Paperboard production batch tracking management method based on RFID

Abstract

The invention relates to the technical field of information tracking management, and discloses an RFID-based paperboard production batch tracking management method, which comprises the steps of judging whether a label of a target product is damaged or not, if the label is damaged, analyzing the batch attribution of the target product, and supplementing the label to the target product according to an analysis result. Damaged labels are identified and repaired in time, delay caused by label problems in the production process is avoided, batch attribution of products with damaged labels is accurately judged, production batch tracking management is optimized, label information integrity of each product is ensured, accuracy and reliability of production data are improved, quality problems caused by batch confusion are avoided, and production efficiency is improved. The inventory and production plan management is optimized, the resource utilization efficiency is improved, the judgment accuracy and the system reliability are improved, the processing capacity of the system for abnormal conditions is enhanced, the stability of the production process is ensured, and the time and errors of manual inspection are reduced.

Description

A method for batch tracking and management of cardboard production based on RFID Technical Field

[0001] This invention relates to the field of information tracking and management technology, specifically to an RFID-based method for tracking and managing cardboard production batches. Background Technology

[0002] Cardboard, as an important packaging material, is widely used in logistics, warehousing, and manufacturing. Effective batch tracking and management during cardboard production is crucial for ensuring product quality, tracing the source of quality problems, optimizing production processes, and meeting compliance requirements (such as food and drug safety). Traditional cardboard batch management typically relies on manual recording, barcode scanning, or simple ERP systems. The core of RFID-based cardboard production batch tracking and management lies in utilizing Radio Frequency Identification (RFID) technology to attach RFID tags containing unique identifiers (EPC - Electronic Product Code) to key entities in the production process (such as raw material rolls, cardboard rolls / sheets in production, finished packaging, and production equipment). RFID readers automatically and non-contactly read this tag information via radio waves and upload the data to the back-end management system. The backend system analyzes, integrates, and correlates the collected data to achieve refined tracking, monitoring, and management of cardboard production batches. By introducing RFID technology at key nodes such as cardboard, raw materials, and production equipment, it enables autonomous data collection, real-time updates, and end-to-end connectivity, thereby constructing an efficient, accurate, and visualized cardboard production batch tracking and management system. However, this RFID-based cardboard production batch tracking and management method has limitations. When tags are missing or damaged, it cannot determine whether the product is located at the boundary between two batches during the production process. It cannot analyze and determine the batch information of the product based on its location during the production process, nor can it infer the true information of the product and regenerate tags. This can easily lead to delays in the production process due to tag issues and quality problems due to batch confusion, reducing the accuracy and reliability of production data. Its practicality has certain limitations. Summary of the Invention

[0003] This invention provides an RFID-based method for batch tracking and management of cardboard production, which helps to solve the problems mentioned in the background art.

[0004] This invention provides the following technical solution: a method for batch tracking and management of cardboard production based on RFID, comprising: acquiring a label for a target product and designating it as a target label; acquiring a database. Determine if the label of the target product is damaged; if the label is damaged, analyze the batch attribution of the target product; supplement the label of the target product according to the analysis results; if the label is normal, use it normally.

[0005] As an optional solution to the RFID-based batch tracking and management method for cardboard production described in this invention, determining whether the label of the target product is damaged specifically involves: acquiring an image of the target label, designated as the target label image; preprocessing the acquired target label image; extracting the image edges to obtain an edge binary image, denoted as... Initialize a two-dimensional array, denoted as . Traversing the edge binary graph Each pixel in Find connected edge pixel regions; count the number of connected regions, denoted as . Count the number of pixels in each connected region, denoted as . Define a continuity decision function. Determine whether the image edges are continuous; if If the edge is continuous, then the edge is determined to be continuous; if If the edge is discontinuous, then the edge is determined to be discontinuous; a contour is fitted to the binary edge image; the difference between the fitted curve and the actual edge pixels is calculated and denoted as . ; Determine the function by rules Determine if the image edges are regular; if Then determine the edge rule; if If the edge is irregular, then the edge is determined to be irregular; define a complete determination function. Determine if the image edges are complete; if If the image edges are incomplete, the label is considered damaged. If the image edges are intact, then the image content of the target label image is determined.

[0006] As an optional solution to the RFID-based paperboard production batch tracking and management method of the present invention, the image content determination of the target label image specifically involves: acquiring a preprocessed image. Detect the presence of barcodes in the image; detect the presence of QR codes in the image; recognize text information in the image; for barcodes and QR codes, check if they can be successfully decoded and the complete information can be obtained; if a barcode exists in the image and is successfully and completely recognized, output... If the image does not contain a barcode, or the barcode cannot be successfully recognized, or the barcode is not fully recognized, then output: If a QR code exists in the image, and the QR code is successfully and completely recognized, then output: If the image does not contain a QR code, or the QR code cannot be successfully recognized, or the QR code is not fully recognized, then output: For text information, check if the key text content can be identified; if the key text content can be identified in the image and the text content is complete, then output the result. If the key text content cannot be identified in the image, or the text content is not fully identified, then output: By identifying the decision function Determine if the image content is complete; if If the image content is incomplete, the label is damaged. If the image content is found to be complete, then scanning and information verification will be performed.

[0007] As an optional solution to the RFID-based paperboard production batch tracking and management method of the present invention, the scanning and information verification are performed specifically as follows: scanning the target tag to obtain the storage information of the target product, which is defined as the tag information and denoted as [label information]. ; Retrieve database Extract the correct entry format for the target product when entering it into the database, and designate it as the reference format, denoted as . Extract tag information The format is defined as the information format, denoted as... ; using format comparison functions Determine if the entered tag information data is complete; if If so, the entered label information is considered complete; if If the data entered is incomplete, it is determined that the label information is incomplete; this is determined by a unique judgment function. It determines whether there are multiple entries for the tag information in the database; among them, For database The number of data entries that match the tag information; if If so, it is determined that there are multiple entries for the label information in the database; If the label information is determined to exist only once in the database, this entry will be designated as the comparison data, denoted as [reference data]. ; through data comparison functions Determine whether the label information matches the comparison data; if If the label information matches the comparison data, then it is determined that the label information is consistent with the comparison data; if If the label information is inconsistent with the comparison data, it is determined that the label information is inconsistent with the comparison data; the determination is made through a comprehensive function. Determine if the label is normal; if If the label is damaged, then it is determined that the label is damaged; if If so, the label is considered normal.

[0008] As an optional solution to the RFID-based cardboard production batch tracking and management method described in this invention, the following steps are included: analyzing the batch affiliation of the target product, specifically: determining whether the batch of the target product is in a special position; if the batch of the target product is in a special position, then analyzing the actual information of the target product; if the batch of the target product is in a normal position, then extracting the two products adjacent to the target product and designating them as the left and right products, respectively; and acquiring the database. Extract batch information for the products on the left and right sides from the database, and denot them as follows: and ; using batch comparison function Determine if the batch information of the two products is consistent; if If the batch information of the two products is consistent, then the batch information of the target product is determined and denoted as... Combining the product information of the products on the left and right sides with the batch information of the target product, infer the product information of the target product and regenerate the label; if If the batch information of the two products is inconsistent, then the true information of the target product will be analyzed.

[0009] As an optional solution to the RFID-based paperboard production batch tracking and management method of the present invention, the method for determining whether the batch of the target product is in a special position specifically involves obtaining the timestamp and serial number of the target product, which are denoted as follows: and ; Obtain the batch switching time point, denoted as Calculate the time difference between the target product timestamp and the batch switching time point, denoted as . ; determined by time function Inspection time difference Does it meet the requirements? Then determine the time difference. Meets the requirements; if Then determine the time difference. Not in compliance with requirements; obtain the starting serial number of the known batch and record it as... ; Obtain the end serial number of a known batch, and record it as . Calculate the difference between the target product serial number and the starting serial number, and denote it as . Calculate the difference between the target product serial number and the ending serial number, and denote it as . ; through information determination function Check the difference and Does it meet the requirements? If the difference is found to meet the requirements, then the difference is considered to be acceptable; if If the difference does not meet the requirements, then the determination function based on the parameter synthesis is used. Determine whether the batch of the target product is in a special position; if If so, the target product is determined to be in a special position.

[0010] like If so, the target product is determined to be in a normal position.

[0011] As an optional solution to the RFID-based paperboard production batch tracking and management method of the present invention, the following steps are taken: analyzing the true information of the target product specifically involves obtaining the timestamp of the target product, denoted as... ; Obtain the start and end timestamps of the preceding and following batches, and record them as follows: , , as well as ; Calculate the overlap range between the timestamp of the target product and the time of the previous batch, denoted as . ; Calculate the overlap range between the timestamp of the target product and the time of the subsequent batch, denoted as . Obtain the serial number of the target product. Obtain the start and end serial numbers of the preceding and following batches, and record them as follows: , , as well as Calculate the similarity between the target product's serial number and the previous batch, and record it as follows: Calculate the similarity between the target product's serial number and the subsequent batch, and record it as follows: .

[0012] As an optional solution to the RFID-based cardboard production batch tracking and management method of the present invention, the method further includes: analyzing the real information of the target product, and obtaining the warehousing time of the target product, denoted as... ; Obtain the allocation record of the target product, and record it as ; Obtain the inbound time and allocation records of the previous and subsequent batches, and record them as follows: , , as well as Compare the matching degree between the target product's warehousing time and allocation records and the previous batch, and record it as follows: Compare the matching degree between the target product's warehousing time and allocation records and the subsequent batch, and record it as follows: ; Obtain the physical, technological, and dimensional characteristics of the target product, and denot them as ; Obtain the features of the upper and lower batches of samples, and denot them as follows: and Compare the characteristics of the target product with those of the previous batch, and record the similarity as follows: Compare the characteristics of the target product with the similarity of the subsequent batch, and record the result as follows: Calculate the combined score of the target product and the previous batch, and record it as follows: Calculate the combined score of the target product and the previous batch, and record it as follows: ;; through the attribution determination function Determine the batch information of the target product; if If so, the target product is determined to belong to the previous batch. ;like If so, the target product is determined to belong to the later batch. ;like If not, manual verification is performed; by combining the product information of the products on the left and right sides with the batch information of the target product, the product information of the target product is inferred, and a new label is generated.

[0013] As an optional solution to the RFID-based cardboard production batch tracking and management method of the present invention, the following steps are taken: combining the product information of the products on the left and right sides with the batch information of the target product, the product information of the target product is inferred, and a new tag is generated; specifically, the product identifier and timestamp of the target product are obtained. ,serial number and batch information Retrieve batch information from the database The corresponding general information is denoted as The known information about the target product is integrated with the general information of the batch to form a complete product information structure, denoted as... The integrated and complete product information will be output, tags will be regenerated, and the database will be updated.

[0014] This invention has the following beneficial effects: 1. This RFID-based batch tracking and management method for cardboard production extracts image edges using edge detection algorithms (such as the Canny algorithm), determines whether the edges are continuous and regular, and checks the completeness of image content using barcode, QR code, and text information recognition algorithms. If the image edges or content are incomplete, the label is deemed damaged. If the image is complete, the label is scanned to obtain product information. If complete information cannot be scanned, the label is deemed damaged. If complete information can be scanned, the scanned information is compared with standard information in the database. If they match and there are no multiple corresponding data in the database, the label is normal; otherwise, the label is damaged. If the label is normal, the product can be used directly. If the label is damaged, the label of the target product will be regenerated based on the analysis results. The system can quickly identify whether there are obvious defects in the image, reduce the amount of unnecessary work in subsequent processing, and ensure that the image processed in subsequent steps is complete through edge and content integrity checks. Multi-level verification through image integrity, scan results and database comparison ensures the accuracy of label status judgment, quickly identifies damaged labels, provides a basis for subsequent batch attribution analysis and label repair, and regenerates accurate labels for products with damaged labels to ensure the integrity and traceability of product information. For products with normal labels, they can be directly put into the normal production process to improve production efficiency.

[0015] 2. This RFID-based batch tracking management method for cardboard production compares the target product's timestamp with the batch switching time to check if it falls within a certain range before and after the switching time. It also compares the target product's serial number with the start and end serial numbers of known batches to check if it is close to the batch boundary. The method consults the equipment logs to confirm the specific time record of the batch switching. If there are conflicts or ambiguities in the information, the method consults manual records or asks the operator to determine if the target product is in a special position during the production process, i.e., the boundary between two batches. By analyzing the timestamp, serial number, and equipment logs, the method accurately identifies products at the batch boundary. Combining multi-dimensional information such as time, serial number, inventory, and manual verification improves the accuracy of batch attribution determination. The method flexibly selects analysis methods based on product location and information completeness to improve the system's adaptability and efficiency.

[0016] 3. This RFID-based batch tracking management method for cardboard production, if the target product is in a special position during production, identifies the overlap between the product's production time and the time of the preceding and following batches. Combining equipment logs and serial numbers, it analyzes the close correlation between the product and a specific batch, examines the product's final warehousing and allocation information, infers the batch it most likely belongs to, selects samples from upstream and downstream batches, and confirms the affiliation through comparison of physical, technological, and specification characteristics. It integrates time, serial number, equipment logs, and inventory classification information to verify batch affiliation and analyze the true information. If the target product is in a normal position during production, it obtains the batch information of products on both sides of the target product and compares the batch information. If they match, it infers that the target product belongs to that batch; if they do not match, it implements measures to indicate that the target product is in a special position during production and analyzes the true information. Through timestamp, serial number, and equipment log analysis, it accurately identifies products at the batch boundary. Combining multi-dimensional information such as time, serial number, inventory, and manual verification improves the accuracy of batch affiliation judgment. Based on product location and information completeness, it flexibly selects analysis methods to improve the system's adaptability and efficiency. Attached Figure Description

[0017] Figure 1 is a flowchart of the RFID-based batch tracking and management method for cardboard production according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0019] Example 1: A method for batch tracking and management of cardboard production based on RFID, referring to Figure 1, includes: acquiring the tag of the target product and designating it as the target tag; acquiring the database. The database stores relevant data for each product during the production process, such as production time, production line location, warehousing and allocation information, for tracking and management; it determines whether the label of the target product is damaged; if the label is damaged, it analyzes the batch affiliation of the target product; it replenishes the label of the target product based on the analysis results; if the label is normal, it is used normally.

[0020] The analysis of target product batch attribution specifically involves: determining whether the target product batch is in a special position, where a special position is the boundary between two batches, such as the last product of the previous batch and the first product of the next batch; if the target product batch is in a special position, then analyzing the target product's actual information; if the target product batch is in a normal position, then extracting the two products adjacent to the target product and designating them as the left and right products, respectively; and acquiring the database. Extract batch information for the products on the left and right sides from the database, and denot them as follows: and : ; ;in, This refers to the product to the left of the target product. This refers to the product to the right of the target product. This indicates retrieving batch information of products from the database; using a batch comparison function. Determine if the batch information of the two products is consistent: ;like If the batch information of the two products is consistent, then the batch information of the target product is determined and denoted as... : Combining the product information of the products on the left and right sides with the batch information of the target product, infer the product information of the target product and regenerate the label; if If the batch information of the two products is inconsistent, then the true information of the target product will be analyzed.

[0021] This embodiment also provides a method for determining whether the batch of the target product is in a special position, specifically: obtaining the timestamp and serial number of the target product from the database or production records, and recording them as follows: and ; Obtain the batch switching time point, denoted as Calculate the time difference between the target product timestamp and the batch switching time point, denoted as . : ; determined by time function Inspection time difference Does it meet the requirements? ;in, A set time range threshold, such as 10 minutes, is used to determine whether the time difference between the target product's timestamp and the batch switching time is within an allowable range. This is to determine whether the target product is at the boundary between two batches. If the time difference between the target product's timestamp and the batch switching time is less than or equal to... If so, it is assumed that the target product may be at the batch boundary; if Then determine the time difference. Meets the requirements; if Then determine the time difference. Not in compliance with requirements; obtain the starting serial number of the known batch and record it as... ; Obtain the end serial number of a known batch, and record it as . Calculate the difference between the target product serial number and the starting serial number, and denote it as . : Calculate the difference between the target product serial number and the ending serial number, and denote it as . : ; through information determination function Check the difference and Does it meet the requirements? ;in, This indicates the set threshold range for serial numbers, such as 10 serial numbers. It is used to determine whether the difference between the target product's serial number and the start or end serial number of a known batch is within an allowable range, thus determining whether the target product is at the boundary between two batches. If the difference between the target product's serial number and the start or end serial number of a known batch is less than or equal to... If so, it is assumed that the target product may be at the batch boundary; if If the difference is found to meet the requirements, then the difference is considered to be acceptable; if If the difference does not meet the requirements, then the determination function based on the parameter synthesis is used. Determine if the target product batch is in a special position: ;like time difference If the requirements are met or the difference is met, the target product is determined to be in a special position.

[0022] like time difference If the requirements are not met and the difference does not meet the requirements, the target product is determined to be in a normal position.

[0023] This embodiment also provides analyzing the real information of the target product, specifically: obtaining the timestamp of the target product, denoted as... ; Obtain the start and end timestamps of the preceding and following batches, and record them as follows: , , as well as ; Calculate the overlap range between the timestamp of the target product and the time of the previous batch, denoted as . : ;in, This indicates retrieving the timestamp of the target product. End time of the previous batch The smaller value in the interval is used to calculate the overlap between the target product's timestamp and the previous batch's end time. By comparing the target product's timestamp with the previous batch's end time, it is determined whether the target product might belong to the previous batch. This indicates retrieving the timestamp of the target product. and the start time of the previous batch The larger value in the interval is used to calculate the overlap between the target product's timestamp and the start time of the previous batch. By comparing the target product's timestamp with the start time of the previous batch, it is determined whether the target product might belong to the previous batch. The time overlap between the target product's timestamp and the next batch is calculated and denoted as . : ;in, This indicates retrieving the timestamp of the target product. and the end time of the next batch The smaller value in the interval is used to calculate the overlap between the target product's timestamp and the end time of the next batch. By comparing the target product's timestamp with the end time of the next batch, it is determined whether the target product might belong to the next batch. This indicates retrieving the timestamp of the target product. and the start time of the next batch The larger value in the interval is used to calculate the overlap between the target product's timestamp and the start time of the next batch. By comparing the target product's timestamp with the start time of the next batch, it is determined whether the target product might belong to the next batch; the serial number of the target product is then obtained. Obtain the start and end serial numbers of the preceding and following batches, and record them as follows: , , as well as Calculate the similarity between the target product's serial number and the previous batch, and record it as follows: : Calculate the similarity between the target product's serial number and the subsequent batch, and record it as follows: : .

[0024] This embodiment also provides that, in analyzing the true information of the target product, the method further includes: obtaining the warehousing time of the target product, denoted as... ; Obtain the allocation record of the target product, and record it as ; Obtain the inbound time and allocation records of the previous and subsequent batches, and record them as follows: , , as well as Compare the matching degree between the target product's warehousing time and allocation records and the previous batch, and record it as follows: : ;in, A matching function is used to compare the similarity between the allocation record of a target product and the allocation record of the previous batch. This function determines whether the target product's allocation record matches the previous batch's. The degree of matching in the allocation records helps determine the batch affiliation of the target product. The matching degree between the target product's entry time and allocation record and the subsequent batch is denoted as... : ;in, A matching function is used to compare the similarity between the allocation record of a target product and the allocation record of a subsequent batch. This function determines whether the allocation record of the target product is consistent with that of the subsequent batch. The degree of matching of the allocation records helps to determine the batch affiliation of the target product. Physical, technological, and specification characteristics of the target product are obtained and denoted as follows: ; Obtain the features of the upper and lower batches of samples, and denot them as follows: and Compare the characteristics of the target product with those of the previous batch, and record the similarity as follows: : ;in, A similarity function is used to compare the physical, technological, and specification characteristics of a target product with those of a previous batch of samples. This function determines whether the target product shares the same characteristics as the previous batch and helps to identify the batch to which the target product belongs. The similarity between the characteristics of the target product and the characteristics of a subsequent batch is denoted as... : ;in, A similarity function is used to compare the physical, technological, and specification characteristics of a target product with those of a subsequent batch of samples. This function determines whether the target product's characteristics are consistent with those of the subsequent batch, and the feature similarity helps in identifying the batch affiliation of the target product. A comprehensive score is calculated between the target product and the previous batch, denoted as [score here]. : ;in, Timestamp weighting is used to determine the importance of the target product's timestamp information in batch attribution. Timestamps reflect a product's position on the production timeline; if the target product's timestamp highly overlaps with the time range of a certain batch, its weight is higher. This can strengthen the influence of the time factor, making the judgment more inclined towards this batch. When other factors (such as serial number or feature similarity) are unclear or contradictory, reasonable settings can be implemented. This approach balances the impact of time factors, avoids relying on a single factor for decision-making, and adjusts decisions based on the time stability of the production process and the clarity of time intervals between batches. It allows for a more detailed analysis of the contribution of time factors to the results, assigning higher priority to production processes with strict time management. It can highlight the role of timestamps and accurately locate product batches. The serial number weight is used to measure the importance of the target product's serial number information in batch attribution determination. The serial number reflects the product's sequential position on the production line. If the target product's serial number highly matches the serial number range of a certain batch, the weight is higher. This can enhance the influence of the serial number, making batch determination more reliant on this factor. If the timestamp or other information is unclear, adjustments can be made. This can balance the influence of serial numbers, prevent the judgment process from being swayed by individual factors, and adjust the timing based on the continuity of serial numbers and the intervals between batches. The contribution of the serial number to the results can be adjusted, improving performance in production scenarios where the serial number is strictly continuous and has few interruptions. This can highlight the value of the serial number and allow it to play a greater role in the judgment process. Information weights are assigned to incoming goods to assess the importance of target product incoming and allocation records in batch attribution determination. Product incoming and allocation records reflect the flow and attribution of products in logistics and warehousing. If the allocation records of a target product are highly consistent with a certain batch, the weight is higher. This can highlight the role of inbound allocation information, making the judgment more inclined towards this batch. When production process information such as timestamps and serial numbers is unclear, appropriate settings can be used. It can balance the impact of inbound allocation information, avoid judgments that solely depend on production information, and adjust based on the accuracy and completeness of allocation records. It can alter the contribution of inbound allocation information to the result, assigning a higher value in scenarios where inbound allocation records are accurate and detailed. It can fully play its role in judgment. Feature similarity weights are used to determine the importance of the degree of similarity between the target product and the sample product in terms of physical, technological, and specification characteristics in batch attribution. These characteristics directly reflect the product's production features and quality attributes. If the target product is highly similar to a batch of sample products, its similarity is considered more significant. This can enhance the influence of feature similarity, making the judgment focus more on the characteristics of the product itself. When other factors such as timestamps, serial numbers, and warehousing allocation information are unclear or contradictory, reasonable settings can be implemented. It can balance the influence of feature similarity, avoid misleading judgments by inaccurate or incomplete factors, and adjust based on feature stability and the significance of batch-to-batch differences. The contribution of feature similarity to the results can be modified, improving performance when product features are stable and there are significant batch-to-batch differences. This highlights the role of product characteristics, making them more decisive in the judgment; calculate the comprehensive score of the target product and the previous batch, and record it as... : ; through the attribution determination function Determine the batch information of the target product: ;like If so, the target product is determined to belong to the previous batch. ;like If so, the target product is determined to belong to the later batch. ;like If not, manual verification is performed; by combining the product information of the products on the left and right sides with the batch information of the target product, the product information of the target product is inferred, and a new label is generated.

[0025] This embodiment also provides that, by combining the product information of the products on the left and right sides with the batch information of the target product, the product information of the target product is inferred, and the label is regenerated. Specifically, this involves obtaining the product identifier of the target product, such as product ID and timestamp. ,serial number and batch information Retrieve batch information from the database The corresponding general information, such as production date range, production line number, operator, etc., is recorded as follows: The known information about the target product is integrated with the general information of the batch to form a complete product information structure, denoted as... : The integrated and complete product information will be output, tags will be regenerated, and the database will be updated.

[0026] The above methods automate label integrity detection through image processing and scanning technologies, reducing the time and errors of manual inspection, promptly identifying and repairing damaged labels, and avoiding delays caused by label issues during production. Multi-dimensional information analysis accurately determines the batch affiliation of products with damaged labels, optimizing production batch tracking management, ensuring complete label information for each product, and improving the accuracy and reliability of production data. Detailed analysis and verification reduce batch affiliation errors, preventing quality problems caused by batch confusion, ensuring each product meets production standards, verifying product quality through feature comparison, and reducing product waste caused by label issues through precise batch tracking and label repair. This optimizes inventory and production planning management, improves resource utilization efficiency, and enhances the accuracy of judgment and system reliability by combining multi-dimensional information such as timestamps, serial numbers, equipment logs, and manual records. Dynamic repair and manual verification strengthen the system's ability to handle abnormal situations, ensuring the stability of the production process.

[0027] Example 2 is an improvement upon Example 1. This RFID-based batch tracking and management method for cardboard production determines whether the label of the target product is damaged. Specifically, it involves: acquiring an image of the target label and designating it as the target label image; and preprocessing the acquired target label image, such as grayscale conversion and noise reduction, to improve the accuracy of edge detection. ;in, The collected label images, The image after preprocessing. The preprocessing function includes operations such as grayscale conversion and denoising; the Canny edge detection algorithm is used to extract image edges, resulting in a binary edge map, denoted as . : ;in, Represents edge pixels, Indicates non-edge pixels, This indicates that the Canny algorithm is used for edge detection in the preprocessed image; a two-dimensional array is initialized, denoted as... The two-dimensional array With binary graph The same size is used to mark visited pixels, with initial values ​​all of 0, indicating unvisited pixels; traverse the edge binary map. Each pixel in Find connected edge pixel regions; count the number of connected regions, denoted as . Count the number of pixels in each connected region, denoted as . Define a continuity decision function. Determine if the edges of an image are continuous: ;in, This is a threshold for the number of connected components used to determine the continuity of edges. If the number of connected components exceeds this threshold, it indicates that the edge may be divided into too many small segments and may be incomplete; if... If the edge is continuous, then the edge is determined to be continuous; if If the edge is discontinuous, then the edge is determined to be discontinuous; a contour fit is performed on the binary edge image, such as using polynomial fitting or curve fitting; the difference between the fitted curve and the actual edge pixels is calculated and denoted as . ; Determine the function by rules Determine if the image edges are regular: ;in, This is the difference threshold, used to determine the regularity of the edges. If the difference between the fitted curve and the actual edge pixels exceeds this threshold, it indicates that the edge shape may be irregular; if... Then determine the edge rule; if If the edge is irregular, then the edge is determined to be irregular; define a complete determination function. Determine if the image edges are intact: ;in, This indicates "or", meaning that either condition must be met. This indicates that all of the conditions must be met simultaneously; if If the image edges are incomplete, the label is considered damaged. If the image edges are intact, then the image content of the target label image is determined; specifically, the number of connected regions and the number of pixels in each connected region are counted, for the edge binary image. Each pixel in If the pixel and If a new connected region is found, a depth-first search (DFS) or breadth-first search (BFS) can be used to search the pixel. and edge binary map Perform a search, marking all pixels belonging to the connected region; count the number of pixels in the connected region and store them in an array. ,in Index the connected region; index the pixel Marked as visited, i.e. Each time a new connected region is discovered, Add 1; count all connected regions and the pixels of each connected region to get the number of connected regions. and the number of pixels in each connected region Specifically, calculating the difference between the fitted curve and the actual edge pixels involves: for the edge binary map... Extract the coordinates of all edge pixels and store them in a list. : Choose the appropriate fitting curve type based on the edge shape, such as polynomial, circle, ellipse, etc. Assume the fitting curve is... order polynomial: The parameters of the fitted curve are obtained by performing polynomial fitting on the edge pixel coordinates using the least squares method. ;in, To design the matrix, For edge pixels Coordinate vector; for each edge pixel Calculate its distance to the fitted curve, denoted as . : Calculate the mean squared error (MSE) of the distance between all pixels as the degree of difference: ;in, This represents the number of edge pixels.

[0028] Specifically, determining the image content of the target label image involves: obtaining a preprocessed image. The algorithm uses barcode recognition algorithms (such as Zxing, OpenCV, etc.) to detect the presence of barcodes in the image; it also uses QR code recognition algorithms (such as QR code decoders) to detect the presence of QR codes in the image; and it uses optical character recognition (OCR) technology (such as Tesseract, Google Vision API, etc.) to recognize text information in the image. For both barcodes and QR codes, it checks whether they can be successfully decoded and the complete information can be obtained. If a barcode exists in the image and is successfully and completely recognized, then the algorithm outputs the result. If the image does not contain a barcode, or the barcode cannot be successfully recognized, or the barcode is not fully recognized, then output: If a QR code exists in the image, and the QR code is successfully and completely recognized, then output: If the image does not contain a QR code, or the QR code cannot be successfully recognized, or the QR code is not fully recognized, then output: For text information, check if key text content, such as product name and batch number, can be identified; if the key text content can be identified in the image and the text content is complete, then output the result. If the key text content cannot be identified in the image, or the text content is not fully identified, then output: By identifying the decision function Determine if the image content is complete: ;like If any key feature of the image is missing, such as a barcode, QR code, or text information, the image is considered incomplete; if the label is damaged, it is considered damaged. If all key features in an image are complete and identifiable, the image content is deemed complete; then scanning and information verification are performed.

[0029] Specifically, the scanning and information verification process involves: scanning the target label to obtain the target product's storage information, which is designated as the label information and denoted as... ; Retrieve database Extract the correct entry format for the target product when entering it into the database, and designate it as the reference format, denoted as . : ;in, Indicates the target product. This indicates the extraction of the correct entry format for the target product in the database; extraction of label information. The format is defined as the information format, denoted as... ; using format comparison functions Determine if the entered tag information data is complete: ;in, Information format representation The A format, Reference format The A format; if If so, the entered label information is considered complete; if If the data entered is incomplete, it is determined that the label information is incomplete; this is determined by a unique judgment function. Determine if there are multiple entries for the tag information in the database: ;in, For database The number of data entries that match the tag information; if If so, it is determined that there are multiple entries for the label information in the database; If the label information is determined to exist only once in the database, this entry will be designated as the comparison data, denoted as [reference data]. ; through data comparison functions Determine whether the label information matches the comparison data: ;like If the label information matches the comparison data, then it is determined that the label information is consistent with the comparison data; if If the label information is inconsistent with the comparison data, it is determined that the label information is inconsistent with the comparison data; the determination is made through a comprehensive function. Determine if the label is normal: ;like If the entered label information is incomplete, or there are multiple entries for the label information in the database, or the label information is inconsistent with the comparison data, then the label is determined to be damaged; if If the tag information is complete, there is only one record of the tag information in the database, and the tag information matches the comparison data, then the tag is considered normal.

[0030] This embodiment automates label integrity detection through image processing and scanning technology, reducing the time and errors of manual inspection, promptly identifying and repairing damaged labels, and avoiding delays caused by label issues during production. Through multi-dimensional information analysis, it accurately determines the batch affiliation of products with damaged labels, optimizing production batch tracking management, ensuring complete label information for each product, and improving the accuracy and reliability of production data. Detailed analysis and verification reduce batch affiliation errors, preventing quality problems caused by batch confusion, ensuring each product meets production standards, verifying product quality through feature comparison, and reducing product waste caused by label issues through precise batch tracking and label repair. It also optimizes inventory and production planning management, improving resource utilization efficiency. Combining multi-dimensional information such as timestamps, serial numbers, equipment logs, and manual records improves the accuracy of judgment and the reliability of the system. Dynamic repair and manual verification enhance the system's ability to handle abnormal situations, ensuring the stability of the production process.

Claims

1. A method for batch tracking and management of cardboard production based on RFID, characterized in that: include: Obtain the tags for the target product and designate them as target tags; Get database ； Determine if the label of the target product is damaged; if the label is damaged, analyze the batch of the target product; replace the label with a new label based on the analysis results; if the label is normal, use the product as usual.

2. The RFID-based batch tracking and management method for cardboard production according to claim 1, characterized in that: To determine whether the label of a target product is damaged, the following steps are taken: acquire an image of the target label, designated as the target label image; preprocess the acquired target label image; extract the image edges to obtain an edge binary map, denoted as... Initialize a two-dimensional array, denoted as . Traversing the edge binary graph Each pixel in Find connected edge pixel regions; The number of connected components is denoted as . Count the number of pixels in each connected region, denoted as . Define a continuity decision function. Determine whether the image edges are continuous; if If the edge is continuous, then the edge is determined to be continuous; if If the edge is discontinuous, then the edge is determined to be discontinuous; a contour is fitted to the binary edge image; the difference between the fitted curve and the actual edge pixels is calculated and denoted as . ; Determine the function by rules Determine if the image edges are regular; if Then determine the edge rule; if If the edge is irregular, then the edge is determined to be irregular; define a complete determination function. Determine if the image edges are complete; if If the image edges are incomplete, the label is considered damaged. If the image edges are intact, then the image content of the target label image is determined.

3. The RFID-based batch tracking and management method for cardboard production according to claim 2, characterized in that: Image content determination is performed on the image of the target label, specifically by obtaining the preprocessed image. Detect the presence of barcodes in the image; detect the presence of QR codes in the image; recognize text information in the image; for barcodes and QR codes, check if they can be successfully decoded and the complete information can be obtained; if a barcode exists in the image and is successfully and completely recognized, output... If the image does not contain a barcode, or the barcode cannot be successfully recognized, or the barcode is not fully recognized, then output: If a QR code exists in the image, and the QR code is successfully and completely recognized, then output: If the image does not contain a QR code, or the QR code cannot be successfully recognized, or the QR code is not fully recognized, then output: For text information, check if the key text content can be identified; if the key text content can be identified in the image and the text content is complete, then output the result. If the key text content cannot be identified in the image, or the text content is not fully identified, then output: ； By identifying the decision function Determine if the image content is complete; if If the image content is incomplete, the label is damaged. If the image content is found to be complete, then scanning and information verification will be performed.

4. The RFID-based batch tracking and management method for cardboard production according to claim 3, characterized in that: The scanning and information verification process involves: scanning the target label to obtain the storage information of the target product, which is then designated as the label information and denoted as [label information]. ; Retrieve database ； Extract the correct entry format for the target product when entering it into the database, and designate it as the reference format, denoted as . Extract tag information The format is defined as the information format, denoted as... ; using format comparison functions Determine if the entered tag information data is complete; if If so, the entered label information is deemed complete; like If so, the entered label information is deemed incomplete; Through the unique decision function It determines whether there are multiple entries for the tag information in the database; among them, For database The number of data entries that match the tag information; if If so, it is determined that there are multiple entries for the label information in the database; If the label information is determined to exist only once in the database, this entry will be designated as the comparison data, denoted as [reference data]. ; through data comparison functions Determine whether the label information matches the comparison data; if If the label information matches the comparison data, then it is determined that the label information is consistent with the comparison data; if If the label information is inconsistent with the comparison data, it is determined that the label information is inconsistent with the comparison data; the determination is made through a comprehensive function. Determine if the label is normal; if If the label is damaged, then it is determined that the label is damaged; if If so, the label is considered normal.

5. The RFID-based batch tracking and management method for cardboard production according to claim 1, characterized in that: The analysis of the target product batch attribution specifically involves: determining whether the target product batch is in a special position; if so, analyzing the target product's actual information; if in a normal position, extracting the two products adjacent to the target product and designating them as the left and right products, respectively; and then retrieving the database. Extract batch information for the products on the left and right sides from the database, and denot them as follows: and ; using batch comparison function To determine whether the batch information of the two products is consistent; like If the batch information of the two products is consistent, then the batch information of the target product is determined and denoted as... By combining the product information of the products on the left and right sides with the batch information of the target product, the product information of the target product is inferred, and the label is regenerated. like If the batch information of the two products is inconsistent, then the true information of the target product will be analyzed.

6. The RFID-based batch tracking and management method for cardboard production according to claim 5, characterized in that: To determine whether the target product's batch is in a special position, specifically: obtain the target product's timestamp and serial number, and record them as follows: and ; Obtain the batch switching time point, denoted as ； Calculate the time difference between the target product timestamp and the batch switch time, denoted as . ; determined by time function Inspection time difference Does it meet the requirements? Then determine the time difference. Meets the requirements; if Then determine the time difference. Not in compliance with requirements; obtain the starting serial number of the known batch and record it as... ; Obtain the end serial number of a known batch, and record it as . Calculate the difference between the target product serial number and the starting serial number, and denote it as . Calculate the difference between the target product serial number and the ending serial number, and denote it as . ; through information determination function Check the difference and Does it meet the requirements? If the difference is found to meet the requirements, then the difference is deemed to be acceptable. like If the difference does not meet the requirements, then the difference is determined to be unacceptable. Determining function based on parameters Determine whether the batch of the target product is in a special position; like If so, the target product is determined to be in a special position; like If so, the target product is determined to be in a normal position.

7. The RFID-based batch tracking and management method for cardboard production according to claim 6, characterized in that: Analyze the true information of the target product, specifically: obtain the timestamp of the target product, denoted as... ; Obtain the start and end timestamps of the preceding and following batches, and record them as follows: 、 、 as well as ； Calculate the overlap range between the timestamp of the target product and the time of the previous batch, denoted as . ； Calculate the overlap range between the timestamp of the target product and the time of the subsequent batch, denoted as . ； Obtain the serial number of the target product Obtain the start and end serial numbers of the preceding and following batches, and record them as follows: 、 、 as well as Calculate the similarity between the target product's serial number and the previous batch, and record it as follows: Calculate the similarity between the target product's serial number and the subsequent batch, and record it as follows: 。 8. The RFID-based batch tracking and management method for cardboard production according to claim 7, characterized in that: Analyzing the true information of the target product also includes: obtaining the warehousing time of the target product, denoted as... ; Obtain the allocation record of the target product, and record it as ; Obtain the inbound time and allocation records of the previous and subsequent batches, and record them as follows: 、 、 as well as Compare the matching degree between the target product's warehousing time and allocation records and the previous batch, and record it as follows: Compare the matching degree between the target product's warehousing time and allocation records and the subsequent batch, and record it as follows: ; Obtain the physical, technological, and dimensional characteristics of the target product, and denot them as ; Obtain the features of the upper and lower batches of samples, and denot them as follows: and Compare the characteristics of the target product with those of the previous batch, and record the similarity as follows: Compare the characteristics of the target product with the similarity of the subsequent batch, and record the result as follows: Calculate the combined score of the target product and the previous batch, and record it as follows: Calculate the combined score of the target product and the previous batch, and record it as follows: ;; through the attribution determination function Determine the batch information of the target product; if If so, the target product is determined to belong to the previous batch. ;like If so, the target product is determined to belong to the later batch. ;like If not, manual verification is performed; combining the product information of the products on the left and right sides with the batch information of the target product, the product information of the target product is inferred, and a new label is generated.

9. The RFID-based batch tracking and management method for cardboard production according to claim 8, characterized in that: By combining the product information of the products on the left and right sides with the batch information of the target product, the product information of the target product is inferred, and the label is regenerated. Specifically, this involves obtaining the product identifier and timestamp of the target product. ,serial number and batch information Retrieve batch information from the database The corresponding general information is denoted as The known information about the target product is integrated with the general information of the batch to form a complete product information structure, denoted as... The integrated and complete product information will be output, tags will be regenerated, and the database will be updated.

Images