Component-level extra-large suspension bridge engineering information model lightweight three-dimensional tile conversion method

Abstract

The invention discloses a component-level extra-large suspension bridge engineering information model lightweight three-dimensional tile conversion method, and the method comprises the steps: simplifying an LOD4 hierarchical model with the most abundant detail levels into an LOD3 hierarchical model according to a display demand through the semantic information extraction of a bridge engineering model; further simplifying an LOD3 hierarchical model through a spatial partition binary tree, and then realizing LOD2 hierarchical bridge model rendering through bridge symmetry and the thought of constructing a geometric entity; and establishing an OBB bounding box to extract the most basic bridge model from the LOD4 hierarchical model. The method has the advantages of being simple and high in execution efficiency, effectively solves the problem that the oversize suspension bridge engineering information model loading speed is low on a common computer, is suitable for loading suspension bridge engineering information models under different scene requirements, and is particularly suitable for loading models of different detail levels under multiple scene requirements.

Description

Technical Field

[0001] The invention relates to the technical field of geographic space information systems, and in particular to a lightweight three-dimensional tile conversion method for a component-level extra-large suspension bridge engineering information model. Background Art

[0002] Engineering information modeling technology is a new technology that has flourished at the beginning of this century and can digitally express the physical properties and functional characteristics of engineering projects. The engineering information model covers the geometric information, spatial relationships, properties and quantity of building components. The engineering information model can be used to display and manage the entire life cycle of the entire project from the survey and design stage to the operation and maintenance stage. The engineering information model is a good carrier of relevant engineering project information. The engineering information model is based on various three-dimensional geometric models stored in the computer, so that modern computer technology can be used to analyze and express engineering information. It is part of the engineering construction big data. The use of engineering information models in modern engineering construction can greatly improve and upgrade the construction mode and management methods of the traditional construction industry, so that it can further realize industrialization and standardization, improve the production efficiency of buildings, and save construction costs.

[0003] As the scale of modern super-large building scenes continues to expand, the requirements for resolution continue to increase, and the amount and size of display data used to represent realistic building scene models also grow exponentially, making the rendering difficulty of engineering scene models increasingly difficult and the rendering efficiency continuously reduced. Existing hardware equipment is difficult to meet the rendering requirements of super-large building scenes. For example, the rendering display of super-large bridge engineering information models is very inefficient due to the large number of bridge components and the complex geometric relationships between the components. As a result, existing hardware has a very low rendering display efficiency for super-large suspension bridge models, and problems such as freezes, frame drops, and even missing bridge display are prone to occur during the loading process.

[0004] More importantly, in the process of building a super-large suspension bridge, problems such as low model display efficiency and incomplete display are likely to bring serious safety hazards to the construction project. Therefore, it is particularly important to simplify the display of super-large suspension bridges in the entire life cycle of their construction and operation.

[0005] At present, there are two main methods for lightweighting information models of large-scale building projects: 1) Simplification methods based on component semantic information: This type of method starts from the basic components of the building, uses semantic extraction methods to extract the various components in the building, and then obtains models of different levels of detail, and finally simplifies and displays them. However, the disadvantage of this method is that it relies on user subjectivity and lacks automatic simplification functions, which is not conducive to large-scale use in engineering scenes. 2) Geometric simplification methods based on computer graphics: This type of method starts from the perspective of computer graphics, uses graphics processing methods, regards the building as a whole, uses graphics methods for overall processing, and realizes simplified display. However, the disadvantage of this method is that it is not conducive to the analysis of engineering scenes and focuses on overall analysis. Therefore, there is an urgent need for a method that is targeted at the characteristics of information data of super-large suspension bridge projects to solve the problem of rendering and displaying super-large suspension bridge project information models. Summary of the invention

[0006] In order to solve the problems existing in the prior art, the present invention provides a lightweight three-dimensional tile conversion method for a component-level extra-large suspension bridge engineering information model, which solves the problems mentioned in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: a lightweight three-dimensional tile conversion method for a component-level super-large suspension bridge engineering information model, comprising the following steps:

[0008] S1, reading the data of the suspension bridge engineering information model, converting the suspension bridge engineering information model into a triangular mesh model to obtain a triangular mesh model, and associating the obtained triangular mesh model with the component semantic attribute information to obtain an LOD4 hierarchical model;

[0009] S2, obtaining external setting parameters, semantically filtering the LOD4 hierarchical model according to the display scene and user needs, and removing the specified bridge entity components to obtain the LOD3 hierarchical model;

[0010] S3, based on the space partition binary tree BSP principle, convert the LOD3 hierarchical model into the LOD2 hierarchical model;

[0011] S4, establishing an OOB bounding box tree on the LOD4 level model, and calculating the coordinate information of the geometric center point under the bounding box tree;

[0012] S5. By calculating the coordinates of the geometric center point of each bounding box, the basic component information in each bounding box is extracted to obtain the LOD1 level model with the lowest level of detail description;

[0013] S6. Convert all hierarchical models obtained in the above steps into b3dm format data, organize them in TileSet.josn format, and finally convert TileSet.josn format into 3d-tiles structure data to complete the conversion.

[0014] Preferably, in step S2, the external setting parameters include viewing angle height and display ratio; the semantic filtering removes the specified bridge entity components including the parallel joints and main trusses in the stiffening beams.

[0015] Preferably, in step S3, the following steps are specifically included:

[0016] Step S31, obtaining bridge deck information of the suspension bridge, including basic parameters of the bridge deck centerline;

[0017] Step S32, dividing the bridge into positive and negative spaces along the bridge deck centerline according to the left-right symmetrical central axis distribution obtained in step S31, storing the bridge deck part by using the repetitiveness of the components, and selecting the positive or negative plane to continue the BSP tree division simplification;

[0018] Step S33, simplifying the bridge tower part by using the idea of ​​constructive solid geometry;

[0019] Step S34: Flip the bridge body part in the positive space or negative space and obtain the LOD2 level model by fitting and splicing the positive and negative planes.

[0020] Preferably, the BSP is an algorithm that effectively determines the positional relationship between a three-dimensional model with dense distribution in some areas and a polyhedron. It divides the partitioning plane into a tree, and each partitioning plane divides the space into two front and back areas, which are called positive and negative planes. The two planes obtained are further divided into smaller spaces until the division meets the set requirements; each storage node of the BSP tree represents a partitioning plane, the left subtree of the node represents the positive space of the partitioning plane, and the right subtree of the node represents the negative space of the partitioning plane.

[0021] Preferably, in step S4, the bounding box tree is established by adopting a top-down method to construct a bounding box tree of a complex bridge body model, by continuously taking the center point of the longest axis to locate the splitting plane, dividing the lines, and then establishing a basic component bounding box tree.

[0022] The beneficial effects of the present invention are as follows: the present invention classifies the suspension bridge engineering model data into four detail levels, and finally converts the data of different detail levels into 3d-tiles data, so as to realize the rapid rendering and loading of the engineering model; by extracting the semantic information of the bridge engineering model, the LOD4 level model with the richest detail levels is simplified to the LOD3 level model according to the display requirements; by implementing the spatial partition binary tree to further simplify the LOD3 level model, and then implementing the LOD2 level bridge model rendering through the bridge symmetry and the idea of ​​constructing geometric entities; by establishing the OBB bounding box, the most basic bridge model is extracted from the LOD4 level model. The present invention has the advantages of simple method and high execution efficiency, and effectively solves the problem of slow loading speed of super-large suspension bridge engineering information model on ordinary computers, and is suitable for loading suspension bridge engineering information models under different scene requirements, and is particularly suitable for loading models of different detail levels for multiple scene requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a flow chart of the method of the present invention;

[0024] Figure 2 This is a schematic diagram of the BSP tree space partitioning principle in an embodiment of the present invention;

[0025] Figure 3 This is a schematic diagram of an OBB bounding box in an embodiment of the present invention;

[0026] Figure 4 Constructing a geometric entity schematic diagram in an embodiment of the present invention;

[0027] Figure 5 This is a schematic diagram of the frustum culling principle in an embodiment of the present invention;

[0028] Figure 6 This is a flow chart of BSP tree segmentation in an embodiment of the present invention;

[0029] Figure 7 This is a flow chart of OBB bounding box segmentation in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] See also Figure 1-Figure 7,The present invention provides a technical solution: a lightweight three-dimensional tile conversion method for a component-level super-large suspension bridge engineering information model, the implementation process uses multiple detail levels to achieve lightweight processing in different scenarios, such as Figure 1 As shown, the following steps are included:

[0032] Step 1, open the engineering information data, convert the suspension bridge engineering information model into a triangular mesh model, and obtain a triangular mesh model after processing, and use the processed triangular mesh model as the LOD4 detail level model. Other detail level models are processed based on the LOD4 detail level model.

[0033] Step 2, obtain external constraint parameters, such as viewing angle height, display scale, etc., and perform semantic filtering on the LOD4 hierarchical model obtained in step 1 according to the display scene and user needs, that is, frustum culling, such as Figure 5 As shown in the figure. The model can be qualitatively simplified by semantic filtering, and the LOD3 model can be obtained by removing the specified bridge entity components.

[0034] Step 3: Based on the LOD3 hierarchical model, use the idea of ​​space partition binary tree to perform further simplification.

[0035] like Figure 2 As shown, the principle of the conversion from LOD3 level to LOD2 level of the present invention is the space partition binary tree (BSP). BSP can effectively determine the positional relationship between the three-dimensional model and the polyhedron in some areas with dense distribution. It can divide the partition plane into a tree. Each partition plane divides the space into two areas in front and back, which are called positive and negative planes. The two planes obtained continue to be divided into smaller spaces according to this idea until they are divided to the set requirements. Each storage node of the BSP tree represents a partition plane. The left subtree of the node represents the positive space of the partition plane, and the right subtree of the node represents the negative space of the partition plane. Figure 6 As shown in the figure, it is the BSP tree segmentation process, which specifically includes the following:

[0036] (1) Obtain all splitting plane data and three-dimensional model data; (2) Establish a root node to store the initial plane and three-dimensional data; (3) Traverse the current node to select a suitable splitting plane; (4) Determine whether there is a suitable splitting plane; (5) If the judgment is yes, establish positive and negative nodes under the current node; (6) Traverse the current node, determine the positive and negative directions of each splitting plane, and place the planes of each direction in the corresponding position; (7) Traverse all structured surfaces of the current node, remove the references to the splitting plane and the model structured surface, retain the splitting plane and the two lower-level node pointers, and end the process; if the judgment in step (4) is no, return the current node as a leaf node, and end the process.

[0037] Using the principle of space partition binary tree (BSP) tree, the uneven distribution characteristics of suspension bridge components in space can be well fitted and stored using a reasonable space segmentation method.

[0038] Step 3.1, since the bridge model is an axisymmetric object after simplification, the central axis of the bridge is used as the division. The task of this step is to obtain the two central axes of the bridge deck;

[0039] Step 3.2, dividing the left and right parts into positive and negative planes according to the left-right symmetrical central axis distribution obtained in step 3.1, and then storing one of the parts;

[0040] Step 3.3, using the bridge body obtained in step 3.3 to continue to simplify the BSP tree partition;

[0041] Step 3.4, since the deck of a suspension bridge is composed of components with different positions but basically the same composition, it is only necessary to record the number of components and their basic characteristics when simplifying the display, and then repeat the display when rendering, that is, geometrically reuse the relevant components;

[0042] Geometry instancing is an important technology in computer graphics. It allows multiple instances to share the same geometric model, so that these instances can be reused at different positions, rotation angles, or scaling ratios, thereby greatly reducing repeated calculations and rendering redundancy and improving rendering efficiency.

[0043] Step 3.5, since the bridge tower is also composed of components with different positions but basically the same composition, the basic principle of constructing geometric entities is used to perform structural transformation, and finally the obtained component modules are spliced ​​to obtain a simplified bridge tower model;

[0044] See also Figure 4 Constructive solid geometry (CSG) is a technology for computer solid modeling display applications. In CSG, simple geometric bodies can be combined using logical operators to form complex surfaces or objects. CSG can construct or express very complex models using simple geometric bodies. The simplest geometric body is called a voxel, which is usually the simplest geometric body in shape, such as a cube, cylinder, prism, pyramid, sphere, cone, etc. Constructing an object is to combine voxels together according to the Boolean logic of set theory. These operations include: union, intersection, and complement.

[0045] Step 4, the model obtained in step 3 is spliced ​​by positive and negative plane fitting to obtain a LOD2 hierarchical model;

[0046] Step 5: Build an OBB bounding box tree based on the LOD4 model and further simplify it using the OBB bounding box concept.

[0047] See also Figure 3 The principle of the conversion from LOD4 to LOD1 in the present invention is the OBB bounding box concept. The bounding box tree of the complex bridge model is constructed in a top-down manner. The OBB bounding box segmentation process is as follows: Figure 7 As shown, the basic method is to continuously take the center point of the longest axis to locate the split plane, divide the line, and establish the basic component bounding box. Specifically include:

[0048] 1) Obtain model surface data; 2) Calculate the center point of the root node of the surface model OOB bounding box tree; 3) Calculate the root node OBB bounding box; 4) Determine the splitting axis and splitting point of the OBB bounding box; 5) Locate the bounding box splitting plane; 6) Use the splitting plane to split the geometric elements into two subsets; 7) Determine whether the segmentation requirements are met; 8) If the judgment result is "yes", end the process; if the judgment is "no", return to step 2) and continue with the following steps until the end.

[0049] Step 5.1, after constructing the bounding box in step 4, calculate the coordinates of the geometric center point of each layer of the bounding box, extract the basic component information in each bounding box, and obtain the LOD1 level model with the lowest level of detail description;

[0050] Step 5.2, determine whether the model of each level has been converted. If not, repeat steps 2 to 4.1 to convert again;

[0051] Step 6. After the conversion of each level of the model is completed, all the models of the levels are converted into b3dm format data, and then they are organized in TileSet.josn format. Finally, the TileSet.josn format is converted into 3d-tiles structure data: the conversion is completed.

[0052] The lightweight three-dimensional tile conversion method of the component-level extra-large suspension bridge engineering information model of the present invention has the advantages of simple method and high execution efficiency, and effectively solves the problem of slow loading speed of the extra-large suspension bridge engineering information model on ordinary computers. It is suitable for loading suspension bridge engineering information models under different scenario requirements, and is particularly suitable for loading models of different detail levels for multiple scenario requirements.

[0053] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A lightweight three-dimensional tile conversion method for a component-level super-large suspension bridge engineering information model, characterized in that: The steps include: S1, reading the data of the suspension bridge engineering information model, converting the suspension bridge engineering information model into a triangular mesh model to obtain a triangular mesh model, and associating the obtained triangular mesh model with the component semantic attribute information to obtain an LOD4 hierarchical model; S2, obtaining external setting parameters, semantically filtering the LOD4 hierarchical model according to the display scene and user needs, and removing the specified bridge entity components to obtain the LOD3 hierarchical model; S3, based on the space partition binary tree BSP principle, convert the LOD3 hierarchical model into the LOD2 hierarchical model; S4, establishing an OOB bounding box tree on the LOD4 level model, and calculating the coordinate information of the geometric center point under the bounding box tree; S5. By calculating the coordinates of the geometric center point of each bounding box, the basic component information in each bounding box is extracted to obtain the LOD1 level model with the lowest level of detail description; S6. Convert all hierarchical models obtained in the above steps into b3dm format data, organize them in TileSet.josn format, and finally convert TileSet.josn format into 3d-tiles structure data to complete the conversion.

2. The lightweight three-dimensional tile conversion method of the component-level super-large suspension bridge engineering information model according to claim 1 is characterized by: In step S2, the external setting parameters include viewing angle height and display ratio; the semantic filtering removes the specified bridge entity components including the parallel joints and main trusses in the stiffening beams.

3. The lightweight three-dimensional tile conversion method of the component-level super-large suspension bridge engineering information model according to claim 1 is characterized by: In step S3, the following steps are specifically included: Step S31, obtaining bridge deck information of the suspension bridge, including basic parameters of the bridge deck centerline; Step S32, dividing the bridge into positive and negative spaces along the bridge deck centerline according to the left-right symmetrical central axis distribution obtained in step S31, storing the bridge deck part by using the repetitiveness of the components, and selecting the positive or negative plane to continue the BSP tree division simplification; Step S33, simplifying the bridge tower part by using the idea of ​​constructive solid geometry; Step S34: Flip the bridge body part in the positive space or negative space and obtain the LOD2 level model by fitting and splicing the positive and negative planes.

4. The lightweight three-dimensional tile conversion method of the component-level super-large suspension bridge engineering information model according to claim 1 is characterized by: The BSP is an algorithm that effectively determines the positional relationship between a three-dimensional model and a polyhedron with dense distribution in some areas. It divides the partitioning plane into a tree, and each partitioning plane divides the space into two front and back areas, which are called positive and negative planes. The two planes obtained are further divided into smaller spaces until the division meets the set requirements; each storage node of the BSP tree represents a partitioning plane, the left subtree of the node represents the positive space of the partitioning plane, and the right subtree of the node represents the negative space of the partitioning plane.

5. The lightweight three-dimensional tile conversion method of the component-level super-large suspension bridge engineering information model according to claim 1 is characterized by: In step S4, the bounding box tree is established by adopting a top-down method to construct a bounding box tree of a complex bridge body model, by continuously taking the center point of the longest axis to locate the split plane, divide the line, and then establish a basic component bounding box tree.

Images