Threaded hole feature recognition method, system and equipment for cartridge receiver part and medium

Abstract

The invention relates to a threaded hole feature recognition method, system and equipment for a cartridge receiver part and a medium, and relates to the field of intelligent manufacturing. According to the technical scheme, the basic constituent surface and the constituent line of the threaded hole are cut into, graph structure modeling is achieved by constructing the model adjacent graph and the feature sub-graph, precise matching is completed in combination with isomorphic search of the sub-graph, complete geometric parameters can be extracted, and the problem that the recognition effect of a traditional method on complex cartridge receiver parts is poor is effectively solved; moreover, aiming at model topology fragmentation, interference is eliminated through fragmentation surface preprocessing and surface cluster reconstruction, and accurate and stable identification is guaranteed; and besides, various types of threaded holes can be covered, extracted parameters such as nominal diameter, thread pitch, thread length and total depth meet manufacturing requirements, subsequent process design, tool selection and numerical control machining path planning of the cartridge receiver part can be directly supported, CAD-CAPP-CAM integrated landing can be assisted, and the machining automation efficiency and process accuracy of the complex cartridge receiver part are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, and in particular to a method, system, device and medium for identifying threaded hole features of casing parts. Background Technology

[0002] In modern manufacturing, the widespread application of computer-aided design (CAD) and computer-aided manufacturing (CAM) technologies has significantly improved the overall efficiency of product development and production. Computer-aided process planning (CAPP), as a core link between CAD and CAM, needs to extract machining features with clear engineering semantics (such as holes, cavities, and threaded holes) from the 3D model of a part. This supports the integrated data flow of CAD-CAPP-CAM and automated programming for CNC machining. This requirement is particularly critical in the manufacturing of complex and precision parts (such as casing parts in the aerospace field).

[0003] Among them, accurate identification of machining features is the core foundation for realizing intelligent CNC programming. Although the current mainstream CAD / CAPP systems can stably identify and apply simple geometric features such as holes and chamfers, for threaded hole features, due to their complex structural morphology (including combinations of multiple surfaces such as helical surfaces, end faces, and conical surfaces) and diverse core parameters (such as nominal diameter, pitch, thread profile, and depth), there is still a lack of universal, accurate, and robust identification technology solutions. As a result, threaded hole features have become a key bottleneck restricting the semantic parsing of part models and the implementation of automated machining processes. This problem is particularly prominent for complex machine casing parts with a large number and variety of threaded holes.

[0004] In existing threaded hole identification technologies, some solutions are based on model boundary information matching or deep learning algorithms to locate threaded holes. However, these methods are mostly limited to identifying the location or macroscopic type of threaded holes (such as blind holes / through holes), and cannot fully extract the geometric attribute parameters that meet manufacturing requirements. At the same time, these methods generally lack the ability to adapt to actual engineering scenarios such as model topology fragmentation (such as surface splitting caused by complex modeling processes or format conversions), and are difficult to deal with the common topological irregularity problem in the modeling of casing parts.

[0005] For example, Chinese patent CN120298747A discloses a method for identifying threaded holes in a 3D lightweight model based on a cloud server. Although it can achieve basic identification of threaded holes, it is only suitable for the visualization display scenario of threaded holes in the national standard. It cannot construct the topological relationship between the threaded hole and the overall casing part, nor can it meet the actual production needs of local offline identification of complex casing parts. Another example is Chinese patent CN116485773A, which discloses a threaded hole locking and guiding method based on image positioning. It relies on image positioning tooling to realize the dynamic tracking and locking guidance of threaded holes, but the processing object is two-dimensional image data. It does not have the ability to reconstruct the geometric features of threaded holes from three-dimensional CAD models and cannot be integrated into the CAD-CAPP-CAM processing link of casing parts.

[0006] In summary, the existing technologies have the following four main shortcomings: (1) It is difficult to accurately identify the multi-morphological geometric composition of threaded holes in complex structural parts such as casings (e.g., conical bottom blind holes, flat bottom blind holes with small holes, threaded through holes, etc.), and cannot cover the structural diversity of threaded holes; (2) It lacks a mechanism for processing broken surfaces and reconstructing surface clusters for model topological fragmentation, which is prone to deviation in recognition results due to interference from broken surfaces, reducing the accuracy and stability of the recognition of casing part models; (3) Most methods rely on lightweight models or two-dimensional image information, and have not established the ability to analyze the complete topological features of the three-dimensional model of the part (face-edge connection relationship, face / edge attribute association, etc.), and cannot support the deep semantic analysis of threaded holes in casing parts; (4) It cannot output core geometric parameters such as nominal diameter, pitch, thread length, and total depth to meet manufacturing requirements, which directly limits the automation of subsequent machining process design, tool selection and CNC machining path planning for casing parts. Summary of the Invention

[0007] The purpose of this invention is to provide a method, system, device, and medium for identifying threaded hole features of casing parts, so as to solve the problems existing in the prior art.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a method for identifying threaded hole features of a casing part, the method being applied in a computer device, the method comprising: Input a 3D geometric model of the casing that includes threaded hole features; Extract the face and edge topology information of the three-dimensional geometric model and construct the model adjacency graph; the face and edge topology information includes the unique identifier of the face, the unique identifier of the edge, and the connection relationship between the face and the edge; Load at least one threaded hole feature template from a preset threaded hole feature library, extract the face-edge topological relationship of the feature template, and construct the corresponding feature sub-graph; The model adjacency graph and the feature subgraph are preprocessed with fragmented surfaces. Based on geometric continuity and topological adjacency, the fragmented surfaces are reconstructed into surface cluster units to form a surface cluster adjacency graph. Define the matching criteria between nodes and directed edges in the adjacency graph of the face cluster, take each feature subgraph as the retrieval target, perform subgraph isomorphic search in the adjacency graph of the model face cluster, collect node mapping results and restore the original fragmented face after the search is completed; Based on the node mapping results and the restored fragmented surfaces, the key geometric attribute parameters of the threaded hole are extracted; the key geometric attribute parameters include nominal diameter, pitch, thread length and total depth. Output the identifiers of all threaded hole feature surfaces in the three-dimensional geometric model and their corresponding key geometric attribute parameters.

[0009] In some implementations, extracting the face and edge topology information of the three-dimensional geometric model and constructing the model adjacency graph includes: Each face of the three-dimensional geometric model is marked with a unique face ID, and each face ID corresponds to a node in the adjacency graph of the model. The ID of the node is the same as the face ID. Obtain all edges of the three-dimensional geometric model. Each edge connects two different faces, generating two directed edges accordingly. That is, a first directed edge pointing from the node corresponding to the first face ID to the node corresponding to the second face ID, and a second directed edge pointing from the node corresponding to the second face ID to the node corresponding to the first face ID. Add face attributes to the nodes of the adjacency graph of the model. The face attributes include the area of ​​the face, the normal vector, and the face type. Add edge attributes to the directed edges of the adjacency graph of the model. The edge attributes include edge length, concavity / convexity, inner / outer edge identifiers, and edge type.

[0010] In some implementations, the preset threaded hole feature library includes five typical threaded hole feature templates, and the feature sub-graph parameters corresponding to each template include: The blind hole template for tapered bottom threaded holes has a feature subgraph containing 12 nodes and 18 directed edges; The flat-bottomed threaded blind hole template has a feature subgraph containing 12 nodes and 18 directed edges; A template with a tapered bottom threaded hole and a small hole has a feature subgraph containing 14 nodes and 20 directed edges; A flat-bottomed threaded hole template with small holes has a feature subgraph containing 14 nodes and 22 directed edges. The threaded hole through-hole template has a feature subgraph containing 8 nodes and 8 directed edges.

[0011] In some implementations, the step of performing fragmented surface preprocessing on the model adjacency graph and the feature subgraph, and reconstructing the fragmented surfaces into face clusters based on geometric continuity and topological adjacency relationships to form a face cluster adjacency graph, includes: The criteria for determining the fragmented surfaces of adjacent faces in the model's adjacency graph and the feature subgraph are: the adjacent faces have the same face type and the connecting edge between adjacent faces is a smooth edge. For fragmented nodes that meet the fragmented surface determination conditions, a reconstruction operation is performed. Starting from any fragmented node, all its connection relationships are traversed, and all nodes connected through fragmented surfaces are identified to form a set of node face clusters. A representative node is selected from the set of node face clusters, the representative node is retained and the remaining nodes in the set are deleted, and all topological relationships of the remaining nodes are merged into the representative node. Repeat the above reconstruction operation until all fragmented nodes are merged into face clusters, thus obtaining the face cluster adjacency graph.

[0012] In some implementations, defining the matching criteria between nodes and directed edges in the face cluster adjacency graph, using each of the feature subgraphs as the retrieval target, performing a subgraph isomorphic search in the model face cluster adjacency graph, and collecting node mapping results and restoring the original fragmented faces after the search is completed includes: The matching criteria are that the number of nodes in the retrieved subgraph is the same as that in the feature subgraph, the adjacency relationship between the nodes is consistent, and the face attribute of the node and the edge attribute of the directed edge in the subgraph match the face attribute of the corresponding node and the edge attribute of the corresponding directed edge in the feature subgraph, respectively. After performing the subgraph isomorphic search, the mapping relationship between the feature subgraph nodes and the model face cluster adjacency graph nodes is recorded as the node mapping result; Based on the association between representative nodes and original fragment nodes in the node mapping results, the face IDs of all original fragments are restored.

[0013] In some implementations, extracting the key geometric attribute parameters of the threaded hole based on the node mapping result and the restored fragments includes: Extract the subset of cylindrical surfaces from the face cluster corresponding to the node mapping result, obtain the radius of the cylindrical surface, and multiply the radius by two to obtain the nominal diameter of the threaded hole; Three-dimensional points are uniformly sampled on all surfaces of the restored fragmented surface. The three-dimensional points are projected along the main axis of the threaded hole, and the maximum distance between the projected points is calculated as the total depth of the threaded hole. Three-dimensional points are uniformly sampled on the boundary of the cylindrical helical surface in the restored fragmented surface. The three-dimensional points are then transformed into a local cylindrical coordinate system with the threaded hole axis as the reference, and the functional relationship between the axial coordinate z of the sampling point and the circumferential angle θ is established. By fitting the function relationship with least squares linearly, the fitted line z = aθ + b is obtained, where a is the axial slope; The pitch of the threaded hole is calculated as p = 2πa based on the axial slope a. Obtain the circumferential angle span Δθ of the sampling point, and calculate the thread length L=a·Δθ of the threaded hole based on the axial slope a and the angle span Δθ.

[0014] In some implementations, the step of outputting the identifiers of all threaded hole feature surfaces in the three-dimensional geometric model and their corresponding key geometric attribute parameters includes: For each thread hole feature template in the preset thread hole feature library, repeat the steps of fragment preprocessing, subgraph isomorphic search, node mapping result collection and fragment restoration, and key geometric attribute parameter extraction. All threaded hole feature faces are visualized and marked in the three-dimensional geometric model, and the nominal diameter, pitch, thread length and total depth corresponding to each threaded hole feature face are displayed in association.

[0015] In a second aspect, the present invention provides a threaded hole feature recognition system for a casing part, the system comprising: The model input module is used to input a three-dimensional geometric model of the casing, including the threaded hole feature; The topology information extraction and adjacency graph construction module is used to extract the face and edge topology information of the three-dimensional geometric model and construct the model adjacency graph; the face and edge topology information includes the unique identifier of the face, the unique identifier of the edge, and the connection relationship between the face and the edge; The feature template loading and subgraph construction module is used to load at least one threaded hole feature template from a preset threaded hole feature library, extract the face-edge topological relationship of the feature template, and construct the corresponding feature subgraph. The fragmented surface preprocessing and surface cluster adjacency graph generation module is used to perform fragmented surface preprocessing on the model adjacency graph and the feature subgraph, and reconstruct the fragmented surfaces into surface cluster units according to geometric continuity and topological adjacency relationship to form a surface cluster adjacency graph. The subgraph isomorphic search and fragmented surface restoration module is used to define the matching criteria between nodes and directed edges in the adjacency graph of the face cluster, take each feature subgraph as the retrieval target, perform subgraph isomorphic search in the adjacency graph of the model face cluster, collect node mapping results and restore the original fragmented surface after the search is completed; The key geometric attribute parameter extraction module is used to extract the key geometric attribute parameters of the threaded hole based on the node mapping result and the restored fragmented surface; the key geometric attribute parameters include nominal diameter, pitch, thread length and total depth. The identification result output module is used to output the identifiers of all threaded hole feature surfaces in the three-dimensional geometric model and their corresponding key geometric attribute parameters.

[0016] Thirdly, the present invention provides a computer device including a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the processor can load and execute at least one instruction, at least one program, code set or instruction set to realize the thread hole feature identification method for casing parts provided above.

[0017] Fourthly, the present invention provides a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein a processor can load and execute at least one instruction, at least one program, code set, or instruction set to implement the threaded hole feature identification method for casing parts provided above.

[0018] Fifthly, the present invention provides a computer program product or computer program including computer program instructions stored in a computer-readable storage medium. A processor reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the threaded hole feature identification method for a casing part as described above.

[0019] The beneficial effects of the technical solution provided by this invention include at least the following: This technical solution starts with the basic constituent surfaces and lines of threaded holes, and achieves graph structure modeling by constructing a model adjacency graph and feature subgraphs. Combined with subgraph isomorphic search, it achieves accurate matching and can extract complete geometric parameters. It effectively solves the problem of poor recognition results of traditional methods on complex casing parts. Furthermore, it eliminates interference through fragmented surface preprocessing and surface cluster reconstruction to ensure accurate and stable recognition, especially for model topological fragmentation. In addition, it can cover various threaded hole types such as conical bottom blind holes and flat bottom blind holes with small holes. The extracted parameters such as nominal diameter, pitch, thread length, and total depth meet manufacturing requirements and can directly support the subsequent process design, tool selection, and CNC machining path planning of casing parts. This facilitates the implementation of CAD-CAPP-CAM integration and significantly improves the automation efficiency and process accuracy of complex casing parts. Attached Figure Description

[0020] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0021] Figure 1 The diagram shows a flowchart illustrating a method for identifying threaded hole features of a casing part according to an exemplary embodiment of the present invention.

[0022] Figure 2A three-dimensional model of a casing containing threaded holes and through holes, provided by an exemplary embodiment of the present invention, is shown.

[0023] Figure 3 An adjacency graph of a three-dimensional model of a casing provided by an exemplary embodiment of the present invention is shown.

[0024] Figure 4(a) shows a conical bottom threaded blind hole and its feature sub-diagram provided by an exemplary embodiment of the present invention.

[0025] Figure 4(b) shows a flat-bottomed threaded blind hole and its feature sub-diagram provided by an exemplary embodiment of the present invention.

[0026] Figure 4(c) shows a tapered bottom threaded hole with a small hole and its feature sub-diagram provided in an exemplary embodiment of the present invention.

[0027] Figure 4(d) shows a flat-bottomed threaded hole with a small hole and its feature sub-graph provided in an exemplary embodiment of the present invention.

[0028] Figure 4(e) shows a threaded through hole and its feature sub-diagram provided in an exemplary embodiment of the present invention.

[0029] Figure 5 This illustration shows a schematic diagram of the merging of helical cylindrical fragments in a blind hole with a tapered bottom threaded hole provided by an exemplary embodiment of the present invention.

[0030] Figure 6 The diagram illustrates a flowchart of merging the model face cluster adjacency graph and the feature face cluster subgraph of a method for identifying threaded hole features of a casing part according to an exemplary embodiment of the present invention.

[0031] Figure 7 This diagram illustrates the visualization results of threaded hole feature recognition in a partial sub-model of a three-dimensional casing model provided by an exemplary embodiment of the present invention.

[0032] Figure 8 The diagram shows the final identification result of all threaded hole features in the three-dimensional model of the casing provided by an exemplary embodiment of the present invention.

[0033] Figure 9 This illustration shows a three-dimensional model of a casing with various different threaded hole machining features, provided as an exemplary application example of the present invention.

[0034] Figure 10 An adjacency graph of a 3D model of a casing provided in an exemplary application example of the present invention is shown.

[0035] Figure 11 A type diagram of the threaded hole feature library provided in an exemplary application example of the present invention is shown.

[0036] Figure 12The illustration shows a diagram of threaded hole features identified in a 3D model of a casing, provided as an exemplary application of the present invention. The corresponding geometric attribute information is obtained by clicking on the interactive feature.

[0037] Figure 13 This diagram shows an overview of the final identification results of all threaded hole features in a 3D model of a casing provided in an exemplary application example of the present invention.

[0038] Figure 14(a) shows a specific illustration of the first type of threaded hole feature present in a casing model provided in an exemplary application example of the present invention.

[0039] Figure 14(b) shows a detailed illustration of the second type of threaded hole feature present in a casing model provided in an exemplary application example of the present invention.

[0040] Figure 14(c) shows a detailed illustration of the third type of threaded hole feature present in a casing model provided in an exemplary application example of the present invention.

[0041] Figure 14(d) shows a specific illustration of the fourth type of threaded hole feature present in a casing model provided in an exemplary application example of the present invention.

[0042] Figure 15 The diagram shows a structural block diagram of a threaded hole feature recognition system for a casing part provided by an exemplary embodiment of the present invention.

[0043] Figure 16 This diagram illustrates the structure of a computer device for identifying threaded hole features of a casing part according to an exemplary embodiment of the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0046] Figure 1 The diagram illustrates a flowchart of a method for identifying threaded hole features of a casing part according to an exemplary embodiment of the present invention. This method for identifying threaded hole features of a casing part is applied in a computer device, and the method includes: Step 101: Input the three-dimensional geometric model of the casing, which includes the threaded hole feature.

[0047] In this embodiment of the application, the input data is a three-dimensional geometric model of the casing containing threaded hole features. This data is the basis for all subsequent topology extraction, subgraph matching and other operations.

[0048] Step 102: Extract the face and edge topology information of the 3D geometric model and construct the model adjacency graph; the face and edge topology information includes the unique identifier of the face, the unique identifier of the edge, and the connection relationship between the face and the edge.

[0049] In this embodiment, the intuitive face and edge elements in the 3D model are transformed into a computer-computable adjacency graph structure, solving the problem of ambiguous relationships when directly processing the 3D model; the specific content contained in the face and edge topology information is clearly defined to ensure the completeness of the extracted information.

[0050] In some embodiments, the above-described extraction of face and edge topology information of the 3D geometric model and construction of the model adjacency graph includes: marking each face of the 3D geometric model as a unique face ID, each face ID corresponding to a node in the model adjacency graph, and the node ID being consistent with the face ID; obtaining all edges of the 3D geometric model, each edge connecting two different faces, corresponding to the generation of two directed edges; that is, a first directed edge from the node corresponding to the first face ID to the node corresponding to the second face ID, and a second directed edge from the node corresponding to the second face ID to the node corresponding to the first face ID; adding face attributes to the nodes of the model adjacency graph, the face attributes including the face area, normal vector, and face type; adding edge attributes to the directed edges of the model adjacency graph, the edge attributes including the edge length, concavity / convexity, internal / external edge identifier, and edge type.

[0051] In this embodiment, by adding face attributes (area, normal vector, etc.) and edge attributes (length, concavity, convexity, etc.), a judgment basis is provided for subsequent subgraph matching, thereby improving the practicality of the adjacency graph.

[0052] In one example, each face of the 3D geometric model is labeled with a unique number (i.e., face ID), with the numbering sequence [0, 1, 2, ..., n]. Simultaneously, n adjacency graph nodes are constructed to correspond to these n faces, and the node ID is completely identical to the face ID of the corresponding face. All edges are extracted from the 3D geometric model, totaling K edges. Since each edge connects two different faces, there is a clear correspondence between edges and faces. Taking edge L as an example, if it connects face i and face j (i ≠ j), then two directed edges are generated: the directed edge {Lij} from node i to node j, and the directed edge {Lji} from node j to node i. When constructing the model adjacency graph according to the above rules, a 3D model will ultimately form n nodes and 2K directed edges. Figure 2As shown, the casing model contains multiple threaded holes of the same type. To simplify the explanation of the principle, only a portion of the threaded hole structure in the model is analyzed. This sub-model consists of 71 faces and 180 edges. After extracting the face and edge topology information of this sub-model, the resulting adjacency graph is as follows. Figure 3 As shown.

[0053] Step 103: Load at least one threaded hole feature template from the preset threaded hole feature library, extract the face-edge topology relationship of the feature template, and construct the corresponding feature subgraph.

[0054] In this embodiment, by loading a threaded hole feature template from a preset feature library, extracting its face-edge topological relationship and constructing a feature subgraph, the abstract threaded hole feature is transformed into a specific topological structure. This not only connects the preset feature library with the actual recognition process, but also provides a clear target reference for searching for threaded holes in the model adjacency graph.

[0055] In some embodiments, the aforementioned preset threaded hole feature library includes five typical threaded hole feature templates, classified according to the topological differences between the basic constituent surfaces and constituent lines of the threaded hole. The feature sub-graph parameters corresponding to each template include: the blind hole template of the conical bottom threaded hole in Figure 4(a), whose feature sub-graph contains 12 nodes and 18 directed edges; the blind hole template of the flat bottom threaded hole in Figure 4(b), whose feature sub-graph contains 12 nodes and 18 directed edges; the template of the conical bottom threaded hole with a small hole in Figure 4(c), whose feature sub-graph contains 14 nodes and 20 directed edges; the template of the flat bottom threaded hole with a small hole in Figure 4(d), whose feature sub-graph contains 14 nodes and 22 directed edges; and the through hole template of the threaded hole in Figure 4(e), whose feature sub-graph contains 8 nodes and 8 directed edges.

[0056] In this embodiment of the application, five threaded hole templates covering common industrial scenarios and their corresponding feature sub-graph parameters are used to make the feature library more visual, ensuring that the method can adapt to the threaded hole recognition needs of most casing parts.

[0057] In some embodiments, the method for constructing the above-mentioned preset threaded hole feature library includes: generating basic parameters for five typical threaded hole feature templates based on industry standard threaded hole parameters (including GB, ISO or ANSI standards); providing a user-defined interface to allow users to extend other irregular threaded hole models to the threaded hole feature library.

[0058] In this embodiment, a basic template is generated based on industry standards to ensure that the template conforms to general industry specifications and avoid recognition deviations caused by non-standard templates. At the same time, through the user-defined interface function, users can extend other irregular threaded hole models to the threaded hole feature library. Then, during recognition, the irregular threaded holes in the feature library are extracted as feature sub-graphs and matched with the model surface cluster adjacency graph.

[0059] Step 104: Perform fragmented surface preprocessing on the model adjacency graph and feature subgraph. Based on geometric continuity and topological adjacency relationship, reconstruct the fragmented surfaces into surface cluster units to form a surface cluster adjacency graph.

[0060] In this embodiment, the fragmented surfaces are reconstructed into surface clusters based on geometric continuity and topological adjacency, forming a simplified surface cluster adjacency graph. This reduces node redundancy and topological complexity in the model adjacency graph and avoids invalid computations in subsequent subgraph matching.

[0061] In some embodiments, the above-described preprocessing of the model adjacency graph and feature subgraph, and the reconstruction of the fragmented surfaces into surface clusters based on geometric continuity and topological adjacency relationships to form a surface cluster adjacency graph, includes: determining the fragmented surface determination conditions of adjacent surfaces in the model adjacency graph and feature subgraph; the determination conditions are that the adjacent surfaces have the same surface type and the connecting edges between adjacent surfaces are smooth edges; the connecting edges between adjacent surfaces are specifically defined as the angle between the normal vectors of adjacent surfaces being within the range of 0° to 8° or within the range of 172° to 180°; performing a reconstruction operation on the fragmented surface nodes that meet the fragmented surface determination conditions; starting from any fragmented surface node, traversing all its connection relationships, identifying all nodes connected through fragmented surfaces, and forming a node surface cluster set; selecting a representative node in the node surface cluster set, retaining the representative node and deleting the remaining nodes in the set, and merging all topological relationships of the remaining nodes into the representative node; repeating the above reconstruction operation until all fragmented surface nodes are merged into surface clusters to obtain a surface cluster adjacency graph.

[0062] In this embodiment of the application, by clearly defining the criteria for determining which faces are considered fragmented, a quantitative basis is provided to avoid subjective misjudgment. Furthermore, by repeating the operation until all fragmented faces are merged, the final face cluster adjacency graph is guaranteed to be free of redundant fragmented faces, thereby improving the efficiency and accuracy of subsequent subgraph matching.

[0063] In one example, during the CAD modeling of a part and its interaction with downstream systems, to meet the requirements of subsequent processing, assembly, and inspection processes for model accuracy and semantic information, the part surface often exhibits a cluster of fragmented surfaces formed by refining the original continuous surface. For example... Figure 5As shown, the selected area was originally a complete helical cylindrical surface on a blind hole with a tapered bottom thread. However, due to discretization processing, this area was ultimately divided into 13 independent facets after multiple format conversions or platform imports. This example addresses this type of fragmented surface by reconstructing it, merging it into a single facet cluster unit based on geometric continuity and topological adjacency, thereby constructing a concise and coherent facet cluster adjacency graph. In the machining feature recognition process based on attribute adjacency graphs, the presence of fragmented surfaces directly alters the original topological relationships, causing problems such as blurred feature boundaries and interrupted recognition paths, thus significantly reducing the accuracy and robustness of recognition. Therefore, effectively merging and restoring the structure of fragmented surfaces is a crucial prerequisite for improving the recognition effect of complex geometric features.

[0064] In the example above, the steps to reconstruct the adjacency graph of the face cluster are as follows: Figure 6 As shown, starting from any fragment node in the graph structure, by traversing its connections, all nodes connected to it through fragments are identified, forming a set of node-face clusters. Figure 6 Starting with node {2}, nodes {2, 3, 4} form a node-face cluster set, where solid lines represent normal connections and dashed lines represent fragmented connections. A representative node (e.g., node 4) is selected from this cluster set, retained, and the remaining nodes are deleted. Simultaneously, all topological relationships of these nodes are merged into the representative node. Figure 6 Select node {4} as the representative node, and inherit the adjacency relationships of other nodes in the set to node {4}; repeat the above process, selecting new fragment nodes as starting points in turn, identifying and merging all fragment cluster sets, as can be seen. Figure 6 The next identified face cluster set {5, 6, 7, 8} is used, and node {5} is selected as the representative node for merging. After multiple rounds of face cluster identification and merging, the adjacency structure of the original image, which consists of multiple fragmented face nodes, is simplified to an adjacency structure containing only the representative node. Figure 6 The eight nodes {1, 2, 3, 4, 5, 6, 7, 8} are reconstructed into three nodes {1, 4, 5}. Therefore, before performing the subgraph isomorphic search, the adjacency graph of the fragmented surface model and the fragmented surface feature subgraph are preprocessed into fragmented surfaces and reconstructed into a face cluster adjacency graph.

[0065] Step 105: Define the matching criteria between nodes and directed edges in the face cluster adjacency graph. Using each feature subgraph as the retrieval target, perform subgraph isomorphic search in the model face cluster adjacency graph. After the search is completed, collect the node mapping results and restore the original fragmented surfaces.

[0066] In this embodiment, the matching criteria between nodes and directed edges are defined to ensure that the subgraph search has clear standards. The subgraph isomorphic search is performed in the adjacency graph of the model face cluster with the feature subgraph as the retrieval target, so as to achieve accurate positioning from the overall model to the local threaded hole. At the same time, the matching relationship is recorded by collecting node mapping results, and the complete geometric information is preserved by restoring the original fragments, so as to avoid information loss due to fragment merging.

[0067] In some embodiments, the matching criteria for nodes and directed edges in the defined face cluster adjacency graph are used as the retrieval targets. Subgraph isomorphic search is performed in the model face cluster adjacency graph. After the search is completed, node mapping results are collected and the original fragmented surfaces are restored. This includes: the matching criteria are that the number of nodes in the retrieved subgraph is the same as that in the feature subgraph, the adjacency relationship between the nodes is consistent, and the face attributes of the nodes and the edge attributes of the directed edges in the subgraph match the face attributes of the corresponding nodes and the edge attributes of the corresponding directed edges in the feature subgraph, respectively; after performing the subgraph isomorphic search, the mapping relationship between the nodes in the feature subgraph and the nodes in the model face cluster adjacency graph is recorded as the node mapping result; based on the association relationship between the representative nodes in the node mapping result and the original fragmented surface nodes, the face IDs of all original fragmented surfaces are restored.

[0068] In this embodiment, the accuracy of subgraph matching is ensured from both structural and attribute perspectives by using a three-dimensional matching criterion of identical number of nodes, consistent adjacency relationship, and matching attributes, thus avoiding misidentification caused by similar structures but inconsistent attributes.

[0069] In one example, a subgraph with the same number of nodes as the feature subgraph needs to be selected from the global model adjacency graph, and the node adjacency relationships and face-edge attributes of the subgraph must completely match those of the feature subgraph. Using the model face cluster adjacency graph as the model graph G_model and the feature face cluster subgraph as the retrieval target G_template, a subgraph isomorphic search targeting G_template is performed in G_model according to the aforementioned isomorphism criteria. After the search is completed, the node mapping results are collected first, and then the original fragmented surfaces are restored based on the results. The specific information is as follows: The node mapping relationship of the first matching feature is {0:9, 2:3, 3:8, 1:7, 4:10, 5:4}, which yields 18 original fragmented surfaces after restoration, with surface IDs of {3, 4, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22}; The node mapping relationship of the second matching feature is {0:13, 2:5, 3:12, 1:11, 4:14, 5:6}, which yields 18 original fragmented surfaces after restoration, with surface IDs of {5, 6, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38}. Therefore, the 3D model contains two G_template-type processing features, each corresponding to six faces. However, since a face merging operation has already been performed, it is now necessary to restore the face fragments. The core purpose is to transform the geometric fragments into structural semantic units, providing a reliable geometric foundation for subsequent graph structure recognition, key attribute extraction, and ensuring recognition accuracy. The final visualization of the recognition result is shown below. Figure 7 As shown in the figure, the areas marked by arrows represent the identified target features.

[0070] Step 106: Based on the node mapping results and the restored fragmented surfaces, extract the key geometric attribute parameters of the threaded hole; the key geometric attribute parameters include nominal diameter, pitch, thread length and total depth.

[0071] In this embodiment, key parameters are extracted based on node mapping results and the restored fragmented surfaces, directly meeting the actual needs of industrial scenarios such as casing part processing, inspection, and assembly for thread hole specifications. At the same time, the key parameters are clearly defined to include nominal diameter, pitch, thread length, and total depth, ensuring that the extracted parameters cover industry requirements.

[0072] In some embodiments, the extraction of key geometric attribute parameters of the threaded hole based on the node mapping results and the restored fragmented surfaces includes: extracting a subset of cylindrical surfaces from the face cluster corresponding to the node mapping results, obtaining the radius of the cylindrical surfaces, multiplying the radius by two to obtain the nominal diameter of the threaded hole; uniformly sampling three-dimensional points on all surfaces of the restored fragmented surfaces, projecting the three-dimensional points along the main axis of the threaded hole, calculating the maximum distance between the projected points as the total depth of the threaded hole; uniformly sampling three-dimensional points on the boundary of the cylindrical helical surface in the restored fragmented surfaces, transforming the three-dimensional points to a local cylindrical coordinate system with the threaded hole axis as the reference, establishing a functional relationship between the axial coordinate z of the sampling points and the circumferential angle θ; obtaining the fitted line z=aθ+b through least squares linear fitting of the functional relationship, where a is the axial slope; calculating the thread pitch p=2πa of the threaded hole based on the axial slope a; obtaining the circumferential angle span Δθ of the sampling points, and calculating the thread length L=a·Δθ of the threaded hole based on the axial slope a and the angle span Δθ.

[0073] In some embodiments, the above-mentioned uniform sampling of three-dimensional points on all surfaces of the restored fragmented surface includes: sampling one three-dimensional point at 1° circumferential angle intervals on the cylindrical helical surface boundary of the threaded hole, with a total number of sampling points of not less than 128 for each thread tooth cycle; performing Gaussian filtering or smoothing filtering preprocessing on the sampled three-dimensional points to remove measurement noise before parameter calculation.

[0074] In this embodiment, instead of relying on the displayed thread modeling features, it is directly based on geometric surface fitting and topological information, which has good versatility and model adaptability, and is especially suitable for engineering models that have caused local degradation or fragmentation during data conversion.

[0075] Step 107: Output the identifiers of all threaded hole feature surfaces in the 3D geometric model and their corresponding key geometric attribute parameters.

[0076] In this embodiment, by outputting the threaded hole feature surface identifier and key geometric attribute parameters, complete information on the location and specifications is provided to the user, which meets the user's needs for threaded hole location and the needs for specification parameters in processing, inspection and other scenarios.

[0077] In some embodiments, the above-mentioned output of the identifiers of all threaded hole feature surfaces and their corresponding key geometric attribute parameters in the three-dimensional geometric model includes: repeatedly performing the steps of fragment preprocessing, subgraph isomorphic search, node mapping result collection and fragment restoration, and key geometric attribute parameter extraction for each threaded hole feature template in the preset threaded hole feature library; visually marking all threaded hole feature surfaces in the three-dimensional geometric model, and displaying the nominal diameter, pitch, thread length, and total depth corresponding to each threaded hole feature surface. Figure 8As shown, this represents the final recognition result and corresponding geometric attribute information.

[0078] In this embodiment, by repeatedly performing the preceding steps on each template, it is ensured that all types of threaded holes in the model can be identified, avoiding missed detections due to template omissions. At the same time, the location of the threaded holes in the 3D model can be intuitively seen by the visual markers, and the corresponding threaded hole specification information can be obtained by the associated display parameters without additional queries, greatly reducing the complexity of subsequent processing, inspection and other operations.

[0079] Next, to better understand the threaded hole feature identification method for casing parts provided in this application, a casing model containing three different threaded hole machining features is used as an application example to illustrate the implementation process of this method, such as... Figure 9 The 3D model shown contains 9464 faces and 22923 edges. The method includes: The first step is to extract the face and edge topology information from the chassis model and construct the model's adjacency graph. Due to the large amount of face and edge information, only the first 100 nodes are shown here for demonstration. The adjacency graph is as follows: Figure 10 As shown; The second step involves loading feature templates from a pre-defined threaded hole feature library, extracting their face-edge topological relationships, and constructing a feature sub-graph. The types of threaded hole feature libraries are as follows: Figure 11 As shown; The third step is to address the issue of fragmented surfaces that may occur during the format conversion of the 3D model by performing fragmented surface preprocessing on the model's adjacency graph and feature subgraph, and reconstructing it into a face cluster adjacency graph. The fourth step is to define the matching criteria between nodes and directed edges in the adjacency graph, take each feature face cluster subgraph as the retrieval target, perform subgraph isomorphic search in the model face cluster adjacency graph, collect the mapping results and restore the fragmented surfaces after the search is completed; Fifth, based on the above identification results, extract the key geometric attribute parameters of the threaded hole, including nominal diameter, pitch, thread length and total depth, etc. Step 6: Repeat steps 3 to 5 above, using each feature in the preset feature library as the retrieval target in turn to perform a subgraph isomorphic search, such as... Figure 12 As shown, clicking on the identified threaded hole features yields the corresponding geometric attribute information, and the final identification result is as follows. Figure 13 As shown, each type of feature is specifically displayed in Figures 14(a), 14(b), 14(c), and 14(d). Since one of the five types of features does not appear in this chassis model, it cannot be displayed.

[0080] Figure 15 The diagram illustrates a structural block diagram of a threaded hole feature recognition system for a casing part according to an exemplary embodiment of the present invention. The system includes: Model input module 1501 is used to input a three-dimensional geometric model of the casing that includes threaded hole features; The topology information extraction and adjacency graph construction module 1502 is used to extract the face and edge topology information of the 3D geometric model and construct the model adjacency graph; the face and edge topology information includes the unique identifier of the face, the unique identifier of the edge, and the connection relationship between the face and the edge; The feature template loading and subgraph construction module 1503 is used to load at least one threaded hole feature template from a preset threaded hole feature library, extract the face and edge topological relationship of the feature template, and construct the corresponding feature subgraph. The fragmented surface preprocessing and surface cluster adjacency graph generation module 1504 is used to perform fragmented surface preprocessing on the model adjacency graph and feature subgraph, and reconstruct the fragmented surfaces into surface cluster units based on geometric continuity and topological adjacency relationship to form a surface cluster adjacency graph. The subgraph isomorphic search and fragmented surface restoration module 1505 is used to define the matching criteria between nodes and directed edges in the face cluster adjacency graph. It uses each feature subgraph as the retrieval target, performs subgraph isomorphic search in the model face cluster adjacency graph, collects node mapping results and restores the original fragmented surface after the search is completed. The key geometric attribute parameter extraction module 1506 is used to extract the key geometric attribute parameters of the threaded hole based on the node mapping results and the restored fragments; the key geometric attribute parameters include nominal diameter, pitch, thread length and total depth. The recognition result output module 1507 is used to output the identifiers of all threaded hole feature surfaces in the three-dimensional geometric model and their corresponding key geometric attribute parameters.

[0081] In some embodiments, the topological information of the face edges of the three-dimensional geometric model is extracted to construct a model adjacency graph, including: Each face of the 3D geometric model is marked with a unique face ID. Each face ID corresponds to a node in the model's adjacency graph, and the node's ID is the same as the face ID. Obtain all edges of the 3D geometric model. Each edge connects two different faces, generating two directed edges accordingly. That is, the first directed edge from the node corresponding to the first face ID to the node corresponding to the second face ID, and the second directed edge from the node corresponding to the second face ID to the node corresponding to the first face ID. Add face attributes to the nodes of the model's adjacency graph. Face attributes include face area, normal vector, and face type. Add edge attributes to the directed edges of the model's adjacency graph. Edge attributes include edge length, concavity / convexity, inner / outer edge identifiers, and edge type.

[0082] In some embodiments, the preset threaded hole feature library includes five typical threaded hole feature templates, and the feature sub-map parameters corresponding to each template include: The blind hole template for tapered bottom threaded holes has a feature subgraph containing 12 nodes and 18 directed edges; The flat-bottomed threaded blind hole template has a feature subgraph containing 12 nodes and 18 directed edges; A template with a tapered bottom threaded hole and a small hole has a feature subgraph containing 14 nodes and 20 directed edges; A flat-bottomed threaded hole template with small holes has a feature subgraph containing 14 nodes and 22 directed edges. The threaded hole through-hole template has a feature subgraph containing 8 nodes and 8 directed edges.

[0083] In some embodiments, the model adjacency graph and feature subgraph undergo fragmentation preprocessing. Based on geometric continuity and topological adjacency relationships, the fragments are reconstructed into surface clusters to form a surface cluster adjacency graph, including: The criteria for determining the fragmented surfaces of adjacent faces in the adjacency graph and feature subgraph of the model are that the adjacent faces have the same face type and the connecting edge between the adjacent faces is a smooth edge. For fragmented nodes that meet the fragmented surface determination criteria, a reconstruction operation is performed. Starting from any fragmented node, all its connection relationships are traversed, and all nodes connected through fragmented surfaces are identified to form a set of node face clusters. A representative node is selected from the set of node face clusters, the representative node is retained and the remaining nodes in the set are deleted, and all topological relationships of the remaining nodes are merged into the representative node. Repeat the above reconstruction operation until all fragmented nodes are merged into face cluster elements, resulting in a face cluster adjacency graph.

[0084] In some embodiments, matching criteria between nodes and directed edges in the face cluster adjacency graph are defined. Using each feature subgraph as the retrieval target, a subgraph isomorphic search is performed in the model face cluster adjacency graph. After the search is completed, node mapping results are collected and the original fragmented faces are restored, including: The matching criteria are that the number of nodes in the retrieved subgraph is the same as that in the feature subgraph, the adjacency relationship between the nodes is consistent, and the face attribute of the node and the edge attribute of the directed edge in the subgraph match the face attribute of the corresponding node and the edge attribute of the corresponding directed edge in the feature subgraph, respectively. After performing the subgraph isomorphic search, the mapping relationship between the feature subgraph nodes and the model face cluster adjacency graph nodes is recorded as the node mapping result; Based on the association between representative nodes and original fragment nodes in the node mapping results, the face IDs of all original fragments are restored.

[0085] In some embodiments, based on the node mapping results and the restored fragments, key geometric attribute parameters of the threaded hole are extracted, including: Extract the subset of cylindrical surfaces in the face cluster corresponding to the node mapping result, obtain the radius of the cylindrical surface, multiply the radius by two, and obtain the nominal diameter of the threaded hole; Uniformly sample three-dimensional points on all surfaces of the restored fragmented surface, project the three-dimensional points along the main axis of the threaded hole, and calculate the maximum distance between the projected points as the total depth of the threaded hole. Three-dimensional points are uniformly sampled on the boundary of the cylindrical helical surface in the restored fragmented surface. The three-dimensional points are then transformed into a local cylindrical coordinate system with the threaded hole axis as the reference, and the functional relationship between the axial coordinate z of the sampling point and the circumferential angle θ is established. By fitting the function relationship using least squares linear fitting, the fitted line z = aθ + b is obtained, where a is the axial slope; The pitch of the threaded hole is calculated as p = 2πa based on the axial slope a. Obtain the circumferential angle span Δθ of the sampling point, and calculate the thread length L=a·Δθ of the threaded hole based on the axial slope a and the angle span Δθ.

[0086] In some embodiments, the identifiers of all threaded hole feature surfaces in the three-dimensional geometric model and their corresponding key geometric attribute parameters are output, including: For each thread hole feature template in the preset thread hole feature library, repeat the steps of fragment preprocessing, subgraph isomorphic search, node mapping result collection and fragment restoration, and key geometric attribute parameter extraction. All threaded hole feature faces are visualized and marked in the 3D geometric model, and the nominal diameter, pitch, thread length and total depth corresponding to each threaded hole feature face are displayed in association.

[0087] It should be noted that the thread hole feature recognition system for casing parts provided in the above embodiments is only an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the equipment can be divided into different functional modules to complete all or part of the functions described above.

[0088] Figure 16 This diagram illustrates a computer device for identifying threaded hole features of a casing part according to an exemplary embodiment of the present invention. The computer device includes: The processor 1601 includes one or more processing cores, and the processor 1601 executes various functional applications and data processing by running software programs and modules.

[0089] The receiver 1602 and transmitter 1603 can be implemented as a communication component, which can be a communication chip. Optionally, the communication component can include signal transmission functionality. That is, the transmitter 1603 can be used to transmit control signals to the image acquisition device and the scanning device, and the receiver 1602 can be used to receive corresponding feedback commands.

[0090] The memory 1604 is connected to the processor 1601 via the bus 1605.

[0091] The memory 1604 can be used to store at least one instruction, and the processor 1601 is used to execute the at least one instruction to implement the various steps in the above method embodiments.

[0092] This invention also provides a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set for loading and execution by a processor to implement the above-described method for identifying threaded hole features of a casing part.

[0093] The present invention also provides a computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the threaded hole feature identification method for a casing part as described in any of the above embodiments.

[0094] Optionally, the computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), solid-state drives (SSDs), or optical discs, etc. The random access memory may include resistive random access memory (ReRAM) and dynamic random access memory (DRAM). The sequence numbers of the above embodiments are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0095] It is understood that the specific examples in this document are only intended to help those skilled in the art better understand this disclosure, and are not intended to limit the scope of the invention.

[0096] It is understood that in the various embodiments of this specification, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of this disclosure.

[0097] It is understood that the various implementation methods described in this specification can be implemented individually or in combination, and this disclosure does not limit them.

[0098] Unless otherwise stated, all technical and scientific terms used in this disclosure have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of this specification. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items. The singular forms "a," "the," and "the" as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0099] It is understood that the processor disclosed herein can be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method implementation can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in this disclosure. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed herein can be directly implemented by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0100] It is understood that the memory in this disclosure can be volatile memory or non-volatile memory, or may include both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM). It should be noted that the memory in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0101] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this specification.

[0102] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the aforementioned method implementations, and will not be repeated here.

[0103] In the several embodiments provided in this specification, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments of the systems described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0104] 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 this embodiment, depending on actual needs.

[0105] In addition, the functional units in the various embodiments of this specification 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.

[0106] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of this specification, in essence, or the parts that contribute to the prior art, or parts of the technical solutions, can be embodied in the form of software products. These computer software products are stored in a storage medium and include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this specification. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0107] The above description is merely a specific embodiment of this specification, but the scope of protection of this invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this specification should be included within the scope of protection of this specification. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A method for identifying threaded hole features in a casing part, characterized in that, The method is applied to a computer device, and the method includes: Input a 3D geometric model of the casing that includes threaded hole features; Extract the face and edge topology information of the three-dimensional geometric model and construct the model adjacency graph; the face and edge topology information includes the unique identifier of the face, the unique identifier of the edge, and the connection relationship between the face and the edge; Load at least one threaded hole feature template from a preset threaded hole feature library, extract the face-edge topological relationship of the feature template, and construct the corresponding feature sub-graph; The model adjacency graph and the feature subgraph are preprocessed with fragmented surfaces. Based on geometric continuity and topological adjacency, the fragmented surfaces are reconstructed into surface cluster units to form a surface cluster adjacency graph. Define the matching criteria between nodes and directed edges in the adjacency graph of the face cluster, take each feature subgraph as the retrieval target, perform subgraph isomorphic search in the adjacency graph of the model face cluster, collect node mapping results and restore the original fragmented face after the search is completed; Based on the node mapping results and the restored fragmented surfaces, the key geometric attribute parameters of the threaded hole are extracted; the key geometric attribute parameters include nominal diameter, pitch, thread length and total depth. Output the identifiers of all threaded hole feature surfaces in the three-dimensional geometric model and their corresponding key geometric attribute parameters.

2. The method for identifying threaded hole features of a casing part according to claim 1, characterized in that, The step of extracting the face and edge topology information of the three-dimensional geometric model and constructing the model adjacency graph includes: Each face of the three-dimensional geometric model is marked with a unique face ID, and each face ID corresponds to a node in the adjacency graph of the model. The ID of the node is the same as the face ID. Obtain all edges of the three-dimensional geometric model. Each edge connects two different faces, generating two directed edges accordingly. That is, a first directed edge pointing from the node corresponding to the first face ID to the node corresponding to the second face ID, and a second directed edge pointing from the node corresponding to the second face ID to the node corresponding to the first face ID. Add face attributes to the nodes of the adjacency graph of the model. The face attributes include the area of ​​the face, the normal vector, and the face type. Add edge attributes to the directed edges of the adjacency graph of the model. The edge attributes include edge length, concavity / convexity, inner / outer edge identifiers, and edge type.

3. The method for identifying threaded hole features of a casing part according to claim 1, characterized in that, The preset threaded hole feature library includes five typical threaded hole feature templates, and the feature sub-graph parameters corresponding to each template include: The blind hole template for tapered bottom threaded holes has a feature subgraph containing 12 nodes and 18 directed edges; The flat-bottomed threaded blind hole template has a feature subgraph containing 12 nodes and 18 directed edges; A template with a tapered bottom threaded hole and a small hole has a feature subgraph containing 14 nodes and 20 directed edges; A flat-bottomed threaded hole template with small holes has a feature subgraph containing 14 nodes and 22 directed edges. The threaded hole through-hole template has a feature subgraph containing 8 nodes and 8 directed edges.

4. The method for identifying threaded hole features of a casing part according to claim 1, characterized in that, The step of performing fragmented surface preprocessing on the model adjacency graph and the feature subgraph, and reconstructing the fragmented surfaces into surface clusters based on geometric continuity and topological adjacency relationships to form a surface cluster adjacency graph includes: The criteria for determining the fragmented surfaces of adjacent faces in the model's adjacency graph and the feature subgraph are: the adjacent faces have the same face type and the connecting edge between adjacent faces is a smooth edge. For fragmented nodes that meet the fragmented surface determination conditions, a reconstruction operation is performed. Starting from any fragmented node, all its connection relationships are traversed, and all nodes connected through fragmented surfaces are identified to form a set of node face clusters. A representative node is selected from the set of node face clusters, the representative node is retained and the remaining nodes in the set are deleted, and all topological relationships of the remaining nodes are merged into the representative node. Repeat the above reconstruction operation until all fragmented nodes are merged into face clusters, thus obtaining the face cluster adjacency graph.

5. The method for identifying threaded hole features of a casing part according to claim 1, characterized in that, The definition of the matching criteria between nodes and directed edges in the face cluster adjacency graph, using each of the feature subgraphs as the retrieval target, performs a subgraph isomorphic search in the model face cluster adjacency graph, and after the search is completed, collects the node mapping results and restores the original fragmented faces, including: The matching criteria are that the number of nodes in the retrieved subgraph is the same as that in the feature subgraph, the adjacency relationship between the nodes is consistent, and the face attribute of the node and the edge attribute of the directed edge in the subgraph match the face attribute of the corresponding node and the edge attribute of the corresponding directed edge in the feature subgraph, respectively. After performing the subgraph isomorphic search, the mapping relationship between the feature subgraph nodes and the model face cluster adjacency graph nodes is recorded as the node mapping result; Based on the association between representative nodes and original fragment nodes in the node mapping results, the face IDs of all original fragments are restored.

6. The method for identifying threaded hole features of a casing part according to claim 1, characterized in that, Based on the node mapping results and the restored fragmented surfaces, the key geometric attribute parameters of the threaded hole are extracted, including: Extract the subset of cylindrical surfaces from the face cluster corresponding to the node mapping result, obtain the radius of the cylindrical surface, and multiply the radius by two to obtain the nominal diameter of the threaded hole; Three-dimensional points are uniformly sampled on all surfaces of the restored fragmented surface. The three-dimensional points are projected along the main axis of the threaded hole, and the maximum distance between the projected points is calculated as the total depth of the threaded hole. Three-dimensional points are uniformly sampled on the boundary of the cylindrical helical surface in the restored fragmented surface. The three-dimensional points are then transformed into a local cylindrical coordinate system with the threaded hole axis as the reference, and the functional relationship between the axial coordinate z of the sampling point and the circumferential angle θ is established. By fitting the function relationship with least squares linearly, the fitted line z = aθ + b is obtained, where a is the axial slope; The pitch of the threaded hole is calculated as p = 2πa based on the axial slope a. Obtain the circumferential angle span Δθ of the sampling point, and calculate the thread length L=a·Δθ of the threaded hole based on the axial slope a and the angle span Δθ.

7. The method for identifying threaded hole features of a casing part according to claim 1, characterized in that, The output includes the identifiers of all threaded hole feature surfaces in the three-dimensional geometric model and their corresponding key geometric attribute parameters, including: For each thread hole feature template in the preset thread hole feature library, repeat the steps of fragment preprocessing, subgraph isomorphic search, node mapping result collection and fragment restoration, and key geometric attribute parameter extraction. All threaded hole feature faces are visualized and marked in the three-dimensional geometric model, and the nominal diameter, pitch, thread length and total depth corresponding to each threaded hole feature face are displayed in association.

8. A threaded hole feature recognition system for a casing part, characterized in that, The system includes: The model input module is used to input a three-dimensional geometric model of the casing, including the threaded hole feature; The topology information extraction and adjacency graph construction module is used to extract the face and edge topology information of the three-dimensional geometric model and construct the model adjacency graph; the face and edge topology information includes the unique identifier of the face, the unique identifier of the edge, and the connection relationship between the face and the edge; The feature template loading and subgraph construction module is used to load at least one threaded hole feature template from a preset threaded hole feature library, extract the face-edge topological relationship of the feature template, and construct the corresponding feature subgraph. The fragmented surface preprocessing and surface cluster adjacency graph generation module is used to perform fragmented surface preprocessing on the model adjacency graph and the feature subgraph, and reconstruct the fragmented surfaces into surface cluster units according to geometric continuity and topological adjacency relationship to form a surface cluster adjacency graph. The subgraph isomorphic search and fragmented surface restoration module is used to define the matching criteria between nodes and directed edges in the adjacency graph of the face cluster, take each feature subgraph as the retrieval target, perform subgraph isomorphic search in the adjacency graph of the model face cluster, collect node mapping results and restore the original fragmented surface after the search is completed; The key geometric attribute parameter extraction module is used to extract the key geometric attribute parameters of the threaded hole based on the node mapping result and the restored fragmented surface; the key geometric attribute parameters include nominal diameter, pitch, thread length and total depth. The identification result output module is used to output the identifiers of all threaded hole feature surfaces in the three-dimensional geometric model and their corresponding key geometric attribute parameters.

9. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing at least one instruction, at least one program, code set, or instruction set, the at least one instruction, at least one program, code set, or instruction set being loaded and executed by the processor to implement the threaded hole feature identification method for casing parts as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The readable storage medium stores at least one instruction, at least one program, code set, or instruction set, which is loaded and executed by the processor to implement the threaded hole feature identification method for casing parts as described in any one of claims 1 to 7.

Images