result corresponding to the cigarette box. The second detection result determination module 320 is used to acquire images of the cigarette box using a machine vision inspection module to obtain the appearance image corresponding to the cigarette box, and to use the appearance image for defect identification to obtain the defect description and the second detection result corresponding to the cigarette box. The third detection result determination module 330 is used to determine the third detection result corresponding to the cigarette box by cross-validation and comprehensive analysis based on the photoelectric features and defect description when the first detection result and the second detection result meet the preset fine inspection conditions. The target detection result determination module 340 is used to determine the target detection result corresponding to the cigarette box based on the third detection result and the historical packaging data corresponding to the cigarette box. The historical packaging data refers to the historical packaging data corresponding to the equipment that performs transparent paper packaging operations on the cigarette box.

[0087] The technical solution of this invention uses a photoelectric dynamic scanning module to perform photoelectric scanning on a cigarette box packaged in transparent paper, obtaining a waveform signal corresponding to the cigarette box. This waveform signal is then used for defect detection, yielding photoelectric features and a first detection result. A machine vision inspection module acquires an image of the cigarette box, obtaining an appearance image. This appearance image is then used for defect identification, yielding a defect description and a second detection result. By combining photoelectric dynamic scanning with machine vision recognition, simultaneous and comprehensive detection of physical morphology defects (such as missing parts or wrinkles) and visual appearance defects (such as misalignment or crooked teeth) in the transparent paper is achieved. This fundamentally improves the comprehensiveness of the detection and overcomes the blind spots inherent in single detection technologies. When the first and second detection results meet preset precision inspection conditions, cross-validation and comprehensive analysis are performed based on the photoelectric features and defect descriptions to determine a third detection result for the cigarette box. By comparing and synthesizing underlying features, the accuracy of identifying and classifying complex and ambiguous defects is significantly improved, and false alarms caused by interference from a single sensor are effectively filtered out. This ensures a high detection rate while reducing the false rejection rate during the detection process. Based on the third test results and the historical packaging data corresponding to the cigarette boxes, the target test results corresponding to the cigarette boxes are determined, which improves the accuracy of packaging quality testing and further improves product quality; among them, the historical packaging data refers to the historical packaging data corresponding to the equipment that performs transparent paper packaging operations on cigarette boxes.

[0088] Based on the above technical solution, the device also includes:

[0089] The first cigarette box rejection module is used to perform different types of rejection operations on the cigarette box based on the defect type and defect level corresponding to the cigarette box when the first detection result and the second detection result meet the first rejection condition.

[0090] The first cigarette box release module is used to release the cigarette box when the first detection result and the second detection result meet the first release condition.

[0091] Based on the above technical solution, the device also includes:

[0092] The second cigarette box rejection module is used to perform different types of rejection operations on the cigarette box based on the defect type and defect level of the cigarette box when the target detection result meets the second rejection condition.

[0093] The second cigarette box release module is used to release the cigarette box when the target detection result meets the second release condition;

[0094] The equipment early warning module is used to generate an early warning command for the equipment that performs transparent paper packaging on cigarette boxes when the target detection result meets the preset equipment fault conditions, so as to facilitate equipment maintenance.

[0095] Based on the above technical solution, the photoelectric dynamic scanning module includes: multiple sets of reflective photoelectric sensor arrays arranged along the cigarette box conveying direction; the photoelectric dynamic scanning module is used to continuously and dynamically scan the edge and surface of the transparent paper; the reflective photoelectric sensor array includes: four sets of photoelectric sensors; each set of photoelectric sensors includes: a transmitter and a receiver.

[0096] Based on the above technical solution, the machine vision inspection module includes: at least two industrial cameras, an adjustable light source, and an image processing unit; the machine vision inspection module is used to acquire multi-angle appearance images of cigarette boxes and perform defect identification on the appearance images; at least two industrial cameras are arranged at an angle; the adjustable light source is an LED diffuse reflection light source with adjustable brightness and angle.

[0097] Based on the above technical solution, the device also includes:

[0098] The adaptive adjustment module is used to periodically and adaptively adjust the photoelectric signal threshold in the photoelectric dynamic scanning module and the detection model parameters in the machine vision inspection module without stopping the machine.

[0099] The cigarette box transparent paper packaging quality inspection device provided in this embodiment of the invention can perform the cigarette box transparent paper packaging quality inspection method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for performing the cigarette box transparent paper packaging quality inspection method.

[0100] It is worth noting that in the above embodiments of quality inspection of cigarette box transparent paper packaging, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0101] Example 4

[0102] Figure 4 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0103] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0104] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0105] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of 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 suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the quality inspection method for transparent paper packaging of cigarette boxes.

[0106] In some embodiments, the method for inspecting the quality of cigarette box transparent paper packaging can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the cigarette box transparent paper packaging quality inspection method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the cigarette box transparent paper packaging quality inspection method by any other suitable means (e.g., by means of firmware).

[0107] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0108] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0109] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0110] To provide interaction with a user, the systems and techniques described herein 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 pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; 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 sound input, voice input, or tactile input).

[0111] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0112] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0113] This application also discloses a computer program product, which includes a computer program that, when executed by a processor, implements the cigarette box transparent paper packaging quality inspection method provided in any embodiment of this application. This program product belongs to the same inventive concept as the cigarette box transparent paper packaging quality inspection method disclosed in the various embodiments of this application, and therefore will not be described in detail here.

[0114] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0115] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. 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 substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for quality inspection of transparent paper packaging for cigarette boxes, characterized in that, include: The photoelectric dynamic scanning module performs photoelectric scanning on the cigarette box after it is wrapped in transparent paper to obtain the waveform signal corresponding to the cigarette box. The waveform signal is then used to perform defect detection to obtain the photoelectric features and the first detection result corresponding to the cigarette box. The cigarette box is image acquired by the machine vision inspection module to obtain the appearance image of the cigarette box, and the appearance image is used to identify defects to obtain the defect description and the second inspection result of the cigarette box. When the first detection result and the second detection result meet the preset fine inspection conditions, cross-validation and comprehensive analysis are performed based on the photoelectric features and the defect description to determine the third detection result corresponding to the cigarette box. Based on the third detection result and the historical packaging data corresponding to the cigarette box, the target detection result corresponding to the cigarette box is determined; wherein, the historical packaging data is the historical packaging data corresponding to the equipment that performs transparent paper packaging operation on the cigarette box.

2. The method according to claim 1, characterized in that, The method further includes: When the first detection result and the second detection result meet the first rejection condition, different types of rejection operations are performed on the cigarette box based on the defect type and defect level corresponding to the cigarette box. When the first detection result and the second detection result meet the first release condition, the cigarette box is released.

3. The method according to claim 1, characterized in that, The method further includes: When the target detection result meets the second rejection condition, different types of rejection operations are performed on the cigarette box based on the defect type and defect level corresponding to the cigarette box. When the target detection result meets the second release condition, the cigarette box is released; When the target detection result meets the preset equipment fault conditions, an early warning instruction is generated for the equipment that performs transparent paper packaging operation on the cigarette box, so as to maintain the equipment.

4. The method according to claim 1, characterized in that, The photoelectric dynamic scanning module includes: multiple sets of reflective photoelectric sensor arrays arranged along the cigarette box conveying direction; the photoelectric dynamic scanning module is used to continuously and dynamically scan the edge and surface of the transparent paper; The reflective photoelectric sensor array includes four groups of photoelectric sensors; each group of photoelectric sensors includes a transmitter and a receiver.

5. The method according to claim 1, characterized in that, The machine vision inspection module includes: at least two industrial cameras, an adjustable light source, and an image processing unit; the machine vision inspection module is used to acquire multi-angle appearance images of cigarette boxes and perform defect identification on the appearance images; The at least two industrial cameras are arranged at an angle to each other; the adjustable light source is an LED diffuse reflection light source with adjustable brightness and angle.

6. The method according to claim 1, characterized in that, The method further includes: Without shutting down the machine, the photoelectric signal threshold in the photoelectric dynamic scanning module and the detection model parameters in the machine vision detection module are periodically and adaptively adjusted.

7. A quality inspection device for transparent paper packaging of cigarette boxes, characterized in that, The device includes: The first detection result determination module is used to perform photoelectric scanning on the cigarette box packaged in transparent paper through the photoelectric dynamic scanning module, obtain the waveform signal corresponding to the cigarette box, and use the waveform signal to perform defect detection, thereby obtaining the photoelectric features and the first detection result corresponding to the cigarette box. The second detection result determination module is used to acquire images of the cigarette box through the machine vision detection module, obtain the appearance image of the cigarette box, and use the appearance image to identify defects, thereby obtaining a defect description and a second detection result for the cigarette box. The third detection result determination module is used to determine the third detection result corresponding to the cigarette box by performing cross-validation and comprehensive analysis based on the photoelectric features and the defect description when the first detection result and the second detection result meet the preset fine inspection conditions. The target detection result determination module is used to determine the target detection result corresponding to the cigarette box based on the third detection result and the historical packaging data corresponding to the cigarette box; wherein, the historical packaging data is the historical packaging data corresponding to the equipment that performs transparent paper packaging operation on the cigarette box.

8. An electronic device, characterized in that, The electronic device includes: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the cigarette box transparent paper packaging quality inspection method as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the method for quality inspection of transparent paper packaging for cigarette boxes as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for quality inspection of transparent paper packaging of cigarette boxes as described in any one of claims 1-6.

Images