Intelligent paper pulp tallying method and system

Abstract

The invention discloses an intelligent paper pulp tallying method and system, and belongs to the technical field of port tallying. The method comprises the following steps: acquiring a paper pulp image by utilizing field equipment and acquiring an automatic counting value by adopting a recognition algorithm; combining a shipping mark verification image uploaded by the mobile terminal to generate electronic receipts with the binding number and vouchers; and transmitting the electronic receipt to a central control terminal for remote comparison and re-checking, and filing and generating a report form after confirming that the electronic receipt is correct. According to the invention, the encryption operation serial number is utilized to ensure that data cannot be tampered, abnormal real-time alarm and on-site correction are supported, and automatic matching with transportation system information can be realized. According to the invention, through an automatic counting and manual verification mode, digitalization and whole-course traceability of tallying operation are realized, and the operation efficiency and accuracy are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of port tallying technology, and more specifically, to an intelligent tallying method and system for pulp. Background Technology

[0002] Currently, port pulp logistics tallying operations primarily rely on the traditional model of manual on-site counting and paper-based document recording. This method is not only labor-intensive and inefficient, but also highly susceptible to counting errors or omissions due to visual fatigue or subjective negligence of tallying personnel, severely impacting data accuracy. Furthermore, in the existing tallying process, the counting of goods quantity and verification of goods attributes (such as shipping marks) are often difficult to synchronize in real time. The lack of effective digital image retention methods on-site prevents central control personnel from conducting real-time visual monitoring and verification of on-site operations, relying solely on passive statistics based on subsequently reported documents, resulting in significant information lag. In addition, existing technologies lack robust data tamper-proof mechanisms and a full-chain traceability system. In the event of cargo damage or quantity disputes, it is difficult to quickly retrieve electronic vouchers containing original images, timestamps, and operator information for liability determination. Moreover, tallying data typically fails to form an automated closed-loop matching with external supply chain or transport vehicle information, leading to a significant information silo effect and failing to meet the high-precision, traceable, and intelligent management requirements of modern smart ports for tallying operations. Summary of the Invention

[0003] In view of the above-mentioned problems existing in the prior art, the purpose of the present invention is to provide a method and system for intelligent pulp sorting.

[0004] To solve the above problems, the technical solution adopted by the present invention is as follows: A smart pulp sorting method, comprising the following steps: Step S1: Receive a sorting operation instruction containing order information and deploy image acquisition equipment at the operation site.

[0005] Step S2: The image acquisition device acquires images of pulp cargo at the dock in real time, and the images are processed using a pre-trained pulp recognition and counting algorithm to obtain an automatic count value of the pulp cargo.

[0006] Step S3: Receive the verification image containing the markings of the pulp goods, taken on-site by the warehouse clerk via the mobile terminal operated by the warehouse clerk.

[0007] Step S4: Bind the automatic counting value to the verification image to generate an electronic receipt containing cargo information, quantity, and shipping mark.

[0008] Step S5: The electronic ticket is transmitted to the central control terminal, so that the central control personnel can compare and verify the mark information in the verification image with the order information recorded on the electronic ticket on the verification interface.

[0009] Step S6: After the central control personnel confirm that the verification is correct, the status of the electronic ticket is updated to "confirmed", and a sorting report is generated based on this. At the same time, the electronic ticket is archived to achieve operation traceability.

[0010] Preferably, the pulp identification and counting algorithm in step S2 performs image segmentation processing on the acquired image to identify and mark the outline boundary of each independent pulp cargo unit in the image, and determines the automatic counting value based on the number of outline boundaries.

[0011] Preferably, in step S4, while generating the electronic receipt, a unique and tamper-proof operation serial number is generated for each electronic receipt, and the automatic counting value, the hash value of the verification image, the warehouse clerk's identity information and timestamp are encrypted and stored together with the operation serial number to ensure the originality and traceability of the data.

[0012] Preferably, the comparison and verification step in S5 further includes, when the central control personnel find that the shipping mark information in the verification image does not match the order information, setting the data status of the electronic receipt to pending processing lock through the central control terminal to prevent it from being included in the subsequent statistical process, and pushing an alarm containing the specific reason for the abnormality to the mobile terminal of the stock clerk, requiring the stock clerk to retake the photo on site or add text notes for clarification.

[0013] Preferably, the sorting operation instruction in step S1 is automatically obtained from an external supply chain management system; the method further includes: using a shared order number or transport task number as the association primary key, automatically matching the sorting result confirmed in step S6 with the transport vehicle information obtained from the transport system to form a complete logistics information closed loop including order, sorting, and transport.

[0014] Preferably, the step of generating the inventory report in step S6 includes, before performing aggregate statistics, first filtering the electronic receipt records in the database by status, extracting only the data with the status field "confirmed", thereby eliminating the interference of data in pending or abnormal status on the accuracy of statistics, and then aggregating the filtered data to generate the statistical report.

[0015] A smart pulp sorting system, comprising: Image acquisition equipment is installed at the sorting and handling site to collect images of pulp goods in real time.

[0016] The mobile terminal is for use by warehouse workers and is equipped with a warehouse worker application for taking and uploading verification images that include the markings on the goods.

[0017] The central control terminal is used by central control personnel to display electronic receipts and provide a human-computer interaction verification interface. The processing server is communicatively connected to the image acquisition device, the mobile terminal, and the central control terminal. The processing server is equipped with: The identification and counting module is used to receive and process images sent by the image acquisition device to obtain automatic counting values ​​of pulp goods.

[0018] The electronic receipt generation and management module is used to bind the automatic count value with the verification image received from the mobile terminal to generate and manage electronic receipts.

[0019] The review and reporting module is used to push the electronic ticket to the central control terminal for review, and generate a sorting report after the review is passed.

[0020] Preferably, the core of the recognition and counting module is a convolutional neural network model for instance segmentation or object detection that has been augmented by a pulp cargo image dataset. The model runs in a graphics processor accelerated environment and is used to accurately identify and locate each instance of pulp cargo in the image.

[0021] Preferably, the inventory management application on the mobile terminal is for receiving interactive alarms from the processing server. The alarm has a shortcut entry for on-site handling operations, which allows inventory managers to directly start the function of retaking photos or submitting text notes on the alarm notification interface, and submit the handling results in association with the original abnormal electronic receipt.

[0022] Preferably, the verification interface of the central control terminal is used to display the verification image and the order text information in the electronic receipt side by side or in columns, and provides interactive buttons for "Confirm Passed" and "Mark as Abnormal"; when a confirmation instruction is received through the interactive button, the processing server updates the status field of the corresponding electronic receipt in the database to "Confirmed".

[0023] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) This invention significantly improves the efficiency and accuracy of pulp sorting operations. By adopting an operation mode that combines automatic counting by machine vision with remote manual verification of marking images, it replaces the traditional tedious manual on-site counting, which not only greatly reduces the labor intensity of sorting staff, but also effectively avoids human errors such as omissions and errors through the dual confirmation mechanism of "automatic counting + manual verification".

[0024] (2) This invention ensures the authenticity, immutability, and full traceability of cargo handling data. By generating electronic receipts containing unique operation serial numbers, image hash values, and timestamps, a robust digital evidence chain is constructed, technically eliminating the risk of data forgery. At the same time, the electronic archiving method solves the pain points of traditional paper documents being easily lost and difficult to query, providing a reliable basis for subsequent operation traceability and dispute resolution.

[0025] (3) This invention constructs an efficient anomaly handling mechanism and a closed loop of logistics information. It realizes real-time linkage between the mobile terminal and the central control terminal at the work site. Once a discrepancy in the cargo order is found, an alarm can be pushed immediately and on-site correction can be guided, avoiding the entry of erroneous data. In addition, by connecting and matching with the external supply chain and transportation system, the data silos of the three links of order, cargo handling and transportation are broken down, and the overall coordination level of port logistics management is improved. Attached Figure Description

[0026] Figure 1 This is a schematic diagram illustrating the exemplary steps of the intelligent inventory management method of the present invention; Figure 2 This is a schematic diagram of the module structure of the cargo handling system of the present invention; Detailed Implementation The present invention will be further described below with reference to specific embodiments.

[0027] Example 1 like Figure 1 As shown in Figure 1, a smart pulp sorting method includes the following steps: Step S1 involves receiving a tallying operation instruction containing order information and deploying image acquisition equipment at the work site. The tallying operation instruction typically originates from the port's TOS system or the cargo owner's ERP system and includes key information such as vessel name, voyage number, bill of lading number, cargo specifications, and planned quantity. The image acquisition equipment mainly consists of high-definition industrial cameras or PTZ cameras, installed at the front of the boom of a gantry crane, above the grab bucket, or on the gantry frame of the tallying passage, ensuring a complete overhead view of each load of pulp cargo. The cameras must have autofocus, strong light suppression, and waterproof / dustproof capabilities to adapt to the complex and variable lighting and weather conditions at the terminal.

[0028] Step S2 involves acquiring real-time images of pulp cargo at the dock using image acquisition equipment. A pre-trained pulp recognition and counting algorithm is then used to process the images to obtain an automatic count of the pulp cargo. The acquisition process is typically triggered when the spreader is lifted or moved across a specific area. The system automatically captures multiple frames of video streams and selects the clearest keyframe image with the least obstruction. The pulp recognition and counting algorithm is deployed on edge computing nodes or a high-performance GPU server in the background, enabling millisecond-level response. It automatically identifies the characteristics of each bundle of pulp in the image, removes background interference such as spreader and wire rope, and counts the specific number of bundles.

[0029] Step S3: The warehouse clerk receives a verification image, captured on-site, showing the markings on the pulp goods, via a mobile terminal. This mobile terminal is typically a rugged industrial-grade PDA or a smartphone with a dedicated warehouse clerk app installed. When a batch of goods is unloaded or loaded onto a truck, the warehouse clerk uses the terminal to take a close-up photo of the markings on the side of the goods. The app's pre-detection feature, with OCR character recognition, prompts the warehouse clerk to ensure the markings are clearly visible, guaranteeing that the uploaded verification image contains clear information such as the brand, batch number, and weight of the goods.

[0030] Step S4 involves binding the automatic count value with the verification image to generate an electronic tally slip containing cargo information, quantity, and shipping marks. This data binding is an automatic association based on the operation timestamp and operation instruction ID. The system combines the "quantity" identified by machine vision with the "attributes (shipping marks)" captured manually within the same lifting operation, generating a complete electronic tally slip. This slip is stored in a structured format and includes fields such as vessel name, bill of lading number, operation time, operation location, lifting number, automatic count result, and path to the manual verification photo, replacing the traditional paper tally slip.

[0031] Step S5: The electronic ticket is transmitted to the central control terminal, so that the central control personnel can compare and verify the shipping mark information in the verification image with the order information recorded on the electronic ticket on the verification interface. Step S6: After the central control personnel confirm and verify the information, the status of the electronic ticket is updated to "Confirmed," and a tallying report is generated based on this. Simultaneously, the electronic ticket is archived to enable operational traceability. The status update operation is an atomic transaction, ensuring data consistency. Confirmed tickets are locked and cannot be modified arbitrarily. Based on these confirmed tickets, the system automatically summarizes and generates daily tallying reports for different bills of lading and cargo owners, as well as completion certificates. The archiving process includes storing structured data in a database, storing image files in an object storage system, and creating indexes for easy retrieval and querying by bill of lading number or time, resolving disputes related to cargo damage or discrepancies.

[0032] The pulp identification and counting algorithm in step S2 performs image segmentation processing on the acquired image to identify and mark the contour boundaries of each independent pulp cargo unit in the image, and determines the automatic count value based on the number of contour boundaries. Specifically, the algorithm adopts instance segmentation technology based on deep learning. The model has been trained on a large number of pulp images with different lighting, angles, and packaging, and can accurately segment the surface area of ​​each bundle of pulp at the pixel level, and render and label it with masks of different colors. The counting logic simply counts the number of masks. Compared with traditional object detection boxes, the segmentation algorithm is more effective in handling cases where goods are closely stacked or slightly occluded, and the counting accuracy can reach over 99.9%.

[0033] In step S4, while generating the electronic receipt, a unique and tamper-proof operation serial number is generated for each electronic receipt. The automatic count value, the hash value of the verification image, the warehouse clerk's identity information, and the timestamp are encrypted and stored together with the operation serial number to ensure the originality and traceability of the data.

[0034] The comparison and verification steps in S5 also include the following: when the central control personnel find that the shipping mark information in the verification image does not match the order information, they lock the electronic receipt data status to "pending processing" via the central control terminal to prevent it from being included in subsequent statistical processes. An alarm containing the specific reason for the anomaly is pushed to the warehouse clerk's mobile terminal, requiring the clerk to retake the photo on-site or add text notes for clarification. The reason for the anomaly might be "blurred shipping mark," "goods not matching the order," or "damaged packaging." In the locked state, the data for that shipment will not be included in the final warehouse report, preventing the spread of erroneous data. After receiving the vibration or ringtone alarm from the APP, the warehouse clerk can access the details page, where they can see the central control personnel's annotations. They can then retake a clear photo and upload it, or enter a text description (such as "The actual shipping mark does not match the pre-declaration, please confirm"), and resubmit it for central control review, forming a closed-loop anomaly handling process.

[0035] The tallying instructions in step S1 are automatically obtained from the external supply chain management system. The method also includes automatically matching the tallying results confirmed in step S6 with the transport vehicle information obtained from the transportation system using a shared order number or transport task number as the primary key, thus forming a complete closed-loop logistics information system encompassing order, tallying, and transportation. For example, when a truck enters the port to pick up goods, the system already knows the goods corresponding to bill of lading number A. After the tallying system confirms the tallying of a specific shipment of pulp under that bill of lading, it uses the vehicle identification system to determine that the goods have been loaded onto the truck. The system automatically pushes the tallying results to the transportation management system as the basis for vehicle exit and freight settlement, eliminating information silos.

[0036] Step S6, which generates the inventory report, includes the following steps: before performing aggregate statistics, the electronic receipt records in the database are first filtered by status, and only data with the status field "confirmed" is extracted to eliminate the interference of data in pending or abnormal status on the accuracy of statistics. Then, the filtered data is aggregated to generate a statistical report.

[0037] like Figure 2 As shown, this embodiment provides an intelligent pulp sorting system, including: Image acquisition equipment is installed at the sorting and handling site to collect images of pulp goods in real time.

[0038] The mobile terminal is for use by warehouse workers and is equipped with a warehouse worker application for taking and uploading verification images that include the markings on the goods.

[0039] The central control terminal is used by central control personnel to display electronic receipts and provide a human-computer interaction verification interface.

[0040] The processing server communicates with the image acquisition equipment, mobile terminals, and central control terminal. The processing server contains: The identification and counting module is used to receive and process images sent by the image acquisition device to obtain automatic counting values ​​for pulp goods.

[0041] The electronic receipt generation and management module is used to bind the automatic count value with the verification image received from the mobile terminal to generate and manage electronic receipts.

[0042] The review and reporting module is used to push electronic receipts to the central control terminal for review and generate inventory reports after the review is passed.

[0043] The core of the recognition and counting module is a convolutional neural network model for instance segmentation or object detection, which has been augmented and trained on a pulp cargo image dataset. The model runs in a graphics processor accelerated environment and is used to accurately identify and locate each instance of pulp cargo in the image.

[0044] The inventory management application on the mobile terminal receives interactive alarms from the processing server. The alarms have embedded quick access to on-site handling operations, allowing inventory managers to directly initiate the function of retaking photos or submitting text notes on the alarm notification interface, and to submit the handling results in association with the original abnormal electronic receipt.

[0045] The central control terminal's verification interface is used to display the verification image and the order text information in the electronic ticket side by side or in columns, and provides interactive buttons for "Confirm Passed" and "Mark as Abnormal". When a confirmation command is received through the interactive button, the processing server updates the status field of the corresponding electronic ticket in the database to "Confirmed".

[0046] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for intelligent pulp sorting, characterized in that, Includes the following steps: Step S1: Receive a sorting operation instruction containing order information and deploy image acquisition equipment at the operation site; Step S2: The image acquisition device acquires images of pulp cargo at the dock in real time, and the images are processed using a pre-trained pulp recognition and counting algorithm to obtain an automatic count value of the pulp cargo. Step S3: Receive the verification image containing the markings of the pulp goods, taken on-site by the warehouse clerk via the mobile terminal operated by the warehouse clerk. Step S4: Bind the automatic counting value to the verification image to generate an electronic receipt containing goods information, quantity, and shipping mark. Step S5: The electronic receipt is transmitted to the central control terminal, so that the central control personnel can compare and verify the mark information in the verification image with the order information recorded on the electronic receipt on the verification interface. Step S6: After the central control personnel confirm that the verification is correct, the status of the electronic ticket is updated to "confirmed", and a sorting report is generated based on this. At the same time, the electronic ticket is archived to achieve operation traceability.

2. The intelligent pulp sorting method according to claim 1, characterized in that: The pulp identification and counting algorithm in step S2 performs image segmentation processing on the acquired image to identify and mark the outline boundary of each independent pulp cargo unit in the image, and determines the automatic counting value based on the number of outline boundaries.

3. The intelligent pulp sorting method according to claim 1, characterized in that: In step S4, while generating the electronic receipt, a unique and tamper-proof operation serial number is generated for each electronic receipt. The automatic counting value, the hash value of the verification image, the warehouse clerk's identity information, and the timestamp are encrypted and stored together with the operation serial number to ensure the originality and traceability of the data.

4. The intelligent pulp sorting method according to claim 1, characterized in that: The comparison and verification step in S5 also includes, when the central control personnel find that the shipping mark information in the verification image does not match the order information, setting the data status of the electronic ticket to pending processing lock through the central control terminal to prevent it from being included in the subsequent statistical process, and pushing an alarm containing the specific reason for the abnormality to the mobile terminal of the stock clerk, requiring the stock clerk to retake the photo on site or add text notes for clarification.

5. The intelligent pulp sorting method according to claim 1, characterized in that: The sorting operation instructions in step S1 are automatically obtained from the external supply chain management system; the method also includes: using the shared order number or transportation task number as the association primary key, automatically matching the sorting results confirmed in step S6 with the transportation vehicle information obtained from the transportation system, so as to form a complete logistics information closed loop including the three links of order, sorting, and transportation.

6. The intelligent pulp sorting method according to claim 1, characterized in that: The step of generating the inventory report in step S6 includes, before performing aggregate statistics, first filtering the electronic receipt records in the database by status, extracting only the data with the status field "confirmed", thereby eliminating the interference of pending or abnormal data on the accuracy of statistics, and then aggregating the filtered data to generate the statistical report.

7. A smart pulp sorting system, characterized in that, include: Image acquisition equipment is installed at the sorting and handling site to collect images of pulp goods in real time; A mobile terminal for warehouse workers, equipped with a warehouse worker application, used to take and upload verification images containing the goods' markings; The central control terminal is used by central control personnel to display electronic receipts and provide a human-computer interaction verification interface. The processing server is communicatively connected to the image acquisition device, the mobile terminal, and the central control terminal. The processing server is equipped with: The identification and counting module is used to receive and process images sent by the image acquisition device to obtain automatic counting values ​​of pulp goods; The electronic receipt generation and management module is used to bind the automatic count value with the verification image received from the mobile terminal to generate and manage electronic receipts. The review and reporting module is used to push the electronic ticket to the central control terminal for review, and generate a sorting report after the review is passed.

8. The intelligent pulp sorting system according to claim 1, characterized in that: The core of the recognition and counting module is a convolutional neural network model for instance segmentation or object detection, which has been augmented and trained on a pulp cargo image dataset. The model runs in a graphics processor accelerated environment and is used to accurately identify and locate each instance of pulp cargo in the image.

9. The intelligent pulp sorting system according to claim 8, characterized in that: The inventory management application on the mobile terminal receives interactive alarms from the processing server. The alarms include a shortcut to on-site handling operations, allowing inventory managers to directly initiate functions such as retaking photos or submitting text notes on the alarm notification interface, and to associate the handling results with the original abnormal electronic receipt.

10. The intelligent pulp sorting system according to claim 9, characterized in that: The verification interface of the central control terminal is used to display the verification image and the order text information in the electronic ticket side by side or in columns, and provides interactive buttons for "Confirm Passed" and "Mark as Abnormal"; when a confirmation instruction is received through the interactive button, the processing server updates the status field of the corresponding electronic ticket in the database to "Confirmed".

Images