Feature automatic identification method and device based on extended attribute adjacency graph and medium

Abstract

The application discloses a feature automatic identification method and device based on an extended attribute adjacency graph and a medium, relates to the technical field of three-dimensional model graph feature identification, and comprises the following steps: acquiring XML information dataset of an extended attribute adjacency graph of a three-dimensional model and XML information dataset of a self-defined feature structure to be identified; obtaining a full graph structure corresponding to the three-dimensional model and a standard graph structure corresponding to the feature structure to be identified based on the two kinds of XML information dataset; and performing graph structure similarity matching according to the full graph structure and the standard graph structure, and identifying the standard graph structure in the full graph structure. The application can improve the efficiency and accuracy of the structure feature to be identified in the three-dimensional model.

Description

Technical Field

[0001] This application relates to the field of 3D model graph feature recognition technology, and in particular to a method, apparatus and medium for automatic feature recognition based on extended attribute adjacency graph. Background Technology

[0002] The emergence of Model-Based Definition (MBD) technology has significantly improved the collaboration between product design, manufacturing, and inspection, overcoming the previous disparate approaches used in different stages. The MBD model unifies design, manufacturing, and inspection information within a single model. From the initial design stage to the manufacturing and inspection stages, all processes are based on the product's MBD model. This greatly enhances the collaboration between design, manufacturing, and inspection, resolving the difficulty of tracing the root cause when discrepancies arise between these stages. It also reduces and avoids the need for iterative updates to engineering drawings and process specifications caused by differences in design, manufacturing, and inspection. However, currently, the method for obtaining MBD models during the transition from the design stage to the manufacturing and inspection stages still relies on manual methods, which are inefficient. The identification of key features also relies on manual judgment, resulting in poor accuracy. Summary of the Invention

[0003] The purpose of this application is to provide a method, apparatus, and medium for automatic feature recognition based on extended attribute adjacency graphs, which can improve the efficiency and accuracy of identifying structural features in 3D models.

[0004] To achieve the above objectives, this application provides the following solution:

[0005] Firstly, this application provides a method for automatic feature recognition based on extended attribute adjacency graphs, including:

[0006] Obtain the XML information dataset of the extended attribute adjacency graph of the 3D model;

[0007] Obtain a custom XML information dataset containing the features to be identified;

[0008] Based on the XML information dataset of the extended attribute adjacency graph and the XML information dataset of the feature structure to be identified, the full graph structure corresponding to the 3D model and the standard graph structure corresponding to the feature structure to be identified are derived; both the full graph structure and the standard graph structure include several nodes and edges between nodes; each node represents a face; each node contains face information attributes; each edge represents the relationship between faces; each edge contains face-to-face relationship attributes;

[0009] Graph structure similarity matching is performed based on the full graph structure and the standard graph structure to identify the standard graph structure in the full graph structure.

[0010] Secondly, this application provides an automatic feature recognition device based on an extended attribute adjacency graph, comprising:

[0011] The 3D model attribute information acquisition module is used to acquire the XML information dataset of the extended attribute adjacency graph of the 3D model.

[0012] The feature structure customization module is used to obtain a dataset of XML information about custom feature structures to be identified.

[0013] The graph structure generation module is used to derive the full graph structure corresponding to the 3D model and the standard graph structure corresponding to the feature structure to be identified based on the XML information dataset of the extended attribute adjacency graph and the XML information dataset of the feature structure to be identified. Both the full graph structure and the standard graph structure include several nodes and edges between nodes. Each node represents a face. Each node contains face information attributes. Each edge represents the relationship between faces. Each edge contains face-to-face relationship attributes.

[0014] The graph structure matching module is used to perform graph structure similarity matching based on the full graph structure and the standard graph structure, and to identify the standard graph structure in the full graph structure.

[0015] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the feature automatic identification method based on extended attribute adjacency graph.

[0016] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the described automatic feature recognition method based on extended attribute adjacency graphs.

[0017] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0018] This application provides a method, apparatus, and medium for automatic feature recognition based on extended attribute adjacency graphs. It allows for the customization of features to be recognized, thereby enabling the identification of relevant graph structure features in a 3D model according to requirements. Furthermore, it applies the concept of similarity matching to perform feature matching between graph structures based on the full graph structure of the 3D model and the standard graph structure corresponding to the features to be recognized, quickly obtaining the features to be recognized from the 3D model. This ensures both the effectiveness and accuracy of feature recognition. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is an application environment diagram of an automatic feature recognition method based on extended attribute adjacency graph in one embodiment of this application;

[0021] Figure 2 A flowchart illustrating an automatic feature recognition method based on an extended attribute adjacency graph, provided in an embodiment of this application;

[0022] Figure 3 This is a schematic diagram illustrating the process of information parsing based on a three-dimensional model, provided in an embodiment of this application.

[0023] Figure 4 This is a schematic diagram of the feature model and graph structure of a square blind hole provided in an embodiment of this application;

[0024] Figure 5 A cylindrical double-step countersunk hole model and schematic diagram provided in an embodiment of this application;

[0025] Figure 6 This application provides a model and schematic diagram of a tapered countersunk hole structure according to an embodiment of the present application.

[0026] Figure 7 A prototype XML diagram of a custom geometric feature structure provided in one embodiment of this application;

[0027] Figure 8 This is a schematic diagram of a feature recognition process provided in an embodiment of this application;

[0028] Figure 9 This is a schematic flowchart of a subgraph isomorphic matching method provided in an embodiment of this application;

[0029] Figure 10 This is a schematic flowchart of a similarity graph matching method provided in another embodiment of this application;

[0030] Figure 11 A functional module schematic diagram of an automatic feature recognition device based on an extended attribute adjacency graph provided in another embodiment of this application;

[0031] Figure 12 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0032] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0033] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] The automatic feature recognition method based on extended attribute adjacency graphs provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send the XML information dataset of the extended attribute adjacency graph of the 3D model and the XML information dataset of the custom feature structure to be identified to server 104. After receiving the XML information dataset of the extended attribute adjacency graph of the 3D model and the XML information dataset of the custom feature structure to be identified, server 104, based on the XML information dataset of the extended attribute adjacency graph and the XML information dataset of the feature structure to be identified, derives the full graph structure corresponding to the 3D model and the standard graph structure corresponding to the feature structure to be identified; it then performs graph structure similarity matching based on the full graph structure and the standard graph structure to identify the standard graph structure in the full graph structure. Server 104 can then feed back the obtained standard graph structure in the full graph structure to terminal 102. Furthermore, in some embodiments, the automatic feature recognition method based on extended attribute adjacency graphs can also be implemented separately by server 104 or terminal 102. For example, terminal 102 can directly perform graph structure construction and graph structure similarity matching on the XML information dataset of the extended attribute adjacency graph of the 3D model and the XML information dataset of the custom feature structure to be identified. Alternatively, server 104 can obtain the XML information dataset of the extended attribute adjacency graph of the 3D model and the XML information dataset of the custom feature structure to be identified from the data storage system, and perform graph structure construction and graph structure similarity matching on the XML information dataset of the extended attribute adjacency graph of the 3D model and the XML information dataset of the custom feature structure to be identified.

[0035] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0036] In one exemplary embodiment, such as Figure 2 As shown, an automatic feature recognition method based on extended attribute adjacency graphs is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 204.

[0037] Step 201: Obtain the XML information dataset of the extended attribute adjacency graph of the 3D model.

[0038] Step 202: Obtain the XML information dataset of the custom feature structure to be identified.

[0039] Step 203: Based on the XML information dataset of the extended attribute adjacency graph and the XML information dataset of the feature structure to be identified, derive the full graph structure corresponding to the 3D model and the standard graph structure corresponding to the feature structure to be identified; both the full graph structure and the standard graph structure include several nodes and edges between nodes; each node represents a face; each node contains face information attributes; each edge represents the relationship between faces; each edge contains face-to-face relationship attributes.

[0040] Step 204: Perform graph structure similarity matching based on the full graph structure and the standard graph structure to identify the standard graph structure in the full graph structure.

[0041] To implement steps 201 to 204 above, firstly, information is parsed based on the 3D model of the part, and its geometry and topology are decomposed and processed, outputting the extended attribute adjacency of the relationships between faces. Figure X ML information dataset. Then, based on a custom standard feature structure, a prototype of the 3D geometric feature structure to be identified is set. Using graph structure similarity matching, the key feature structure of the part's 3D model is quickly identified, ensuring both the effectiveness and accuracy of feature recognition. Furthermore, the feature recognition results can be fed back to a 3D CAD system (such as CATIA modeling software), realizing feature recognition and feature localization of the 3D model.

[0042] In another exemplary embodiment of this application, surface geometric and topological information of the part's 3D model is obtained through an interface in the CATIA secondary development toolset, including topological points, topological lines, and topological surface element information. A set of surface information and a set of surface-to-surface relationships are obtained using relevant interfaces, and all information is combined and stored as a FACE set and a Relationship set. An extended attribute adjacency based on the 3D model is created using the CATIDOMImplementation interface. Figure X ML information dataset. By obtaining the FACE and Relationship sets and outputting document information according to the XML markup language specification, the extended attribute adjacency is derived. Figure X ML information dataset. Specifically, step 201 involves obtaining the XML information dataset of the extended attribute adjacency graph of the 3D model, including:

[0043] (a1) Obtain the geometric and topological information of the surface of the three-dimensional model.

[0044] The CATIA secondary development toolkit provides numerous interfaces for acquiring the surface geometry and topology information of 3D parts. These include CATFrmEditor (document editor interface), CATIMOcicalTool (mechanical solid feature interface), CATIGeometricalElement (geometric feature management interface), CATBody (part topology interface), CATCell (the most basic topological unit interface for parts, including topological points CATVertex, topological lines CATEdge, and topological faces CATFace), CATFace (CATIA topological face information interface), and CATSurface (geometric information interface in topological elements). These interfaces are used to obtain a set of face information, which is then used to obtain a set of face-to-face relationships. Finally, by integrating all the information, this invention uses the FACE set and the Relationship set to represent and store the extended attribute adjacency graph information parsed from the 3D part model. The extended attribute adjacency graph information includes face information and face-to-face relationship information.

[0045] The process of information parsing based on 3D models is as follows: Figure 3As shown. First, the CATFrmEditor interface is used to access the currently opened part window to obtain its document information, and the CATIMIchanicalTool is used to obtain the mechanical part features. Second, the CATIGIgeometricalElement and CATBody interfaces are used to access the topology set in the part's 3D model. This topology set contains information on all one-dimensional topological points (CATVertex), two-dimensional topological lines (CATEdge), and three-dimensional topological surfaces (CATFace) in the part's 3D model. Then, based on the topology set, GetAllCells is called to obtain all topological surface elements CATLISTP(CATCell) / CATFace in the part's 3D model. The obtained CATFace is then used with the CATSurface interface to obtain the geometric element information in the topological surface and the relevant surface attribute information. Finally, the surface topology and geometric information of the part's 3D model are extracted.

[0046] (a2) Obtain the set of surface information attributes and the set of surface-to-surface relationship attributes of the three-dimensional model based on the geometric information and the topological information.

[0047] Construction of the face node information attribute set in the extended attribute adjacency graph: Geometric and topological face information is obtained based on the CATFace and CATSurface interfaces. The constructed FACE set represents the face node information of the extended attribute adjacency graph, containing the following specific attribute information: unique ID of the face (UniqueID, CATUnicodeString), face number (FaceNo, int), face label (FaceTag, CATLONG32), face area (FaceArea, double), face type (FaceType, CATUnicodeString), face name (FaceTypeName, CATUnicodeString), face inner loop (InnerLoopSize, int), face outer loop (OutterLoopSize, int), face center point (EstimateCenterPoint, CATMathPoint), face normal (FaceVector, CATMathVector), whether it is a repeating Tag face (IsRepeatTagFace, int), repeating Tag face label (RepeatTagNo, int), number of repeating Tag faces (RepeatTagSize, int), and repeating face FACE container (vector). <repeattagface>Finally, the acquired surface information is stored to prepare for subsequent surface-to-surface relationships, information output, and feature recognition.

[0048] Construction of the face-to-face relation attribute set in the extended attribute adjacency graph:

[0049] Based on the aforementioned set of face information attributes, and according to the face geometry and topological relationships in the 3D part model, a RelationShip set is constructed to represent the face relationship information of the extended attribute adjacency graph. The specific attribute information included is: the unique ID of the source face (UniqueIDSourceFace, CATUnicodeString), the unique ID of the target face (UniqueIDTargetFace, CATUnicodeString), the tag identifier of the source face (TagSourceFace, CATLONG32), the tag identifier of the target face (TagTargetFace, CATLONG32), the type of the source face (TypeSourceFace, CATUnicodeString), and the target face's... Type (TypeTargetFace, CATUnicodeString), whether faces intersect (IsIntersection, int), whether faces are parallel (IsIntersection, int), face concavity / convexity (IsConvexity, int), whether faces are perpendicular (IsVertical, int), number of face intersections (SizeEdgeIntersection, int), face intersection type (RelationShipType, CATUnicodeString), face intersection length (EdgeLength, double), face angle (AngleDegree, double), face intersection container (vectorEdge, vector...)<CATEdge*> Finally, the acquired face-to-face relationship information is stored for information output and feature recognition. Among the two faces with an edge relationship, one face is called the source face, and the other face is called the target face.

[0050] (a3) Based on the set of face information attributes and the set of face-to-face relationship attributes of the three-dimensional model, the XML information dataset of the extended attribute adjacency graph is obtained; the nodes in the extended attribute adjacency graph contain the face information attributes of the three-dimensional model, and the edges between the nodes in the extended attribute adjacency graph contain the face-to-face relationship attributes of the three-dimensional model.

[0051] This invention uses CATIA secondary development to parse information from the 3D model of a part, and utilizes CATIDOMImplementation (XML Document Object Model Interface) to create an XML information dataset based on an extended attribute adjacency graph of the 3D model. Through information parsing in steps (a1) and (a2), a set of face information attributes and a set of face-to-face relationship attributes are obtained. Document information is output according to the XML tag language specification, representing and storing the face information attributes and face-to-face relationship attributes. This process allows the dataset to be separated from the 3D CAD system, making it more convenient for subsequent feature recognition research and applications.

[0052] The output specification for the XML information dataset is as follows: A root node named RootFaceFeatureAnalysisInformation is constructed. Under the root node are two child nodes: AllFace and AllFaceRelationShip. The AllFace node stores the data of the face objects in the entire 3D model, while the AllFaceRelationShip node stores the data of the RelationShip objects representing the face relationships in the entire 3D model. Each AllFace node contains multiple FACE nodes, each representing a face object and containing extended attribute information such as FaceNo, FaceTag, FaceArea, FaceType, UniqueID, RepeatTagNo, InnerLoopSize, RepeatTagSize, OutterLoopSize, and EstimateCenterPoint. Each AllFaceRelationShip node contains multiple RelationShipSourceFace nodes, representing the source face information in the face relationships, including SourceFaceNo, SourceFaceTag, and SourceFaceType attributes. Each RelationShipSourceFace node contains multiple RelationShipTargetFace nodes, representing the target face information in the face-to-face relationship. Each RelationShipTargetFace node contains extended attribute face-to-face relationship information data between faces, such as UniqueIDSourceFace, UniqueIDTargetFace, TypeSourceFace, TypeTargetFace, IsIntersection, IsParallel, IsConvexity, IsVertical, SizeEdgeIntersection, RelationShipType, EdgeLength, AngleDegree, etc.

[0053] In this embodiment, the geometric and topological information of the 3D model of the part can be analyzed by CATIA software to obtain the extended attribute adjacency graph data of the entire part feature surface. This enables the extraction of information on 3D surface features and surface-to-surface relationship features. The extended attribute adjacency graph information can be completely saved locally independently of the 3D CAD software, so that the XML information of the extended attribute adjacency graph can be read by Python software for feature recognition.

[0054] In one exemplary embodiment of this application, a custom geometric feature structure prototype to be identified can be defined. In this case, the geometry and topology of the feature to be identified need to be set in advance. The feature structure prototype includes: feature structure name (StructureName, String), feature structure number (StructureNo, int), number of feature structure constituent faces (FaceSize, int), number of each constituent feature face (StructureFaceNo, int), attribute set of each constituent feature face (StructureFaceAttr, Object), and relationship set between each constituent feature face (StructureFaceRelationShipAttr, Object).

[0055] This invention determines the geometric topology to be identified by using a user-defined geometric feature structure prototype, such as... Figure 4 , Figure 5 and Figure 6 Examples of common feature structure models and specific surface-to-surface relationship diagrams are shown. For example... Figure 4 As shown, a typical square blind hole structure consists of five feature planes and their surface-to-surface relationships. The specific surface information attributes and surface-to-surface relationship attributes are shown in Table 1 and Table 2.

[0056] Table 1. Characteristic surface information attributes of the square blind hole

[0057]

[0058] Table 2. Attributes of the surface relationships of the characteristic surfaces of the square blind hole

[0059]

[0060] like Figure 5 As shown, a typical cylindrical double-step countersunk hole structure consists of three characteristic planes and two characteristic cylindrical surfaces, plus their surface-to-surface relationships. The specific surface information attributes are shown in Table 3, and the surface-to-surface relationship attributes are shown in Table 4.

[0061] Table 3. Characteristic surface information attributes of cylindrical double-step countersunk holes

[0062]

[0063] Table 4. Attributes of the surface relationships of the characteristic surfaces of cylindrical double-step countersunk holes.

[0064]

[0065] like Figure 6 As shown, a typical conical countersunk hole structure consists of a characteristic conical surface, a characteristic cylindrical surface, and two characteristic planes, plus their surface-to-surface relationships. The specific surface information attributes are shown in Table 5, and the surface-to-surface relationship attributes are shown in Table 6.

[0066] Table 5. Characteristic Surface Information Attributes of Conical Countersunk Holes

[0067]

[0068] Table 6. Surface Relationship Attributes of Characteristic Surfaces of Conical Countersunk Holes

[0069]

[0070] Based on the custom geometric feature structure prototype, an XML information dataset of the feature structure to be identified can be further derived. Specifically, step 202 involves obtaining the custom XML information dataset of the feature structure to be identified, including:

[0071] (b1) Obtain the attribute information Excel format file; the attribute information Excel format file refers to the file generated by organizing the surface information attributes and surface-to-surface relationship attributes of the feature structure to be identified in an Excel spreadsheet.

[0072] The surface information attributes and surface-to-surface relationship attributes of the feature structure to be identified can be manually filled into an Excel spreadsheet to form an attribute information Excel file.

[0073] (b2) Convert the attribute information Excel format file into an XML format file to obtain the XML information dataset of the feature structure to be identified.

[0074] You can use Python software to import an Excel file containing attribute information, use the openpyxl interface to read the Excel file and convert it to an XML file, and then use the XML file to generate a graph structure.

[0075] In this embodiment, the required feature structure to be identified can be set according to the needs, so as to identify the custom feature structure in the 3D model. Since the feature structure to be identified can be customized, more feature recognition needs can be met.

[0076] In another exemplary embodiment of this application, step 202, obtaining a custom XML information dataset of the feature structure to be identified, specifically includes:

[0077] In an existing XML information dataset with a feature structure, the face relation attributes and face-to-face relation attributes of the feature structure to be identified are configured by adding, deleting, and modifying operations, thus obtaining the XML information dataset of the feature structure to be identified.

[0078] This invention can provide user-defined geometric feature structure prototypes. The feature structure to be identified is configured by adding to and modifying an existing feature structure's XML information dataset. The custom geometric feature structure prototype XML file is as follows: Figure 7 As shown.

[0079] If a user wants to add a new feature structure to be identified, they only need to add it according to the XML style of the existing feature structure. Adding a new Structure node under the StandardFeatureStructure root node is considered adding a new feature structure to be identified. The Structure node's field attributes must include the feature structure number (StructureNo), the feature structure name (StructureName), and the feature structure's English name (StructureEnglishName). The Structure node contains FaceList feature face set nodes and RelationShipList feature face relationship nodes. The FaceList feature face set node contains the Face sections that make up the feature structure. The Face node contains fields such as Face No., Face Type, Number of Outer Loops (OutterLoopSize), Number of Inner Loops (InnerLoopSize), and Convexity (IsConvexSurface). The RelationShipList node contains RelationShip nodes for each face-to-face relationship, with fields including SourceFace No., TargetFace No., Intersection, Parallel, Vertical, and Convexity (IsConvexity). If the feature to be identified has multiple matching attribute fields, simply set the field content to the "multi_value" prefix (e.g., some faces in a typical slot may have multiple inner loop edges, so the InnerLoopSize of that face in the custom feature structure is multi_0). Users can customize new feature structures to be identified by adding content under the Structure node.

[0080] If a user wants to modify a custom feature structure to be identified, they only need to modify the field attributes of the corresponding face nodes or face-to-face relationships (edge ​​relationships between faces) to correct the geometric and topological information of the feature structure to be identified.

[0081] In one exemplary embodiment of this application, the subgraph isomorphic matching method uses the standard graph structure to be identified to search and match in the global graph structure information. If one or more graph structures to be identified are found, they are saved and the search continues until the entire search process is completed. Figure 8 and Figure 9 As shown. First, the two input XML data files are determined. The face and face-to-face relationship data of the extended attribute adjacency graph of the entire 3D model are obtained from the feature parsing data of the 3D model, and the AllGraph dataset, representing the full graph structure of the 3D model, is constructed using this data. Then, the standard graph structure data to be identified is obtained from the user-defined standard feature structure, and the StandardGraph dataset, representing the standard graph structure to be identified, is constructed using this data. Second, the matching attribute conditions for face nodes are configured, incorporating fields such as FaceType and OutterLoopSize of face nodes into the graph node judgment criteria. The matching attribute conditions for node relationships are also configured, incorporating fields such as IsIntersection, IsParallel, IsConvexity, and SizeEdgeIntersection of node relationships into the graph node judgment criteria. Additionally, when "multi" appears before a field attribute, that field is set as a universal matching field. Next, matching and identification are performed starting with a single standard graph structure and the full graph structure. The similarity between each feature subgraph structure in the full graph structure and each individual standard graph structure is traversed. A feature subgraph structure refers to the graph structure formed by a node in the full graph structure and nodes with edge relationships to that node. A set of candidate node pairs (feature subgraph-standard graph structure) is generated by combining the nodes and relationships between nodes in the feature subgraph structure and the standard graph structure. Next, a syntactic feasibility condition is determined, judging whether the degree of nodes and the mapping of adjacent nodes in the feature subgraph-standard graph structure combination are similar. Simultaneously, a semantic feasibility condition is determined, judging whether the matched node attributes and edge attributes are similar. If neither syntactic nor semantic feasibility conditions are met, a depth-first search is performed based on the current feature subgraph to generate new candidate nodes, thus generating new adjacent candidate node pairs (i.e., generating a new feature subgraph), and the feasibility judgment continues. If the syntactic and semantic feasibility conditions are met, the matched candidate node pairs are recorded. It is then confirmed whether all feature subgraphs to be identified have been completely matched. Here, a feature subgraph refers to a feature subgraph with the same number of nodes. If not all feature subgraphs with the same number of nodes have been traversed, the process backtracks to the previous decision point (the decision point is equivalent to adding a node using depth-first search; if no suitable node is found, the process returns to the original state of the feature subgraph) and continues generating and judging new candidate node pairs. Afterward, if all feature subgraphs with the same number of nodes have been traversed, it is determined whether all standard graph structures have been completely traversed. If not, the next standard graph structure is switched to perform the same full-graph recognition search until all standard graph structures have been traversed. Finally, all matching result pairs are saved, and the feature recognition result set file is output.

[0082] Therefore, as an optional implementation, step 204, performing graph structure similarity matching based on the full graph structure and the standard graph structure to identify the standard graph structure in the full graph structure, specifically includes: applying a subgraph isomorphism matching method to perform graph structure similarity matching on the full graph structure and the standard graph structure to identify the standard graph structure in the full graph structure, specifically:

[0083] (c1) Randomly select a standard graph structure and randomly determine any target feature subgraph in the full graph structure; the target feature subgraph is composed of multiple nodes with edge relationships; the number of nodes in the target feature subgraph is the same as the number of nodes in the currently selected standard graph structure.

[0084] (c2) Combine the current target feature subgraph with the currently selected standard graph structure into a graph structure combination.

[0085] (c3) Determine whether the two graph structures in the current graph structure combination meet the matching conditions to obtain the first judgment result; the matching conditions include the degree of the nodes of the two graph structures, the mapping of the adjacent nodes of the nodes, and the same node attributes and edge attributes.

[0086] (c4) If the first judgment result is yes, then record the successfully matched graph structure combination.

[0087] (c5) If the first judgment result is negative, then determine whether all target feature subgraphs contained in the current full graph structure have been traversed to obtain the second judgment result.

[0088] (c6) If the second judgment result is negative, then based on the current target feature subgraph, the current target feature subgraph is updated using a depth-first search algorithm to obtain a new target feature subgraph; the number of nodes in the new target feature subgraph is the same as the number of nodes in the currently selected standard graph structure.

[0089] (c7) Using the new target feature subgraph as the current target feature subgraph, return to the step "Combining the current target feature subgraph with the currently selected standard graph structure into a graph structure combination".

[0090] (c8) If the second judgment result is yes, then determine whether all standard graph structures have been traversed, and obtain the third judgment result.

[0091] (c9) If the third judgment result is negative, a new standard graph structure is randomly selected from the untraversed standard graph structures, and the new standard graph structure is set as the currently selected standard graph structure. Then, the process returns to the step "randomly determine any target feature subgraph in the full graph structure".

[0092] (c10) If the third judgment result is yes, then all the standard graph structures in the full graph structure are obtained.

[0093] In another exemplary embodiment of this application, the similarity graph matching method achieves the search and matching of the graph structure to be identified in the global graph structure by calculating the number of steps for mutual transformation between graphs. If one or more graph structures to be identified are found, they are saved and the search continues until the entire search process is completed. Figure 10 As shown, the process of the similarity graph matching method is as follows: Based on the previously established StandardGraph structure, it is used as the query graph, and the constructed full graph structure AllGraph is used as the object graph. First, all face nodes in the full graph structure AllGraph are traversed, and all face nodes in the full graph structure are called source nodes. Then, nodes associated with all source nodes are obtained to form a second-level node subgraph. Then, each adjacent node of the second-level node is used as an object to form a third-level node subgraph. Using this traversal rule, the full graph structure AllGraph is decomposed into a set of third-level node subgraphs. Specifically, a second-level node is all adjacent nodes that intersect with the source node, and a third-level node is a specific adjacent node that intersects with a second-level node. The decomposed third-level node subgraphs are collected to form a dataset. Using this dataset as the query object, the similarity score between each standard graph structure and each third-level node subgraph in the dataset is calculated. When calculating the similarity score, face information attributes and face-to-face relationship attributes are considered as conditions for similarity calculation. The edit distance between graphs is used as the final similarity score. The similarity score between all tertiary node subgraphs and each standard graph structure is obtained, and the one with the highest score is the final identified object. The index of the standard graph structure with the highest score and the face number corresponding to the tertiary node subgraph are output to form a result set file.

[0094] Therefore, as another optional implementation, step 204, performing graph structure similarity matching based on the full graph structure and the standard graph structures to identify each of the standard graph structures in the full graph structure, specifically includes: applying a similarity graph matching method to perform graph structure similarity matching on the full graph structure and the standard graph structures to identify the standard graph structures in the full graph structure, specifically:

[0095] (d1) The full graph structure is decomposed into several tertiary node subgraphs; each tertiary node subgraph includes a source node, at least one second-level node and a tertiary node; each node in the full graph structure is regarded as a source node; the second-level node is a node that has an edge relationship with the source node in the full graph structure; the tertiary node is a node that has an edge relationship with the second-level node in the full graph structure.

[0096] (d2) For each of the three-level node subgraphs, calculate the edit distance between the three-level node subgraph and each of the standard graph structures according to the surface relationship attribute and the surface-to-surface relationship attribute, and obtain the similarity score between the three-level node subgraph and each of the standard graph structures; when multiple types of features to be identified are defined, one type of feature to be identified corresponds to one type of standard graph structure.

[0097] (d3) The standard graph structure with the highest similarity score is regarded as the feature recognition result of the corresponding third-level node subgraph.

[0098] This application also provides an application scenario in which the above-mentioned automatic feature recognition method based on extended attribute adjacency graphs is applied. Specifically, the automatic feature recognition method based on extended attribute adjacency graphs provided in this embodiment can be applied to the structural feature dimension annotation scenario of a 3D model in CATIA software. This scenario includes a 3D model feature recognition stage, a feature recognition structural information processing stage, an annotation view and annotation datum selection stage, and a 3D annotation stage of structural features. Automatic 3D model feature recognition is performed to obtain a feature recognition result set. Information is bound to the XML result dataset obtained after feature recognition. The 3D model of the part to be recognized is opened in CATIA software. By reading the XML result dataset file, multiple recognized feature structures are displayed and bound one by one to the feature surfaces of the 3D model of the part. Based on the view and datum, the surfaces contained in the structure are annotated, quickly realizing the 3D dimension annotation of the feature structure under different datums and different views. The automatic feature recognition method based on extended attribute adjacency graphs provided in this embodiment belongs to the 3D model feature recognition stage.

[0099] In this embodiment, a feature recognition system for 3D models was developed based on the CATIA secondary development tool and the Python programming language. The set of surface information and the set of surface-to-surface relationships of the 3D model were obtained by parsing using the CATIA software, and the similarity matching between the full-image structure and the standard image structure in step 204 was implemented in the Python software.

[0100] Based on the same inventive concept, this application also provides an automatic feature recognition device based on extended attribute adjacency graphs for implementing the above-mentioned automatic feature recognition method based on extended attribute adjacency graphs. The solution provided by this device is similar to the implementation described in the above method. Therefore, the specific limitations in one or more embodiments of the automatic feature recognition device based on extended attribute adjacency graphs provided below can be found in the limitations of the automatic feature recognition method based on extended attribute adjacency graphs described above, and will not be repeated here.

[0101] In one exemplary embodiment, such as Figure 11 As shown, an automatic feature recognition device based on an extended attribute adjacency graph is provided, comprising:

[0102] The 3D model attribute information acquisition module M1 is used to acquire the XML information dataset of the extended attribute adjacency graph of the 3D model.

[0103] The custom feature structure module M2 is used to obtain a dataset of XML information about custom feature structures to be identified.

[0104] The graph structure generation module M3 is used to derive the full graph structure corresponding to the 3D model and the standard graph structure corresponding to the feature structure to be identified based on the XML information dataset of the extended attribute adjacency graph and the XML information dataset of the standard graph structure. Both the full graph structure and the standard graph structure include several nodes and edges between nodes. Each node represents a face. Each node contains face information attributes. Each edge represents the relationship between faces. Each edge contains face-to-face relationship attributes.

[0105] The graph structure matching module M4 is used to perform graph structure similarity matching based on the full graph structure and the standard graph structure, and to identify the standard graph structure in the full graph structure.

[0106] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 12 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores feature-based automatic identification data based on extended attribute adjacency graphs. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When executed by the processor, the computer program implements a feature-based automatic identification method based on extended attribute adjacency graphs.

[0107] Those skilled in the art will understand that Figure 12 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0108] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0109] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0110] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0111] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0112] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0113] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0114] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0115] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.< / repeattagface>

Claims

1. A method for automatic feature recognition based on extended attribute adjacency graphs, characterized in that, include: Obtain the XML information dataset of the extended attribute adjacency graph of the 3D model; Obtain a custom XML data set containing the features to be identified; Based on the XML information dataset of the extended attribute adjacency graph and the XML information dataset of the feature structure to be identified, the full graph structure corresponding to the 3D model and the standard graph structure corresponding to the feature structure to be identified are derived; both the full graph structure and the standard graph structure include several nodes and edges between nodes; each node represents a face; each node contains face information attributes; each edge represents the relationship between faces; each edge contains face-to-face relationship attributes; Graph structure similarity matching is performed based on the full graph structure and the standard graph structure to identify the standard graph structure in the full graph structure; Specifically, the process of performing graph structure similarity matching based on the full graph structure and the standard graph structures to identify each of the standard graph structures in the full graph structure includes: applying a similarity graph matching method to perform graph structure similarity matching on the full graph structure and the standard graph structures to identify the standard graph structures in the full graph structure. The full graph structure is decomposed into several tertiary node subgraphs; each tertiary node subgraph includes a source node, at least one second-level node, and one tertiary node; each node in the full graph structure is considered a source node; the second-level node is a node that has an edge relationship with the source node in the full graph structure; the tertiary node is a node that has an edge relationship with the second-level node in the full graph structure. For each of the three-level node subgraphs, the edit distance between the three-level node subgraph and each of the standard graph structures is calculated based on the face relationship attribute and the face-to-face relationship attribute, respectively, and the similarity score between the three-level node subgraph and each of the standard graph structures is obtained; when multiple types of features to be identified are defined, one type of feature to be identified corresponds to one type of standard graph structure; The standard graph structure with the highest similarity score is regarded as the feature recognition result of the corresponding third-level node subgraph.

2. The automatic feature recognition method based on extended attribute adjacency graph according to claim 1, characterized in that, Obtain the XML information dataset of the extended attribute adjacency graph of the 3D model, specifically including: Obtain the geometric and topological information of the surface of the 3D model; The set of surface information attributes and the set of surface-to-surface relationship attributes of the three-dimensional model are obtained based on the geometric information and the topological information. The XML information dataset of the extended attribute adjacency graph is derived based on the set of face information attributes and the set of face-to-face relationship attributes of the 3D model; the nodes in the extended attribute adjacency graph contain the face information attributes of the 3D model, and the edges between the nodes in the extended attribute adjacency graph contain the face-to-face relationship attributes of the 3D model.

3. The automatic feature recognition method based on extended attribute adjacency graph according to claim 1, characterized in that, Obtain a custom XML information dataset containing the features to be identified, specifically including: Obtain an Excel file containing attribute information; the Excel file containing attribute information refers to a file generated by organizing the surface information attributes and surface-to-surface relationship attributes of the feature structure to be identified in an Excel spreadsheet. The attribute information Excel file is converted into an XML file to obtain the XML information dataset of the feature structure to be identified.

4. The automatic feature recognition method based on extended attribute adjacency graph according to claim 3, characterized in that, Obtain a custom XML information dataset containing the features to be identified, specifically including: In an existing XML information dataset with a feature structure, the face relation attributes and face-to-face relation attributes of the feature structure to be identified are configured by adding, deleting, and modifying operations, thus obtaining the XML information dataset of the feature structure to be identified.

5. The automatic feature recognition method based on extended attribute adjacency graph according to claim 1, characterized in that, Based on the full graph structure and the standard graph structure, graph structure similarity matching is performed to identify the standard graph structure within the full graph structure. Specifically, this includes applying a subgraph isomorphism matching method to perform graph structure similarity matching on the full graph structure and the standard graph structure to identify the standard graph structure within the full graph structure. A standard graph structure is randomly selected, and any target feature subgraph in the full graph structure is randomly determined; the target feature subgraph is composed of multiple nodes with edge relationships; the number of nodes in the target feature subgraph is the same as the number of nodes in the currently selected standard graph structure. Combine the current target feature subgraph with the currently selected standard graph structure into a graph structure combination; Determine whether the two graph structures in the current graph structure combination meet the matching conditions to obtain the first judgment result; the matching conditions include the degree of the nodes of the two graph structures, the mapping of the adjacent nodes of the nodes, and the same node attributes and edge attributes; If the first judgment result is yes, then record the successfully matched graph structure combination; If the first judgment result is negative, then it is determined whether all target feature subgraphs contained in the current full graph structure have been traversed to obtain the second judgment result; If the second judgment result is negative, then based on the current target feature subgraph, the current target feature subgraph is updated using a depth-first search algorithm to obtain a new target feature subgraph; the number of nodes in the new target feature subgraph is the same as the number of nodes in the currently selected standard graph structure; Using the new target feature subgraph as the current target feature subgraph, return to the step "Combining the current target feature subgraph with the currently selected standard graph structure into a graph structure combination"; If the second judgment result is yes, then determine whether all standard graph structures have been traversed to obtain the third judgment result; If the third judgment result is negative, a new standard graph structure is randomly selected from the untraversed standard graph structures, and the new standard graph structure is set as the currently selected standard graph structure. Then, the process returns to the step "randomly determine any target feature subgraph in the full graph structure". If the third judgment result is yes, then all the standard graph structures in the full graph structure are obtained.

6. An automatic feature recognition device based on an extended attribute adjacency graph, characterized in that, include: The 3D model attribute information acquisition module is used to acquire the XML information dataset of the extended attribute adjacency graph of the 3D model. The feature structure customization module is used to obtain a dataset of XML information about custom feature structures to be identified. The graph structure generation module is used to derive the full graph structure corresponding to the 3D model and the standard graph structure corresponding to the feature structure to be identified based on the XML information dataset of the extended attribute adjacency graph and the XML information dataset of the feature structure to be identified. Both the full graph structure and the standard graph structure include several nodes and edges between nodes. Each node represents a face. Each node contains face information attributes. Each edge represents the relationship between faces. Each edge contains face-to-face relationship attributes. The graph structure matching module is used to perform graph structure similarity matching based on the full graph structure and the standard graph structure, and to identify the standard graph structure in the full graph structure. Specifically, the process of performing graph structure similarity matching based on the full graph structure and the standard graph structures to identify each of the standard graph structures in the full graph structure includes: applying a similarity graph matching method to perform graph structure similarity matching on the full graph structure and the standard graph structures to identify the standard graph structures in the full graph structure. The full graph structure is decomposed into several tertiary node subgraphs; each tertiary node subgraph includes a source node, at least one second-level node, and one tertiary node; each node in the full graph structure is considered a source node; the second-level node is a node that has an edge relationship with the source node in the full graph structure; the tertiary node is a node that has an edge relationship with the second-level node in the full graph structure. For each of the three-level node subgraphs, the edit distance between the three-level node subgraph and each of the standard graph structures is calculated based on the face relationship attribute and the face-to-face relationship attribute, respectively, and the similarity score between the three-level node subgraph and each of the standard graph structures is obtained; when multiple types of features to be identified are defined, one type of feature to be identified corresponds to one type of standard graph structure; The standard graph structure with the highest similarity score is regarded as the feature recognition result of the corresponding third-level node subgraph.

7. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the automatic feature identification method based on an extended attribute adjacency graph as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the automatic feature recognition method based on extended attribute adjacency graphs as described in any one of claims 1-5.

Images