Cast pipe end face two-dimensional code dynamic identification and warehouse payment automation system and method

Abstract

The invention discloses a cast pipe end face two-dimensional code dynamic identification and warehouse payment automation system and method, and belongs to the technical field of industrial automation, and the system comprises a visual trigger positioning module, a multi-code reader networking identification module, a data processing module, a quantity verification module, an abnormity processing module, and a warehouse payment sheet generation module. According to the invention, through the networking design of multiple code readers, the industrial code readers are arranged according to the preset spatial layout to form an identification area covering the end face of the cast pipe with the full specification of DN300mm-DN1000mm, the cast pipes with different specifications can be adapted without adjusting the positions of the code readers, and the problem that the coverage range of the existing fixed code readers is limited is solved; meanwhile, real-time monitoring and accurate triggering of the in-place state of the mobile transportation equipment are achieved through the visual triggering positioning module, equipment parking is not needed, and the transportation efficiency and the recognition continuity of the cast pipe payment warehouse are improved.

Description

Technical Field

[0001] This invention relates to the field of industrial automation technology, and in particular to an automated system and method for dynamic identification and warehousing of QR codes on the end face of cast iron pipes. Background Technology

[0002] In the large-scale production of cast iron pipes, the warehousing process is a crucial link between production and storage, directly impacting the accuracy and efficiency of subsequent warehousing, storage, outbound processes, and production ledger statistics. Currently, the industry widely uses AGVs (Automated Guided Vehicles) as the core equipment for transporting cast iron pipes, but the existing warehousing process has the following technical shortcomings: Firstly, the existing process relies on manual querying and selection of pipe numbers for the cast iron pipes to be delivered from the Manufacturing Execution System (MES), manually printing the delivery slip, and then associating it with the cast iron pipes on the AGV cart. Manual operation is prone to problems such as incorrect pipe number selection, omissions, and duplicate selections, resulting in inconsistencies between the delivery slip information and the actual cast iron pipe information. This leads to a chain of problems such as chaotic warehousing and incorrect delivery, seriously affecting the smoothness of the production process.

[0003] Secondly, the QR codes on the end faces of cast iron pipes are characterized by their small size (typically 12mm×12mm) and high module density (the smallest module is only 0.38mm). Furthermore, the AGV (Automated Guided Vehicle) is in continuous motion during transport (speed 0.3m / s-1.0m / s), and the cast iron pipes vary widely in size (DN300mm-DN1000mm), resulting in significant differences in the end face positions of different pipe sizes. Existing recognition solutions struggle to handle these complex scenarios: fixed readers have limited recognition range, covering only a single or a small number of pipe sizes, failing to achieve full coverage; handheld readers require manual operation after the AGV stops, reducing transport efficiency and exhibiting poor recognition stability during movement; simultaneously, existing recognition technologies lack dedicated algorithms for moving targets, leading to significant blurring of the QR code images caused by AGV movement and low decoding success rates.

[0004] Third, existing solutions rely solely on QR code recognition results to count the number of cast iron pipes. If QR codes are missed, misread, or there are human errors, quantity discrepancies cannot be detected in a timely manner, further exacerbating the risk of errors in the deposit information. With the increasing demands for automation, informatization, and precision in intelligent manufacturing processes, the existing cast iron pipe deposit identification method, which relies on manual intervention, has limited recognition capabilities, and lacks verification mechanisms, can no longer meet the actual needs of cast iron pipe manufacturers. There is an urgent need for an integrated solution that can achieve full-specification coverage, high-accuracy recognition, automatic verification, and document generation in mobile environments. Summary of the Invention

[0005] To address the problems of traditional cast iron pipe warehousing identification methods, such as reliance on manual intervention, limited identification capabilities, and lack of verification mechanisms, this invention provides a dynamic QR code identification and automated warehousing system and method for cast iron pipe end faces. This system is applicable to the warehousing process in cast iron pipe production, enabling rapid QR code identification, data verification, and automatic generation of warehousing slips when mobile transport equipment such as AGVs carries cast iron pipes of different specifications, achieving real-time synchronization of material and information flows in cast iron pipe production.

[0006] The technical solution adopted by the present invention, a dynamic QR code recognition and automated warehousing system and method for cast iron pipe end faces, is as follows: An automated system for dynamic QR code recognition and warehousing of cast iron pipe ends includes a visual trigger positioning module, a multi-reader network recognition module, a data processing module, a quantity verification module, an anomaly handling module, and a warehousing slip generation module. The visual trigger positioning module monitors the operating status of the mobile transport equipment carrying the cast iron pipes in real time, generating and outputting a recognition trigger signal when the equipment enters a preset recognition area. The multi-reader network recognition module consists of multiple industrial QR codes arranged in a preset spatial layout, forming a complete recognition area covering the cast iron pipe ends within a preset specification range. The data processing module responds to the recognition trigger signal and controls the multi-reader network recognition module to synchronously capture images of the cast iron pipe ends. The QR code image on the pipe end face is decoded using a dedicated moving target recognition algorithm to extract the pipe number information and generate pipe queue information according to preset rules. The quantity verification module collects visual data of the pipes through an independent visual data model, calculates the total number of pipes on the mobile transportation equipment, and performs real-time matching and verification with the number of pipes in the pipe queue information. When the quantity verification module determines that the verification is abnormal, the anomaly handling module triggers an alarm mechanism and records the relevant abnormal data. After the verification is passed, the delivery slip generation module pushes the pipe queue information and associated pipe attribute information to the manufacturing execution system, triggering the manufacturing execution system to automatically generate and print the pipe delivery slip.

[0007] A further improvement of the technical solution of the present invention is that the multi-reader network identification module includes multiple industrial readers, which are evenly arranged in a vertical direction perpendicular to the axis of the cast pipe, and the complete identification area covers the end face of the cast pipe with specifications of DN300mm-DN1000mm.

[0008] A further improvement of the technical solution of the present invention is that: the preset rules include sorting the pipe number information according to the recognition time of the industrial barcode reader, and removing duplicate pipe number data through a deduplication algorithm; the delivery note generation module also includes a data storage unit for recording the casting pipe queue information, quantity verification results, delivery note generation time and printing status, forming a traceable delivery note data archive.

[0009] A further improvement of the technical solution of the present invention is that: the visual trigger positioning module includes several industrial cameras and a visual detection model trained based on deep learning algorithms; wherein, the industrial cameras are deployed on the side or top of the preset recognition area, and the visual detection model is used to extract features and identify targets from the image data collected by the industrial cameras to determine whether the mobile transportation equipment has entered the preset recognition area.

[0010] A further improvement of the technical solution of the present invention is that: the data processing module includes an edge computing host that is communicatively connected to the multi-reader network identification module; the moving target dedicated identification algorithm includes an image deblurring sub-algorithm, a QR code region positioning sub-algorithm, and a decoding sub-algorithm; wherein, the image deblurring sub-algorithm is used to eliminate image blurring caused by the movement of mobile transportation equipment, the QR code region positioning sub-algorithm is used to accurately locate the QR code position in the image, and the decoding sub-algorithm is used to extract the tube number information in the QR code.

[0011] A further improvement of the technical solution of the present invention is that: the visual data model continuously acquires a sequence of images of the end face of the cast pipe, and performs inter-frame comparison and counting based on the contour features, size parameters and spacing information of the end face of the cast pipe and adjacent cast pipes; the quantity verification module determines that the verification is abnormal when the absolute value of the difference between the total number of cast pipes and the number of pipe numbers is greater than 0.

[0012] A further improvement of the technical solution of the present invention is that the alarm mechanism includes audible and visual alarms, abnormal information recording, and abnormal remarks on the delivery slip; wherein, the abnormal information recording includes the location information of the mobile transportation equipment, image data, quantity comparison results, and the time of abnormal occurrence, and the abnormal remarks on the delivery slip are used to mark abnormal prompt information on the printed delivery slip.

[0013] A method for dynamic recognition and automated warehousing of QR codes on the end face of cast iron pipes, using the aforementioned system, includes the following steps: S1. The visual trigger positioning module collects image data of the mobile transportation equipment and the cast pipe in real time through an industrial camera. The visual detection model analyzes the image data to determine whether the mobile transportation equipment has entered the preset recognition area. S2. If it is determined that the mobile transportation equipment has entered the preset identification area, the visual trigger positioning module sends an identification trigger signal to the multi-reader network identification module, and at the same time the quantity verification module starts the visual data model. S3. The multi-reader network identification module responds to the identification trigger signal, synchronously captures the QR code image on the end face of the cast pipe, and uploads the QR code image to the data processing module. S4. The data processing module uses a dedicated moving target recognition algorithm to deblur, locate, and decode the QR code image, extract the pipe number information, sort it according to preset rules, remove duplicates, and generate the casting pipe queue information. S5. The quantity verification module analyzes the continuously acquired sequence of images of the cast pipe end face through a visual data model, calculates the total number of cast pipes on the mobile transportation equipment, and matches and verifies the total number of cast pipes with the number of pipe numbers in the cast pipe queue information. S6. If the verification passes, the delivery order generation module pushes the cast pipe queue information and cast pipe attribute information to the manufacturing execution system. The manufacturing execution system generates an electronic delivery order and drives the printer to print a paper delivery order. At the same time, the data storage unit records the relevant data. If the verification fails, the exception handling module triggers the alarm mechanism, records the exception information and makes an exception note in the delivery order, and waits for manual intervention.

[0014] The technological advancements achieved by this invention due to the adoption of the above technical solutions are as follows: This invention utilizes a multi-reader network design, arranging industrial barcode readers in a preset spatial layout to form an identification area covering the end faces of cast iron pipes of all sizes from DN300mm to DN1000mm. It adapts to different sizes of cast iron pipes without requiring adjustments to the reader positions, solving the problem of limited coverage of existing fixed barcode readers. Simultaneously, the visual trigger positioning module enables real-time monitoring and precise triggering of the arrival status of mobile transportation equipment, eliminating the need for equipment shutdown and improving the transportation efficiency and identification continuity of cast iron pipes upon arrival at the warehouse.

[0015] The data processing module of this invention adopts a dedicated recognition algorithm for moving targets, specifically addressing the image blurring problem caused by the movement of mobile transportation equipment. Through deblurring, precise positioning, and efficient decoding, it improves the decoding success rate of QR codes. Combined with the high resolution and fast capture capability of industrial barcode readers, it ensures that the QR code recognition accuracy is no less than 99% at a movement speed of 0.3m / s-1.0m / s, solving the problem of poor adaptability of existing technologies to moving targets.

[0016] The quantity verification module of this invention calculates the total number of cast pipes through an independent visual counting model and performs real-time matching and verification with the number of pipes generated by QR code recognition. This enables timely detection of quantity deviations caused by missed readings, misreadings, or manual intervention. The anomaly handling module simultaneously triggers alarms and records relevant information, reducing the risk of errors in the deposit information.

[0017] This invention eliminates the need for manual intervention throughout the entire process, from pipe identification, data processing, quantity verification to delivery slip generation and printing. It not only eliminates errors caused by manual operation but also reduces the labor intensity of staff, shortening the delivery time for a single batch of pipes from the current 10-15 minutes to 1-2 minutes, thus improving the overall efficiency of the delivery process. Attached Figure Description

[0018] Figure 1 This is a flowchart of a method for dynamic identification and automated warehousing of QR codes on the end face of cast iron pipes according to the present invention; Figure 2and Figure 3 This is a schematic diagram of the structure of a dynamic QR code recognition and automated warehousing system for cast pipe end faces according to the present invention.

[0019] In the attached diagram: 1. Mobile transportation equipment; 2. Multi-reader network identification module; 21. Industrial barcode reader; 3. Industrial camera; 4. QR code; 5. Cast pipe. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings. In the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily obscuring the concept of this invention.

[0021] Example 1 like Figure 2 and Figure 3 As shown, this embodiment provides a dynamic QR code recognition and automated deposit system for cast pipe end faces, including a visual trigger positioning module, a multi-reader network recognition module 2, a data processing module, a quantity verification module, an anomaly handling module, and a deposit slip generation module.

[0022] The aforementioned multi-reader network identification module 2 includes four industrial barcode readers 21. The four industrial barcode readers 21 are evenly arranged in a vertical direction perpendicular to the axis of the cast pipe 5. The complete identification area covers the end face of the cast pipe 5 with specifications of DN300mm-DN1000mm. The minimum module identification capability of the industrial barcode reader 21 is not less than 0.38mm. The speed range of the adapted mobile transportation equipment 1 is 0.3m / s-1.0m / s.

[0023] Meanwhile, the aforementioned preset rules include sorting the pipe number information according to the recognition time of the industrial barcode reader 21, and removing duplicate pipe number data through a deduplication algorithm; the deposit slip generation module also includes a data storage unit for recording the casting pipe 5 queue information, quantity verification results, deposit slip generation time and printing status, forming a traceable deposit data archive.

[0024] In this embodiment, the visual trigger positioning module includes at least one industrial camera 3 and a visual detection model trained based on a deep learning algorithm; the industrial camera 3 is deployed on the side or top of the preset recognition area, with a shooting frame rate of not less than 30fps and an image resolution of not less than 1920×1080; the visual detection model is used to extract features and identify targets from the image data collected by the industrial camera 3, and to determine whether the mobile transportation device 1 has entered the preset recognition area, with a recognition response time of not more than 50ms.

[0025] The lens axis of the industrial camera 3 is perpendicular to the end face of the cast pipe 5, and the distance between the lens and the end face of the cast pipe 5 is 300mm-500mm; the visual inspection model is trained based on the YOLO series algorithm or the SSD algorithm.

[0026] In this embodiment, the data processing module includes an edge computing host, which is communicatively connected to the multi-reader network identification module 2; the moving target dedicated identification algorithm includes an image deblurring sub-algorithm, a QR code region positioning sub-algorithm, and a decoding sub-algorithm. The image deblurring sub-algorithm is used to eliminate image blurring caused by the movement of the mobile transportation device 1, the QR code region positioning sub-algorithm is used to accurately locate the position of the QR code 4 in the image, and the decoding sub-algorithm is used to extract the tube number information in the QR code 4.

[0027] In this embodiment, the visual data model continuously acquires a sequence of images of the end faces of the cast pipes 5. Based on the contour features, dimensional parameters, and spacing information of adjacent cast pipes 5, it performs inter-frame comparison and counting. The quantity verification module determines an anomaly when the absolute value of the difference between the total number of cast pipes 5 and the number of pipe numbers is greater than 0. Simultaneously, the alarm mechanism in this embodiment includes audible and visual alarms, anomaly information recording, and anomaly notes on the delivery slip. The anomaly information recording includes the location information of the mobile transport equipment 1, image data, quantity comparison results, and the time of the anomaly. The anomaly notes on the delivery slip are used to clearly indicate the anomaly message on the printed delivery slip.

[0028] Example 2 like Figure 1 As shown, this embodiment provides a method for dynamic recognition and automated warehousing of QR codes on the end face of cast iron pipes, including the following steps: S1. Initialization Phase: Staff configure the working parameters of each module through the system backend, including the shooting frame rate and resolution of the industrial camera 3, the arrangement position and capture frequency of the barcode reader, the data interface protocol type, and the boundary parameters of the preset recognition area.

[0029] S2. Position Monitoring Stage: The AGV carrying the cast pipe 5 travels along a preset path. The industrial camera 3 of the vision-triggered positioning module collects image data in real time, and each frame of the image is transmitted to the vision detection model for analysis. The vision detection model extracts features such as the outline of the AGV body and the outline of the end face of the cast pipe 5, compares them with preset thresholds, and determines whether the AGV has entered the preset recognition area.

[0030] S3. Triggering and Start-up Phase: When the visual detection model determines that the AGV has completely entered the preset recognition area, it immediately sends a recognition trigger signal to the multi-reader network recognition module 2, and at the same time sends a start signal to the quantity verification module to start the visual data model.

[0031] S4. Image Acquisition Stage: After receiving the trigger signal, the multi-reader network identification module 2 simultaneously captures images of the QR code 4 on the end face of the cast pipe 5. Each reader captures one image at a time, for a total of four images, which are then simultaneously uploaded to the edge computing host.

[0032] S5. Data Processing Stage: After the edge computing host receives four QR code images, the dedicated moving target recognition algorithm sequentially performs deblurring, QR code positioning, and decoding operations to extract the pipe number information from each image. The decoding results of the four readers are summarized, sorted according to the capture time, and duplicate pipe numbers are removed by a deduplication algorithm to generate a casting pipe queue information containing a unique pipe number list.

[0033] S6. Quantity Verification Stage: The visual data support model of the quantity verification module receives 10 frames of images of the end face of the cast pipe 5. Through inter-frame alignment, feature extraction, and counting analysis, the total quantity of cast pipe 5 is obtained. The total quantity of cast pipe 5 is compared with the number of pipe numbers in the queue information of cast pipe 5, and the absolute value of the difference is calculated.

[0034] S7. Normal Processing Flow: If the absolute value of the difference is 0, the verification is considered successful; the delivery note generation module associates the cast pipe 5 queue information (pipe number list) with the cast pipe 5 attribute information (specifications, materials, production batches, etc.) pre-stored in the MES system, packages and pushes it to the MES system; the MES system calls the delivery note template to generate an electronic delivery note, which is stored in the system database on one hand, and sends a print command to the industrial printer on the other hand to print a paper delivery note; at the same time, the data storage unit records the cast pipe 5 queue information, quantity verification results, delivery note generation time, printing status and other data.

[0035] S8. Abnormal Handling Process: If the absolute value of the difference is >0, the verification is determined to be abnormal; the abnormal handling module activates the audible and visual alarm, which continues to alarm for 30 seconds (or is turned off after manual confirmation); the abnormal information is recorded to the abnormal database simultaneously; an abnormal marker signal is sent to the delivery note generation module. The electronic and paper delivery notes generated by the MES system clearly note "Quantity verification abnormal, the actual quantity does not match the pipe number quantity" and related abnormal parameters. After receiving the alarm, the staff manually reviews the cast pipe 5 on the AGV cart. After the processing is completed, the abnormal status is cleared through the system background.

[0036] In the above embodiments, an automated system and method for dynamic QR code recognition and warehousing of cast iron pipe ends is provided. This invention utilizes a multi-reader network design, arranging industrial QR codes in a preset spatial layout to form a recognition area covering the end faces of all cast iron pipe sizes from DN300mm to DN1000mm. This allows for adaptation to different pipe sizes without adjusting the reader positions, solving the problem of limited coverage of existing fixed QR code readers. Simultaneously, a visual trigger positioning module enables real-time monitoring and precise triggering of the mobile transport equipment's arrival status, eliminating the need for equipment shutdown and improving the transport efficiency and recognition continuity of cast iron pipe warehousing. The data processing module employs a dedicated moving target recognition algorithm to specifically address image blurring caused by the movement of mobile transport equipment. Through deblurring, precise positioning, and efficient decoding, the QR code decoding success rate is improved. Combined with the high efficiency of industrial QR codes… With its high resolution and fast capture capability, this invention ensures a QR code recognition accuracy of no less than 99% at movement speeds of 0.3m / s-1.0m / s, solving the problem of poor adaptability to moving targets in existing technologies. The quantity verification module calculates the total number of cast iron pipes using an independent visual data model and performs real-time matching and verification with the pipe number generated by QR code recognition, enabling timely detection of quantity deviations caused by missed readings, misreadings, or manual intervention. The anomaly handling module simultaneously triggers alarms and records relevant information, reducing the risk of errors in the deposit information. From cast iron pipe identification, data processing, and quantity verification to deposit slip generation and printing, this invention requires no manual intervention throughout the entire process. This not only eliminates errors caused by manual operation but also reduces the workload of staff, shortening the deposit time for a single batch of cast iron pipes from the current 10-15 minutes to 1-2 minutes, thus improving the overall efficiency of the deposit process.

[0037] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the concept and scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the inventive concept should fall within the protection scope of the present invention. All technical contents for which protection is sought in this invention are fully described in the claims.

Claims

1. A dynamic QR code recognition and automated warehousing system for cast iron pipe ends, characterized in that: The system includes a visual trigger positioning module, a multi-reader network identification module (2), a data processing module, a quantity verification module, an anomaly handling module, and a delivery note generation module. The visual trigger positioning module monitors the operating status of the mobile transport equipment (1) carrying the cast iron pipe (5) in real time, generating and outputting an identification trigger signal when the mobile transport equipment (1) enters the preset identification area. The multi-reader network identification module (2) consists of multiple industrial barcode readers (21) arranged in a preset spatial layout, forming a complete identification area covering the end face of the cast iron pipe (5) within a preset specification range. The data processing module responds to the identification trigger signal, controlling the multi-reader network identification module (2) to simultaneously capture images of the QR code (4) on the end face of the cast iron pipe (5), and transmits the data. The dedicated recognition algorithm for moving targets decodes the QR code (4) image, extracts the pipe number information of the cast pipe (5), and generates the cast pipe (5) queue information according to preset rules; the quantity verification module collects visual data of the cast pipe (5) through an independent visual data branch model, calculates the total number of cast pipes (5) on the mobile transportation equipment (1), and performs real-time matching and verification with the number of pipe numbers in the cast pipe (5) queue information; when the quantity verification module determines that the verification is abnormal, the anomaly handling module triggers an alarm mechanism and records the abnormal data; after the verification is passed, the delivery order generation module pushes the cast pipe (5) queue information and the associated cast pipe (5) attribute information to the manufacturing execution system, triggering the manufacturing execution system to automatically generate and print the cast pipe (5) delivery order.

2. The automated system for dynamic identification and warehousing of QR codes on the end face of cast iron pipes according to claim 1, characterized in that: The multi-reader network identification module (2) includes multiple industrial readers (21). The multiple industrial readers (21) are evenly arranged in a vertical direction perpendicular to the axis of the cast pipe (5), and the complete identification area covers the end face of the cast pipe (5) with specifications of DN300mm-DN1000mm.

3. The automated system for dynamic identification and warehousing of QR codes on the end face of cast iron pipes according to claim 2, characterized in that: The preset rules include sorting the pipe number information according to the recognition time of the industrial barcode reader (21) and removing duplicate pipe number data through a deduplication algorithm; the deposit slip generation module also includes a data storage unit for recording the casting pipe (5) queue information, quantity verification results, deposit slip generation time and printing status, forming a traceable deposit data archive.

4. The automated system for dynamic identification and warehousing of QR codes on the end face of cast iron pipes according to claim 1, characterized in that: The visual trigger positioning module includes several industrial cameras (3) and a visual detection model trained based on deep learning algorithms; wherein, the industrial cameras (3) are deployed on the side or top of the preset recognition area, and the visual detection model is used to extract features and identify targets from the image data collected by the industrial cameras (3) to determine whether the mobile transportation equipment (1) has entered the preset recognition area.

5. The automated system for dynamic identification and warehousing of QR codes on the end face of cast iron pipes according to claim 1, characterized in that: The data processing module includes an edge computing host that is communicatively connected to the multi-reader network identification module (2); the moving target dedicated identification algorithm includes an image deblurring sub-algorithm, a QR code region positioning sub-algorithm, and a decoding sub-algorithm; wherein, the image deblurring sub-algorithm is used to eliminate image blurring caused by the movement of the mobile transportation equipment (1), the QR code region positioning sub-algorithm is used to accurately locate the position of the QR code (4) in the image, and the decoding sub-algorithm is used to extract the tube number information in the QR code (4).

6. The automated system for dynamic identification and warehousing of QR codes on the end face of cast iron pipes according to claim 1, characterized in that: The visual data branch model continuously acquires a sequence of images of the end face of the cast pipe (5), and performs inter-frame comparison and counting based on the contour features, size parameters and spacing information of the end face of the cast pipe (5) and adjacent cast pipes (5); the quantity verification module determines that the verification is abnormal when the absolute value of the difference between the total number of cast pipes (5) and the number of pipe numbers is greater than 0.

7. The automated system for dynamic identification and warehousing of QR codes on the end face of cast iron pipes according to claim 6, characterized in that: The alarm mechanism includes audible and visual alarms, abnormal information recording, and abnormal remarks on the delivery slip; among which, abnormal information recording includes the location information of the mobile transportation equipment (1), image data, quantity comparison results, and the time of abnormal occurrence, and abnormal remarks on the delivery slip are used to mark abnormal prompt information on the printed delivery slip.

8. A method for dynamic identification and automated warehousing of QR codes on the end face of cast iron pipes, characterized in that, Using the system according to any one of claims 1-7 includes the following steps: S1. The visual trigger positioning module collects image data of the mobile transportation equipment (1) and the cast pipe (5) in real time through the industrial camera (3). The visual detection model analyzes the image data to determine whether the mobile transportation equipment (1) has entered the preset recognition area. S2. If it is determined that the mobile transportation equipment (1) has entered the preset identification area, the visual trigger positioning module sends an identification trigger signal to the multi-reader network identification module (2), and at the same time the quantity verification module starts the visual data branch model. S3. The multi-reader network identification module (2) responds to the identification trigger signal, synchronously captures the QR code (4) image on the end face of the cast pipe (5), and uploads the QR code (4) image to the data processing module. S4. The data processing module uses a dedicated moving target recognition algorithm to deblur, locate and decode the QR code (4) image, extract the pipe number information, sort and deduplicate according to preset rules, and generate the casting pipe (5) queue information. S5. The quantity verification module analyzes the continuously collected end face sequence images of the cast pipe (5) through the visual data branch model, calculates the total number of cast pipes (5) on the mobile transportation equipment (1), and matches and verifies the total number of cast pipes (5) with the number of pipe numbers in the cast pipe (5) queue information. S6. If the verification passes, the delivery order generation module will push the casting pipe (5) queue information and casting pipe (5) attribute information to the manufacturing execution system. The manufacturing execution system will generate an electronic delivery order and drive the printer to print a paper delivery order. At the same time, the data storage unit will record the relevant data. If the verification fails, the exception handling module will trigger an alarm mechanism, record the exception information and make an exception note in the delivery order, and wait for manual intervention.

Images