Engineering drawing processing method and device, electronic equipment and storage medium

Abstract

The invention provides an engineering drawing processing method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence, the method comprises the following steps: analyzing user input information to obtain an analysis result, and generating a structured instruction based on the analysis result; determining a target processing task based on the structured instruction, and checking whether the structured instruction meets a preset executable condition or not; when the structured instruction meets a preset executable condition, executing the target processing task based on the structured instruction: generating an operation instruction according to the structured instruction, and executing a drawing task based on the operation instruction; or, screening out a target rule corresponding to the structured instruction from a preset rule base, and executing the verification task by using the target rule. Based on the natural language information and / or the engineering drawing and other user input information, automatic execution of a drawing task or a verification task of the engineering drawing can be rapidly driven, multiple processing links of the engineering drawing are covered, and the engineering drawing processing efficiency and reliability are comprehensively improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and more specifically, to an engineering drawing processing method, apparatus, electronic device, and storage medium. Background Technology

[0002] Engineering drawings, as key technical documents in engineering projects, serve as an important reference for communication between design, manufacturing, and assembly stages. Whether it's large-scale construction projects, complex machinery manufacturing, or precision electronic equipment production, the accuracy and reliability of engineering drawings directly determine the quality and performance of the final product or project.

[0003] Currently, the drafting process of engineering drawings mainly relies on manual labor, and after the drawings are completed, they also rely on manual verification and review. The processing methods for each stage of engineering drawings from drawing to verification are cumbersome and inefficient, and have obvious limitations. Summary of the Invention

[0004] The problem addressed by this invention is how to improve the efficiency of engineering drawing processing.

[0005] To address the above problems, the present invention provides an engineering drawing processing method, comprising: The received user input information is parsed to obtain the parsing result, and a structured instruction is generated based on the parsing result; wherein, the user input information includes natural language information and / or engineering drawings; The target processing task is determined based on the structured instructions, and the structured instructions are checked to see if they meet the preset executable conditions; wherein, the target processing task includes one of a drawing task and a verification task; the preset executable conditions include that the structured instructions include the target elements pre-associated with the target processing task; When the structured instruction meets the preset executable condition, the target processing task is executed based on the structured instruction, including: generating an operation instruction according to the structured instruction and executing the drawing task based on the operation instruction; or, selecting the target rule corresponding to the structured instruction from a preset rule base and executing the verification task using the target rule.

[0006] Optionally, the step of parsing the received user input information to obtain a parsing result, and generating structured instructions based on the parsing result, includes: When the user input information includes the natural language information, entity recognition is performed on the natural language information using a preset entity library to obtain entity information; wherein, the preset entity library includes intent entities, object entities, and parameter entities; When the user input information includes the engineering drawing, graphic element information is extracted from the engineering drawing; wherein, the graphic element information includes element information and annotation information; The structured instructions are obtained by filling a preset structured template with the entity information and / or the graphic element information.

[0007] Optionally, determining the target processing task based on the structured instructions includes: Determine whether the structured instruction includes the intent entity; If so, the target processing task is determined based on the processing task pre-associated with the intent entity; If not, an intent confirmation prompt message is generated, and the process returns to the step of parsing the received user input information to obtain the parsing result, until the target processing task is determined.

[0008] Optionally, the target elements pre-associated with the drawing task include the object entity and the parameter entity; the verification task includes a geometric verification task or a specification verification task; the target elements pre-associated with the geometric verification task include the object entity and the primitive information; the target elements pre-associated with the specification verification task include the object entity and the annotation information; after verifying whether the structured instruction meets the preset executable conditions, the method further includes: When the structured instruction does not meet the preset executable conditions, a prompt message for supplementary elements is generated based on the target elements not included in the structured instruction.

[0009] Optionally, the step of generating operation instructions based on the structured instructions and executing the drawing task based on the operation instructions includes: Obtain the object entity and the parameter entity from the structured instruction, extract the storage address corresponding to the 3D model that matches the object entity from the preset 3D model library, and obtain the default parameter configuration corresponding to the parameter entity in the 3D model; The storage address, the parameter entity, and the default parameter configuration are input into a preset instruction generation model, so that the preset instruction generation model generates operation semantic instructions according to a preset instruction template, and converts the operation semantic instructions into operation instructions corresponding to a preset drawing model; wherein, the preset instruction template includes calling the corresponding three-dimensional model based on the storage address, replacing the default parameter configuration with the parameter entity, updating the three-dimensional model, and creating a target engineering drawing corresponding to the updated three-dimensional model; The preset drawing model is driven according to the operation instructions, so that the preset drawing model generates the target engineering drawing.

[0010] Optionally, the preset rule base includes a geometric verification rule base; the primitive information includes color information and line type information corresponding to each primitive; the step of filtering out the target rule corresponding to the structured instruction from the preset rule base and using the target rule to perform the verification task includes: The graphic element type corresponding to each graphic element is determined according to the preset mapping relationship, the color information and the line type information, and the graphic element type is associated with the corresponding color information and line type information to obtain mapping information; wherein, the preset mapping relationship represents the mapping relationship between each preset graphic element and the preset color information and the preset line type information; The target rule is obtained by retrieving the geometric verification rules pre-associated with the object entity in the structured instruction from the geometric verification rule base; The target rule, the mapping information, and the engineering drawing are input into a preset geometric verification model, so that the preset geometric verification model can identify each of the graphic element types in the engineering drawing according to the mapping information, and verify whether each of the graphic element types satisfies the corresponding target rule.

[0011] Optionally, the preset rule base further includes an annotation rule base; the step of filtering out the target rule corresponding to the structured instruction from the preset rule base and using the target rule to perform the verification task further includes: The target rule is obtained by retrieving the annotation rules that are pre-associated with the object entity in the structured instruction from the annotation rule base; The target rule and the annotation information are input into a preset standard verification model so that the preset standard verification model can verify whether the annotation information meets the target rule.

[0012] In this invention, parsing received user input information, such as natural language information and / or engineering drawings, yields parsing results. This facilitates the rapid extraction of valid information from the user input, providing reliable guidance for subsequent automated processing of engineering drawings based on user input. Generating structured instructions based on the parsing results helps integrate fragmented information into an ordered and clear instruction set, providing a clear and reliable reference for determining and executing target processing tasks. Furthermore, this invention determines target processing tasks based on the generated structured instructions, clarifying whether the user expects a drawing or verification task, and defining the task type. This allows for targeted operations on different types of processing tasks, improving processing efficiency and ensuring the processing meets user expectations. Further, this invention verifies whether the structured instructions include all pre-associated target elements for the target processing task, facilitating the identification of the structured instructions' completeness. When a structured instruction meets preset executable conditions, it indicates that the structured instruction possesses all the key target elements for executing the corresponding target processing task, ensuring the smooth execution of the target processing task based on the structured instructions. In this invention, when the target processing task is a drawing task, operation instructions are generated based on structured instructions, and the drawing task is executed based on these instructions. This not only enables automated drawing of engineering drawings driven by user-generated natural language information but also helps ensure the accuracy of the drawings. When the target processing task is a verification task, target rules corresponding to the structured instructions are selected from a preset rule base, and the verification task is executed using these rules. This not only enables automated verification of engineering drawings but also helps ensure the rigor of the verification process. Thus, this invention, based on user input information such as natural language information and / or engineering drawings, can quickly drive the automated execution of drawing or verification tasks, covering multiple processing stages of engineering drawings and comprehensively improving the efficiency and reliability of engineering drawing processing.

[0013] The present invention also provides an engineering drawing processing apparatus, comprising: A parsing module is used to parse the received user input information to obtain the parsing result, and generate structured instructions based on the parsing result; wherein, the user input information includes natural language information and / or engineering drawings; The verification module is used to determine the target processing task based on the structured instructions and to verify whether the structured instructions meet the preset executable conditions; wherein, the target processing task includes one of a drawing task and a verification task; the preset executable conditions include that the structured instructions include each target element pre-associated with the target processing task; An execution module is configured to execute the target processing task based on the structured instruction when the structured instruction meets the preset executable conditions, including: generating an operation instruction based on the structured instruction and executing the drawing task based on the operation instruction; or, selecting the target rule corresponding to the structured instruction from a preset rule base and executing the verification task using the target rule.

[0014] The engineering drawing processing device and the engineering drawing processing method provided by this invention have essentially the same advantages as the prior art, and will not be repeated here.

[0015] The present invention also provides an electronic device, including a memory and a processor; The memory is used to store computer programs; The processor is used to implement the engineering drawing processing method as described above when executing the computer program.

[0016] The electronic device provided by this invention has essentially the same advantages as the engineering drawing processing method compared to the prior art, and will not be repeated here.

[0017] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the engineering drawing processing method described above.

[0018] The advantages of the computer-readable storage medium and the engineering drawing processing method provided by this invention compared to the prior art are basically the same, and will not be repeated here. Attached Figure Description

[0019] Figure 1 This is a schematic flowchart of the engineering drawing processing method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the engineering drawing processing device according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0020] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0021] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0022] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0023] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0024] It should be noted that the information (including but not limited to user device information, user personal information, voice information, etc.), data (including but not limited to data used for analysis, data stored, data displayed, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties. The collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0025] like Figure 1 As shown in the figure, an engineering drawing processing method provided by an embodiment of the present invention includes the following steps: S1: Parse the received user input information to obtain the parsing result, and generate structured instructions based on the parsing result; wherein, the user input information includes natural language information and / or engineering drawings.

[0026] Specifically, in this embodiment, the user input information includes natural language information and / or engineering drawings. Natural language information can be acquired by a voice acquisition device and converted into text information, or it can be text information input by the user through an interactive operation device. The engineering drawing can be a pre-drawn and annotated engineering drawing, which can be uploaded to the interactive platform of the engineering drawing processing method in this embodiment. The parsing result referred to in this embodiment represents the key content extracted from the user input information, providing a basis for subsequent generation of structured instructions. The structured instruction referred to in this embodiment represents an instruction formed after the parsing result is organized according to a specific format.

[0027] In one embodiment, after receiving user input information, it can be parsed to obtain the parsing results. For example, suppose the received natural language information is: "Draw a mechanical engineering drawing of a cylinder with a base radius of 3 cm and a height of 10 cm." Natural language processing methods such as lexical analysis and syntactic analysis can be used to identify keywords such as "cylinder," "radius of 3 cm," and "height of 10 cm" as parsing results. Similarly, when an engineering drawing is received, the information in the drawing's title bar can be identified to obtain parameters such as part names and materials as parsing results. Based on this, the parsing results can be integrated according to preset format requirements (such as sorting requirements for different types of information) to obtain corresponding structured instructions.

[0028] S2: Determine the target processing task based on structured instructions and verify whether the structured instructions meet the preset executable conditions; wherein, the target processing task includes one of the drawing task and the verification task; the preset executable conditions include the structured instructions including the target elements pre-associated with the target processing task.

[0029] Specifically, the target processing task referred to in this embodiment is a specific task to be performed on an engineering drawing, determined according to structured instructions, including either a drawing task or a verification task. A drawing task refers to the task of creating an engineering drawing according to the requirements of structured instructions, such as creating an engineering drawing with a specific shape and size. A verification task, on the other hand, refers to the task of checking an existing engineering drawing according to certain rules, such as checking whether the annotations on the engineering drawings conform to standard specifications.

[0030] In one embodiment, after generating structured instructions, the desired target processing task based on the user input information can be determined. For example, existing large language models can be used to perform semantic parsing on the structured instructions to determine whether they match a drawing task or a verification task. Based on this, the pre-associated target elements for the target processing task can be obtained, and it can be determined whether the structured instructions contain each target element. The target elements can be pre-defined according to the type of the target processing task. For example, for a drawing task, the target elements may include the name of the drawing object, key dimension parameters, etc. For a verification task, the target elements may include the acquisition method of the engineering drawing (such as storage address or file name), and textual information such as the technical requirements corresponding to the engineering drawing.

[0031] S3: When the structured instruction meets the preset executable conditions, the target processing task is executed based on the structured instruction, including: generating operation instructions based on the structured instruction and executing drawing tasks based on the operation instructions; or, selecting the target rules corresponding to the structured instruction from the preset rule base and using the target rules to execute verification tasks.

[0032] Specifically, in this embodiment, the operation instructions refer to specific instructions that can be recognized and executed by a preset drawing model (such as CAD or Solidworks), driving the preset drawing model to complete the drawing task. The preset rule base in this embodiment refers to a database that pre-stores various engineering drawing verification rules, covering verification rules for engineering drawings from different fields or of different types of parts. The target rule in this embodiment refers to a rule selected from the preset rule base that matches the structured instructions and provides a reference for subsequent verification tasks.

[0033] In one embodiment, when the target processing task is a drawing task, operation instructions can be generated based on structured instructions, and the drawing task can be executed based on these operation instructions. For example, suppose the structured instructions are: Object type: cylinder; Size requirements: radius: 3 cm; height: 10 cm. Then, the drawing steps can be decomposed based on the structured instructions, including: creating a circular sketch with a radius of 3 cm; extruding the cross-section of the sketch by 10 cm; creating the three views of the cylinder and outputting the engineering drawing. Based on this, the above drawing steps can be converted into corresponding API sequences, and the interface of a preset drawing model can be called to execute the drawing task. In this process, both the decomposition of drawing steps based on structured instructions and the conversion of drawing steps into corresponding API sequences can be implemented using a pre-trained large model. The components of various commonly used standard parts can be classified (such as cylinders, screws, cubes, etc.). The components of each standard part, the corresponding drawing steps for each element, and the corresponding API sequence can be associated to construct a training dataset. This training dataset can then be used to train and fine-tune an existing large model, resulting in a model capable of parsing structured instructions and generating operational instructions (referred to as the drawing model for ease of understanding and description). In actual use, inputting structured instructions into the drawing model will output the corresponding operational instructions.

[0034] In one embodiment, rule information (such as annotation rules, drawing specifications, etc.) corresponding to different types of parts can be obtained, and these rules can be associated with the corresponding parts to construct a dataset. In actual use, the part type can be determined based on structured instructions, and the corresponding rules can be filtered out from the dataset to obtain the target rules. On this basis, the structured instructions and the corresponding target rules can be input into an existing large model (such as a large-scale language model), and the semantic parsing and reasoning capabilities of the large model can be used to achieve the verification task.

[0035] In this embodiment, the received user input information, such as natural language information and / or engineering drawings, is parsed to obtain parsing results. This facilitates the rapid extraction of valid information contained in the user input, providing reliable guidance for subsequent automated processing of engineering drawings based on user input. Generating structured instructions based on the parsing results helps integrate the fragmented parsed information into an ordered and clear instruction set, providing a clear and reliable reference for subsequently determining and executing target processing tasks. Based on this, this embodiment determines the target processing task based on the generated structured instructions, clarifying whether the user expects a drawing task or a verification task, and defining the task type. This facilitates targeted operations for different types of processing tasks, improving processing efficiency and ensuring the processing meets user expectations. Furthermore, this embodiment verifies whether the structured instructions include the pre-associated target elements of the target processing task, which helps identify the completeness of the structured instructions. When a structured instruction meets preset executable conditions, it indicates that the structured instruction possesses all the key target elements for executing the corresponding target processing task, ensuring the smooth execution of the target processing task based on the structured instructions. In this embodiment, when the target processing task is a drawing task, operation instructions are generated based on structured instructions, and the drawing task is executed based on these instructions. This not only enables automated drawing of engineering drawings driven by user-generated natural language information but also helps ensure the accuracy of the drawings. When the target processing task is a verification task, target rules corresponding to the structured instructions are selected from a preset rule base, and the verification task is executed using these rules. This not only enables automated verification of engineering drawings but also helps ensure the rigor of the verification process. Thus, this embodiment can quickly drive the automated execution of drawing or verification tasks of engineering drawings based on user input information such as natural language information and / or engineering drawings, covering multiple processing stages of engineering drawings and comprehensively improving the efficiency and reliability of engineering drawing processing.

[0036] Optionally, the received user input information is parsed to obtain a parsing result, and a structured instruction is generated based on the parsing result, including: When user input includes natural language information, entity recognition is performed on the natural language information using a preset entity library to obtain entity information; the preset entity library includes intent entities, object entities, and parameter entities; When the user input includes an engineering drawing, the graphic element information is extracted from the engineering drawing; the graphic element information includes element information and annotation information. Fill a preset structured template with entity information and / or graphic element information to obtain structured instructions.

[0037] Specifically, the preset entity library referred to in this embodiment refers to a pre-built database used to identify different types of entities in natural language information, which may include intent entities, object entities, and parameter entities. In this embodiment, an intent entity represents the user's intention to perform a certain operation on the engineering drawing, expressed through natural language, such as "draw" or "verify." An object entity refers to the object represented in the engineering drawing in natural language, such as "a certain part." A parameter entity refers to specific parameters related to the object, such as "side length," "radius," or "angle." Graphic element information refers to important information related to the graphics extracted from the engineering drawing, which may include primitive information and annotation information. Primitive information can represent the basic graphic elements constituting the engineering drawing, such as the relevant data (e.g., line type, color) of points, lines, circles, etc. Annotation information can represent textual information describing the dimensions, tolerances, and technical requirements of the graphics in the engineering drawing, which can be extracted using methods such as OCR.

[0038] In one embodiment, when the user input includes natural language information, entity recognition is performed on the natural language information using a preset entity library to obtain entity information. For example, assuming the user inputs the natural language information as "draw an engineering drawing of a cube with a side length of 4 centimeters," entity recognition can be performed using the preset entity library. Through sentence analysis, the intent entity is identified as "draw," the object entity as "cube," and the parameter entity as "side length 4 centimeters," thus obtaining entity information. When the user input includes an engineering drawing, existing DWG parsing tools can be used to extract line types, colors, and other information corresponding to various graphic elements from the engineering drawing to obtain graphic element information. Based on this, entity information and / or graphic element information can be used to fill a preset structured template to obtain structured instructions. The preset structured template represents a template with a specific format framework, which can be pre-set and used to integrate entity information and graphic element information to generate structured instructions. For example, the preset structured template can contain a specific order of different information fields such as intent, object, parameter, graphic element, and annotation. Corresponding information from entity information and / or graphic element information can be used to fill the corresponding slots to obtain structured instructions.

[0039] In this embodiment, when user input includes natural language information, entity recognition is performed on the natural language information using a preset entity library. This accurately extracts entities corresponding to user intent, involved objects, and related parameters, thereby obtaining entity information and ensuring the comprehensiveness and accuracy of natural language understanding. When user input includes engineering drawings, graphic element information is extracted from the engineering drawings, which helps to comprehensively obtain the key component information of the engineering drawings and provides a detailed data foundation for subsequent processing. Based on this, entity information and / or graphic element information are used to fill a preset structured template to generate structured instructions. The parsing results are then systematically integrated to form a standardized and unified instruction format, effectively improving the accuracy and efficiency of information processing and providing a foundation for the reliable execution of the target processing task.

[0040] Optionally, the target processing task is determined based on structured instructions, including: Determine whether the structured instruction contains an intent entity; If so, the target processing task is determined based on the processing tasks pre-associated with the intent entity; If not, generate an intent confirmation message and return to the step of parsing the received user input to obtain the parsing result, until the target processing task is determined.

[0041] Specifically, in this embodiment, different intent entities can be pre-associated with corresponding processing tasks and stored accordingly. For example, intent entities such as drawing, painting, and generating can be pre-associated with drawing tasks, while intent entities such as checking, verifying, and validating can be pre-associated with verification tasks.

[0042] In this embodiment, by determining whether the structured instruction includes an intent entity, it is possible to quickly identify whether the user input information has a clear actual processing intent. If so, the target processing task is determined based on the processing task pre-associated with the intent entity, which is beneficial for quickly and accurately associating the user intent with the target processing task to be automatically executed. If not, an intent confirmation prompt message can be generated (such as "Do you want to draw an engineering drawing or do you want me to help you verify the engineering drawing?"), and the steps of parsing the received user input information to obtain the parsing result can be returned. Thus, the specific target processing task is determined based on the user's feedback, which maximizes the satisfaction of the engineering drawing processing needs under diverse and non-standard user input conditions, and improves the flexibility and reliability of engineering drawing processing.

[0043] Optionally, the target elements pre-associated with the drawing task include object entities and parameter entities; the verification task includes a geometric verification task or a specification verification task; the target elements pre-associated with the geometric verification task include object entities and primitive information; the target elements pre-associated with the specification verification task include object entities and annotation information; after verifying whether the structured instructions meet the preset executable conditions, it also includes: When a structured instruction does not meet the preset executable conditions, a prompt message for supplementary elements is generated based on the target elements not included in the structured instruction.

[0044] Specifically, in this embodiment, the target elements pre-associated with the drawing task include object entities and parameter entities, which helps to clarify the main structure and key dimensions of the engineering drawing. In this embodiment, the verification task includes a geometric verification task or a specification verification task. The target elements pre-associated with the geometric verification task include object entities and primitive information, while the target elements pre-associated with the specification verification task include object entities and annotation information. This helps to extract the geometric or annotation information that needs to be verified in the engineering drawing and provides it with specification references.

[0045] In this embodiment, when a structured instruction does not meet the preset executable conditions, target elements not included in the structured instruction can be obtained, and supplementary element prompts can be generated based on them. For example, assuming the structured instruction corresponding to the drawing task includes object entities but not parameter entities, the supplementary element prompt can be: Please provide the parameter entities (such as radius, length, etc.) corresponding to the object entity (such as a part). This helps users quickly understand the key information missing in the instruction, supplement it in a timely manner, and improve interaction efficiency. It avoids repeated communication and error handling caused by incomplete instructions, ensuring that engineering drawing processing tasks can proceed smoothly.

[0046] Optionally, operation instructions are generated based on structured instructions, and drawing tasks are performed based on the operation instructions, including: Obtain the object entity and parameter entity from the structured instructions, extract the storage address of the 3D model that matches the object entity from the preset 3D model library, and obtain the default parameter configuration corresponding to the parameter entity in the 3D model. The storage address, parameter entity, and default parameter configuration are input into the preset instruction generation model, so that the preset instruction generation model generates operation semantic instructions according to the preset instruction template, and converts the operation semantic instructions into operation instructions corresponding to the preset drawing model; wherein, the preset instruction template includes calling the corresponding 3D model based on the storage address, replacing the default parameter configuration with the parameter entity, updating the 3D model, and creating the target engineering drawing corresponding to the updated 3D model; The preset drawing model is driven according to the operation instructions, so that the preset drawing model can generate the target engineering drawing.

[0047] Specifically, the preset 3D model library referred to in this embodiment is a database that pre-stores information on various 3D models. Each 3D model is modeled parametrically and has default parameter configurations. 3D models of various commonly used standard parts in engineering can be pre-built, and the names of the 3D models, the default parameter configurations for each parameter, and other information such as their storage addresses can be associated and stored to construct the 3D model library.

[0048] In one embodiment, the object entity and parameter entity in the structured instruction can be obtained, and the storage address corresponding to the 3D model matching the object entity can be extracted from a preset 3D model library, along with the default parameter configuration of the 3D model. For example, suppose the structured instruction is "Draw a cylindrical mechanical part with a base radius of 3 cm and a height of 8 cm", where the object entity is "cylinder" and the parameter entity is "base radius 3 cm, height 8 cm". Then, the preset 3D model library can be searched for to find the 3D model matching "cylinder", and the storage address of the 3D model in the preset 3D model library, as well as the default parameter configuration corresponding to the parameter entity, can be extracted.

[0049] In one embodiment, the preset instruction generation model represents a pre-trained large model used to generate corresponding instructions based on input information, while the preset instruction template specifies the logic and steps of the preset instruction generation model in generating instructions. A corresponding training dataset can be constructed to train and fine-tune the existing large language model (the specific method has been clearly defined previously and will not be repeated here). Simultaneously, the preset instruction template serves as a prompt template, constraining the preset instruction generation model to generate operational semantic instructions based on storage addresses, parameter entities, and default parameter configurations. These operational semantic instructions are then converted into executable API instructions for a preset drawing model (such as Solidworks), resulting in operational instructions. Constraining the model to generate operational instructions step-by-step through the preset instruction template helps avoid the model directly performing inference, which could affect the reliability of the operational instructions. In this embodiment, the preset instruction template includes: first, calling the corresponding 3D model based on the storage address to provide a basic 3D model reference for the drawing task. Based on this, the default parameter configuration is replaced with parameter entities to update the 3D model, which helps convert the default configuration parameters of the 3D model into user-specified parameters, thereby quickly meeting the user's actual needs. Furthermore, after updating the 3D model, creating the target engineering drawing (such as three views) corresponding to the updated 3D model yields the engineering drawing desired by the user. Preset instruction templates cover the entire process of calling the 3D model, modifying 3D model parameters, and creating 2D engineering drawings based on the 3D model, ensuring that the generated operation instructions can be accurately executed by the preset drawing model, improving the efficiency and accuracy of drawing task processing.

[0050] Optionally, the preset rule base includes a geometric verification rule base; the primitive information includes the color information and line type information corresponding to each primitive; the target rules corresponding to the structured instructions are filtered from the preset rule base, and the verification task is performed using the target rules, including: The graphic element type corresponding to each graphic element is determined based on the preset mapping relationship, color information, and line type information. The graphic element type is then associated with the corresponding color information and line type information to obtain the mapping information. The preset mapping relationship represents the mapping relationship between each preset graphic element and the preset color information and preset line type information. The target rule is obtained by retrieving the geometric validation rules that are pre-associated with the object entities in the structured instructions from the geometric validation rule base; The target rules, mapping information, and engineering drawings are input into the preset geometric verification model, so that the preset geometric verification model can identify the types of each element in the engineering drawing based on the mapping information and check whether each element type meets the corresponding target rules.

[0051] Specifically, the preset rule base referred to in this embodiment includes a geometric verification rule base, which can include requirements for various geometric elements of engineering drawings (such as the requirement that circles have center lines and closed outlines). The preset mapping relationship referred to in this embodiment represents a series of pre-defined correspondences, which clarify the relationship between different preset graphic elements (such as lines, circles, rectangles, etc.) and specific preset color information (such as black, red, blue, etc.) and preset line type information (such as solid lines, dashed lines, dotted lines, etc.). It can be established in advance according to the general rules of engineering drawing (such as industry standards, enterprise standards, etc.).

[0052] In one embodiment, the element type corresponding to each element can be determined based on a preset mapping relationship, color information, and line type information. The element type is then associated with its corresponding color and line type information to obtain mapping information. For example, if an element has a thick solid line type and a black color, its element type can be determined as an outline line based on the preset mapping relationship. Another element has a dashed line type and a green color, and its element type can be determined as a center line based on the preset mapping relationship. Based on this, the determined element type can be associated with its corresponding line type and color information to obtain mapping information.

[0053] In one embodiment, the preset rule base includes a geometric verification rule base. Geometric verification rules pre-associated with object entities in structured instructions can be obtained from this base to generate target rules. The target rules, mapping information, and engineering drawings are then input into a preset geometric verification model. This model identifies the types of graphic elements in the engineering drawing based on the mapping information and verifies whether each type satisfies the corresponding target rule. In this embodiment, the preset geometric verification model can be an existing multimodal model, which possesses both natural language understanding and image recognition capabilities. Engineering drawings of different standard parts can be created based on the line types and colors of each graphic element type specified by the preset mapping relationship. The engineering drawings are then manually verified based on the pre-associated geometric verification rules for each type of standard part, yielding verification results. Furthermore, the preset mapping relationship, engineering drawings, geometric verification rules, and corresponding verification results can be associated to form a training dataset. This training dataset is then used to train and optimize the existing multimodal model. In practical use, the target rules, mapping information, and engineering drawings are input into the preset geometric verification model. The preset geometric verification model identifies different primitive types based on visual recognition and mapping information. On this basis, it can determine whether different source types meet the corresponding target rules (such as contour line closure requirements, circles need to have center lines, etc.), thereby realizing automated geometric verification of engineering drawings.

[0054] Optionally, the preset rule base also includes an annotation rule base; selecting target rules corresponding to structured instructions from the preset rule base and performing verification tasks using the target rules also includes: The target rule is obtained by retrieving the annotation rules that are pre-associated with the object entities in the structured instructions from the annotation rule base; The target rules and annotation information are input into the preset specification verification model so that the preset specification verification model can check whether the annotation information meets the target rules.

[0055] Specifically, in this embodiment, the preset rule base also includes an annotation rule base, which can be a database specifically storing rules related to annotations of various engineering drawings. These rules cover the specification requirements for annotations of different types of engineering drawings (such as tolerance annotation, surface roughness annotation, technical requirement annotation, etc.). The corresponding annotation rules can be determined according to the type of the object entity to obtain the target rule.

[0056] In one embodiment, the preset specification verification model can be a large-scale language model. This model can associate annotation information, corresponding annotation rules, and manual verification results from preset engineering drawings of different object types to construct a training dataset. The training dataset is then used to train and optimize an existing large-scale language model, resulting in the preset specification verification model. In practical use, target rules and annotation information can be input into the preset specification verification model, enabling it to perform semantic understanding and reasoning on the annotation information and target rules, verifying whether the annotation information meets the target rules, thereby achieving automated specification verification of engineering drawings.

[0057] like Figure 2 As shown, an engineering drawing processing device 200 provided in an embodiment of the present invention includes: The parsing module 210 is used to parse the received user input information to obtain a parsing result, and generate structured instructions based on the parsing result; wherein, the user input information includes natural language information and / or engineering diagrams; The verification module 220 is used to determine the target processing task based on the structured instructions and to verify whether the structured instructions meet the preset executable conditions; wherein, the target processing task includes one of a drawing task and a verification task; the preset executable conditions include that the structured instructions include each target element pre-associated with the target processing task; The execution module 230 is used to execute the target processing task based on the structured instruction when the structured instruction meets the preset executable condition, including: generating an operation instruction according to the structured instruction and executing the drawing task based on the operation instruction; or, filtering out the target rule corresponding to the structured instruction from a preset rule base and using the target rule to execute the verification task.

[0058] The engineering drawing processing device and engineering drawing processing method provided in this embodiment can produce basically the same technical effects, and will not be described in detail here.

[0059] like Figure 3 As shown, an electronic device 300 provided in this embodiment of the invention includes a memory 310 and a processor 320; the memory 310 is used to store a computer program; the processor 320 is used to implement the engineering drawing processing method as described above when the computer program is executed.

[0060] Alternatively, an electronic device 300 includes a memory 310 and a processor 320 coupled to the memory 310; the memory 310 is configured to store a computer program; and the processor 320 is configured to perform the following operations when the computer program is executed: The received user input information is parsed to obtain the parsing result, and a structured instruction is generated based on the parsing result; wherein, the user input information includes natural language information and / or engineering drawings; The target processing task is determined based on the structured instructions, and the structured instructions are checked to see if they meet the preset executable conditions; wherein, the target processing task includes one of a drawing task and a verification task; the preset executable conditions include that the structured instructions include the target elements pre-associated with the target processing task; When the structured instruction meets the preset executable condition, the target processing task is executed based on the structured instruction, including: generating an operation instruction according to the structured instruction and executing the drawing task based on the operation instruction; or, selecting the target rule corresponding to the structured instruction from a preset rule base and executing the verification task using the target rule.

[0061] The electronic device and the engineering drawing processing method provided in this embodiment can produce basically the same technical effects, and will not be described again here.

[0062] This invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the engineering drawing processing method described above.

[0063] Alternatively, a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the following operations: The received user input information is parsed to obtain the parsing result, and a structured instruction is generated based on the parsing result; wherein, the user input information includes natural language information and / or engineering drawings; The target processing task is determined based on the structured instructions, and the structured instructions are checked to see if they meet the preset executable conditions; wherein, the target processing task includes one of a drawing task and a verification task; the preset executable conditions include that the structured instructions include the target elements pre-associated with the target processing task; When the structured instruction meets the preset executable condition, the target processing task is executed based on the structured instruction, including: generating an operation instruction according to the structured instruction and executing the drawing task based on the operation instruction; or, selecting the target rule corresponding to the structured instruction from a preset rule base and executing the verification task using the target rule.

[0064] The computer-readable storage medium and the engineering drawing processing method provided in this embodiment can produce basically the same technical effects, and will not be described again here.

[0065] The present invention will now be described an electronic device 300 that can serve as a server or client of the present invention, which is an example of a hardware device that can be applied to various aspects of the present invention. Electronic device 300 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 300 can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0066] Electronic device 300 includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) or a computer program loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0067] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. In this application, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention according to actual needs. Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units can be implemented in hardware or as software functional units.

[0068] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A method for processing engineering drawings, characterized in that, include: The received user input information is parsed to obtain the parsing result, and a structured instruction is generated based on the parsing result; wherein, the user input information includes natural language information and / or engineering drawings; The target processing task is determined based on the structured instructions, and the structured instructions are checked to see if they meet the preset executable conditions; wherein, the target processing task includes one of a drawing task and a verification task; the preset executable conditions include that the structured instructions include the target elements pre-associated with the target processing task; When the structured instruction meets the preset executable condition, the target processing task is executed based on the structured instruction, including: generating an operation instruction according to the structured instruction and executing the drawing task based on the operation instruction; or, selecting the target rule corresponding to the structured instruction from a preset rule base and executing the verification task using the target rule.

2. The engineering drawing processing method according to claim 1, characterized in that, The step of parsing the received user input information to obtain a parsing result, and generating structured instructions based on the parsing result, includes: When the user input information includes the natural language information, entity recognition is performed on the natural language information using a preset entity library to obtain entity information; wherein, the preset entity library includes intent entities, object entities, and parameter entities; When the user input information includes the engineering drawing, graphic element information is extracted from the engineering drawing; wherein, the graphic element information includes element information and annotation information; The structured instructions are obtained by filling a preset structured template with the entity information and / or the graphic element information.

3. The engineering drawing processing method according to claim 2, characterized in that, The step of determining the target processing task based on the structured instructions includes: Determine whether the structured instruction includes the intent entity; If so, the target processing task is determined based on the processing task pre-associated with the intent entity; If not, an intent confirmation prompt message is generated, and the process returns to the step of parsing the received user input information to obtain the parsing result, until the target processing task is determined.

4. The engineering drawing processing method according to claim 2, characterized in that, The target elements pre-associated with the drawing task include the object entity and the parameter entity; the verification task includes a geometric verification task or a specification verification task; the target elements pre-associated with the geometric verification task include the object entity and the primitive information; the target elements pre-associated with the specification verification task include the object entity and the annotation information; After verifying whether the structured instructions meet the preset executable conditions, the method further includes: When the structured instruction does not meet the preset executable conditions, a prompt message for supplementary elements is generated based on the target elements not included in the structured instruction.

5. The engineering drawing processing method according to claim 4, characterized in that, The step of generating operation instructions based on the structured instructions and executing the drawing task based on the operation instructions includes: Obtain the object entity and the parameter entity from the structured instruction, extract the storage address corresponding to the 3D model that matches the object entity from the preset 3D model library, and obtain the default parameter configuration corresponding to the parameter entity in the 3D model; The storage address, the parameter entity, and the default parameter configuration are input into a preset instruction generation model, so that the preset instruction generation model generates operation semantic instructions according to a preset instruction template, and converts the operation semantic instructions into operation instructions corresponding to a preset drawing model; wherein, the preset instruction template includes calling the corresponding three-dimensional model based on the storage address, replacing the default parameter configuration with the parameter entity, updating the three-dimensional model, and creating a target engineering drawing corresponding to the updated three-dimensional model; The preset drawing model is driven according to the operation instructions, so that the preset drawing model generates the target engineering drawing.

6. The engineering drawing processing method according to claim 4, characterized in that, The preset rule base includes a geometric verification rule base; the primitive information includes color information and line type information corresponding to each primitive; the step of filtering out the target rule corresponding to the structured instruction from the preset rule base and using the target rule to perform the verification task includes: The graphic element type corresponding to each graphic element is determined according to the preset mapping relationship, the color information and the line type information, and the graphic element type is associated with the corresponding color information and line type information to obtain mapping information; wherein, the preset mapping relationship represents the mapping relationship between each preset graphic element and the preset color information and the preset line type information; The target rule is obtained by retrieving the geometric verification rules pre-associated with the object entity in the structured instruction from the geometric verification rule base; The target rule, the mapping information, and the engineering drawing are input into a preset geometric verification model, so that the preset geometric verification model can identify each of the graphic element types in the engineering drawing according to the mapping information, and verify whether each of the graphic element types satisfies the corresponding target rule.

7. The engineering drawing processing method according to claim 6, characterized in that, The preset rule base also includes an annotation rule base; the step of filtering out the target rule corresponding to the structured instruction from the preset rule base and using the target rule to perform the verification task further includes: The target rule is obtained by retrieving the annotation rules that are pre-associated with the object entity in the structured instruction from the annotation rule base; The target rule and the annotation information are input into a preset standard verification model so that the preset standard verification model can verify whether the annotation information meets the target rule.

8. An engineering drawing processing device, characterized in that, include: A parsing module is used to parse the received user input information to obtain the parsing result, and generate structured instructions based on the parsing result; wherein, the user input information includes natural language information and / or engineering drawings; The verification module is used to determine the target processing task based on the structured instructions and to verify whether the structured instructions meet the preset executable conditions; wherein, the target processing task includes one of a drawing task and a verification task; the preset executable conditions include that the structured instructions include each target element pre-associated with the target processing task; An execution module is configured to execute the target processing task based on the structured instruction when the structured instruction meets the preset executable conditions, including: generating an operation instruction based on the structured instruction and executing the drawing task based on the operation instruction; or, selecting the target rule corresponding to the structured instruction from a preset rule base and executing the verification task using the target rule.

9. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to implement the drawing processing method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the engineering drawing processing method as described in any one of claims 1 to 7.

Images