A method for controlling the production quality of electronic labels

Abstract

The present invention relates to a method for controlling the production quality of electronic labels in the field of information technology, comprising: in a second image, a multi-scale denoising algorithm is used to perform noise reduction processing on the image for vibration interference and dust interference to obtain a third image; an online detection system judges the printing quality of the electronic label based on the OCR recognition result, and if an abnormality is detected, an abnormality signal is generated and transmitted to a process control system; the process control system receives the abnormality signal and adjusts the operating parameters of the printing, laminating, and cutting processes to ensure smooth process connection and avoid imbalance in the production rhythm; the adjusted process parameters are compared with preset product consistency standards, and if the parameter deviation exceeds a preset range, an automatic calibration module is activated to ensure product consistency; and the operating status of the entire production process is monitored in real time by a system monitoring module. If it is found that the system stability has decreased, an early warning mechanism is triggered, a maintenance signal is generated, and the maintenance signal is transmitted to the maintenance system.

Description

Technical Field

[0001] The present invention relates to the field of information technology, specifically to the field of label production technology, and in particular to a method for controlling the production quality of electronic labels. Background Art

[0002] In the production of electronic labels, optical character recognition (OCR) and online inspection technologies are often introduced to improve efficiency and quality while reducing manual intervention. However, in practical applications, the introduction of these technologies also presents new challenges. First, OCR technology requires high image quality. During the production of electronic labels, the complexities of the production environment, such as lighting fluctuations, vibration, and dust, can degrade the quality of captured images, thereby affecting OCR recognition accuracy. Second, online inspection technology requires real-time processing of large amounts of data, placing higher demands on the system's data transmission and processing capabilities. If data transmission speeds cannot keep up with the production pace, or if data processing delays occur, normal production lines may be impacted. Furthermore, electronic label production involves multiple processes, such as printing, lamination, and cutting, and the integration and coordination between these processes is a critical issue. If data between processes cannot be transmitted promptly and accurately, production efficiency may decline and even quality issues may occur. Finally, electronic label production requires high product consistency and stability. Ensuring consistent product quality while maintaining production efficiency is also a pressing issue. Summary of the Invention

[0003] The present invention provides a method for controlling the production quality of electronic labels, comprising the following steps:

[0004] S101. Acquire a first image of an electronic label production process using an image acquisition device, and process the first image using an adaptive illumination compensation algorithm to account for interference from light variations, thereby obtaining a second image.

[0005] S102, using a multi-scale denoising algorithm to perform noise reduction processing on the second image in response to vibration interference and dust interference, to obtain a third image;

[0006] S103: Input the third image into a preset OCR recognition model. If there is an error in the OCR recognition result, locally optimize the third image using an image enhancement algorithm to obtain a fourth image, and perform OCR recognition again.

[0007] S104, transmitting the OCR recognition results to the online detection system via a high-speed data transmission channel. If the data transmission delay exceeds a preset threshold, a data compression algorithm is activated to reduce the data volume to ensure real-time processing capability;

[0008] S105, the online detection system determines the printing quality of the electronic label based on the OCR recognition result. If an abnormality is detected, an abnormality signal is generated and transmitted to the process control system;

[0009] S106: The process control system receives abnormal signals and adjusts the operating parameters of the printing, laminating, and cutting processes to ensure smooth process connection and avoid imbalance in production rhythm;

[0010] S107. The operating status data of each process is acquired in real time through the data acquisition module. A time series analysis method is used to determine whether the production rhythm is stable. If fluctuations are found, the process parameters are dynamically adjusted.

[0011] S108, comparing the adjusted process parameters with the preset product consistency standards. If the parameter deviation exceeds the preset range, the automatic calibration module is activated to ensure product consistency;

[0012] S109. The operating status of the entire production process is monitored in real time through the system monitoring module. If the system stability is found to be degraded, the early warning mechanism is triggered, and a maintenance signal is generated and transmitted to the maintenance system.

[0013] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0014] The present invention discloses a method for controlling the production quality of electronic labels. The method acquires images during the production process through an image acquisition device, and processes the images using adaptive illumination compensation and multi-scale denoising algorithms to cope with light changes and vibration interference. The processed image is input into an OCR recognition model, and local optimization is performed if there is an error. The recognition result is transmitted to an online detection system through a high-speed data transmission channel, and the printing quality is judged based on the result. If an abnormality is detected, a signal is generated and transmitted to the process control system to adjust the relevant process parameters. The present invention also obtains the operating status data of each process in real time, judges the stability of the production rhythm through time series analysis, and dynamically adjusts the parameters to ensure product consistency. At the same time, the present invention monitors the entire production process in real time and triggers an early warning mechanism when the system stability decreases. Through these measures, the present invention effectively improves the quality control level and production efficiency of electronic label production. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The figure is a flow chart of a method for controlling the production quality of electronic labels according to the present invention.

[0016] Figure 2 Schematic diagram of a method for controlling the production quality of electronic tags according to the present invention.

[0017] Figure 3 This is another schematic diagram of a method for controlling the production quality of electronic tags according to the present invention. DETAILED DESCRIPTION

[0018] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0019] like Figure 1-3 In this embodiment, a method for controlling the production quality of electronic tags specifically includes the following steps:

[0020] Step S101: A first image of an electronic label production process is acquired by an image acquisition device, and an adaptive illumination compensation algorithm is used to process the first image to obtain a second image in view of interference from light changes.

[0021] The invention obtains an original image during the electronic label production process; determines whether the original image is subject to light interference based on a preset light change threshold; triggers an adaptive light compensation algorithm if the original image is subject to light interference, and the adaptive light compensation algorithm includes: analyzing the brightness distribution characteristics of the original image and determining light compensation parameters; using a Ret i nex algorithm combined with an adaptive histogram equalization method to adaptively adjust the brightness and contrast of a local area of ​​the original image according to the light compensation parameters to obtain a compensated image; performing a quality assessment on the compensated image through a convolutional neural network to obtain an assessment result; dynamically optimizing the parameters of the adaptive light compensation algorithm based on the assessment result; and transmitting the light-compensated image to the electronic label defect detection module.

[0022] For example, during the production of electronic labels, changes in illumination can significantly impact image quality. To address this issue, an adaptive illumination compensation algorithm is used to pre-process the raw images. First, a high-speed camera captures images of electronic labels on the production line in real time and transmits them to the image processing unit. The system pre-sets an illumination change threshold; for example, a brightness change exceeding 20% ​​is considered interference. When illumination interference is detected, the adaptive illumination compensation algorithm is activated. This algorithm analyzes the image brightness histogram to calculate the optimal brightness adjustment parameters. For example, if the image is generally dark, the algorithm will increase the brightness appropriately; otherwise, it will decrease it. Conversely, contrast stretching is used to enhance image details, making the text and barcode on the electronic label more legible. The Ret i nex algorithm, combined with adaptive histogram equalization, effectively addresses uneven illumination. It simulates the human visual system, decomposing images into illumination and reflection components. By adjusting the illumination components, interference such as shadows and light spots can be eliminated. For example, for an electronic label image that is partially overexposed, the algorithm will reduce the brightness of the highlight areas while enhancing the details in the shadow areas, resulting in a more balanced visual effect. To evaluate the effectiveness of illumination compensation, the system uses a convolutional neural network for image quality assessment. This network model, trained with a large amount of annotated data, can rapidly assess image clarity, contrast, and other metrics. For example, if the compensated image quality score falls below a threshold, the system automatically adjusts compensation parameters, such as increasing the contrast enhancement or varying the brightness adjustment, forming a closed-loop feedback control system. The high-quality image after illumination compensation is then transmitted to the defect detection module. This improved image quality significantly enhances defect detection accuracy. For example, poorly printed text or blurred barcodes are more easily identified in the compensated image. This not only improves production efficiency but also ensures the quality standards of electronic labels. The entire process demonstrates the important application of computer vision technology in industrial production. Through real-time image processing and adaptive algorithms, it effectively overcomes the challenges of varying illumination and provides reliable assurance for electronic label quality control. This approach is not only applicable to electronic label production but can also be applied to other industrial scenarios requiring precise image analysis, such as PCB inspection and packaging printing quality control.

[0023] Step S102 : In the second image, a multi-scale denoising algorithm is used to perform denoising on the image for vibration interference and dust interference to obtain a third image.

[0024] Acquire a second image to be processed; determine to use a multi-scale denoising algorithm for denoising based on the noise type of the second image; obtain image features at different scales by performing multi-scale decomposition on the second image; obtain a preset denoising algorithm for vibration interference and dust interference; adaptively adjust the parameters of the denoising algorithm based on the image features at different scales to obtain an adaptive denoising algorithm at different scales; use the adaptive denoising algorithm at different scales to perform denoising on the image features at different scales to obtain denoised images at different scales; fuse the denoised images at different scales to obtain a denoised third image; obtain a preset image quality threshold; determine whether the quality of the third image reaches the preset image quality threshold; if so, output the third image as the final denoising result; if not, return to the step of adaptively adjusting the parameters of the denoising algorithm and re-perform the multi-scale denoising until the quality of the third image reaches the preset image quality threshold.

[0025] For example, a multiscale denoising algorithm is an effective image denoising method. By performing multiscale decomposition on an image, noise can be analyzed and processed at different scales. For example, the wavelet transform decomposes an image into low-frequency approximation coefficients and high-frequency detail coefficients. The low-frequency coefficients contain the main structural information of the image, while the high-frequency coefficients contain details such as edges and texture, as well as noise. In practical applications, a bilateral filter can be used as the basis for an adaptive denoising algorithm. Bilateral filters consider both spatial distance and pixel value differences, smoothing noise while preserving edge information. To address vibration interference, an adaptive filter based on motion estimation can be designed. This filter first estimates the motion vector of a local region of the image and then adjusts the filter parameters based on this motion information to better suppress the blurring effect caused by vibration. To address dust interference, a morphological operation combined with a median filter can be used. An opening operation is first used to remove small bright spots (dust particles), followed by a median filter to further smooth the image. This combined approach effectively removes dust interference while preserving the overall image structure. Adaptive adjustment of algorithm parameters is crucial in the denoising process. Local variance can be used as an estimate of noise intensity. For areas with strong noise, the filter strength is increased; for areas with weak noise, the filter strength is reduced to preserve more detail. For example, in bilateral filtering, the standard deviation parameters of spatial and grayscale values ​​can be dynamically adjusted based on local variance. Multi-scale image fusion is a key step in the denoising process. A weighted averaging method can be used to assign weights to each scale based on its importance. For example, low-frequency components containing more structural information are given higher weights, while high-frequency components that may contain more noise are given lower weights. This method can preserve the main features of the image while denoising. Image quality assessment is crucial for ensuring denoising effectiveness. Peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) can be used as evaluation metrics. PSNR reflects the overall quality of the image, while SSIM focuses more on preserving structural information. For example, PSNR greater than 30dB and SSIM greater than 0.9 can be set as quality thresholds. If the evaluation results do not meet the standards, it is necessary to adjust the denoising parameters, such as increasing the number of iterations or adjusting the filter strength. This multi-scale denoising method can effectively handle different types of noise interference and improve image quality. During the electronic tag production process, high-quality images are crucial for subsequent defect detection. Denoised images can more accurately reflect the actual condition of the electronic tag, helping to improve the accuracy and reliability of defect detection, thereby ensuring product quality and production efficiency.

[0026] Step S103: Input the third image into a preset OCR recognition model. If there is an error in the OCR recognition result, locally optimize the third image using an image enhancement algorithm to obtain a fourth image, and perform OCR recognition again.

[0027] Obtain a target image, input the target image into a preset OCR recognition model for text recognition processing, and obtain the OCR recognition result of the target image; determine whether there is a recognition error in the OCR recognition result of the target image, and if there is a recognition error, determine that the target image needs to be optimized; use an image enhancement algorithm to optimize the local area of ​​the target image where the OCR recognition error exists, and obtain an optimized image through image sharpening and contrast enhancement operations; re-input the optimized image into the preset OCR recognition model for text recognition processing, and obtain the OCR recognition result of the optimized image; determine whether the recognition accuracy of the OCR recognition result of the optimized image is improved compared with the OCR recognition result of the target image, and if the recognition accuracy is improved, determine that the optimization processing is effective; extract the text information in the image based on the OCR recognition result of the optimized image, structure the text information, and obtain structured text recognition result data; store the structured text recognition result data in a designated database.

[0028] For example, after acquiring the third image, it is input into a preset OCR recognition model for text recognition. OCR models typically utilize deep learning algorithms, such as a combination of convolutional neural networks (CNNs) and recurrent neural networks (RNNs), to improve recognition accuracy. For example, in an image containing handwritten digits, an OCR model may mistakenly recognize a "5" as a "6." This recognition error may be due to factors such as poor image quality, font distortion, or background interference. Image enhancement algorithms are then applied to optimize the local areas where recognition errors occur. Image sharpening can use the Laplacian operator or Gaussian sharpening filter to enhance image edges and details. Contrast enhancement can be achieved through histogram equalization or adaptive contrast enhancement (CLAHE). For example, applying a sharpening algorithm to a blurred document image can make text edges clearer, facilitating accurate character shape recognition by the OCR model. The optimized fourth image is then input into the OCR model for recognition. The recognition results of the third and fourth images are compared to evaluate the effectiveness of the optimization process. For example, an image with an original recognition rate of 85% increases to 95% after optimization, demonstrating the effectiveness of the optimization process. This improvement is not only reflected in overall accuracy but may also manifest itself in improved recognition of specific difficult-to-recognize characters (such as the similar-looking numerals "1" and "7"). Extracting and structuring text information involves converting unstructured OCR recognition results into an organized, easily analyzable data format. For example, for a recognized invoice image, information such as product name, quantity, unit price, and total price can be extracted and organized into a table. This structuring makes subsequent data analysis and information retrieval more efficient. The structured text recognition results are stored in a designated database to provide data support for subsequent business applications. Either a relational database (such as MySQL) or a non-relational database (such as MongoDB) can be selected based on data characteristics and query requirements. For example, a relational database may be more suitable for business scenarios requiring frequent updates and complex queries, while a non-relational database may be more advantageous for storing and quickly accessing large amounts of unstructured data. This series of processing steps constitutes the complete process for image-based text recognition and storage. From image acquisition, OCR recognition, image optimization, to data structuring and storage, each step is designed to improve recognition accuracy and data usability. This process is widely used in various fields, such as document digitization in automated office systems, bill processing in the financial industry, and package information recognition in the logistics industry. By continuously optimizing algorithms and models, the system's recognition capabilities and efficiency can be continuously improved, reducing labor costs for enterprises and increasing data processing speed and accuracy.

[0029] Step S104: The OCR recognition result is transmitted to the online detection system via a high-speed data transmission channel. If the data transmission delay exceeds a preset threshold, a data compression algorithm is activated to reduce the data volume and ensure real-time processing capability.

[0030] Obtaining an OCR recognition result, packaging the OCR recognition result data into a preset format, and sending it to the online detection system via a high-speed data transmission channel; the online detection system receives the OCR recognition result data, parses the OCR recognition result data, extracts the OCR recognition result information, and stores the OCR recognition result information in a result cache queue; starting a data transmission monitoring thread, obtaining a current data transmission delay, and triggering a data compression algorithm if the data transmission delay exceeds a preset threshold; the data compression algorithm compresses the OCR recognition result data in the result cache queue using a lossless compression algorithm to reduce the amount of transmitted data while ensuring data integrity; repackaging the compressed OCR recognition result data, sending it to the online detection system via the high-speed data channel, and updating the delay threshold in the data transmission monitoring thread; the online detection system receives the compressed OCR recognition result data, decompresses it, restores the original OCR recognition result, and enters a subsequent real-time analysis and processing process; the analysis result is associated with the original OCR recognition result and stored, persisted through a distributed cache system, and an index of the OCR recognition result and the analysis result is established based on a time series database.

[0031] For example, the data transmission and processing of OCR recognition results is a complex process involving multiple technical links. First, the recognition results are packaged into a standard format, such as JSON or XML, to ensure data structure and readability. High-speed data transmission channels can utilize dedicated fiber optic networks or 5G technology to ensure efficient data transmission. After the online detection system receives the data, the parsing process may involve a JSON or XML parser to extract key information such as text content and location coordinates. The result cache queue can be implemented using an in-memory database such as Redis, supporting high-concurrency read and write operations. The data transmission monitoring thread calculates transmission latency in real time by calculating the difference between the packet send and receive time. If the latency exceeds a preset threshold, such as 100 milliseconds, data compression is triggered. The compression algorithm used is Huffman coding, which assigns codes of different lengths based on character frequency, using short codes for high-frequency characters and long codes for low-frequency characters, thereby reducing the overall data volume. The compressed data is repackaged and transmitted, and the latency threshold in the transmission monitoring thread is updated simultaneously. This dynamic adjustment mechanism adapts to network conditions. For example, when the network is congested, the threshold is adjusted to 150 milliseconds, and when the network is unobstructed, it is adjusted back to 100 milliseconds, balancing transmission efficiency and system load. After receiving compressed data, the online detection system decompresses it and restores the original OCR recognition results. This process may involve Huffman tree reconstruction and encoding parsing. Subsequently, the system may perform real-time analysis and processing such as text classification and entity recognition. The analysis results and the original OCR recognition results can be stored in a key-value pair format, where the key is the unique identifier of the OCR result and the value is the corresponding analysis result. Distributed caching systems such as Memcached can be used for data persistence, providing high-speed read and write speeds and fault tolerance. Time series databases such as InfluxDB are suitable for storing timestamps and can create indexes that link OCR recognition time and result content. This index structure facilitates subsequent time range queries and trend analysis, such as statistically analyzing changes in text recognition accuracy over a certain period of time. The entire process is designed to balance real-time performance, accuracy, and system resource utilization. Through dynamic compression and distributed storage, the system can maintain stability despite changing network conditions while facilitating subsequent data mining and analysis. This architecture is not only applicable to OCR systems, but can also be extended to other scenarios that require large-scale real-time data processing, such as video stream analysis or IoT data processing.

[0032] In step S105 , the online detection system determines the printing quality of the electronic label based on the OCR recognition result. If an abnormality is detected, an abnormality signal is generated and transmitted to the process control system.

[0033] Acquire image data of the electronic label and preprocess the image data, wherein the preprocessing includes image enhancement and noise removal, to obtain a preprocessed electronic label image; use OCR technology to perform text recognition on the preprocessed electronic label image, extract the text content printed on the electronic label image, and convert the text content into text format data; obtain preset electronic label quality assessment rules and abnormality judgment thresholds, and perform quality analysis on the text format data according to the electronic label quality assessment rules and abnormality judgment thresholds to determine whether the electronic label image has printing abnormalities; if the electronic label image has printing abnormalities, generate abnormal signal data, wherein the abnormal signal data includes the abnormality type and detection time; transmit the abnormal signal data to the process control system in real time through a predefined communication protocol and interface; after receiving the abnormal signal data, the process control system takes corresponding measures according to a preset abnormality handling process, wherein the measures include suspending the production line and isolating abnormal products.

[0034] For example, capturing electronic label images is the starting point for quality inspection. High-resolution industrial cameras can capture label details, ensuring the quality of the underlying data for subsequent processing. During image preprocessing, Gaussian filtering is used to remove noise, improve contrast, and enhance text clarity. For example, applying a 3x3 Gaussian kernel to a 1920x1080 resolution label image for smoothing effectively reduces image noise while preserving edge information. Optical character recognition (OCR) technology is the core of text recognition. Open-source OCR engines such as Tesseract are highly efficient at recognizing printed text. Customized models can be trained to improve recognition accuracy based on the characteristics of electronic labels. For example, for labels containing information such as product numbers and production dates, predefined character sets and format templates can increase recognition accuracy to over 99%. Designing quality assessment rules is crucial. Scoring criteria can be developed based on multiple dimensions, such as character integrity, spacing uniformity, and contrast. A score below 85 is considered the passing score, with an exception alert triggered. For example, if a recognized 12-digit product number contains one missing character or two or more characters are blurred, the product is considered defective. Abnormal alert mechanisms must be fast and accurate. Message queues such as RabbitMQ can be used to transmit abnormal signals in real time, ensuring millisecond-level feedback to the process control system. Signal data should include key information such as the abnormality type, detection time, and image location to facilitate subsequent location and processing. Upon receiving the abnormality signal, the process control system takes appropriate action according to pre-set procedures. For example, if three consecutive defective products are detected, the production line is automatically paused. For a single abnormality, a robotic arm can move the product to an isolation area. This intelligent processing ensures production efficiency while minimizing the risk of defective products being released. Real-time monitoring and quality inspection form a closed-loop management system. By establishing an analytical model linking production parameters with quality data, potential quality issues can be predicted. For example, if a trend of declining print quality is detected, the system can proactively adjust printing pressure or replace ink, preventing defective products from occurring at the source. This proactive control significantly improves production efficiency and product quality. Throughout the entire process, real-time and accurate data are crucial. High-speed data transmission channels ensure timely information delivery, while distributed storage systems guarantee data reliability and traceability. This not only meets the needs of immediate quality control but also provides valuable data support for subsequent production optimization and quality improvement.

[0035] In step S106, the process control system receives the abnormal signal and adjusts the operating parameters of the printing, laminating, and cutting processes to ensure smooth process connection and avoid imbalance in production rhythm.

[0036] The real-time operation status data of each process is obtained, and when an abnormal signal is received, the automatic adjustment mechanism is triggered; according to the preset abnormal handling rules and process parameter range, combined with the association model established by the machine learning algorithm, the optimal adjustment parameters of the printing, laminating and cutting processes are calculated; the calculated adjustment parameters are sent to the equipment control unit of the corresponding process, and the operating parameters of the equipment, including printing speed, laminating pressure, and cutting force, are modified in real time; image recognition technology is used to judge the smoothness of the transfer of semi-finished products between processes after parameter adjustment. If any connection abnormality is found, the relevant parameters are further fine-tuned; according to the real-time production efficiency and material consumption rate of each process, the production rhythm is dynamically predicted, and in the event of a rhythm imbalance, measures such as adjusting the production speed of upstream and downstream processes and increasing buffer inventory are taken in advance; the production process data is transmitted to the big data platform in real time, and the control strategy and abnormal handling model are optimized through data mining and statistical analysis, so as to continuously improve the system's adaptability and control accuracy, and ensure the long-term stable operation of the production line.

[0037] For example, the process control system is the core of an electronic label production line, monitoring the operating status of each process in real time to ensure production quality and efficiency. When the system receives an abnormality signal, it triggers an automatic adjustment mechanism to address various production process issues. For example, in the printing process, the system may detect blurred characters. In this case, the control system calculates the optimal printing parameter adjustment based on preset rules. This may include reducing the printing speed, increasing the ink supply, or adjusting the printing pressure. For example, reducing the printing speed from 100 to 80 sheets per minute while increasing the printing pressure by 5% to ensure character clarity. During the laminating process, if the bond strength between the electronic chip and the label substrate is insufficient, the system may increase the laminating pressure and time. Specific adjustments may include increasing the laminating pressure from 2 MPa to 2.5 MPa and extending the laminating time from 1.5 seconds to 2 seconds to improve bond strength. Adjustments in the cutting process may involve the frequency of cutting tool replacement and the cutting force. For example, if an uneven cut is detected, the system may reduce the cutting speed and increase the cutting force to improve cutting accuracy. Machine learning algorithms play an important role in parameter optimization. By analyzing historical production data, the system can establish a correlation model between process parameters and product quality. For example, support vector machine algorithms can help identify key parameter combinations that lead to product defects, while decision tree algorithms can be used to develop parameter adjustment strategies for different scenarios. The system also focuses on the connection between processes. Using image recognition technology, the system monitors the transfer of semi-finished products between processes in real time. If any problems are detected, such as labels shifting on the conveyor, the system will fine-tune the conveyor speed or adjust the guides to ensure production line continuity. Dynamic prediction of the production rhythm is crucial for maintaining a balanced production line. The system analyzes the production efficiency and material consumption rates of each process to predict potential bottlenecks. For example, if it predicts that the speed of the printing process may cause material shortages in the subsequent lamination process, the system will increase the printing speed in advance or add buffer inventory between the two processes. Big data analysis plays a key role in continuous optimization. The system transmits various production process data, such as equipment operating parameters, product quality indicators, and material consumption, to a big data platform in real time. Using data mining techniques, the system can identify potential optimization areas. For example, analysis may reveal that a specific parameter combination can improve production capacity while maintaining quality, thus guiding future parameter adjustment strategies. This intelligent process control system can not only quickly respond to production anomalies, but also continuously improve the overall efficiency of the production line through continuous learning and optimization, achieving long-term stable and high-quality electronic label production.

[0038] In step S107, the operating status data of each process is obtained in real time through the data acquisition module, and a time series analysis method is used to determine whether the production rhythm is stable. If fluctuations are found, the process parameters are dynamically adjusted.

[0039] Based on pre-established rules for collecting operational status data for each process on the production line, operational status data for each process on the production line is acquired in real time and stored in chronological order in a production database. A time series analysis algorithm is used to calculate the data feature values ​​at each time point for each process status data in the production database. The calculated results are compared with a preset production rhythm stability threshold to determine whether the current production rhythm is stable. If the time series analysis results indicate fluctuations in the current production rhythm, a dynamic process parameter adjustment mechanism is triggered. Based on the correlations between the processes, the target process to be adjusted and its corresponding parameter range are determined. Using a deep reinforcement learning algorithm, combining historical production data with real-time collected status data, the parameter combination for the target process is continuously tested and optimized until the optimal parameter value that restores the production rhythm to stability is found. The optimized process parameter values ​​are distributed to the production line control system, which adjusts the corresponding equipment operating parameters to ensure that the target process operates according to the optimized parameter values. During the dynamic process parameter adjustment process, the operational status data of each process is continuously monitored, and a time series anomaly detection algorithm is used to determine whether the adjusted production rhythm has returned to a stable state, thereby evaluating the effectiveness of the parameter adjustment. If the production rhythm has not returned to stability within the preset time, the process of dynamic adjustment of process parameters will be repeated until the production process returns to a stable state.

[0040] For example, the collection of operating status data of each process on the production line is the basis for realizing intelligent manufacturing. Taking the printing and packaging production line as an example, the ink volume, temperature, pressure and other parameters of the printing machine, the speed, tension, temperature and other data of the laminating machine, and the tool position, cutting force and other information of the cutting machine can be collected in real time through the sensor network. These data are stored in the production database at a frequency of seconds or milliseconds to form a continuous time series. Time series analysis algorithms can extract valuable features from these data. For example, the short-term trend of the ink volume of the printing machine is calculated by the moving average method, or the changing trend of the tension of the laminating machine is predicted by the exponential smoothing method. These characteristic values ​​are compared with the preset thresholds. For example, if the fluctuation of the ink volume of the printing machine exceeds ±

[0041] If the production rhythm fluctuates by more than 5%, the system will be deemed to be experiencing an anomaly. When an anomaly is detected, the system triggers a dynamic adjustment mechanism for process parameters. For example, in the printing, laminating, and cutting processes, if unstable ink levels in the printing process lead to reduced laminating quality, the system will prioritize adjusting the printing process parameters. A deep reinforcement learning algorithm continuously tries different combinations of parameters, such as ink supply speed and printing pressure, to identify the optimal solution for consistent printing quality. During parameter optimization, the algorithm considers successful historical production data. For example, if a combination of increasing the ink supply speed by 10% and reducing the printing pressure by 5% has previously yielded good results under similar circumstances, this parameter combination will be prioritized and validated. The algorithm will also explore new parameter combinations to adapt to the current situation. The optimized parameters are then distributed to the production line control system. For example, in the case of a printing press, the control system will adjust specific equipment parameters, such as the ink roller speed and impression cylinder pressure. These adjustments aim to stabilize the printing process, thereby restoring a smooth rhythm for the entire production line. During the parameter adjustment process, the system continuously monitors data from each process. Time series anomaly detection algorithms, such as the Local Outlier Factor (LOF) method, can determine whether abnormal fluctuations in adjusted production data persist. If the production rhythm fails to stabilize within a preset time window (e.g., 15 minutes), the system restarts the parameter optimization process until a more suitable parameter combination is found. The advantage of this dynamic adjustment mechanism is that it can quickly respond to production fluctuations, reduce manual intervention, and improve production efficiency and product quality stability. Through continuous learning and optimization, the system can continuously improve its ability to respond to various production anomalies, providing strong support for enterprises to achieve intelligent manufacturing and lean production.

[0042] In step S108, the adjusted process parameters are compared with the preset product consistency standards. If the parameter deviation exceeds the preset range, the automatic calibration module is activated to ensure product consistency.

[0043] Obtain the actual parameter value of the current process, compare it with the standard value of the corresponding parameter in the preset product consistency standard, and calculate the parameter deviation value; determine whether the parameter deviation value exceeds the preset allowable range. If it exceeds the allowable range, trigger the automatic calibration module; the automatic calibration module uses the gradient descent algorithm to adjust the corresponding process parameters in an incremental or decremental manner according to the positive or negative situation of the parameter deviation value until the parameter deviation value is reduced to the preset range; during the parameter adjustment process, obtain the change value of the process parameter in real time, compare it with the preset change threshold, and dynamically control the step size of the parameter adjustment according to the comparison result; after the parameter adjustment is completed, start according to the latest process parameter settings The trial production procedure conducts sampling inspection on product consistency; the support vector machine algorithm is used to comprehensively evaluate the various quality indicators of the sampled products to determine whether the product consistency meets the standards; if the product consistency test results meet the preset standards, the current process parameter settings are saved as new standard parameter values; if they do not meet the standards, the parameter automatic calibration process is re-executed until the product consistency stabilizes at a qualified level; a knowledge base for automatic calibration of process parameters is established to record the optimal parameter combinations under various process conditions and their corresponding product quality data; when the production process or raw materials change, the system automatically searches the knowledge base, matches similar working conditions, and calls the corresponding parameter settings to achieve rapid optimization of process parameters.

[0044] For example, an automatic process parameter calibration system is key to ensuring product consistency. The system first obtains the actual process parameter values, such as the injection molding machine's temperature and pressure, and compares them with preset product standards. For example, the standard mold temperature for a certain plastic product is 180°C, with an allowable deviation of ±5°C. The system monitors the temperature in real time. If the actual temperature is 188°C, exceeding the allowable range, the automatic calibration module is triggered. The automatic calibration module uses a gradient descent algorithm to adjust the parameters. In the above example, due to high temperature, the system gradually reduces the heating power. During the adjustment process, the system monitors the temperature change rate in real time. If the temperature drops too quickly (e.g., by more than 2°C per minute), the adjustment step size is reduced to prevent a sudden temperature drop from affecting product quality. After parameter adjustment is completed, the system initiates trial production. For example, in the production of plastic bottles, samples are taken for testing of dimensions, weight, strength, and other indicators. A support vector machine algorithm is used to comprehensively evaluate these indicators to determine whether product consistency meets the standards. If key indicators such as the bottle mouth diameter and bottle body thickness are found to be within the allowable range and the difference between batches is less than 1%, the consistency is considered to be met and the new temperature setting will be saved as the standard parameter value. The establishment of a knowledge base for automatic calibration of process parameters is crucial for responding to changes in production conditions. For example, when a raw material supplier changes, resulting in a slight change in the melting point of the raw material, the system can quickly search the knowledge base to find the optimal parameter combination for similar situations. This not only reduces debugging time but also ensures the stability of product quality. Through this automated parameter calibration and optimization process, production efficiency and product quality consistency can be greatly improved. It can quickly respond to various changes in the production process, reduce human intervention, and reduce the risk of operational errors. At the same time, through the continuous accumulation and updating of the knowledge base, the system's optimization capabilities will continue to improve, providing strong support for the company's long-term stable production.

[0045] Step S109: The system monitoring module monitors the operating status of the entire production process in real time. If the system stability is found to be degraded, the early warning mechanism is triggered, and a maintenance signal is generated and transmitted to the maintenance system.

[0046] Acquire equipment operating parameter data collected by sensors deployed at various nodes in the production process, including temperature, pressure, and vibration frequency; use an anomaly detection algorithm to determine whether the equipment operating parameters exceed a preset normal range threshold; if so, determine that the equipment operating status is abnormal; if the equipment operating status is abnormal, divide the warning level into at least two levels based on the degree of abnormality of the equipment operating parameters, and generate maintenance signals of corresponding levels; obtain corresponding emergency response strategies from a preconfigured strategy library for the maintenance signals of different warning levels, and generate equipment control instructions based on the emergency response strategies; transmit the maintenance signals and related equipment abnormality information to the maintenance system; After receiving the maintenance signal, the maintenance system obtains the equipment location and abnormal parameter information contained in the maintenance signal, and obtains the historical maintenance record of the equipment; uses a case-based reasoning algorithm to generate an equipment maintenance plan based on the equipment location, abnormal parameter information and historical maintenance records; according to the equipment maintenance plan, guides maintenance personnel to perform maintenance work on the equipment, and feeds back the maintenance results to the maintenance system; the maintenance system updates the equipment maintenance record according to the maintenance results, and transmits the maintenance results to the monitoring module; the monitoring module determines whether the equipment has returned to normal operating status based on the maintenance results. If it has returned to normal, the warning is lifted; otherwise, monitoring continues until the equipment is fully restored.

[0047] For example, real-time collection of production equipment operating parameters is the foundation of intelligent manufacturing. For example, at an automobile manufacturer, various sensors are deployed on the engine assembly line, including temperature sensors to monitor engine cylinder temperature, pressure sensors to detect oil pressure, and vibration sensors to detect abnormal vibrations during assembly. These sensors transmit data in real time to a central monitoring system via the Industrial Internet of Things. The system monitoring module uses statistical anomaly detection algorithms, such as moving average and exponential smoothing, to analyze equipment parameters in real time. For example, the normal operating temperature range of an engine cylinder is 80-120°C. If the temperature exceeds 130°C for a sustained period, the system will identify an abnormal state. The early warning mechanism is designed based on the severity and duration of the abnormality. For example, if the oil pressure drops by 10% for 5 minutes, the system triggers a yellow alert; if it drops by 20% for 10 minutes, it escalates to a red alert. Different alert levels correspond to different response strategies. For example, a yellow alert may simply reduce production speed, while a red alert may require shutdown for maintenance. Upon receiving the alert signal, the maintenance system invokes a case-based reasoning algorithm. The algorithm first searches historical maintenance records for similar cases. For example, abnormal engine oil pressure may be related to oil pump failure, pipeline leaks, and other conditions. The system then matches the most similar historical cases based on the specific parameters of the current abnormality, such as the magnitude of the oil pressure drop and temperature fluctuation, and generates corresponding repair recommendations. During equipment maintenance, maintenance personnel can update the repair progress and results in real time using their mobile devices. For example, if an oil pressure abnormality is found to be caused by a loose oil pipe joint, the maintenance personnel will enter this information into the system after tightening the joint. The system then updates the equipment maintenance record, which not only facilitates future fault diagnosis but also optimizes preventive maintenance plans. The entire process embodies the closed-loop control concept of intelligent manufacturing. From data collection, anomaly detection, early warning triggering, maintenance guidance, and result feedback, a complete information flow is formed. This approach not only enables timely detection and resolution of equipment failures but also enables continuous optimization of production processes through data accumulation, improving equipment utilization and product quality. For example, by analyzing the frequency of temperature anomalies on an engine assembly line, it may be possible to identify certain batches of parts that are more prone to overheating, thereby improving product design or adjusting supplier selection at the source. The ultimate goal of this intelligent equipment monitoring and maintenance system is to achieve predictive maintenance. By analyzing massive amounts of historical data using machine learning algorithms, the system can predict when equipment is likely to fail, proactively scheduling maintenance before a failure occurs, minimizing downtime and improving production efficiency.

[0048] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above and that the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the present application is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A method for controlling the production quality of electronic labels, characterized in that: The method comprises the following steps: S101. Acquire a first image of an electronic label production process using an image acquisition device, and process the first image using an adaptive illumination compensation algorithm to account for interference from light variations, thereby obtaining a second image. S102, using a multi-scale denoising algorithm to perform noise reduction processing on the second image in response to vibration interference and dust interference, to obtain a third image; S103: Input the third image into a preset OCR recognition model. If there is an error in the OCR recognition result, locally optimize the third image using an image enhancement algorithm to obtain a fourth image, and perform OCR recognition again. S104, transmitting the OCR recognition results to the online detection system via a high-speed data transmission channel. If the data transmission delay exceeds a preset threshold, a data compression algorithm is activated to reduce the data volume to ensure real-time processing capability; S105, the online detection system determines the printing quality of the electronic label based on the OCR recognition result. If an abnormality is detected, an abnormality signal is generated and transmitted to the process control system; S106: The process control system receives abnormal signals and adjusts the operating parameters of the printing, laminating, and cutting processes to ensure smooth process connection and avoid imbalance in production rhythm; S107. The operating status data of each process is acquired in real time through the data acquisition module. A time series analysis method is used to determine whether the production rhythm is stable. If fluctuations are found, the process parameters are dynamically adjusted. S108, comparing the adjusted process parameters with the preset product consistency standards. If the parameter deviation exceeds the preset range, the automatic calibration module is activated to ensure product consistency; S109. The operating status of the entire production process is monitored in real time through the system monitoring module. If the system stability is found to be degraded, the early warning mechanism is triggered, and a maintenance signal is generated and transmitted to the maintenance system.

2. The method for controlling the production quality of electronic labels according to claim 1, characterized in that: The S101 includes: Obtain original images during the electronic label production process; Determining whether the original image is subject to illumination interference according to a preset illumination change threshold; If the original image is disturbed by light, an adaptive light compensation algorithm is triggered, wherein the adaptive light compensation algorithm includes: analyzing the brightness distribution characteristics of the original image and determining light compensation parameters; Adopting a Retinex algorithm combined with an adaptive histogram equalization method to adaptively adjust the brightness and contrast of a local area of ​​the original image according to the illumination compensation parameters to obtain a compensated image; Performing quality assessment on the compensated image using a convolutional neural network to obtain an assessment result; Dynamically optimizing the parameters of the adaptive illumination compensation algorithm according to the evaluation results; The illumination-compensated image is transmitted to the electronic tag defect detection module.

3. The method for controlling the production quality of electronic labels according to claim 1, characterized in that: The S102 includes: acquiring a second image to be processed; Determining, according to the noise type of the second image, to adopt a multi-scale denoising algorithm for noise reduction processing; Performing multi-scale decomposition on the second image to obtain image features at different scales; Get the preset denoising algorithm for vibration interference and dust interference; Adaptively adjusting the parameters of the denoising algorithm according to the image features at different scales to obtain adaptive denoising algorithms at different scales; Adopting the adaptive denoising algorithm at different scales to perform denoising on the image features at different scales to obtain denoised images at different scales; fusing the denoised images at different scales to obtain a denoised third image; Get the preset image quality threshold; Determining whether the quality of the third image reaches the preset image quality threshold; If yes, output the third image as the final noise reduction result; If not, return to the step of adaptively adjusting the parameters of the denoising algorithm and re-perform the multi-scale denoising process until the quality of the third image reaches the preset image quality threshold.

4. The method for controlling the production quality of electronic labels according to claim 1, characterized in that: The S103 includes: Obtaining a target image, inputting the target image into a preset OCR recognition model for text recognition processing, and obtaining an OCR recognition result of the target image; Determining whether there is a recognition error in the OCR recognition result of the target image, and if there is a recognition error, determining that the target image needs to be optimized; For the local areas of the target image where OCR recognition errors exist, an image enhancement algorithm is used to optimize the areas, and an optimized image is obtained through image sharpening and contrast enhancement operations; Re-inputting the optimized image into the preset OCR recognition model for text recognition processing to obtain an OCR recognition result of the optimized image; Determining whether the OCR recognition result of the optimized image has an improved recognition accuracy compared to the OCR recognition result of the target image, and if the recognition accuracy is improved, determining that the optimization process is effective; Extracting text information from the image based on the OCR recognition result of the optimized image, and performing structured processing on the text information to obtain structured text recognition result data; The structured text recognition result data is stored in a designated database.

5. The method for controlling the production quality of electronic labels according to claim 4, characterized in that: The S104 includes: Obtaining OCR recognition results, packaging the OCR recognition result data into a preset format, and sending it to the online detection system through a high-speed data transmission channel; The online detection system receives the OCR recognition result data, parses the OCR recognition result data, extracts the OCR recognition result information, and stores the OCR recognition result information in a result cache queue; Start the data transmission monitoring thread to obtain the current data transmission delay. If the data transmission delay exceeds a preset threshold, trigger the data compression algorithm; The data compression algorithm compresses the OCR recognition result data in the result cache queue, using a lossless compression algorithm to reduce the amount of transmitted data while ensuring data integrity; Repackaging the compressed OCR recognition result data, sending it to the online detection system through the high-speed data channel, and updating the delay threshold in the data transmission monitoring thread; The online detection system receives the compressed OCR recognition result data, decompresses it, restores the original OCR recognition result, and enters the subsequent real-time analysis and processing process; The analysis result is associated with the original OCR recognition result and stored, persisted through a distributed cache system, and an index of the OCR recognition result and the analysis result is established based on a time series database.

6. A method for controlling the production quality of electronic labels according to any one of claims 1 to 5, characterized in that: The S105 includes: Acquire image data of the electronic tag, and preprocess the image data, wherein the preprocessing includes image enhancement and noise removal, to obtain a preprocessed electronic tag image; Using OCR technology to perform text recognition on the pre-processed electronic label image, extract the text content printed on the electronic label image, and convert the text content into text format data; Obtaining preset electronic label quality assessment rules and abnormality judgment thresholds, performing quality analysis on the text format data according to the electronic label quality assessment rules and abnormality judgment thresholds, and determining whether the electronic label image has printing abnormalities; If there is a printing abnormality in the electronic label image, abnormal signal data is generated, and the abnormal signal data includes the abnormality type and the detection time; Transmitting the abnormal signal data to the process control system in real time through a predefined communication protocol and interface; After receiving the abnormal signal data, the process control system takes corresponding measures according to a preset abnormality handling process, and the measures include suspending the production line and isolating abnormal products.

7. A method for controlling the production quality of electronic labels according to any one of claims 1 to 5, characterized in that: The S106 includes: Obtain real-time operating status data for each process and trigger an automatic adjustment mechanism when an abnormal signal is received; Based on the preset exception handling rules and process parameter ranges, combined with the correlation model established by the machine learning algorithm, the optimal adjustment parameters for the printing, laminating and cutting processes are calculated; The calculated adjustment parameters are sent to the equipment control unit of the corresponding process to modify the equipment's operating parameters in real time, including printing speed, lamination pressure, and cutting force; Image recognition technology is used to determine the smoothness of the transfer of semi-finished products between processes after parameter adjustment. If any abnormal connection is found, the relevant parameters will be further fine-tuned; Based on the real-time production efficiency and material consumption rate of each process, the production rhythm is dynamically predicted to anticipate any imbalance in the rhythm. Measures such as adjusting the production speed of upstream and downstream processes and increasing buffer inventory are taken in advance. Transmit production process data to the big data platform in real time, optimize control strategies and exception handling models through data mining and statistical analysis, continuously improve the system's adaptability and control accuracy, and ensure the long-term stable operation of the production line.

8. The method for controlling the production quality of electronic labels according to any one of claims 1 to 5, characterized in that: The S109 includes: Obtaining equipment operating parameter data collected by sensors deployed at each node in the production process, including temperature, pressure, and vibration frequency; An abnormality detection algorithm is used to determine whether the operating parameters of the device exceed a preset normal range threshold, and if so, the operating state of the device is determined to be abnormal; If the operating state of the equipment is abnormal, the warning level is divided into at least two levels according to the abnormality degree of the equipment operating parameters, and a maintenance signal of the corresponding level is generated; For the maintenance signals of different warning levels, corresponding emergency processing strategies are obtained from a pre-configured strategy library, and equipment control instructions are generated according to the emergency processing strategies; Transmitting the maintenance signal and related equipment abnormality information to a maintenance system; After receiving the maintenance signal, the maintenance system obtains the device location and abnormal parameter information contained in the maintenance signal, and obtains the historical maintenance record of the device; Using case-based reasoning algorithms, a maintenance plan is generated based on the equipment location, abnormal parameter information, and historical maintenance records. Instruct maintenance personnel to perform maintenance work on the equipment according to the equipment maintenance plan, and feed back maintenance results to the maintenance system; The maintenance system updates the equipment maintenance record according to the maintenance result and transmits the maintenance result to the monitoring module; The monitoring module determines whether the equipment has returned to normal operation based on the maintenance result. If it has returned to normal, the warning is lifted; otherwise, the monitoring module continues until the equipment is fully restored.

Images