Intelligent manufacturing execution system and method based on face recognition and station card management

Abstract

The invention discloses an intelligent manufacturing execution system and method based on face recognition and station card management. The system comprises a face recognition module, a station card management module, a production scheduling module and a data management module. The face recognition module adopts a face recognition algorithm, verifies the on-duty identity of an employee in combination with a camera, and checks or issues a station card in combination with hierarchical authority control; the station card management module is used for generating, updating and managing station cards of all stations; the production scheduling module is used for specifying, adjusting and executing a production plan, performing optimal resource configuration, allocating station tasks and dynamically adjusting the production progress; and the data management module is used for carrying out centralized management on the station card data, and employee information, production tasks, operation records and equipment states associated with the station cards. According to the invention, the technical problem of how to realize personnel authentication, authority control, real-time guidance and production data closed-loop traceability and construct an efficient and intelligent station execution system is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent manufacturing, in particular to an intelligent manufacturing execution system and method based on face recognition and station card management. BACKGROUND

[0002] In traditional manufacturing, station operation guidance mainly relies on paper station cards, and this mode has many management problems. The update of station cards needs to be reprinted and distributed, which is not only inefficient, but also can easily lead to version management confusion, and the on-site personnel may misuse the expired version. At the same time, paper materials are easily damaged by the environment, and the long-term maintenance cost is high, which brings inconvenience to production management. Especially in the production scene of multi-variety, small batch or frequent type change, the change lag problem of paper station cards is particularly prominent, which seriously affects the production preparation efficiency and operation accuracy. In addition, the station card formats of different processes and different equipment are not unified, and the information presentation mode is not standardized, which further increases the understanding difficulty and error risk of the operating personnel.

[0003] In addition, the existing paper system lacks an identity verification mechanism for operating personnel, and cannot effectively confirm the identity information and post qualification of the operating personnel, making it difficult to achieve permission control and difficult to trace specific responsibilities once a problem occurs. Due to the lack of digital recording means, the key information in the actual operation process is often missing or incomplete, making it difficult to locate quality abnormalities and providing reliable data support for subsequent process optimization and problem analysis.

[0004] The patent document with application number CN202411208255.7 discloses an information security management system based on face recognition, which includes a face recognition unit, an operation start-stop unit and a background service center. The face recognition unit includes a high-definition camera, an intrusion identification module, an AI electronic fence server, a switch and a face recognition background server. The face recognition background server includes a face recognition machine, a face recognition comparison module and a feature extraction module. The face recognition systems and station management systems on the current market are mostly independent, and cannot be organically integrated, resulting in serious data island phenomenon, which makes the station instructions, personnel management and production execution disjointed, the station card content often out of sync with the actual production plan, and unable to real-time feedback production status, restricting the flexibility and accuracy of production scheduling. Therefore, it is urgent to propose an intelligent manufacturing execution system and method based on face recognition and station card management. SUMMARY

[0005] The main purpose of the present application is to propose an intelligent manufacturing execution system and method based on face recognition and station card management, aiming to solve the technical problem of how to realize personnel authentication, permission control, real-time guidance and production data closed-loop tracing, and build an efficient and intelligent station execution system.

[0006] To achieve the above object, the application provides an intelligent manufacturing execution system based on face recognition and workstation card management, wherein the intelligent manufacturing execution system based on face recognition and workstation card management comprises:

[0007] a face recognition module, a workstation card management module, a production scheduling module and a data management module;

[0008] The face recognition module is connected with the workstation card management module and the data management module, the workstation card management module is connected with the production scheduling module and the data management module, and the production scheduling module is connected with the data management module.

[0009] The face recognition module adopts a face recognition algorithm, combines a camera to verify the on-duty identity of an employee, and combines a hierarchical permission control to view and / or issue a workstation card.

[0010] The workstation card management module is used to generate, update and manage the workstation cards of various workstations.

[0011] The production scheduling module is used to specify, adjust and execute a production plan, optimally configure resources, allocate workstation tasks and dynamically adjust production progress.

[0012] The data management module is used to centrally manage the workstation card data, employee information associated with the workstation card, production tasks, operation records and equipment states.

[0013] In one of the preferred schemes, the face recognition module comprises a face detection unit, a face recognition unit and a permission control unit.

[0014] The face detection unit is used to analyze the collected real-time images through a deep learning algorithm, automatically locate the face area and extract key facial feature points.

[0015] The face recognition unit is used to map the input face image into a high-dimensional feature vector by calling a deep convolutional neural network model, and calculate the cosine similarity between the feature vector of the face to be recognized and the feature vector of the registered face in the database.

[0016] The permission control unit is used to dynamically allocate the permission domain area access according to the employee post and safety level.

[0017] In one of the preferred schemes, the accuracy index of the face detection unit is:

[0018]

[0019] wherein, is the detection performance pass rate index, are true positive, true negative, false positive and false negative, respectively.

[0020] In one preferred embodiment, the face recognition unit maps the input face image into a high-dimensional feature vector by a deep convolutional neural network model, and the high-dimensional feature vector is:

[0021]

[0022] wherein, is the high-dimensional feature vector, i.e., the face feature fingerprint, is the normalized face image, is the deep convolutional neural network model.

[0023] The cosine similarity between the face feature vector to be recognized and the registered face feature vector in the database is calculated, and the cosine similarity is:

[0024]

[0025] wherein, is the face feature vector to be recognized and the cosine similarity between the registered face feature vector in the database

[0026] In one preferred embodiment, the face recognition unit further comprises a similarity threshold, and the similarity threshold is dynamically configurable, and the similarity threshold is:

[0027] If the cosine similarity is greater than or equal to the similarity threshold, it is determined that the face recognition is successful; otherwise, the recognition fails.

[0028] In one preferred embodiment, the station card management module comprises a PDF analysis unit, a content editing unit, an image generation unit, and a version management unit.

[0029] The PDF analysis unit is configured to extract the content of the uploaded standardized station card PDF file.

[0030] The content editing unit is configured to modify the content of the station card online based on a Web visual editing interface.

[0031] The image generation unit is configured to convert the edited station card content into an image format suitable for display on the station terminal, and adapt to the screen specifications of different models of station tablets.

[0032] The version management unit is configured to manage the entire life cycle of the station card.

[0033] ​​One of the preferred solutions, the PDF analysis unit uses the pdfplumber tool to extract the content of the uploaded standardized station card PDF file, and matches the key information field through the preset region coordinates, the key information field includes process number, operation instruction, process parameter and quality standard, realizes the efficient extraction of structured data, and calculates the extraction accuracy; The extraction accuracy is:

[0034]

[0035] Among them, The extraction accuracy, The number of correctly identified and matched fields, The total number of fields to be extracted.

[0036] One of the preferred solutions, the production scheduling module includes a production progress tracking unit and a station allocation unit;

[0037] The production progress tracking unit is used to collect the job reporting data of each station, combine the time stamp information, calculate the unit hour output in real time, and display it according to the work order, production line or team; The unit hour output is:

[0038]

[0039] Among them, The unit hour output, The number of qualified outputs, The actual effective working time;

[0040] The station allocation unit is used to realize the reasonable allocation of station resources according to the main production plan and work order task, and comprehensively consider the skill level of employees, equipment state and material preparation condition, and uses a priority scoring model for intelligent matching.

[0041] One of the preferred solutions, the priority scoring model is:

[0042]

[0043] Among them, The allocation priority output by the priority scoring model, The weight coefficients of skill matching degree, equipment availability and material preparation state, The skill matching degree, The equipment availability, The material preparation state.

[0044] A method including the intelligent manufacturing execution system based on face recognition and station card management, comprising the following steps:

[0045] S1, the face image of the employee is collected through the camera, and the on-duty identity of the employee is verified and permission is authenticated through face recognition;

[0046] S2, based on the result of identity authentication, automatically generating or dynamically updating the workstation card according to the production plan of the day, writing task content, process requirements, material information and quality standard data, and distributing the workstation based on the skill level of the employee, work load and equipment state;

[0047] S3, according to the production plan, the priority scoring model is used for intelligent allocation of workstation resources, and real-time collection of each workstation operation report data is performed, combined with timestamp calculation of unit hour output, monitoring of production progress and abnormal early warning;

[0048] S4, a unified data management platform is established to centrally manage workstation card data and employee information, production tasks, operation records and equipment states associated with the workstation card.

[0049] In the above technical scheme of the present application, the intelligent manufacturing execution system based on face recognition and workstation card management includes a face recognition module, a workstation card management module, a production scheduling module and a data management module; the face recognition module is connected with the workstation card management module and the data management module respectively, the workstation card management module is connected with the production scheduling module and the data management module respectively, and the production scheduling module is connected with the data management module; the face recognition module uses face recognition algorithm, combines camera to verify the on-duty identity of the employee, and combines hierarchical permission control to view or publish the workstation card; the workstation card management module is used to generate, update and manage the workstation card of each workstation; the production scheduling module is used to specify, adjust and execute the production plan, to optimally configure resources and to allocate workstation tasks and dynamically adjust production progress; the data management module is used to centrally manage workstation card data and employee information, production tasks, operation records and equipment states associated with the workstation card. The present application solves the technical problem of how to realize human authentication, permission control, real-time guidance and production data closed-loop traceability, and builds an efficient and intelligent workstation execution system.

[0050] In the present application, the real-time update and distribution of operation guidance are realized through electronic workstation card, avoiding the cumbersome process of paper workstation card printing, distribution and version management, and improving the production preparation efficiency.

[0051] In the present application, high-precision face recognition technology is used, combined with living body detection and hierarchical permission control, to ensure that only authorized personnel can access the corresponding workstation and operation permission, effectively preventing misoperation and safety hazards.

[0052] In this invention, the system fully records operations such as workstation card creation, modification, transfer, and data entry, forming a traceable operation audit chain that meets quality management and compliance requirements. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0054] Figure 1 This is a schematic diagram of an intelligent manufacturing execution system based on facial recognition and workstation card management according to an embodiment of the present invention;

[0055] Figure 2 This is a schematic diagram of an intelligent manufacturing execution method based on face recognition and workstation card management according to an embodiment of the present invention;

[0056] Figure 3 This is a flowchart illustrating the employee login process according to an embodiment of the present invention.

[0057] Figure 4 This is a flowchart illustrating the login process for workstation card administrators according to an embodiment of the present invention.

[0058] The realization of the objective, functional characteristics and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0059] 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 a part of the embodiments of the present invention, and not all of them. 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.

[0060] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0061] See Figures 1-4 According to one aspect of the present invention, the present invention provides an intelligent manufacturing execution system based on facial recognition and workstation card management, wherein the intelligent manufacturing execution system based on facial recognition and workstation card management includes:

[0062] Facial recognition module, workstation card management module, production scheduling module, and data management module;

[0063] The face recognition module is connected to the workstation card management module and the data management module, respectively. The workstation card management module is connected to the production scheduling module and the data management module, respectively. The production scheduling module is connected to the data management module.

[0064] The facial recognition module uses a facial recognition algorithm, combined with a camera, to verify the employee's identity and, combined with hierarchical access control, allows the employee to view and / or issue workstation cards.

[0065] The workstation card management module is used to generate, update, and manage workstation cards for each workstation.

[0066] The production scheduling module is used to specify, adjust and execute production plans, optimize resource allocation, assign workstation tasks, and dynamically adjust production progress.

[0067] The data management module is used to centrally manage workstation card data, as well as employee information, production tasks, operation records, and equipment status associated with the workstation cards.

[0068] Specifically, in this embodiment, the face recognition module uses a face recognition algorithm combined with a high-definition camera on a terminal tablet to achieve rapid and accurate verification of employee identity. The system automatically verifies employee identity before they start work, ensuring that personnel information matches their employee ID, position, and qualification data in real time. Simultaneously, the face recognition module sets up hierarchical access control. For example, frontline workers can only view their workstation card and employee ID via face recognition on their workstation tablet, while managers can edit, update, and distribute workstation cards by logging in via computer. The face recognition module includes a face detection unit, a face recognition unit, and an access control unit.

[0069] The face detection unit is used to analyze the acquired real-time images using deep learning algorithms, automatically locate the face region, and extract key facial feature points. The face detection unit maintains high robustness even under complex lighting conditions, pose changes, or partial occlusion, and outputs the face location coordinates. The detection performance of the face detection unit is measured by the accuracy index, which is: [The text then lists the image data and its preprocessing, which are not directly related to the face detection unit description.]

[0070]

[0071] in, To test the performance pass rate index, These are respectively true positives, true negatives, false positives, and false negatives;

[0072] The face recognition unit is used to map the input face image into a high-dimensional feature vector by calling a deep convolutional neural network model, and to calculate the cosine similarity between the feature vector of the face to be recognized and the feature vector of the face registered in the database. The face recognition unit uses the Face++ API as the core visual engine to achieve high-precision face detection, key point localization, feature vector extraction, and identity matching. The system maps the input face image into a high-dimensional feature vector (typically a 512-dimensional Euclidean space vector) by calling the deep convolutional neural network provided by Face++. The high-dimensional feature vector is:

[0073]

[0074] in, This is a high-dimensional feature vector, also known as a facial fingerprint. For the normalized face image, It is a deep convolutional neural network model;

[0075] Calculate the cosine similarity between the feature vector of the face to be identified and the feature vectors of faces registered in the database; the cosine similarity is:

[0076]

[0077] in, The facial feature vector to be identified Registered facial feature vectors in the database Cosine similarity between them;

[0078] The face recognition unit further includes setting a similarity threshold, and the similarity threshold supports dynamic configuration. The similarity threshold is as follows: ;

[0079] If the cosine similarity is greater than or equal to the similarity threshold, the face recognition is considered successful; otherwise, the recognition fails. It integrates Face++ liveness detection to prevent forgery attacks, binds the recognition results to the workstation card and records them in the log, and supports self-learning to update feature templates, thereby improving system security and adaptability.

[0080] The access control unit is used to dynamically allocate access areas based on employee positions and security levels. Frontline workers can only view workstation cards and employee numbers through facial recognition on their workstation tablets. Managers can log in via computer to edit, update, and distribute workstation cards. Access is based on role management to ensure both safety and efficiency.

[0081] Specifically, in this embodiment, the workstation card management module is responsible for generating, updating, and distributing workstation cards for each workstation. The content covers key information such as operating procedures, process parameters, quality standards, and safety reminders. It supports real-time push of the latest version of the workstation card to the corresponding terminal based on product model, batch, or process changes, ensuring accurate and timely information for frontline employees. The workstation card management module also has version management and traceability functions, facilitating standardization and continuous optimization of the production process. The workstation card management module aims to achieve automated processing, visual display, and full lifecycle management of workstation cards. The module includes a PDF parsing unit, a content editing unit, an image generation unit, and a version management unit. The PDF parsing unit extracts content from uploaded standardized workstation card PDF files. The content editing unit allows online modification of workstation card content based on a web-based visual editing interface. The image generation unit converts the edited workstation card content into an image format suitable for display on workstation terminals and adapts to the screen specifications of different workstation tablet models. The version management unit manages the entire lifecycle of the workstation cards.

[0082] Specifically, in this embodiment, the PDF parsing unit uses the pdfplumber tool to extract content from the uploaded standard chemical work card PDF file, including text content, coordinate positions, font styles, and table data; and matches key information fields through preset area coordinates. These key information fields include process numbers, operating instructions, process parameters, and quality standards, achieving efficient extraction of structured data and calculating the extraction accuracy rate. The extraction accuracy rate is:

[0083]

[0084] in, To improve accuracy, To ensure the correct number of fields are identified and matched, This represents the total number of fields to be extracted.

[0085] After content extraction, the structured data enters the content editing unit, providing a web-based visual editing interface. This allows workstation card administrators to modify the workstation card content online, including adding or deleting steps, adjusting parameters, and replacing descriptive text. This module supports multi-level access control to ensure that authorized personnel can perform editing operations. All changes are recorded with the operator and timestamp, forming a complete operation log to meet auditing and traceability requirements. The image generation unit is responsible for converting the edited workstation card content into an image format suitable for display on the workstation terminal. The system first renders the structured data into a high-resolution page, and then uses the pdf2image tool to call the Poppler image engine to batch convert the page content into high-quality PNG images. This process supports custom resolution (such as DPI settings) and output size, mainly adapting to the screen specifications of different models of workstation tablets.

[0086] Specifically, in this embodiment, the production scheduling unit, as the core of production management, is responsible for formulating, adjusting, and executing production plans to achieve optimal resource allocation. The system intelligently allocates workstation tasks and dynamically adjusts the production pace based on order demand, equipment status, personnel scheduling, and material supply. Simultaneously, it supports real-time monitoring of the production progress of each workstation, timely warnings of abnormal situations, and improved overall production efficiency and response speed. The production scheduling module includes a production progress tracking unit and a workstation allocation unit. The production progress tracking unit collects work report data from each workstation, combines it with timestamp information, calculates the hourly output in real time, and displays it in summary by work order, production line, or shift. The workstation allocation unit, based on the master production plan and work order tasks, achieves reasonable allocation of workstation resources, comprehensively considers employee skill levels, equipment status, and material availability, and uses a priority scoring model for intelligent matching. The production scheduling module displays the summarized UPH data by work order, production line, or shift on the tablet terminal page.

[0087] Specifically, in this embodiment, the unit hourly output is:

[0088]

[0089] in, Output per unit hour To achieve the required output quantity, This refers to the actual effective working time.

[0090] Specifically, in this embodiment, the priority scoring model is:

[0091]

[0092] in, The priority assignment is output by the priority scoring model. These are the weighting coefficients for skill matching, equipment availability, and material readiness status, respectively. For skill matching, For equipment availability, The material is ready.

[0093] Specifically, in this embodiment, the data management module provides the system with a unified data storage and management platform, centrally storing core data such as workstation cards, operation records, and task execution. The face recognition module, workstation card management module, and production scheduling module establish bidirectional data interaction with the data management module. The face recognition module synchronizes the identity verification results and permission information to the data management module in real time. The workstation card management module stores data such as the version and content updates of electronic workstation cards into the data management module. The production scheduling module uploads data such as task allocation and progress monitoring to the data management module in real time.

[0094] Specifically, in this embodiment, the intelligent manufacturing execution system based on face recognition and workstation card management further includes an image acquisition and preprocessing module, a face recognition communication module, and a latency monitoring module;

[0095] The image acquisition and preprocessing module acquires employee facial images through a front-facing camera and performs standardized processing, including illumination compensation, noise filtering, image enhancement, face alignment, and size normalization, to improve image quality, eliminate environmental interference, and ensure that the images input to the face recognition module are clear and consistent, thereby improving recognition accuracy and robustness.

[0096] The face recognition communication module allows for data interaction between the workstation terminal and cloud or local face recognition services. It receives preprocessed image data, sends recognition requests, and parses the returned identity information and authorization results to achieve real-time identity verification. It also supports functions such as disconnection reconnection and encrypted data transmission to ensure the stability and security of communication.

[0097] The latency monitoring module is used to monitor the response time of the entire face recognition process in real time, recording the time consumed in each stage of client image acquisition, data transmission, server processing, and result return, analyzing system performance bottlenecks, ensuring that the recognition process is efficient and smooth, meeting the real-time requirements of the production site, and providing data support for system optimization and resource scheduling.

[0098] See Figure 2 This invention provides an intelligent manufacturing execution method based on facial recognition and workstation card management, wherein the intelligent manufacturing execution method based on facial recognition and workstation card management includes the following steps:

[0099] S1. Collect employees' facial images through cameras and verify their identity and authorization through facial recognition.

[0100] S2. Based on the identity authentication result, automatically generate or dynamically update workstation cards according to the daily production plan, write task content, process requirements, material information and quality standard data, and allocate workstations based on employee skill level, workload and equipment status.

[0101] S3. Based on the production plan, a priority scoring model is used to intelligently allocate workstation resources, and work report data of each workstation is collected in real time. The hourly output per unit is calculated by combining the timestamp, and the production progress is monitored and anomaly warnings are issued.

[0102] S4. Establish a unified data management platform to centrally manage workstation card data, as well as employee information, production tasks, operation records, and equipment status associated with the workstation cards.

[0103] Specifically, in this embodiment, step S1 is as follows:

[0104] S11. Image Acquisition: When personnel enter the work site, the system uses a camera on a tablet or computer to capture the employee's face image in real time. The captured images are compressed or converted in format during the upload process to ensure transmission efficiency and image quality. At the same time, the system can select single-frame image acquisition or continuous multi-frame acquisition according to the actual scenario to improve the accuracy and stability of face detection in complex environments such as uneven lighting and pose changes.

[0105] S12. Recognition: After receiving a face image, the system calls a face recognition algorithm to extract features and compare identities, outputting the corresponding person's identity information and matching score. To ensure the reliability of recognition, the system introduces a threshold judgment mechanism during the comparison process. If the matching score is higher than the preset threshold, the recognition result is considered valid; if it is lower than the threshold, the system prompts for re-collection.

[0106] S3. Confirmation and Recording: After the identification result meets the threshold requirements, the system further verifies the personnel's identity information by combining it with the current line number, process link, and work location information to confirm the consistency between the personnel and their positions. After the verification is completed, the system generates the final confirmation result and displays it on the terminal device. At the same time, the system can automatically record the identification and confirmation data to form complete log information for subsequent statistical analysis, anomaly tracing, and audit management.

[0107] Specifically, in this embodiment, step S2 is as follows:

[0108] S21. File Upload: The user first uploads the workstation card PDF file. The system receives, verifies, and stores the uploaded file to ensure file integrity and security. The upload operation not only serves as the starting point for parsing and processing but also provides the original basis for subsequent data tracking. The system records the file upload time, uploader, and file attributes for easy auditing and management later.

[0109] S22. PDF parsing: This tool parses the uploaded file content, including table structure, text information, and format style recognition. The parsing process automatically extracts table cells, titles, paragraphs, and key fields, providing an accurate data foundation for subsequent workstation information recognition. Through parsing, the originally static PDF content is transformed into structured digital data, making information extraction, comparison, and storage more efficient and reliable.

[0110] S23. Workstation information extraction: After parsing, the system aligns, standardizes, and intelligently matches key information by combining the built-in job dictionary and precautions library. For example, the system can automatically identify workstation number, process steps, job responsibilities, and precautions, and perform field standardization. This step ensures the consistency and accuracy of workstation information, avoids errors caused by manual input, and supports subsequent data analysis, traceability, and process optimization.

[0111] S24. Page Segmentation and Image Generation: Render each page of the PDF into a high-quality image file, and segment the page according to the content to generate corresponding PNG or JPEG preview images; These images are convenient for front-end interface display and also facilitate manual proofreading and confirmation of parsing results; Through image presentation, users can intuitively view the workstation card content, layout and key process information, while supporting quick location and modification of abnormal data;

[0112] S25. Data entry or update: The system performs intelligent storage processing on the parsed and assembled workstation card data. When the same "table type + process" combination does not exist in the database, the system performs a batch insertion operation to completely store the new workstation card data into the database. When the same combination already exists, an update operation is performed to replace the original record. This processing logic ensures the integrity and real-time performance of the data, while providing a basis for the version iteration of workstation cards.

[0113] S26. Version Management: Upon initial data entry, the system automatically sets a default version number of 1.0 for each workstation card. During subsequent updates, users can input a new version number through the front end for version iteration management. The system uses model, process, and index number to form unique key-value pairs, enabling precise record coverage and version tracking. This mechanism not only ensures the traceability of historical data but also supports process improvement and change management, providing reliable data for production management.

[0114] S27. Front-end display: The parsed workstation card data and corresponding images will be synchronized to the front-end interface, allowing users to intuitively view workstation information, precautions, and historical version records. The front-end display not only supports visual browsing of data, but also allows users to quickly verify, query, and retrieve data, making it convenient for production managers to grasp the process status and workstation arrangement in real time, thereby improving production efficiency and management accuracy.

[0115] When frontline employees log into the system, they will be directly redirected to the workstation card display interface. This interface provides frontline employees with an intuitive view of workstation usage, helping them quickly identify their work location or understand the details of their current workstation allocation. This design simplifies the employee query process and improves work efficiency.

[0116] Specifically, in this embodiment, workstation managers need to create a personal account by registering when using the system for the first time. This process ensures the authenticity of the manager's identity and the security of the system. After registering and logging in, the manager will gain access to the workstation card maintenance module. Here, a series of management tasks can be performed, including but not limited to adding new workstation card information, deleting expired or unused workstation records, and updating relevant information for existing workstations. This not only enhances the flexibility of workstation resource management but also ensures the accuracy and real-time nature of the data.

[0117] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. All equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. An intelligent manufacturing execution system based on face recognition and station card management, characterized in that, The application relates to a face recognition module, a work station card management module, a production scheduling module and a data management module. The face recognition module is connected with the work station card management module and the data management module, the work station card management module is connected with the production scheduling module and the data management module, and the production scheduling module is connected with the data management module. The face recognition module adopts a face recognition algorithm, combines a camera to verify the on-duty identity of an employee, and combines hierarchical permission control to check and / or release a work station card. The work station card management module is used for generating, updating and managing work station cards of various work stations. The production scheduling module is used for specifying, adjusting and executing a production plan, optimally configuring resources, allocating work station tasks and dynamically adjusting production progress. The data management module is used for centrally managing work station card data and employee information, production tasks, operation records and equipment states associated with the work station card. The face recognition module comprises a face detection unit, a face recognition unit and a permission control unit.

2. The intelligent manufacturing execution system based on face recognition and station card management according to claim 1, characterized in that, The face detection unit is used for automatically locating a face area and extracting key facial feature points by analyzing real-time images collected through a deep learning algorithm. The face recognition unit is used for mapping an input face image into a high-dimensional feature vector through a deep convolutional neural network model and calculating the cosine similarity between a feature vector of a face to be recognized and a feature vector of a registered face in a database. The permission control unit is used for dynamically allocating a permission domain area access according to an employee post and a safety level. The accuracy rate of the face detection unit is: 3.The intelligent manufacturing execution system based on face recognition and station card management according to claim 2, characterized in that, The face recognition unit maps an input face image into a high-dimensional feature vector through a deep convolutional neural network model, and the high-dimensional feature vector is: ； wherein, is a detection performance pass indicator, are true positive, true negative, false positive and false negative, respectively.

4. The intelligent manufacturing execution system based on face recognition and station card management according to claim 2, characterized in that, The cosine similarity between a feature vector of a face to be recognized and a feature vector of a registered face in a database is calculated. ； wherein, is a high-dimensional feature vector, i.e., a face feature fingerprint, is a normalized face image, is a deep convolutional neural network model; If the cosine similarity is greater than or equal to a similarity threshold, it is determined that face recognition is successful; otherwise, face recognition fails. ； wherein, is the face feature vector to be identified is the registered face feature vector in the database is the cosine similarity between the two.

5. The intelligent manufacturing execution system based on face recognition and station card management according to claim 4, characterized in that, The face recognition unit further comprises a similarity threshold value, and the similarity threshold value supports dynamic configuration, and the similarity threshold value is: ; The work station card management module comprises a PDF analysis unit, a content editing unit, an image generation unit and a version management unit.

6. The intelligent manufacturing execution system based on face recognition and station card management according to any one of claims 1-5, characterized in that, The PDF analysis unit is used for extracting the content of an uploaded standardized work station card PDF file. The content editing unit modifies the content of a work station card online based on a Web visual editing interface. The image generation unit is used for converting the edited content of a work station card into an image format suitable for display on a work station terminal and adapting the screen specifications of different types of work station tablets. The version management unit is used for managing the whole life cycle of a work station card. The PDF analysis unit extracts the content of an uploaded standardized work station card PDF file through a pdfplumber tool and matches key information fields through preset area coordinates, the key information fields comprise a process number, operation instructions, process parameters and quality standards, efficient extraction of structured data is realized, and an extraction accuracy rate is calculated; the extraction accuracy rate is:

7. The intelligent manufacturing execution system based on face recognition and station card management according to claim 6, characterized in that, The production scheduling module comprises a production progress tracking unit and a work station allocation unit. ； wherein, is the extraction accuracy, is the number of correctly identified and matched fields, is the total number of fields to be extracted.

8. The intelligent manufacturing execution system based on face recognition and station card management according to any one of claims 1-5, characterized in that, ​ The production progress tracking unit is used to collect the work reporting data of each station, combine the time stamp information, calculate the unit hour output in real time, and display the work order, production line or team summary; the unit hour output is: ； wherein, is the hourly output, is the quantity of acceptable output, is the actual effective working time; The station allocation unit is used to realize the reasonable allocation of station resources according to the main production plan and work order task, comprehensively consider the skill level of employees, the state of equipment and the material complete condition, and adopt the priority scoring model for intelligent matching. 9.The intelligent manufacturing execution system based on face recognition and station card management of claim 8, wherein, The priority scoring model is: ； wherein, a priority assigned by a priority score model, a weight coefficient for the skill matching degree, the equipment availability, and the material preparation state, respectively, a skill matching degree, an equipment availability, a material preparation state.

10. A method comprising the intelligent manufacturing execution system based on face recognition and station card management according to any one of claims 1-9, characterized in that, The method comprises the following steps: S1, acquiring the face image of the employee through the camera, and verifying and authenticating the on-duty identity and authority of the employee through face recognition; S2, based on the result of identity authentication, automatically generating or dynamically updating the station card according to the production plan of the day, writing in the task content, process requirement, material information and quality standard data, and distributing the station based on the skill level of the employee, the work load and the state of the equipment; S3, using the priority scoring model to intelligently allocate the station resources according to the production plan, collecting the work reporting data of each station in real time, combining the time stamp to calculate the unit hour output, monitoring the production progress and giving abnormal early warning; S4, establishing a unified data management platform to centrally manage the station card data and the employee information, production task, operation record and equipment state associated with the station card.

Images