76 results about "Boundary representation" patented technology

The invention relates to a medical image segmentation method and system based on residual Mama and multi-scale boundary enhancement. The method comprises the following steps: acquiring and preprocessing a medical image; inputting the image into a segmentation model based on an encoder-decoder architecture; the encoder synchronously extracts local texture features and models long-range spatial dependence through residual error convolution blocks and residual error Mama blocks which are alternately connected; fusing and enhancing the jump connection features between the encoder and the decoder through a boundary enhancement module to optimize boundary characterization; integrating a multi-scale gating attention module in a decoding path, and adaptively selecting and fusing multi-scale context features; and finally outputting the high-precision segmentation mask. The method effectively solves the problems that in the prior art, long-range dependence and local details are difficult to consider, the multi-scale feature fusion capability is insufficient, boundary segmentation is fuzzy and the like, and the segmentation accuracy, the boundary continuity and the clinical practicability are remarkably improved.

The invention discloses a three-dimensional physical field real-time prediction method and system based on geometric deep learning and physical constraint, and relates to the technical field of computer-aided engineering and artificial intelligence. The method comprises the following steps: directly extracting native boundary representation data (B-Rep) of a three-dimensional model from a computer aided design system; constructing a heterogeneous dual graph taking a parameterized curved surface as a graph node, skipping finite element grid division, and aggregating local and global topological features by using a graph neural network; in combination with a physical information driving mechanism, a partial differential equation (PDE) residual error is introduced as a loss function for constraint training, and generalization prediction of a novel geometric structure is realized; and finally, the physical field state quantity is predicted through direct regression and is rendered in real time. An incremental reasoning mechanism based on a local topology subgraph is adopted, millisecond-level physical field real-time feedback under design modification is achieved, and the method is suitable for scheme rapid screening and trend prediction in the initial stage of design.

The invention discloses a three-dimensional CAD model feature recognition method. The method comprises the following steps: analyzing boundary representation data of a CAD model and constructing a topological graph structure; the feature complexity is quantified by calculating a topological entropy value; in combination with a predefined process semantic rule base and a graph neural network model, realizing association identification of geometric features and processing semantics; and outputting the parameterized feature tree and supporting CAM system integration. The technical problems that a traditional feature recognition method is poor in adaptability to a complex topological structure and lacks process semantic association are solved, and the recognition accuracy and the automation degree are remarkably improved.

The invention relates to the field of mechanical design and manufacturing, and provides a three-dimensional model processing method, a training and application method of a system, equipment and a medium, and a target boundary representation three-dimensional model is obtained based on a preset complex manufacturing feature model and a boundary representation three-dimensional model. The complex manufacturing feature model is added to the boundary representation three-dimensional model containing the simple structure manufacturing feature model, so that the boundary representation three-dimensional model containing the simple structure manufacturing feature model and the complex structure manufacturing feature model is synthesized. According to the three-dimensional model processing scheme provided by the invention, the complex manufacturing feature model is added in the boundary representation three-dimensional model, so that when the boundary representation three-dimensional model obtained by the processing scheme is used for training an image classification system, the image classification efficiency is improved; the generalization ability of the image classification system and the classification accuracy of the manufacturing feature model in the boundary representation three-dimensional model can be improved, so that the production efficiency and quality of process products are improved.

The invention discloses a three-dimensional geologic model construction method, a three-dimensional geologic model construction device, three-dimensional geologic model construction equipment and a computer readable storage medium, and the method comprises the steps: carrying out the elevation data fusion processing of collected high-resolution topographic data, and obtaining a digital elevation model; wherein the high-resolution topographic data comprises contour line data, remote sensing image data and field actual measurement elevation point data; performing interpolation processing on the digital elevation model based on an irregular triangulation network interpolation algorithm of geological constraints and interpolation weight coefficients in different directions to obtain stratigraphic interface data; wherein the geological constraints comprise sedimentary facies constraints, structural constraints and trend constraints; performing three-dimensional sealing processing on the stratum interface data based on a boundary representation method and the constructed lithologic attribute database to obtain a three-dimensional entity model; wherein the lithologic attribute database is constructed based on the spatial position information of the drill hole. The modeling precision of a complex structure is improved.

The invention relates to the technical field of computer graphics, and provides a model data adaptation processing method for a graphics rendering engine, which comprises the following steps of: importing and analyzing model files from different geometric modeling engines, and extracting original geometric data, original topological data and auxiliary information in the model files; converting the original geometric data into a unified standard geometric data structure for representation and storage; converting the original topological data into a standard topological data structure based on a boundary representation method for representation and storage; creating, analyzing or setting spatial position and direction information of the model entity and the virtual camera in a world coordinate system; analyzing material attributes, surface textures and animation information in the affiliated information of the model file, and packaging the materials, the surface textures and the animation information into corresponding standard data structures; and combining the standardized geometric data, topological data, coordinate system information and affiliated information into a complete standardized model data packet which can be directly used by a graphic rendering engine.

The invention discloses a boundary representation generation method based on graph diffusion. The method comprises the following steps: step 1, constructing an industrial part data set and preprocessing the industrial part data set; step 2, constructing and training a Brep-GD model based on the preprocessed self-built data set and the two open source data sets; the Brep-GD model comprises a variational auto-encoder, a graph diffusion model, a continuous topology decoupling model and a post-processing module; 3, reasoning and evaluating the trained Brep-GD model by using a test set of a data set so as to verify multiple types of indexes of distribution measurement, CAD measurement and efficiency measurement of the model; and step 4, applying the evaluated Brep-GD model to generate a simplified B-rep model, the method can significantly reduce redundancy calculation, improve generation efficiency, and ensure that the generated B-rep model has higher geometric accuracy and topology effectiveness.

The invention discloses a geometric topology fusion-based part assembly method, which comprises the following steps of: acquiring boundary representation models of at least two to-be-assembled CAD parts, and converting the boundary representation model of each CAD part into a structured heterogeneous geometric topology graph; performing feature coding on each node in the heterogeneous geometric topological graph to form node feature representation; inputting the node feature representation into a graph attention reasoning module, and obtaining a node embedding representation representing a geometrical relationship and a topological relationship in the CAD part through multi-layer residual graph structure information propagation and feature aggregation; based on the node embedding representations of the different parts, calculating association scores of node pairs among the different parts; and performing supervised training on the feature coding module and the graph attention reasoning module by using the labeled assembly constraint sample data. According to the method, fine geometric features can be extracted from a boundary representation B-Rep model, and a complex topological dependency relationship is inferred, so that robust assembly constraint inference is realized.

The invention discloses a multi-modal information guided CAD model automatic assembly method based on a boundary representation method, and belongs to the technical field of computer aided design. The method comprises the following steps: constructing a B-Rep graph of a part to form a joint connection graph; extracting vertex features by using MLP and MPN, and extracting image features through ResNet; carrying out edge convolution analysis on the joint connection diagram after multi-modal features are fused, and predicting a potential joint axis position; and finally, optimizing joint attitude parameters by adopting a neural guidance search algorithm to realize automatic assembly. According to the method, by fusing geometric and image multi-modal information, the problem of single data modal in the prior art is solved, and the assembly precision is remarkably improved; b-Rep representation is combined with neural network guide search, so that the calculation complexity is reduced; the designed feature extraction and fusion mechanism enhances the adaptability of the method to different assembly scenes, is particularly suitable for automatic assembly of complex part models, and can be widely applied to the fields of machine manufacturing, aerospace and the like.

The invention provides a MaskRcnn-based tumor detection method and system, and the method comprises the steps: inputting medical image data into a MaskRcnn network integrated with an edge perception module, generating an edge response graph through a Sobel operator, and enabling the edge response graph to serve as an additional channel injection feature graph, and enhancing the boundary representation; fusing multi-scale features through a feature pyramid network FPN, extracting candidate tumor area features through RoI Align, introducing non-local attention modeling global dependence in the area, extracting directional entropy, texture energy, uniformity and contrast in combination with a gray level co-occurrence matrix GLCM, and splicing to generate structure sensing features; performing classification, bounding box regression and mask segmentation based on the structure perception features, and constructing a joint loss function including classification, regression, mask cross entropy, edge alignment and structure consistency; the positioning capability of the model on the fuzzy tumor contour is effectively improved, the boundary positioning error is remarkably optimized, and the defect that over-segmentation or missing detection is likely to occur in a traditional model is overcome.

The application discloses a point cloud analysis and reconstruction method and system based on a cloud CAD platform, belongs to the technical field of point cloud reconstruction, and is used for solving the technical problems that the existing point cloud data reconstruction mode depends on manual operation, is low in efficiency, poor in precision and consistency, and difficult to adapt to complex scenes and automation requirements. The method comprises the following steps: receiving a point cloud data file and determining a processing mode; determining a corresponding irregular sampling region according to the processing mode, and performing signed distance field prediction on the point cloud data file in the irregular sampling region to obtain unsigned distance field and gradient information; generating a sliding window, identifying local boundary information in the sliding window through a boundary prediction neural network, and performing splicing to obtain global boundary information; dividing a face region of the point cloud data file according to the global boundary information to obtain different types of faces; performing targeted fitting processing on each type of face, and solving a boundary representation relationship between the faces to reconstruct a CAD model.

The application discloses a method for intelligent component identification and segmentation of an aircraft CAD model based on three-mode fusion, comprising: obtaining an aircraft CAD model with a boundary representation structure; constructing a multi-modal training data set that fuses surface images, geometric parameters and component semantic labels; training a multi-modal identification model; using the model for preliminary identification; for low confidence or small area surfaces, starting a geometric rule auxiliary module based on geometric topology and parameterization rules for semantic completion and correction to form a final segmentation result; automatically extracting key geometric feature lines; and converting and outputting into a structured engineering semantic format that can be used by grid generation software. The application realizes full automation of component identification, solves the problems of low efficiency and consistency caused by manual dependence, improves the identification integrity and engineering practicability of complex geometry, and the output result can directly drive subsequent grid generation, thereby significantly improving the automation level of the digital simulation pre-processing process.

The application relates to the technical field of computer-aided engineering and computer-aided design data integration, and discloses a Catia data conversion method and system for CAE preprocessing modeling software, which comprises the following steps: loading a Catia model and traversing a product structure tree to extract parameterized semantic information; converting geometric data into a boundary representation format and detecting defects; automatically repairing serious topological defects by using a deep learning model; constructing a hierarchical intermediate format based on a general digital thread protocol, and associating geometry, semantics and analysis attributes; dynamically adjusting model precision according to simulation types; and exporting a format suitable for CAE software and generating a parameter mapping table. The application can realize lossless conversion of a Catia model to a CAE environment, retain design semantics and parameter association, support precision self-adaptive adjustment of multi-physical field analysis, and significantly improve CAE preprocessing efficiency.

The application provides a symmetric workpiece disorderly grabbing method and system based on hierarchical reasoning, relates to the technical field of industrial robots, and comprises the following steps: acquiring a multi-modal image of a stacking scene, and reconstructing a complete physical mask representation of each workpiece from the multi-modal image. Through multi-modal perception fusion and full-view geometric reconstruction processing, the application obtains complete physical boundary representation and unbalanced load feature data of each workpiece in a shielding state, and on this basis, a directed hierarchical topological model describing the physical pressing and stacking relationship between workpieces is constructed through spatial gradient analysis, a candidate workpiece in an absolute top layer in the current scene is screened out, and then, a grabbability quantitative evaluation is performed to determine a target to be grabbed, so that the robot can identify and eliminate the bottom shielding workpiece which is easy to cause the collapse of the material pile in advance based on the real physical support relationship before action execution, and interference and instability caused by blind grabbing are avoided from the root.

The invention relates to the technical field of computer-aided engineering and computer-aided design data integration, and discloses a Catia data conversion method and system for CAE pretreatment modeling software, and the method comprises the steps: loading a Catia model, and traversing a product structure tree to extract parameterized semantic information; converting the geometric data into a boundary representation format and detecting defects; a deep learning model is used for automatically repairing serious topology defects; constructing a hierarchical intermediate format based on a general digital thread protocol, and associating geometric, semantic and analysis attributes; dynamically adjusting the model precision according to the simulation type; and exporting a format adaptive to the CAE software and generating a parameter mapping table. According to the method, lossless conversion from the Catia model to the CAE environment can be realized, design semantics and parameter association are reserved, precision self-adaptive adjustment of multi-physics field analysis is supported, and the CAE pretreatment efficiency is remarkably improved.

The application discloses a high-dimensional data computing architecture, method, system and device based on recursive projection compression and closed audit. After the method obtains high-dimensional input data, at least one recursive projection compression is performed to form an intermediate audit state representation and / or a decision state representation; a state space deviation function and a state evolution trend prediction operator are calculated on the state representation; dimension hedging processing is performed between different recursive levels, different projection dimensions, different loop representations or different boundary representations, and combined with boundary legality audit, resource closed audit, capacity constraint audit, loop consistency audit, platform flatness audit and stability window audit, a control amount, a compensation amount, a reconstruction amount or a branch adjustment amount is generated; a deterministic early termination is triggered through consistency comparison of parallel audit loops, and in the preferred embodiment, the integrity of the compression result is verified through reverse checking. The application is suitable for industrial monitoring, omics data analysis and other high-dimensional complex system state audit scenarios.

The application belongs to the technical field of artificial intelligence and digital image processing, and discloses an extensible farmland boundary regularization simplification method, system, device and terminal, determines a farmland category list to be segmented; uses an existing labeled training segmentation model to obtain segmentation masks of all farmlands; performs binaryzation processing on the segmentation masks to obtain binaryzation images; performs digital image processing on the binaryzation images to extract the contours of each connected domain; uses the contour information obtained through image processing to perform polygon approximation and remove redundant points to obtain farmland boundary representation. The processing method for extracting local farmland regular boundary uses the farmland segmentation result obtained through deep learning segmentation, uses a traditional image processing method and is supplemented with a polygon approximation algorithm to perform regularization processing on the result, and simultaneously simplifies the contour points of the farmland, so that the number of farmland boundary points and boundary description accuracy can be better balanced while reducing labor cost.

The invention belongs to the field of boundary representation model transition feature simplification, and relates to a B-Rep model simplification method based on a fillet feature network. According to the method, based on the fillet radius of the fillet feature, accurate construction and reasonable classification of the fillet feature network are realized by combining the concavity and convexity of the fillet feature and whether the fillet feature is degraded or not, and the fillet feature network comprises equal-radius and non-equal-radius fillet feature networks. The equal-radius fillet feature network represents fillet surfaces possibly created in the same rounding operation, the variable-radius fillet feature network is an adjacent complex fillet surface set and can be inhibited through extension intersection of supporting surfaces, and the remaining fillet surfaces are inhibited through additional Boolean union and Boolean subtraction operations. And finally, high-precision identification and robust suppression of the fillet features are realized. According to the method, the problem that the design is limited due to insufficient generalization ability and robustness of fillet feature identification on a multi-scale heterogeneous model during the optimization design of the physical field characteristics of the existing product is solved.

This invention discloses a method and system for constructing a data model for a business intelligence agent based on deep learning, relating to the field of deep learning technology. Through continuous processing from steps S1 to S4, the business intelligence agent no longer trains its model solely based on the final adopted decision results. Instead, it incorporates candidate information explicitly rejected during historical decision-making processes into the modeling scope. It retains excluded candidate decisions by constructing a set of rejected information (Una), and transforms the original rejected information into computable multi-dimensional structured data by generating a set of rejected features (Ufc). Then, based on the density of rejected information (Ind), it characterizes the concentration of rejection behavior at different decision stages, forming a rejected boundary representation (Nbd) which is applied to the business intelligence agent. When generating new candidate decisions, it can proactively identify and avoid historically high rejection density regions. Compared to existing methods that rely solely on input and output results for training, this approach overcomes the problem of insufficient learning of decision boundaries.

The invention discloses a lightweight conversion method and system for a CAD (computer-aided design) model. According to the method, a standard STEP format model is converted into a lightweight PBZ format through a multi-stage conversion architecture, and the method comprises the following steps: analyzing a STEP file to extract boundary representation geometry, a topological structure and product assembly data, discretizing an NURBS curved surface into a triangular mesh according to configurable precision, and meanwhile, nondestructively retaining parameterized sideline geometric definition; organizing the extracted data into a structured JSON (JavaScript Object Notation) intermediate representation; carrying out serialization by utilizing an optimized protobuf message structure, applying differential coding to vertex data, and applying triangular stripe coding to a patch index; and finally, based on the file size and the application scene precision requirement, adaptively selecting a GZIP compression level to carry out compression packaging. According to the method, the submicron geometric accuracy is kept, meanwhile, the high compression rate is achieved, the network transmission and mobile terminal rendering efficiency is remarkably improved, and the method is suitable for industrial scenes such as online collaborative design, digital twinning and mobile terminal detection.

The present disclosure provides a system (100) and a method (200) for generating routes for flexible components that connect principal components in a predefined environment. The method (200) comprises converting (202) a boundary representation of the principal components into a geometric representation and defining (204) an enclosed space around the principal components based on user-defined parameters. The method (200) comprises discretizing (206) the enclosed space into a plurality of nodes, assigning (208) one or more parameters to a node to construct a spatial structure within the enclosed space, and identifying (210) neighboring nodes among the plurality of nodes connected to the enclosed space within the spatial structure in order to generate one or more connectors to the one or more neighboring nodes.Furthermore, the procedure (200) includes determining (212) a shortest path by identifying start and end points for each connector and generating (214) routes for the flexible component that connects the main components.

The invention relates to an annotation method for annotating a training data set (LD) for training an environment recognition system (3).The training dataset (LD) comprises environment perception data (4), which is a sensory representation of a vehicle environment containing at least one physical linear road surface boundary (Lim), and an ordered data structure for the computer-readable representation of a physical linear road surface boundary (Lim) in the environment of a vehicle as a virtual linear road surface boundary (10), wherein the ordered data structure (1) defines at least one line object with a list of support points (P1, P2, P3), wherein the line object is assigned a semantic line property (5) that identifies the line object as a whole as a virtual linear road surface boundary, and wherein at least one of the support points (P1, P2, P3) can be assigned a separate semantic point property (A, B, C) by the annotation procedure. The annotation procedure comprises the steps: a.a. Displaying a file of the environment capture data (4) in a visually recognizable representation on a display means (11), in particular on a monitor (11); b. Displaying the at least one line object with the list of vertices (P1, P2, P3), wherein the vertices P1, P2, P3 of the list of vertices (P1, P2, P3) are displayed in a visually recognizable representation overlaid on the environment capture file (4) on the display means (11); c. Capturing a selection of a first vertex P1 from the list of vertices (P1, P2, P3), wherein the first selected vertex P1 is selected with an area segment selection means (12), in particular with a cursor (12), on the display means (11); d. Capturing an assignment of a separate semantic point property (A, B, C) to the first selected vertex P1.

This invention provides a computer implementation method, system, and program for machine learning to assemble mechanical parts based on mating scores and mating axes. [Solution] The method includes the step of providing a dataset of pairs of B-Reps, each containing at least one boundary representation (B-Rep) representing an assembly of mechanical parts. The pairs are labeled with mating compatibility data and, if the B-Reps of the pair are compatible according to the mating compatibility data, with mating axis compatibility data. The method also includes the step of training a neural network on the dataset. The neural network outputs mating scores for pairs of single embeddings, where each single embedding represents a B-Rep, and mating scores representing a score of mating compatibility between the mechanical parts represented by the pair, and data defining the mating axis if the B-Reps are compatible according to the mating score.

Embodiments of the present disclosure provide a method for detecting a virtual effect mounting plane, an electronic device, and a storage medium, and the method includes: acquiring contour line segments, and obtaining corresponding edge corner points based on the contour line segments, in which the contour line segments represent a contour of an object in an image to be detected, and the edge corner points are intersection points between the contour line segments; generating at least two quadrilateral structural borders with the edge corner points, in which each of the structural borders represents a contour of a plane of an object in the image to be detected; and obtaining matching degrees of the structural borders through target vanishing points corresponding to the structural borders, and determining a target border based on the matching degrees.

The invention relates to a kidney tumor image segmentation method and system based on multi-scale feature extraction, and the method comprises the steps: extracting the multi-scale features of a kidney tumor image to be processed, dividing feature similarity groups, carrying out the optimization through combining the texture and shape, and obtaining similar feature groups; generating a complementary information flow, adaptively enhancing an interaction weight between groups by using an attention mechanism, and obtaining an enhanced complementary feature representation; evaluating and denoising noise, establishing association mapping among groups based on features after denoising, and performing multi-scale feature fusion and boundary refining by adopting a U-Net network to obtain tumor boundary representation; and the boundary fuzzy degree is evaluated, iterative optimization is carried out by adopting a graph convolutional network, optimized collaborative hierarchical representation is obtained, and finally a high-precision kidney tumor segmentation result and a pixel-level contour are output. According to the method, through multi-scale feature cooperation, attention interaction and boundary optimization, the problems of boundary blur and noise sensitivity in kidney tumor segmentation are effectively solved, and the accuracy and robustness of segmentation are remarkably improved.

The present application particularly relates to a computer-implemented method of machine learning for assembling mechanical parts based on fit scores and fit axes. The method comprises providing a dataset of B-Rep pairs, each pair comprising at least one B-Rep representing an assembly of the mechanical part, which is labeled with mating compatibility data, and if the B-Rep in the pair of boundary representations is compatible according to the mating compatibility data, labeled with mating axis compatibility data. The method further includes training a neural network based on the data set, the neural network configured to take a pair of B-Rps pairs as inputs. The neural network also outputs a fit score for a pair of single embedded vectors, each single embedded vector representing one B-Rep, the fit score representing a score of fit compatibility between the mechanical parts represented by the pair of B-Rps, and outputs data defining a fit axis if the B-Rps are compatible according to the fit score.

The present invention discloses a feature recognition method based on B-rep steam turbine blades. By extracting attributes from a B-rep (boundary representation) steam turbine blade model, different attributes such as faces, edges, and connection modes are assigned identifiers to construct an extended attribute adjacency graph of the blade. The extended attribute adjacency graph is then trimmed by suppressing invalid features. The VF2p algorithm is then used to calculate and sort the priority access nodes of the target feature subgraph of the knowledge base, and the target feature subgraph is reconstructed into a forest structure according to the attribute identifiers. Finally, a pre-explored extended attribute adjacency graph search is performed using a depth-first search method to identify the features and types of the blades. The present invention realizes the recognition of steam turbine blade features with efficient recognition speed, laying the foundation for the comprehensive digitalization of the steam turbine blade manufacturing field.

The invention is suitable for the technical field of part labeling, and provides an automatic part labeling method which comprises the following steps: analyzing an stp file through Open CASCADE, and extracting B-rep boundary representation; constructing a graph structure: regarding the surface of the part as a node of the graph, extracting the type, direction, area and other business information of the surface as node features, regarding the edge of the part as the edge of the graph, extracting the type of the edge of the part and an included angle between two connecting surfaces of the edge as edge features, and performing normalization processing; according to the automatic part labeling method, the corresponding labels are established between the faces with the reserved relation, the attention mechanism is introduced, the compactness between the faces is automatically judged, the labeling efficiency is improved, the labeling cost is reduced, for different labeling styles of different scenes, good robustness and expansibility are achieved, the algorithm design is simple, the calculation overhead is small, and the method is suitable for popularization and application. And integration into a lightweight CAD plug-in or script system is facilitated.

The application discloses a kind of three-dimensional boundary representation model blanking line generation method and system and electronic equipment, its characteristics are that the method obtains three-dimensional boundary representation model and viewpoint parameter, and the entity in model is discretized into surface grid unit;Surface grid is classified based on the direction of viewpoint, and forward surface is screened out towards the viewpoint;The projection transformation of edge in model is carried out, and the visibility interval of edge is determined by the projection coverage detection of forward surface to edge, and blanking line segment is generated after interval merging and post-processing;System includes: parameter configuration module, entity gridding module, face orientation classification module, occlusion detection module, interval merging module, line segment generation module and output management module.Compared with prior art, the application has the advantages that blanking line segment conforming to engineering drawing specification can be generated from three-dimensional CAD model, the problems of high coupling degree of blanking calculation and low efficiency of occlusion detection are solved, and the application is suitable for three-dimensional computer-aided design, engineering drawing and three-dimensional visualization interaction scenes.

This application provides a complex structure modeling system and method based on 3D surface skeleton extraction. The system acquires initial 3D point cloud data of pipelines and converts it into a 3D voxel mesh. A symbolic distance field is generated based on the pipeline voxel mesh. Local extremum traversal is performed on the symbolic distance field to obtain a discrete skeleton candidate point set. Topological homology extraction is performed based on the discrete skeleton candidate point set to generate an initial pipeline topological skeleton. Semantic topological classification is performed using the pipeline voxel mesh and the initial pipeline topological skeleton to obtain structural attribute labels for each skeleton segment. Local topological optimization is performed on the initial pipeline topological skeleton based on the structural attribute labels of all skeleton segments to obtain an optimized pipeline topological skeleton. A parameterized boundary representation model of the complex pipeline network structure is generated using the optimized pipeline topological skeleton. The technical solution provided in this application can effectively construct the topological skeleton of complex pipeline network structures, thereby improving the modeling accuracy of parametric modeling.