Electrical assembly process report generation method and system, and computer equipment

Abstract

The invention relates to an electrical assembly process technology, and discloses an electrical assembly process report generation method and system, and computer equipment, and the method comprises the steps: obtaining an electrical assembly image; performing image recognition on the electrical assembly image to obtain contour information in the electrical assembly image; performing character recognition on the electrical assembly image to obtain technical requirement information of the electrical assembly image and bit number information of each component; obtaining an adjacent matrix of the electrical assembly image according to the contour information and the information of each bit number; inputting the contour information, the adjacent matrix, the technical requirement information and each bit number information into a graph neural network to generate an assembly logic file; and inputting the preset process rule set and the assembly logic file into a process report generation model to generate an electrical assembly process report. The electrical assembly image information is extracted through the image and character recognition technology, the process report is automatically generated in combination with the process rule, and the report generation efficiency and accuracy are improved.

Description

Technical Field

[0001] This application relates to the technical field of electrical assembly processes, and particularly to a method, a system, and a computer device for generating an electrical assembly process report. Background Art

[0002] The compilation of traditional electrical assembly process reports usually relies on manual identification of drawings, searching for component parameters, and manually writing process reports. Specifically, engineers need to spend a lot of time interpreting circuit diagrams and searching for specific information of components from various materials. This process is not only time-consuming but also may lead to omission of key data or misunderstanding of design intentions due to negligence. Moreover, the manually written process documents often have inconsistent formats and vague descriptions, increasing confusion and the risk of misoperation in actual operation. Thus, it seriously affects the production efficiency and the consistency of product quality. Summary of the Invention

[0003] In view of this, embodiments of this application provide a method, a system, and a computer device for generating an electrical assembly process report, which can effectively solve the problems in the prior art that the compilation of electrical assembly process reports relies on manual identification of drawings, searching for component parameters, and manually writing process reports, resulting in low efficiency, easy errors, poor standardization, and seriously affecting the production efficiency and the consistency of product quality, etc.

[0004] In a first aspect, embodiments of this application provide a method for generating an electrical assembly process report, the method including: Obtain an electrical assembly image; Perform image recognition on the electrical assembly image to obtain contour information in the electrical assembly image; Perform character recognition on the electrical assembly image to obtain technical requirement information and part number information of each component in the electrical assembly image; According to the contour information and each of the part number information, obtain an adjacency matrix of the electrical assembly image; Input the contour information, the adjacency matrix, the technical requirement information, and each of the part number information into a graph neural network to generate an assembly logic file; Input a preset process rule set and the assembly logic file into a process report generation model to generate an electrical assembly process report.

[0005] In some embodiments, after obtaining the adjacency matrix of the electrical assembly image according to the contour information and each of the part number information, the method further includes: Obtain a bill of materials; Query model information of each component corresponding to each of the part number information in the bill of materials respectively; Obtain the three-dimensional spatial structure information of each of the components in a preset component database according to each of the model information; Input the three-dimensional spatial structure information, the contour information, the adjacency matrix, each of the part number information, and the technical requirement information into the graph neural network to generate the assembly logic file.

[0006] In some embodiments, the performing image recognition on the electrical assembly image to obtain the contour information in the electrical assembly image includes: Use OpenCV to extract the contours in the electrical assembly image to obtain the contour information of each connecting line, as well as the contour information of each component and the center coordinates of each component; Use a vision large model to classify the contour information of each component to obtain each contour category, and obtain the connection relationship of each component according to the contour information of each connecting line. The contour information includes each center coordinate, each contour category, and the connection relationship.

[0007] In some embodiments, the obtaining the adjacency matrix of the electrical assembly image according to the contour information and each of the part number information includes: Obtain the adjacency matrix with each component as a node according to each of the part number information, each of the center coordinates, and the connection relationship.

[0008] In some embodiments, the edge weights of the adjacency matrix include: the distance and connection type between the nodes. In some embodiments, after inputting the preset process rule set and the assembly logic file into the process report generation model to generate an electrical assembly process report, the method further includes: Extract the information in the electrical assembly process report using regular expressions according to a preset standard template; Fill the extracted information into the preset standard template to generate a process specification card.

[0009] In some embodiments, before inputting the preset process rule set and the assembly logic file into the process report generation model to generate an electrical assembly process report, the method further includes: Collect historical electrical assembly images, historical bills of materials, and historical electrical assembly process reports to form an electrical assembly data sample set; Preprocess the electrical assembly data sample set to obtain a standard electrical assembly data sample set; Input the standard electrical assembly data sample set into a general language model for optimization training to obtain the trained process report generation model.

[0010] In a second aspect, an electrical assembly process report generation system provided by an embodiment of the present application includes: An image acquisition module for acquiring electrical assembly images; An image recognition module for performing image recognition on the electrical assembly images to obtain contour information in the electrical assembly images; A character recognition module for performing character recognition on the electrical assembly images to obtain technical requirement information and part number information of each component in the electrical assembly images; A matrix acquisition module for obtaining an adjacency matrix of the electrical assembly images according to the contour information and each part number information; An assembly logic file generation module for inputting the contour information, the adjacency matrix, the technical requirement information and each part number information into a graph neural network to generate an assembly logic file; A report generation module for inputting a preset process rule set and the assembly logic file into a process report generation model to generate an electrical assembly process report.

[0011] In a third aspect, an embodiment of the present application provides a computer device, which includes a processor and a memory. The memory stores a computer program, and the processor is used to execute the computer program to implement the above electrical assembly process report generation method.

[0012] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed on a processor, the above electrical assembly process report generation method is implemented.

[0013] The embodiments of the present application have the following beneficial effects: The electrical assembly process report generation method of the embodiment of the present application includes: acquiring electrical assembly images; performing image recognition on the electrical assembly images to obtain contour information in the electrical assembly images; performing character recognition on the electrical assembly images to obtain technical requirement information and part number information of each component in the electrical assembly images; obtaining an adjacency matrix of the electrical assembly images according to the contour information and each part number information; inputting the contour information, the adjacency matrix, the technical requirement information and each part number information into a graph neural network to generate an assembly logic file; inputting a preset process rule set and the assembly logic file into a process report generation model to generate an electrical assembly process report. By extracting electrical assembly image information through character recognition and image recognition and combining process rules, the process report generation model automatically generates an electrical assembly process report, greatly improving the generation efficiency and accuracy of the report. Description of the Drawings

[0014] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0015] Figure 1 Shows the first flowchart of the method for generating an electrical assembly process report according to an embodiment of the present application; Figure 2 Shows the second flowchart of the method for generating an electrical assembly process report according to an embodiment of the present application; Figure 3 Shows the third flowchart of the method for generating an electrical assembly process report according to an embodiment of the present application; Figure 4 Shows the fourth flowchart of the method for generating an electrical assembly process report according to an embodiment of the present application; Figure 5 Shows the fifth flowchart of the method for generating an electrical assembly process report according to an embodiment of the present application; Figure 6 Shows a schematic structural diagram of an electrical assembly process report generation system according to an embodiment of the present application. Detailed implementation manners

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments.

[0017] Generally, the components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0018] As used hereinafter, the terms "including", "having" and their cognates that may be used in various embodiments of the present application are only intended to indicate specific features, numbers, steps, operations, elements, components or combinations of the foregoing items, and should not be construed as first excluding the existence of one or more other features, numbers, steps, operations, elements, components or combinations of the foregoing items or precluding the possibility of adding one or more features, numbers, steps, operations, elements, components or combinations of the foregoing items. In addition, the terms "first", "second", "third", etc. are only used for differentiating descriptions and cannot be construed as indicating or implying relative importance.

[0019] Unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the various embodiments of the present application pertain. The terms (such as those defined in a commonly used dictionary) will be construed to have the same meaning as the contextual meaning in the relevant technical field and will not be construed to have an idealized meaning or an overly formal meaning unless clearly defined in the various embodiments of the present application.

[0020] The following will describe in detail some embodiments of the present application in conjunction with the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0021] Considering the problems in the prior art that the compilation of electrical assembly process reports relies on manual identification of drawings, searching for component parameters, and manually writing process reports, resulting in low efficiency, easy errors, and poor standardization, which seriously affect the production efficiency and the consistency of product quality, etc. The present application provides a method, a system, and a computer device for generating an electrical assembly process report. By character recognition and image recognition, electrical assembly image information is extracted, and combined with process rules, a process report generation model automatically generates an electrical assembly process report, greatly improving the generation efficiency and accuracy of the report.

[0022] The following will describe the method for generating an electrical assembly process report in conjunction with some specific embodiments.

[0023] Figure 1 A flowchart showing a method for generating an electrical assembly process report according to an embodiment of the present application is shown. It can be understood that the electrical assembly process report can be in the form of a process document, and the electrical assembly process report can also be in natural language. The electrical assembly process report includes the assembly sequence, wiring logic, process requirements, etc. of each component. Exemplarily, the report generation method is executed by a processor in a computer device and includes steps S101 to step S106: S101, obtain an electrical assembly image.

[0024] It is understandable that relevant functions of OpenCV can be used to load the drawing and convert it into an image format. The user uploads the electrical assembly image and saves it to the database. The processor obtains the electrical assembly image from the database. It is understandable that the electrical assembly image can be an image in any format, such as DWG format or PDF format, etc.

[0025] In one embodiment, the electrical assembly image can be a two-dimensional planar structure image, and the electrical assembly image can also be a three-dimensional spatial structure image. For example, the two-dimensional planar structure electrical assembly image contains the part numbers, two-dimensional planar structure and angles of each component, as well as text information such as technical requirements, etc., and can describe the two-dimensional planar structure positional relationship, connection relationship and assembly requirements of each component; the three-dimensional spatial structure electrical assembly image includes the part numbers, three-dimensional spatial structure and angles of each component, as well as text information such as technical requirements, etc., and can describe the three-dimensional spatial structure positional relationship, connection relationship and assembly requirements of each component.

[0026] S102. Perform image recognition on the electrical assembly image to obtain the contour information in the electrical assembly image.

[0027] Use an image recognition algorithm to perform image recognition to extract the contour information in the image. Among them, the contour information includes the positions, sizes and shapes of the contours of each component in the image, as well as the positions, sizes and shapes of the contours of the connecting lines, etc.

[0028] For example, in one implementation, as Figure 2 shown, S102 includes the following sub-steps: S201. Use OpenCV to extract the contours in the electrical assembly image to obtain the contour information of each connecting line, as well as the contour information of each component and the center coordinates of each component.

[0029] OpenCV is a computer vision library with rich image processing and computer vision functions. Further, before contour extraction, the image can be preprocessed. Specifically, relevant functions of OpenCV can be used to load the image and convert it into the required image format. Or, before contour extraction, if the electrical assembly drawing received by the processor, the drawing can also be preprocessed. Specifically, relevant functions of OpenCV can be used to load the drawing and convert it into an image format, and then convert the color image into a grayscale image, and use Gaussian blur to remove image noise, and convert the denoised image into a binary image to facilitate contour extraction.

[0030] Furthermore, use an edge detection function (such as Canny edge detection, etc.) to extract all the contours in the image, including the contours of each component and the contours of each connecting line, and calculate geometric features such as the area, perimeter, and center coordinates of each contour. It can be understood that the contours in the electrical assembly image can be two-dimensional planar structure contours or three-dimensional spatial structure solid contours. If they are two-dimensional planar structure contours, the contours of each component represent the two-dimensional planar structure of each component; if they are three-dimensional spatial structure solid contours, the contours of each component represent the three-dimensional spatial structure of each component.

[0031] S202, use a large vision model to classify the contour information of each component to obtain each contour category, and obtain the connection relationship of each component according to the contour information of each connecting line. The contour information includes each center coordinate, each contour category, and the connection relationship.

[0032] Specifically, the large vision model can be selected according to the actual application situation. A model based on a convolutional neural network can be selected, or a model with a Transformer architecture can be selected, or a hybrid model combining the two can be selected, etc.

[0033] Exemplarily, use the template or features of the components to train the model so that the model can obtain the component type corresponding to the contour through the contour information or contour features of each component. Input the contour information of each component into the model, so that the model compares the contour information of each component with the known component templates to identify the type of the component; or input the contour features of each component into the model, so that the model compares the contour features of each component with the features of the components to identify the specific component type, for example, resistors, capacitors, switches, etc.

[0034] A connection line detection model can be configured, and the contour information of each connection line is input into the connection line detection model to extract the path information of the connection line.

[0035] Use OpenCV and a large vision model to perform image recognition on the electrical assembly image to obtain the contour information of the electrical assembly image. OpenCV can quickly and accurately extract key contour information from the electrical assembly image, and the large vision model can further improve the accuracy of contour extraction, greatly improving the processing speed and accuracy, and at the same time reducing the risk caused by human errors.

[0036] S103, perform character recognition on the electrical assembly image to obtain the technical requirement information of the electrical assembly image and the reference designator information of each component.

[0037] Use OCR tools such as Tesseract and PaddleOCR to extract the text in the image, obtain the part number information of each component and the technical requirement information in the image, match the extracted part number information with the contour information of each component, and determine the part number of each component contour information. Use OCR to recognize the characters in the image, obtain the technical requirement information and part number information of the image. OCR technology can quickly scan and recognize a large number of images, greatly reducing the time of manual reading and recording. Moreover, the extracted information can be converted into a standardized data format for subsequent processing and analysis. Thus, it improves work efficiency, reduces errors, and lowers labor costs.

[0038] S104. According to the contour information and each part number information, obtain the adjacency matrix of the electrical assembly image.

[0039] According to each part number information, each center coordinate and the connection relationship, obtain the adjacency matrix with each component as a node. The edge weights of the adjacency matrix can be set according to the actual application situation. Demonstratively, the edge weights of the adjacency matrix include: the distance between nodes and the connection type. If the electrical assembly image is a three-dimensional space structure image, the edge weights of the adjacency matrix can also include the hierarchical relationship between each node.

[0040] Specifically, regard each component as a node, record information such as the type, center coordinate and part number of the component, construct a node list for all components according to the part number, create an initial adjacency matrix, and set the initial value to infinity or zero, indicating that there is no direct connection between nodes.

[0041] Calculate the distance between components according to the path information of the connecting line, match the connecting line and the components, obtain the connection relationship of each component, determine the connection type between each pair of components from the connection relationship data, convert the connection type into a numerical weight coefficient, calculate the distance between each node, combine the distance between nodes and the weight coefficient of the connection type to form a comprehensive weight, write the comprehensive weight into the initial adjacency matrix, and update the initial adjacency matrix until all connection relationships are processed.

[0042] According to the contour information and each part number information, obtain the adjacency matrix of the electrical assembly image. The adjacency matrix is a structured data representation method that can clearly show the connection relationship between components, facilitate computer program processing, and through the adjacency matrix, more reasonable component layout and wiring planning can be carried out, reducing the wiring length and electromagnetic interference.

[0043] S105. Input the contour information, adjacency matrix, technical requirement information and each part number information into the graph neural network to generate an assembly logic file.

[0044] Specifically, if the image is an electrical assembly image with a three-dimensional spatial structure, the contour information includes the three-dimensional spatial structures of various components. The contour information, adjacency matrix, technical requirement information, and each part number information can be directly input into the graph neural network to generate an assembly logic file. If the image is an image with a two-dimensional planar structure, the three-dimensional spatial structures of various components need to be obtained, and then the three-dimensional spatial structure information, contour information, adjacency matrix, technical requirement information, and each part number information are input into the graph neural network to generate an assembly logic file.

[0045] For example, in one implementation, as Figure 3 shown, after obtaining the adjacency matrix, the electrical assembly process report generation method further includes S301 - S303: S301, obtain a bill of materials.

[0046] Specifically, the bill of materials includes the part numbers and models of various components on the electrical assembly image, and the part numbers and models correspond one by one. The user can upload the bill of materials in advance and save it to the database, and the processor obtains the bill of materials from the database. The bill of materials can be a table in any format. Exemplarily, the bill of materials is in Word format.

[0047] S302, respectively query the model information of each component corresponding to each part number information in the bill of materials.

[0048] A hash table can be used to query in the bill of materials to obtain the corresponding models of each component.

[0049] S303, respectively obtain the three-dimensional spatial structure information of each component in the preset component database according to each model information.

[0050] It can be understood that the preset component database stores the three-dimensional spatial structures and models of various components, and the models of the components correspond one by one to the three-dimensional spatial structures of the components. Further, the database can also include information such as the materials and weights of the components.

[0051] Taking the model of each component as the primary key, search for the corresponding record in the preset component database, obtain the three-dimensional spatial structure information of the component, read the three-dimensional spatial structure information from the database, convert the structure of the three-dimensional spatial structure information according to the actual application situation, and output it. Exemplarily, the three-dimensional spatial structure information can be converted into JSON format.

[0052] Search for the model in the bill of materials according to the part number, and obtain the three-dimensional spatial structure information according to the model. Through the three-dimensional spatial structure information, the position of each component in the assembly can be determined more accurately, thereby optimizing the overall layout and improving the assembly efficiency and quality.

[0053] S304, Input the three-dimensional spatial structure information, contour information, adjacency matrix, each part number information, and technical requirement information into the graph neural network to generate an assembly logic file.

[0054] Specifically, each component can be used as a node of the graph to construct a graph structure. The features of the node are set as the three-dimensional spatial structure information, contour information, adjacency matrix, and part number information. The edges of the graph are constructed according to the adjacency matrix. The features of the edges include distance, connection type, hierarchical relationship, and technical requirement information, etc. The technical requirement information can also be used as global features and input into the graph neural network.

[0055] Use graph neural network models such as graph convolutional network or graph attention network to process the graph structure data to obtain the assembly logic. The assembly logic output by the model can be converted into a specific assembly logic file format, such as XML format, JSON format, etc. The assembly logic file can include information such as assembly sequence, assembly path, and assembly tool selection.

[0056] Inputting the three-dimensional spatial structure information, contour information, adjacency matrix, each part number information, and technical requirement information into the graph neural network to generate an assembly logic file can optimize the pre-assembly sequence, improve the assembly accuracy, and also improve the accuracy and consistency of the process report.

[0057] S106, Input the preset process rule set and the assembly logic file into the process report generation model to generate an electrical assembly process report.

[0058] It can be understood that the preset process rule geometry can include process constraints, operation specifications, priorities, and standard processes, etc. Before inputting the preset process rule set and the assembly logic file into the process report generation model to generate an electrical assembly process report, the model can be trained using historical data.

[0059] For example, in one implementation, as Figure 4 shown, the training method includes S401 - S403: S401, Collect historical electrical assembly images, historical bills of materials, and historical electrical assembly process reports to form an electrical assembly data sample set. After classifying each data, extract relevant information to form an electrical assembly data sample set.

[0060] S402, Preprocess the electrical assembly data sample set to obtain a standard electrical assembly data sample set.

[0061] Specifically, the data in the electrical assembly data sample set can be cleaned and transformed. Further, a small amount of noise can be added to the data to improve the robustness of the model.

[0062] S403. Input the standard electrical assembly data sample set into a general language model for optimization training to obtain a trained process report generation model.

[0063] Optionally, optimize and train a model based on rules or machine learning. Demonstratively, use a model based on the Transformer architecture. Input the standard electrical assembly data sample set into the general language model for optimization training, enabling the model to extract key information from the input data and generate a complete report according to a predefined template. The module can be set according to the actual application situation to ensure that the generated report content is accurate, the format is standardized, and it meets the actual requirements.

[0064] Input the preset process rule set and the assembly logic file into the trained model. Extract the constraint conditions and operation requirements from the process rules, and extract the component connection sequence and operation steps from the assembly logic. The model infers based on the input process rules and assembly logic to generate an electrical assembly process report.

[0065] Through the automated process report generation model, the preset process rules and assembly logic can be quickly integrated, thus significantly shortening the time for manually writing process reports. Moreover, the preset process rule set can provide a unified standard and specification for the entire process flow, ensuring that each generated process report follows the same technical requirements and operation specifications, and significantly improving the standardization level of electrical assembly processes.

[0066] Furthermore, after generating the electrical assembly process report, a process specification card can also be generated according to the content of the electrical assembly process report. For example, in one implementation, as Figure 5 shown, after S106, the process specification card is generated through the following sub-steps: S501. Extract the information in the electrical assembly process report using regular expressions according to a preset standard template.

[0067] The preset standard template can be set according to the actual application situation. For example, according to the process standards of different enterprises, obtain the standard template. The standard template can be a formatted process card, etc. Analyze the electrical assembly process report using regular expressions to extract the key fields that need to be filled into the process card, such as process name, equipment parameters, operation steps, etc. S502. Fill the extracted information into the preset standard template to generate a process specification card.

[0068] The output template can be automatically adjusted according to the process standard library of different enterprises to ensure that the generated process cards fully meet the enterprise requirements. Furthermore, format verification and version management can be performed on the process cards. Realize the automated processing from electrical images to process cards, greatly improving the process preparation efficiency.

[0069] AsFigure 6 As shown in Figure 6 , based on the method of the above embodiment, this embodiment provides an electrical assembly process report generation system. Exemplarily, the electrical assembly process report generation system 100 includes: An image acquisition module 110, which acquires electrical assembly images; An image recognition module 120, which performs image recognition on the electrical assembly images to obtain contour information in the electrical assembly images; A character recognition module 130, which performs character recognition on the electrical assembly images to obtain technical requirement information and part number information of each component in the electrical assembly images; A matrix acquisition module 140, which obtains an adjacency matrix of the electrical assembly images according to the contour information and each part number information; An assembly logic file generation module 150, which inputs the contour information, the adjacency matrix, the technical requirement information, and each part number information into a graph neural network to generate an assembly logic file; A report generation module 160, which inputs a preset process rule set and the assembly logic file into a process report generation model to generate an electrical assembly process report.

[0070] It can be understood that the system in this embodiment corresponds to the control method in the above embodiment, and the optional items in the above embodiment are also applicable to this embodiment, so they will not be described repeatedly here.

[0071] This application also provides a computer device. Exemplarily, the computer device includes a processor and a memory. Among them, the memory stores a computer program, and the processor runs the computer program to enable the device to execute the above electrical assembly process report generation method or the functions of each module in the above electrical assembly process report generation system.

[0072] Among them, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including at least one of a central processing unit (CPU), a graphics processing unit (GPU), a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., and can implement or execute the various methods, S, and logic block diagrams disclosed in the embodiments of the present application.

[0073] The memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electric Erasable Programmable Read-Only Memory (EEPROM), etc. Among them, the memory is used to store computer programs, and after receiving the execution instruction, the processor can execute the computer program accordingly.

[0074] This application also provides a computer-readable storage medium for storing the computer program used in the above computer device. For example, the computer-readable storage medium can include, but is not limited to: various media such as USB flash drives, external hard drives, Read-Only Memory (ROM), Random Access Memory (RAM), magnetic disks, or optical discs that can store program codes.

[0075] In several embodiments provided by this application, it should be understood that the disclosed device and method can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and structure diagrams in the drawings show the possible architectures, functions, and operations of the device, method, and computer program product according to multiple embodiments of this application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in the alternative implementation, the functions marked in the blocks can occur in a different order from that marked in the drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the structure diagram and / or flowchart, as well as the combination of blocks in the structure diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0076] In addition, each functional module or unit in various embodiments of this application can be integrated together to form an independent part, or each module can exist alone, or two or more modules can be integrated to form an independent part.

[0077] When the above-mentioned functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a smart phone, a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application.

[0078] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered by the protection scope of this application.

Claims

1. A method for generating an electrical assembly process report, characterized in that, The method includes: Obtain an electrical assembly image; Perform image recognition on the electrical assembly image to obtain the contour information in the electrical assembly image; Perform character recognition on the electrical assembly image to obtain the technical requirement information and the part number information of each component in the electrical assembly image; According to the contour information and each of the part number information, obtain the adjacency matrix of the electrical assembly image; Input the contour information, the adjacency matrix, the technical requirement information, and each of the part number information into a graph neural network to generate an assembly logic file; Input the preset process rule set and the assembly logic file into a process report generation model to generate an electrical assembly process report.

2. The method for generating an electrical assembly process report according to claim 1, wherein, After obtaining the adjacency matrix of the electrical assembly image according to the contour information and each of the part number information, the method further includes: Obtain a bill of materials; Query the model information of each component corresponding to each of the part number information in the bill of materials respectively; Obtain the three-dimensional spatial structure information of each component in the preset component database according to each of the model information respectively; Input the three-dimensional spatial structure information, the contour information, the adjacency matrix, each of the part number information, and the technical requirement information into the graph neural network to generate the assembly logic file.

3. The method for generating an electrical assembly process report according to claim 1, wherein The performing image recognition on the electrical assembly image to obtain the contour information in the electrical assembly image includes: Use OpenCV to extract the contours in the electrical assembly image to obtain the contour information of each connecting line, as well as the contour information of each component and the center coordinates of each component; Use a vision large model to classify the contour information of each component to obtain each contour category, and obtain the connection relationship of each component according to the contour information of each connecting line. The contour information includes each center coordinate, each contour category, and the connection relationship.

4. The method for generating an electrical assembly process report according to claim 3, wherein The obtaining the adjacency matrix of the electrical assembly image according to the contour information and each of the part number information includes: According to each of the part number information, each of the center coordinates, and the connection relationship, obtain the adjacency matrix with each component as a node.

5. The method for generating an electrical assembly process report according to claim 4, wherein The edge weights of the adjacency matrix include: the distance and connection type between the nodes.

6. The method for generating an electrical assembly process report according to claim 1, wherein After inputting the preset process rule set and the assembly logic file into a process report generation model to generate an electrical assembly process report, the method further includes: Extract the information in the electrical assembly process report using regular expressions according to a preset standard template; Fill the extracted information into the preset standard template to generate a process specification card.

7. The method for generating an electrical assembly process report according to claim 1, characterized in that, Before inputting the preset process rule set and the assembly logic file into a process report generation model to generate an electrical assembly process report, the method further includes: Collect historical electrical assembly images, historical bills of materials, and historical electrical assembly process reports to form an electrical assembly data sample set; Preprocess the electrical assembly data sample set to obtain a standard electrical assembly data sample set; Input the standard electrical assembly data sample set into a general language model for optimized training to obtain the trained process report generation model.

8. An electrical assembly process report generation system, characterized in that, It includes: An image acquisition module for acquiring electrical assembly images; An image recognition module for performing image recognition on the electrical assembly images to obtain the contour information in the electrical assembly images; A character recognition module for performing character recognition on the electrical assembly images to obtain the technical requirement information and the part number information of each component in the electrical assembly images; A matrix acquisition module for obtaining the adjacency matrix of the electrical assembly images according to the contour information and each of the part number information; An assembly logic file generation module for inputting the contour information, the adjacency matrix, the technical requirement information and each of the part number information into a graph neural network to generate an assembly logic file; A report generation module for inputting a preset process rule set and the assembly logic file into the process report generation model to generate an electrical assembly process report.

9. A computer device, characterized in that, The computer device includes: a processor and a memory, the memory stores a computer program, and the processor is used to execute the computer program to implement the electrical assembly process report generation method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It stores a computer program, and when the computer program is executed on a processor, it implements the electrical assembly process report generation method according to any one of claims 1-7.

Images