A lightweight method for BIM models with semantic constraints

Abstract

The present application discloses a lightweight method for a BIM model with semantic constraints, which relates to the technical field of data processing. The method includes: constructing sub-high-precision models for each component according to the geometric information of each component in the BIM model to be processed; obtaining the semantic attribute information of the component, mapping the semantic attribute information into the sub-high-precision model, and then generating a high-precision model according to the sub-high-precision model; filtering non-main structure components in the high-precision model according to the semantic attribute information to obtain a low-precision model. The present application can perform lightweight processing on the model by means of semantic filtering.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular, to a lightweight method for BIM models with semantic constraints. Background Art

[0002] The LOD (Levels of Detail) technology can dynamically adjust the precision of the model according to factors such as the viewing distance and screen resolution. When on a low-resolution screen or the viewing point is far from the model, a low-precision model is displayed; while on a high-resolution screen or the viewing point is close to the model, a high-precision model and texture are loaded. By constructing LOD models at different levels, the data processing volume can be reduced as much as possible while ensuring the model quality, and the model loading speed can be accelerated.

[0003] When performing lightweight processing on a BIM model, different levels of LOD models are generally constructed through size filtering. When the set filtering threshold is too high, a large amount of redundant information is included in the retained data, resulting in a large data volume. If the set filtering threshold is too low, although some smaller redundant data can be removed, some important but slightly larger-sized data is misfiltered, thus affecting the integrity and accuracy of the data.

[0004] The above content is only used to assist in understanding the technical solution of the present application, and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of the present application is to provide a lightweight method for BIM models with semantic constraints, aiming to solve the technical problem of how to perform lightweight processing on BIM models.

[0006] To achieve the above object, the present application proposes a lightweight method for BIM models with semantic constraints, and the method includes:

[0007] Construct sub-high-precision models for each component according to the geometric information of each component in the BIM model to be processed;

[0008] Obtain the semantic attribute information of the component, and after mapping the semantic attribute information into the sub-high-precision model, generate a high-precision model according to the sub-high-precision model;

[0009] Filter non-main structure components in the high-precision model according to the semantic attribute information to obtain a low-precision model.

[0010] In an embodiment, the method further includes:

[0011] When receiving a model switching instruction, parse the model switching instruction to obtain the current model information and the target model information;

[0012] Determine the switching method according to the current model information and the target model information.

[0013] In one embodiment, the step of determining the switching method according to the current model information and the target model information includes:

[0014] When switching from the symbol model to the volumetric model, or from the volumetric model to the low-precision model, the switching method is the replacement method;

[0015] When switching from the low-precision model to the high-precision model, the switching method is the addition method.

[0016] In one embodiment, the step of switching from the low-precision model to the high-precision model includes:

[0017] Determine the addition requirement according to the model switching instruction, and the target semantic attribute information corresponding to the addition requirement;

[0018] Determine the component to be added according to the target semantic attribute information, and superimpose the component to be added into the low-precision model to obtain the high-precision model.

[0019] In one embodiment, after the step of filtering non-main structure components in the high-precision model according to the semantic attribute information to obtain the low-precision model, the following steps are further included:

[0020] Obtain the road surface components in the high-precision model according to the semantic attribute information, and obtain the bounding box of the road surface components;

[0021] Obtain the three-dimensional coordinates of the geometric center of the bounding box, and sort the three-dimensional coordinates in the order of the route direction to generate a coordinate point sequence;

[0022] Generate the symbol model according to the coordinate point sequence.

[0023] In one embodiment, after the step of filtering non-main structure components in the high-precision model according to the semantic attribute information to obtain the low-precision model, the following steps are further included:

[0024] Perform triangulation simplification on the low-precision model, delete or merge redundant vertices, and perform mesh simplification;

[0025] Perform texture compression on the low-precision model after mesh simplification to obtain the volumetric model.

[0026] In one embodiment, after obtaining the semantic attribute information of the component and mapping the semantic attribute information into the sub-high-precision model, the step of generating the high-precision model according to the sub-high-precision model includes:

[0027] Obtain the attribute information and semantic descriptions of each of the components in the BIM model, and generate the semantic attribute information based on the attribute information and the semantic descriptions;

[0028] Obtain and associate the semantic attribute information with the components in the sub-high-precision model according to the identifiers of the components;

[0029] Generate a high-precision model based on the sub-high-precision model.

[0030] In one embodiment, the step of filtering non-main-structure components in the high-precision model according to the semantic attribute information to obtain a low-precision model includes:

[0031] Traverse the semantic information. If the semantic information corresponding to the component is not in the preset list of main-structure components, determine that the component is a non-main-structure component;

[0032] Filter the non-main-structure components from the high-precision model to obtain the low-precision model.

[0033] In addition, to achieve the above object, the present application also proposes a semantic-constrained BIM model lightweighting system. The semantic-constrained BIM model lightweighting system includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. The computer program is configured to implement the steps of the semantic-constrained BIM model lightweighting method as described above.

[0034] The present application provides a semantic-constrained BIM model lightweighting method. According to the geometric information of each component in the BIM model to be processed, sub-high-precision models of each of the components are constructed; by obtaining the geometric information of each component in the BIM model, the accuracy of the sub-high-precision models can be ensured. After obtaining the semantic attribute information of the components and mapping the semantic attribute information to the sub-high-precision models, a high-precision model is generated based on the sub-high-precision models; according to the semantic attribute information, non-main-structure components in the high-precision model are filtered to obtain a low-precision model. By mapping the semantic attribute information in the sub-high-precision models and generating a high-precision model, and filtering non-main-structure components in the high-precision model according to the semantic attribute information, a simplified low-precision model is obtained. Through semantic filtering, deeper-level information can be processed, and it can be avoided that redundant information fails to be effectively removed due to improper setting of the filtering threshold, or important information is wrongly filtered out. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings here are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0036] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0037] Figure 1 Schematic flowchart provided for the first embodiment of the BIM model lightweighting method with semantic constraints of the present application;

[0038] Figure 2 Schematic detailed flowchart provided for the first embodiment of the BIM model lightweighting method with semantic constraints of the present application;

[0039] Figure 3 Schematic flowchart provided for the second embodiment of the BIM model lightweighting method with semantic constraints of the present application;

[0040] Figure 4 Schematic flowchart provided for the third embodiment of the BIM model lightweighting method with semantic constraints of the present application;

[0041] Figure 5 Schematic flowchart provided for the fourth embodiment of the BIM model lightweighting method with semantic constraints of the present application;

[0042] Figure 6 Schematic diagram of the model switching method provided for the fourth embodiment of the BIM model lightweighting method with semantic constraints of the present application;

[0043] Figure 7 Schematic flowchart provided for the fourth embodiment of the BIM model lightweighting method with semantic constraints of the present application;

[0044] Figure 8 Schematic diagram of the device structure of the hardware operating environment involved in the BIM model lightweighting method with semantic constraints in the embodiments of the present application. Detailed implementation manners

[0045] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0046] To better understand the technical solutions of the present application, the following will be described in detail in combination with the accompanying drawings of the specification and specific implementation manners.

[0047] The main solution of the embodiment of the present application is as follows: according to the geometric information of each component in the BIM model to be processed, sub-high-precision models of each of the components are constructed; the semantic attribute information of the components is obtained, and after the semantic attribute information is mapped into the sub-high-precision models, high-precision models are generated according to the sub-high-precision models; according to the semantic attribute information, non-main structural components in the high-precision models are filtered to obtain low-precision models.

[0048] The LOD (Levels of Detail) technology can dynamically adjust the precision of the model according to factors such as the viewing distance and screen resolution. When on a low-resolution screen or the viewing point is far from the model, a low-precision model is displayed; while when on a high-resolution screen or the viewing point is close to the model, a high-precision model and texture are loaded. By constructing LOD models at different levels, the amount of data processing can be minimized as much as possible while ensuring the model quality, and the model loading speed can be accelerated.

[0049] When performing lightweight processing on a BIM model, generally different levels of LOD models are constructed through size filtering. When the set filtering threshold is too high, a large amount of redundant information is included in the retained data, resulting in a large data volume. If the set filtering threshold is too low, although some smaller redundant data can be removed, some important but slightly larger-sized data is wrongly filtered out, thus affecting the integrity and accuracy of the data.

[0050] The present application provides a semantic-constrained BIM model lightweighting method. According to the geometric information of each component in the BIM model to be processed, sub-high-precision models of each of the components are constructed; by obtaining the geometric information of each component in the BIM model, the accuracy of the sub-high-precision models can be ensured. The semantic attribute information of the components is obtained, and after the semantic attribute information is mapped into the sub-high-precision models, high-precision models are generated according to the sub-high-precision models; according to the semantic attribute information, non-main structural components in the high-precision models are filtered to obtain low-precision models. By mapping the semantic attribute information in the sub-high-precision models and generating high-precision models, and filtering non-main structural components in the high-precision models according to the semantic attribute information, a simplified low-precision model is obtained. Through semantic filtering, deeper-level information can be processed, and it can be avoided that due to improper setting of the filtering threshold, redundant information cannot be effectively removed, or important information is wrongly filtered out.

[0051] It should be noted that the execution subject of this embodiment can be a computing service device with network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a device, etc. that can implement the above functions. Hereinafter, taking the semantic-constrained BIM model lightweighting system as an example, this embodiment and the following embodiments will be described.

[0052] Based on this, the embodiments of the present application provide a lightweight method for a BIM model with semantic constraints. Refer to Figure 1 , Figure 1 which is a schematic flowchart of the first embodiment of the lightweight method for a BIM model with semantic constraints in the present application.

[0053] In this embodiment, the lightweight method for a BIM model with semantic constraints includes steps S100 to S300:

[0054] Step S100: Construct sub-high-precision models for each component according to the geometric information of each component in the BIM model to be processed.

[0055] It should be noted that BIM (Building Information Modeling) refers to the general term for the process and results of digitally expressing the physical and functional characteristics of construction projects and facilities throughout their life cycles and designing, constructing, and operating based on this. The traffic road BIM model is a three-dimensional digital model constructed based on BIM technology by integrating information in multiple stages such as road design, construction, and operation and maintenance. The high-precision model is the LOD4 (Level of Detail 4) hierarchical model, which is the highest level of detailed and complete component-level model and can meet the requirements of refined business scenario management and high-precision display. LOD (Levels of Detail), the LOD technology refers to determining the resource allocation for object rendering according to the position and importance of the nodes of the object model in the display environment, reducing the number of faces and detail level of unimportant objects, so as to obtain high-efficiency rendering operations.

[0056] In addition, it should be noted that the components in the BIM model refer to the basic units that make up buildings or infrastructure such as traffic roads. In the traffic road BIM model, the components can include road surfaces, roadbeds, bridges, tunnels, drainage facilities, traffic signs, and markings, etc. These components are accurately represented by three-dimensional modeling technology in the BIM model and contain rich information.

[0057] In this embodiment, the BIM model of the traffic road can be obtained from an online platform or a database. For example, the BIM model of the traffic road can be obtained from the CIM platform (platform of city information modeling, the city-wide spatio-temporal information platform). Open the BIM model of the traffic road according to BIM software (such as Revit, Civil 3D, etc.), and use the measurement and query tools of the BIM software to extract the geometric information of each component, such as length, width, height, curvature, position coordinates, etc. Organize the extracted geometric information to form a structured data set. And verify the data to ensure the accuracy and integrity of the information. Select a suitable modeling software (such as AutoCAD, 3ds Max, SketchUp, etc.) to build a high-precision model.

[0058] In this embodiment, the road entity can be divided into a civil engineering structure layer, a traffic facility layer, a drainage facility layer, and a safety facility layer. The civil engineering structure layer includes a roadbed and a road surface. The traffic facility layer includes lanes, sidewalks, a median strip, and traffic islands. The drainage facility layer includes drainage ditches and drainage pipes. The safety facility layer includes traffic signs, traffic lights, guardrails, and lighting systems.

[0059] In this embodiment, when converting the BIM model of the traffic road into a high-precision model, in order to ensure the consistency between the high-precision model and the BIM model of the traffic road, parameters are preset as constraint conditions. The constraint conditions can include: project base point coordinates, azimuth information, and model dimension units. The project base point coordinates and azimuth information are used to determine the position and orientation of the BIM model in the global coordinate system. During the conversion process, it can ensure that the generated triangular mesh model is consistent with the original BIM model in terms of spatial position. For the model dimension units, the BIM model is usually modeled using specific dimension units (such as millimeters, meters, etc.). During the conversion process, it is necessary to ensure that these units are correctly converted or retained to avoid dimensional errors. After setting the conversion constraint conditions, traverse all components in the BIM model of the traffic road and extract the geometric information of each component.

[0060] Optionally, before extracting the geometric information from the BIM model, first verify the BIM model of the traffic road to ensure its accuracy, integrity, and consistency. The verification content includes but is not limited to: the geometric shape, dimensions, position relationship of the model, as well as attribute information such as materials and textures.

[0061] Optionally, for different regions and components, scanned data with different resolutions or mesh divisions with different densities are adopted. According to project requirements and design specifications, key components and important regions are set. For key components and important regions, scanned data with higher resolutions and denser mesh divisions are used to construct high-precision models. Exemplarily, key components can be determined based on structural stability and safety assessments. Bridges and tunnels are key structures in transportation roads, and their stability and safety are crucial. Therefore, more detailed modeling and analysis are carried out. Intersection and overpass regions are prone to traffic accidents due to large traffic flows and diverse traffic modes, and thus more detailed modeling and analysis are required. Key components can also be determined according to geographical environment and climatic conditions. Regions with complex geological conditions and harsh climatic conditions are regarded as important regions, and components in important regions are regarded as key components.

[0062] Optionally, during the process of constructing a sub-high-precision model, the accuracy change of the model is monitored according to a preset cycle, the geometric shape, size, and positional relationship of the model are verified, the accuracy data of each verification is recorded, and compared with the previous data to analyze the trend and cause of the accuracy change. When it is found that the accuracy decreases or does not meet the preset accuracy standard, the modeling parameters are adjusted and optimized.

[0063] In a feasible implementation manner, step S100 may further include the following steps:

[0064] Obtain the texture information of each of the components in the transportation road BIM model;

[0065] Map the texture information to the corresponding components of the high-precision model.

[0066] In this embodiment, while obtaining the geometric information of each component in the transportation road BIM model, the texture information of each component is obtained. After constructing a high-precision model based on the geometric information, the components that need to add textures are selected. For each selected component, the texture mapping tool in the BIM software is used to apply the corresponding texture picture to the surface of the component. The basic model is optimized in detail, such as adding edges, chamfers, textures, etc. High-quality texture materials are used to enhance the visual effect of the model. And the simulation function of the modeling software is used to simulate the effects such as lighting, shadows, and materials of the model. According to the simulation results, the model is further adjusted and optimized.

[0067] In this embodiment, by extracting geometric information from the transportation road BIM model and selecting a suitable modeling software for the construction and optimization of the high-precision model, a transportation road model with high precision and realistic effects can be obtained.

[0068] Step S200: Obtain the semantic attribute information of the component, map the semantic attribute information into the sub-high-precision model, and then generate a high-precision model according to the sub-high-precision model.

[0069] In this embodiment, for the semantic attribute information of each component in the transportation road BIM model, the semantic attribute information may include information such as the type, function, material, and positional relationship of the component. Use BIM software or related tools to extract the semantic attribute information of each component from the model. Verify the extracted semantic attribute information to ensure its accuracy and integrity. At the same time, organize the information for subsequent mapping into the high-precision model. According to the structure and component definition of the high-precision model, establish a mapping relationship with the semantic attribute information of the components in the BIM model. Assign the verified and organized semantic attribute information to the corresponding components in the high-precision model.

[0070] Optionally, verify the mapped high-precision model to ensure that the semantic attribute information has been correctly mapped to each component. Select a suitable BIM verification tool or plugin, such as Autodesk's Navisworks, Solibri Model Checker, etc. Load the mapped high-precision model into the verification tool and check each component in the model one by one to confirm whether its semantic attribute information (such as type, material, size, function, etc.) has been correctly mapped. First, use the batch verification function of the verification tool to quickly check whether the semantic attribute information of a large number of components is consistent and accurate. Compare the semantic attribute information in the model with the reference data to ensure their consistency. Second, check whether there are components in the model that lack semantic attribute information.

[0071] Identify and correct any erroneously mapped semantic attribute information.

[0072] Please refer to Figure 2 , in a feasible implementation manner, step S200 may include steps S210 to S230:

[0073] Step S210: Obtain the attribute information and semantic description of each component in the BIM model, and generate the semantic attribute information according to the attribute information and the semantic description;

[0074] Step S220: Obtain and associate the semantic attribute information with the component in the sub-high-precision model according to the identifier of the component;

[0075] Step S230: Generate the high-precision model according to the sub-high-precision model.

[0076] In this embodiment, each component in the BIM model is traversed to read its attribute information, including type (such as curb, traffic sign), dimensions (such as length, width, height), material (such as concrete, steel), function (such as indicating direction, providing lighting), etc. Referring to the semantic description of the road entity components, a set of semantic categories is predefined. For each component, its attribute information is matched with the predefined semantic categories. One or more semantic labels are assigned to each component, and these labels represent the semantic categories to which it belongs. Ensure that each component has a unique semantic identifier. The generated semantic information is stored in an appropriate data structure. The unique identifier of each component (such as UUID, ID, etc.) is extracted, a mapping table or association table is created to associate the identifier of the component with the semantic attribute information, and the semantic attribute information is applied to the corresponding component according to the mapping table. Finally, the sub-high-precision models are integrated into a complete high-precision model.

[0077] Step S300: Filter the non-main-structure components in the high-precision model according to the semantic attribute information to obtain a low-precision model.

[0078] It should be noted that the main-structure components refer to the skeleton of a building, mainly including load-bearing and supporting structures such as beams, columns, slabs, and walls. The non-main-structure components refer to non-load-bearing and auxiliary structures such as building facades, roofs, doors and windows, pipes, and equipment.

[0079] In addition, it should be noted that the low-precision model (LOD3, Level of Detail 3) represents a relatively detailed model level in BIM or road information models. At this level, the extraction of the main road outline becomes precise, and at the same time, the material texture information is fully expressed to meet the needs of real material texture display.

[0080] In this embodiment, it is determined whether the component belongs to the main-structure component or the non-main-structure component according to the component type in the semantic attribute information. The main structure includes road surfaces, lanes, sidewalks, bridges, and tunnels, etc. The non-main structure includes green belts, traffic signs, street lights, traffic signals, and fences, etc. If the type of a component belongs to the non-main-structure type, it is filtered. After performing a semantic filtering operation on the high-precision model to filter all non-main-structure components in the high-precision model, a low-precision model is obtained. The low-precision model retains the main-structure components, as well as their geometric information and material texture information, and can meet the basic needs of appearance recognition.

[0081] Please refer to Figure 2 , in a feasible embodiment, step S300 may include steps S310 to S320:

[0082] Step S310: Traverse the semantic information. If the semantic information corresponding to the component is not in the preset list of main structure components, determine that the component is a non-main structure component.

[0083] Step S320: Filter the non-main structure components from the high-precision model to obtain the low-precision model.

[0084] In this embodiment, a list containing the types of main structure components is preset. Use BIM software to load the high-precision model to ensure that all components and their semantic attribute information in the model have been correctly loaded. Traverse each component in the model and extract its corresponding semantic attribute information. Compare the semantic information of each component with the preset list of main structure components. If the type of the component is in the list, consider that the component is a main structure component; if not, determine that the component is a non-main structure component. During the traversal, mark the components determined to be non-main structure components.

[0085] According to the marking results, remove or hide all non-main structure components from the high-precision model. Save the model after removing the non-main structure components as the low-precision model.

[0086] In this embodiment, by filtering the non-main structure components in the high-precision model according to the semantic attribute information, a simplified low-precision model is obtained. It can improve the application efficiency and flexibility of the model and provide convenience for subsequent analysis and management. Compared with the size filtering method based on physical dimensions, semantic filtering can process deeper information and can avoid the situation where redundant information fails to be effectively removed due to improper setting of the filtering threshold or important information is wrongly filtered out.

[0087] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar content as in the above-mentioned first embodiment can be referred to the above introduction and will not be repeated hereinafter. On this basis, please refer to Figure 3 , after step S300, steps S400 to S500 are further included:

[0088] Step S400: Simplify the triangular mesh of the low-precision model, delete or merge redundant vertices, and perform mesh simplification.

[0089] Step S500: Compress the texture of the low-precision model after mesh simplification to obtain the volume model.

[0090] In this embodiment, lightweight processing such as triangular mesh simplification and texture merging and compression is performed on the low-precision model to obtain a volume model, reducing the complexity of components and the file size. Triangular mesh simplification is an effective lightweight processing technology that reduces the complexity and file size of a 3D model by reducing the number of triangular faces in the model. When performing triangular mesh simplification, first, vertex downsampling is performed. By using vertex downsampling algorithms such as Quadric Error Metric (QEM) and Vertex Clustering, important vertices in the low-precision model are retained, and other redundant vertices are deleted or merged to reduce the number of vertices. Second, mesh simplification is performed. By merging and collapsing the faces in the mesh, the number of faces is reduced. Finally, unnecessary details are removed, such as removing the internal structure of the model and reducing the overlapping faces between models. The texture merging and compression technology reduces the file size by merging and compressing the textures of the model. When performing texture merging and compression, the original texture image can be downsampled (i.e., reducing the image size), and the texture is simplified by reducing the resolution of the texture image. Texture mapping optimization can also be performed. The UV mapping technology is used to map the texture coordinates to the model surface, reducing texture seams and repeating areas to reduce texture repetition and waste. Mipmapping can also be used to reduce aliasing and jagged edges of texture details at different distances, improving rendering performance.

[0091] Optionally, the mesh simplification method can perform selective simplification based on the curvature of the mesh. By calculating the curvature value of each vertex in the mesh and sorting according to the magnitude of the curvature value, vertices with smaller curvature and their adjacent faces are deleted, thereby achieving mesh simplification. The mesh simplification method can also perform selective simplification based on the normal rate of the mesh. By calculating the normal direction of each vertex in the mesh and sorting according to the degree of change in the normal direction, vertices with smaller changes in the normal direction and their adjacent faces are deleted, thereby achieving mesh simplification.

[0092] Based on the first embodiment of the present application, in the third embodiment of the present application, for the same or similar content as in the above-mentioned embodiment one, reference can be made to the above introduction and will not be repeated hereinafter. On this basis, please refer to Figure 4 , after step S300, steps S600 to S800 are further included:

[0093] Step S600, obtaining a road surface component in the high-precision model according to the semantic attribute information, and obtaining the bounding box of the road surface component;

[0094] It should be noted that the bounding box refers to a simple geometric shape (usually a cube or a cuboid) that completely encloses the road surface component and is consistent with the road surface component in shape and position.

[0095] In this embodiment, in the high-precision model, each element (such as roads, buildings, trees, etc.) has specific semantic attribute information. This information describes the type, location, size, etc. of the element. There are various types of road surface components, including but not limited to: lanes (including main lanes, auxiliary lanes, bus lanes, bicycle lanes, etc.), sidewalks, road shoulders, and intersections (including crossroads, T-shaped intersections, roundabouts, etc.).

[0096] In this embodiment, traverse the high-precision model to obtain the semantic attribute information of all components in the high-precision model. The semantic attribute information includes the type information (such as roads, buildings, trees, etc.) and attribute information (such as the number of lanes, width, and material of the road; the number of floors, height, and use of the building, etc.) of each component in the high-precision model. After obtaining the semantic attribute information, locate the semantic tags related to the road surface components. For example, the semantic tags belonging to the road surface components may include "lane", "sidewalk", "intersection", etc. Check whether each component matches the semantic tags of the road surface components. If it matches, the component is considered a road surface component, and the identified road surface components are classified and sorted. For example, lanes, sidewalks, and road shoulders can be classified separately, and a list or database containing all relevant elements can be created for each category.

[0097] In this embodiment, first, after determining the road surface components in the high-precision model, extract the geometric data of the road surface components from the high-precision model, including vertex coordinates, edge information, face information, etc. Using the extracted geometric data, construct a geometric model of the road surface component. The geometric model can be a polygon mesh, triangular patches, etc., depending on the representation method and accuracy requirements of the high-precision model. Secondly, create an empty bounding box object. The bounding box object is a cube or cuboid with the minimum and maximum coordinate points. The minimum coordinate point of the bounding box can be set to positive infinity, and the maximum coordinate point can be set to negative infinity for subsequent updates. Then, traverse all vertices or edges in the geometric model of the road surface component. For each vertex or edge, update the minimum and maximum coordinate points of the bounding box. If a certain coordinate of the vertex is less than the current minimum coordinate point of the bounding box, update this coordinate point to the new minimum coordinate point. If a certain coordinate of the vertex is greater than the current maximum coordinate point of the bounding box, update this coordinate point to the new maximum coordinate point. After the traversal is completed, the minimum and maximum coordinate points of the bounding box define the bounding box of the road surface component.

[0098] Optionally, the bounding box is verified to ensure that it correctly encloses the road surface component. The boundaries of the bounding box are compared with the boundaries of the geometric model of the target object. Ensure that the boundaries of the bounding box completely contain the boundaries of the target object, and no vertices or edges extend beyond the range of the bounding box. Intersection tests can also be performed to verify the validity of the bounding box. For example, methods such as ray casting or the separating axis theorem are used to detect whether the bounding box intersects the target object. If the result of the intersection test is negative (i.e., no intersection), it indicates that the bounding box does not correctly enclose the target object and needs to be adjusted. And the bounding box is adjusted and optimized according to the verification results until a bounding box that meets the requirements is obtained.

[0099] Step S700: Obtain the three-dimensional coordinates of the geometric center of the bounding box, and sort the three-dimensional coordinates in the order of the route direction to generate a sequence of coordinate points.

[0100] Step S800: Generate the symbolic model according to the sequence of coordinate points.

[0101] It should be noted that the symbolic model (LOD1, Level of Detail 1) is a level used to express the macroscopic characteristics of the model in 3D modeling. At this level, the model is expressed in the form of 3D vector lines, mainly used to show the position and direction of the entire road. The route direction refers to the continuous and ordered spatial arrangement direction of the road surface components in the high-precision model.

[0102] In this embodiment, for each road surface component, the minimum and maximum coordinate points of the bounding box are calculated according to its geometric representation (such as vertex coordinates, edge information, etc.), and the three-dimensional coordinates of its geometric center are calculated according to the average value of the minimum and maximum coordinate points. According to the route information or navigation data of the high-precision model, the route direction of the road surface component is determined. According to the route direction, the three-dimensional coordinate points of the geometric center of the bounding box are sorted. Exemplarily, these coordinates are arranged in the order that a vehicle or pedestrian may pass. The sorted three-dimensional coordinate points are stored in a data structure, such as an array, a list, or a database, etc., to obtain a coordinate sequence.

[0103] In this embodiment, a line-type vector symbolic model is generated by fitting according to the sequence of the three-dimensional coordinates of the geometric center of the bounding box. According to the characteristics of the point set and the requirements of the fitting method, appropriate fitting parameters (such as the order of the polynomial, the number of segments of the spline curve, etc.) are selected. The point set is fitted using the selected fitting method and parameters. The fitting result is converted into a vector symbol, usually one or more continuous lines. Finally, the fitting result is further smoothed or adjusted to ensure that it meets the requirements of the actual application scenario.

[0104] Optionally, perform a least squares fit. Sort the point set according to the route direction and store it as a list of (x, y, z) coordinate points. For a two-dimensional fit (ignoring the z coordinate or assuming the z coordinate is constant), select a straight line or a quadratic curve as the fitting model. For a three-dimensional fit, select a plane or a quadratic surface. Use the least squares algorithm to calculate the parameters of the fitting model. Generate a vector symbol (such as a line segment, a plane, etc.) representing the fitting result based on the calculated fitting model parameters.

[0105] Optionally, perform a polynomial fit. Select an appropriate polynomial order according to the complexity of the point set and the required fitting accuracy. Based on the selected order, construct a polynomial function, which will serve as the fitting model. Use numerical methods (such as Gaussian elimination, QR decomposition, etc.) to solve the coefficients of the polynomial so that the sum of the squared errors of the polynomial function on the point set is minimized. Generate a vector symbol (such as a curve segment) representing the fitting result based on the solved polynomial function.

[0106] Optionally, perform a spline curve fit. Determine the number of segments of the spline curve according to the characteristics of the point set and the required fitting accuracy. Select or calculate the positions of the segment points in the point set. These points will serve as the segment connection points of the spline curve. Within each segment, construct a polynomial function as the fitting model. These polynomial segments need to satisfy the smoothness condition at the segment connection points. Use numerical methods to solve the coefficients of each polynomial segment while ensuring the smoothness at the segment connection. Generate a spline curve vector symbol representing the fitting result based on the solved polynomial segments.

[0107] Based on the first embodiment of the present application, in the fourth embodiment of the present application, the same or similar content as in the above-mentioned embodiment 1 can be referred to the above introduction and will not be repeated hereinafter. On this basis, please refer to Figure 5 , the method may further include steps A100 to A200:

[0108] Step A100, when receiving a model switching instruction, parse the model switching instruction to obtain the current model information and the target model information;

[0109] Step A200, determine the switching method according to the current model information and the target model information.

[0110] In this embodiment, receive a model switching instruction from the user or the upper-layer application. The received instruction contains the current model information and the target model information. Parse the instruction to extract the current model information and the target model information. The current model information includes the identifier, type, version, etc. of the currently used model. The target model information is the identifier, type, version, etc. of the model specified in the instruction to be switched to. Determine the strategy according to the model types before and after the switch.

[0111] Please refer to Figure 6 , in a feasible implementation manner, step A200 may include steps A210 to A220:

[0112] Step A210, when switching from the symbol model to the volume model, or from the volume model to the low-precision model, the switching method is the replacement method;

[0113] Step A220, when switching from the low-precision model to the high-precision model, the switching method is the addition method.

[0114] In this implementation manner, when switching from the symbol model to the volume model, or from the volume model to the low-precision model, since there may be significant differences in the representation methods and data structures of these models, a replacement strategy is adopted. First, unload the current model, and then load and initialize the target model. Release the resources occupied by the current model, such as memory, graphics rendering resources, etc., and according to the target model information, load and initialize the target model. Update the view and data structure, and update the view and data structure to a state matching the target model.

[0115] In this implementation manner, when switching from the low-precision model to the high-precision model, since the high-precision model usually adds more details and precision on the basis of the low-precision model, an overlay strategy is adopted. While keeping the low-precision model unchanged, gradually add the details of the high-precision model. While keeping the low-precision model unchanged, load the detailed parts of the high-precision model. Merge the detailed parts of the high-precision model with the data of the low-precision model to generate a complete high-precision model. Finally, update the view to the state of the high-precision model.

[0116] Please refer to Figure 7 , in this implementation manner, after obtaining the high-precision model, low-precision model, and volume model, according to the 3DTiles data service standard and model visualization characteristics, a combination of ADD (addition) and REPLACE (replacement) is used to organize the multi-level-of-detail 3D tiles. 3DTiles is a geospatial data format used to store and distribute geospatial data. It is used to transmit and load a large amount of heterogeneous 3D geospatial datasets. When the addition strategy is adopted, on the basis of the existing tiles, new tiles are added to increase the level of detail of the model. When the replacement strategy is adopted, the original tiles are replaced with new tiles to achieve model update or replacement of the level of detail.

[0117] In this embodiment, by parsing the semantic description information of the traffic road BIM model, using the semantic information to screen the component geometric information, constructing a multi-level of detail LOD model, and optimizing the data organization of the LOD model, a standard 3DTiles data service is generated. This improves the expression efficiency of the traffic road BIM model in a visualization platform (such as WebGIS).

[0118] In a feasible implementation manner, step A220 may include the following steps:

[0119] Determine the addition requirements according to the model switching instruction, and the target semantic attribute information corresponding to the addition requirements;

[0120] Determine the component to be added according to the target semantic attribute information, and superimpose the component to be added onto the low-precision model to obtain the high-precision model.

[0121] In this implementation manner, through semantic filtering, the components that need to be added to the low-precision model are screened out from the high-precision model. These screened components are superimposed onto the low-precision model at the LOD3 level in an additional manner to form a high-precision model at the LOD4 level.

[0122] In this implementation manner, first, the received switching instruction is parsed to determine the addition requirements such as the type, quantity, and position of the components to be added when switching from the low-precision model to the high-precision model. According to the addition requirements, the target semantic attribute information corresponding to the components to be added is determined, including the type of the components (such as street lights, traffic signs, drainage facilities, etc.), material, size, color, function, etc. According to the target semantic attribute information, the components to be added that meet the conditions are selected from the component library. These components can be predefined models, components, or parametric objects. According to specific requirements, the parameters of the components to be added are adjusted, such as size, position, direction, etc., to ensure that they are perfectly integrated with the low-precision model. The components to be added are added to the low-precision model in a superimposed manner, and the geometric shape, material attributes, and other information of the components are merged with the low-precision model. The view and data structure of the superimposed model are updated to the state of the high-precision model to obtain the high-precision model.

[0123] This application provides a BIM model lightweighting system with semantic constraints. The BIM model lightweighting system with semantic constraints includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the BIM model lightweighting method with semantic constraints in the first embodiment above.

[0124] Next, refer to Figure 8, which shows a schematic structural diagram of a BIM model lightweighting system suitable for implementing the semantic constraints of the embodiments of the present application. The BIM model lightweighting system with semantic constraints in the embodiments of the present application may include, but is not limited to, mobile terminals such as laptop computers, PDAs (Personal Digital Assistant), PADs (portable android devices), etc., and fixed terminals such as desktop computers, etc. Figure 8 The shown BIM model lightweighting system with semantic constraints is merely an example and should not impose any limitations on the functions and usage scope of the embodiments of the present application.

[0125] As Figure 8 shown, the BIM model lightweighting system with semantic constraints may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which may perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 into the random access memory (RAM: Random Access Memory) 1004. In the RAM 1004, various programs and data required for the operation of the BIM model lightweighting system with semantic constraints are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. The input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems may be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 may allow the BIM model lightweighting system with semantic constraints to communicate with other devices wirelessly or wiredly to exchange data. Although the figure shows a BIM model lightweighting system with various systems, it should be understood that it is not required to implement or have all the shown systems. More or fewer systems may be alternatively implemented or had.

[0126] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product that includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by a processing device 1001, the above functions defined in the methods of the embodiments disclosed in the present application are executed.

[0127] The semantic constraint-based BIM model lightweighting system provided by the present application adopts the semantic constraint-based BIM model lightweighting method in the above embodiments, and can solve the technical problem of how to improve the lightweighting process of the BIM model for long texts. Compared with the prior art, the beneficial effects of the semantic constraint-based BIM model lightweighting system provided by the present application are the same as those of the semantic constraint-based BIM model lightweighting method provided by the above embodiments, and other technical features in the semantic constraint-based BIM model lightweighting system are the same as the features disclosed in the method of the previous embodiment, and will not be elaborated here.

[0128] It should be understood that each part disclosed in the present application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0129] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all of them should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0130] The present application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the semantic constraint-based BIM model lightweighting method in the above embodiments.

[0131] The computer-readable storage medium provided by this application can, for example, be a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or components, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or component. The program code contained on the computer-readable storage medium can be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0132] The above computer-readable storage medium can be included in the semantically constrained BIM model lightweighting system; or it can exist independently without being assembled into the semantically constrained BIM model lightweighting system.

[0133] The above computer-readable storage medium carries one or more programs. When the one or more programs are executed by the semantically constrained BIM model lightweighting system, the semantically constrained BIM model lightweighting system is caused to: construct sub-high-precision models of each component according to the geometric information of each component in the BIM model to be processed; obtain the semantic attribute information of the component, map the semantic attribute information into the sub-high-precision model, and then generate a high-precision model according to the sub-high-precision model; filter non-main structural components in the high-precision model according to the semantic attribute information to obtain a low-precision model.

[0134] Computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The above-mentioned programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, by connecting through the Internet using an Internet service provider).

[0135] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of the code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0136] The modules described in the embodiments of this application can be implemented in software or in hardware. Among them, the name of the module does not constitute a limitation to the unit itself in some cases.

[0137] The readable storage medium provided by this application is a computer-readable storage medium. The computer-readable storage medium stores computer-readable program instructions (i.e., computer programs) for performing the above-mentioned BIM model lightweighting method with semantic constraints, and can solve the technical problem of how to improve the lightweighting process of the BIM model with long text. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by this application are the same as those of the BIM model lightweighting method with semantic constraints provided by the above embodiments, and will not be elaborated here.

[0138] The above are only some embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structural transformation made under the technical concept of the present application by using the content of the specification and drawings of the present application, or any direct / indirect application in other related technical fields, is included in the patent protection scope of the present application.

Claims

1. A semantically constrained BIM model lightweight method, characterized in that: The method includes: Taking the project base point coordinates, azimuth information and model size units as constraint conditions, traverse each component in the BIM model to be processed, obtain and construct a sub-high-precision model of each component based on the geometric information of the component, including: determining the grid division strategy for constructing the sub-high-precision model based on whether the component is a key component; detecting the accuracy change of the sub-high-precision model according to a preset period, and adjusting the modeling parameters when the accuracy change does not meet the preset accuracy standard; the grid density of the key component is greater than that of the non-key component; Acquiring semantic attribute information of the component, mapping the semantic attribute information to the sub-high-precision model, and generating a high-precision model according to the sub-high-precision model; According to the semantic attribute information, the non-main structure components in the high-precision model are filtered to obtain a low-precision model, which includes: traversing the semantic attribute information, if the component type in the semantic attribute information corresponding to the component is not in the preset main structure component list, then determining that the component is the non-main structure component; filtering the non-main structure components from the high-precision model to obtain the low-precision model.

2. The semantically constrained BIM model lightweight method according to claim 1, characterized in that: The method further comprises: When a model switching instruction is received, the model switching instruction is parsed to obtain current model information and target model information; A switching mode is determined according to the current model information and the target model information.

3. The semantically constrained BIM model lightweight method according to claim 2, characterized in that: The step of determining the switching mode according to the current model information and the target model information comprises: When switching from a symbolic model to a block model, or from a block model to the low-precision model, the switching mode is a substitution mode; When switching from the low-precision model to the high-precision model, the switching mode is an additional mode.

4. The semantically constrained BIM model lightweight method according to claim 3, characterized in that: The step of switching from the low-precision model to the high-precision model comprises: Determining, according to the model switching instruction, an adding requirement and target semantic attribute information corresponding to the adding requirement; The component to be added is determined according to the target semantic attribute information, and the component to be added is superimposed on the low-precision model to obtain the high-precision model.

5. The semantically constrained BIM model lightweight method according to claim 3, characterized in that: After the step of filtering the non-main structural components in the high-precision model according to the semantic attribute information to obtain the low-precision model, the following step further comprises: According to the semantic attribute information, a pavement component is obtained in the high-precision model, and a bounding box of the pavement component is obtained; Obtaining the three-dimensional coordinates of the geometric center of the bounding box, and sorting the three-dimensional coordinates according to the order of the route to generate a coordinate point sequence; The symbol model is generated according to the coordinate point sequence.

6. The semantically constrained BIM model lightweight method according to claim 3, characterized in that: After the step of filtering the non-main structural components in the high-precision model according to the semantic attribute information to obtain the low-precision model, the following step further comprises: Simplifying the triangulated network of the low-precision model, deleting or merging redundant vertices, and simplifying the mesh; The low-precision model after mesh simplification is texture compressed to obtain the block model.

7. The semantically constrained BIM model lightweight method according to claim 1, characterized in that: After acquiring the semantic attribute information of the component and mapping the semantic attribute information to the sub-high-precision model, the step of generating a high-precision model according to the sub-high-precision model includes: Acquire the attribute information and semantic description of each of the components in the BIM model, and generate the semantic attribute information according to the attribute information and the semantic description; Acquire and associate the semantic attribute information with the component in the sub-high-precision model according to the identifier of the component; The high-precision model is generated according to the sub-high-precision model.

Images