LOD generation method of BIM model

Abstract

The invention relates to the technical field of building information, in particular to an LOD generation method of a BIM model, which comprises the following steps: S10, acquiring information parameters of all edges in a model grid, and acquiring the data volume and secondary error of each edge according to the information parameters, S20, deleting the edge length with the minimum secondary error, and S30, deleting the edge length with the minimum secondary error. S30, simulating a length value of a triangular mesh mapped to a screen according to a screen space error, and obtaining a pixel size of a longest side mapped to the screen and a mesh error according to the length value, and S40, when the model precision is switched, obtaining a new point with the minimum distance to all planes of the deleted side end points, and updating the new point to the model mesh. And if the grid error meets a preset value, generating the LOD level as a rendering target. By the adoption of the method, it can be guaranteed that the LOD contains rich information and complete building element representation, the LOD of the proper level is generated, manual intervention is reduced, and the generation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of building information technology, and in particular to a LOD generation method for a BIM model. Background Art

[0002] Building Information Modeling has become a widely adopted digital modeling method. With BIM, construction professionals can create, manage, and interactively use digital representations of construction projects. BIM models include not only the geometry of the building, but also information related to the building elements, such as materials, dimensions, costs, schedules, etc. This comprehensive information model helps improve the efficiency, accuracy, and collaboration of construction projects. In BIM, Level of Detail is a key concept, also known as LOD, which is used to describe the level of detail of the model. LOD defines the accuracy and completeness of each element in the model, and there are usually different LOD requirements at different stages of the construction project. Therefore, generating the appropriate level of LOD is crucial to ensuring the quality and usability of the BIM model.

[0003] At the current stage, the LOD generation method of BIM models usually requires a lot of manual intervention, including manual adjustment of geometry, adding information and attributes, etc. This leads to an increase in the time and cost of the generation process, and is also prone to human errors. It is also impossible to automatically and appropriately generate LODs for some complex building elements or scenarios, which makes it difficult for the current LOD generation method to effectively deal with complex building elements and scenarios. Summary of the invention

[0004] The present invention aims at the deficiencies in the prior art and provides the following technical solutions:

[0005] A method for generating LOD of a BIM model comprises the following steps:

[0006] S10: obtaining information parameters of all edges in the model grid, and obtaining the data volume and quadratic error of each edge according to the information parameters;

[0007] S20: Delete the edge length with the smallest quadratic error, re-establish a new point with the minimum distance to all planes of the deleted edge endpoint, and update the point to the model grid;

[0008] S30: simulating a length value of the triangular mesh mapped to the screen according to the screen space error, and obtaining a pixel size of the longest side mapped to the screen and a mesh error according to the length value;

[0009] S40: When switching the model accuracy, if the mesh error meets the preset value, the LOD level is generated as a rendering target.

[0010] As an improvement of the above technical solution, step S10 includes the following steps:

[0011] S11: Obtain information parameters of all points in the model grid, and record the length of the edge formed by the line connecting two points in the model grid;

[0012] S12: Calculate the scene bounding box, take the bounding box diameter as the maximum side length, perform interpolation calculation between the minimum side length and the maximum side length, and obtain the maximum folded side lengths of different LOD levels;

[0013] S13: Obtain the Q matrix of all points, and obtain the quadratic error of the edge according to the Q matrix.

[0014] As an improvement of the above technical solution, the method for obtaining the Q matrix depends on the following formula:

[0015]

[0016] Among them, p is the plane where point v is located, pp T is a symmetric matrix.

[0017] As an improvement of the above technical solution, step S20 includes the following steps:

[0018] S21: while satisfying that all edge lengths are less than the maximum folded edge length of the corresponding LOD level, delete the edge with the smallest quadratic error;

[0019] S22: establishing a new point in the model grid, the new point being a point with a minimum distance to the plane where the two endpoints of the deleted edge are located;

[0020] S23: Update the new point to the model grid according to the information parameters of the new point.

[0021] As an improvement of the above technical solution, the acquisition of the length value of the triangular mesh mapped to the screen depends on the following formula:

[0022]

[0023] Among them, θ represents the angle of triangle mesh mapping, ε represents the side length of triangle mesh, d represents the viewing distance from the point to the triangle mesh, and x represents the length of the front plane of the viewing cone.

[0024] As an improvement of the above technical solution, the acquisition of the pixel size of the longest side mapped to the screen depends on the following formula:

[0025]

[0026] Among them, SSE(x) is the length of the horizontal coordinate mapped to the screen, SSE(y) is the length of the vertical coordinate mapped to the screen, x represents the horizontal axis of the plane where the screen is located, y represents the vertical axis of the plane where the screen is located, Weight represents the width of the plane, and Height represents the height of the plane;

[0027] The grid error is the maximum pixel value mapped to the screen by the triangular grid with the maximum folded edge as the side length. The grid error is expressed as follows:

[0028]

[0029] Wherein, SSE(x) represents the length of the horizontal coordinate mapped to the screen.

[0030] As an improvement of the above technical solution, if the mesh error is acceptable, the LOD level with the maximum error within the acceptable range is selected to balance the data volume and the model appearance.

[0031] As an improvement of the above technical solution, when step S40 is executed, the pixel size of the longest side obtained in step S30 mapped to the screen is obtained in sequence from low to high according to the LOD level. If the pixel size is not less than the preset maximum error, a higher LOD level is entered for comparison until the pixel size is less than the preset maximum error, and then the LOD level is generated as the rendering target.

[0032] Beneficial effects of the present invention:

[0033] By adopting the method described in the present invention, it is possible to generate LOD of an appropriate level while ensuring that LOD contains rich information and complete representation of architectural elements, thereby reducing manual intervention and improving generation efficiency, being able to meet the needs of different projects and scenarios, and satisfying the processing requirements of large-scale architectural model data. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 The state change diagram before and after the edge deletion operation of the present invention;

[0035] Figure 2 is a schematic diagram of screen space error of the present invention;

[0036] Figure 3 A flow chart for generating the LOD model of the present invention;

[0037] Figure 4 It is a flowchart of LOD level switching of the present invention. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0039] At the current stage, the LOD generation method of BIM models usually requires a lot of manual intervention, including manual adjustment of geometry, adding information and attributes, etc. This leads to an increase in the time and cost of the generation process, and is also prone to human errors. It is also impossible to automatically and appropriately generate LODs for some complex building elements or scenarios, which makes it difficult for the current LOD generation method to effectively deal with complex building elements and scenarios.

[0040] To solve the above problems, see Figures 1 to 4 , the following embodiments are provided:

[0041] Embodiment 1

[0042] A method for generating LOD of a BIM model comprises the following steps:

[0043] S10: Obtain information parameters of all edges in the model grid, and obtain the data volume and quadratic error of each edge according to the information parameters.

[0044] Specifically, step S10 includes the following steps:

[0045] S11: Obtain information parameters of all points in the model grid, and record the length of the edge formed by the line connecting two points in the model grid.

[0046] S12: Calculate the scene bounding box, take the bounding box diameter as the maximum side length, perform interpolation calculation between the minimum side length and the maximum side length, and obtain the maximum folded side length of different LOD levels.

[0047] The amount of data between different LOD levels needs to present a certain step-like shape, so as to ensure that the amount of data changes to a certain extent when the model is switched. When controlling the amount of data generated by the model LOD, the length of the edge folding is used to control the amount of data for each level of LOD. Therefore, it is necessary to traverse the triangular mesh in the model, obtain the maximum and minimum edge lengths of the entire model, and evenly interpolate between the maximum edge length and the minimum folded edge length to achieve multiple levels of maximum folded edges. Then generate LOD models of different levels based on the maximum folded edge.

[0048] S13: Obtain the Q matrix of all points, and obtain the quadratic error of the edge according to the Q matrix.

[0049] The method for obtaining the Q matrix depends on the following formula:

[0050]

[0051] Among them, p is the plane where point v is located, pp T is a symmetric matrix.

[0052] S20: Delete the edge length with the smallest quadratic error, re-establish a new point with the minimum distance to all planes of the deleted edge endpoint, and update the point to the model grid.

[0053] like Figure 1 As shown, the folded edges v1 and v2 update the maximum folded edge length each time the edge length is folded. In the LOD model of this level, the triangle mesh formed by the maximum folded edge length and the edges less than the maximum folded edge length of this level is an inaccurate mesh triangle facet.

[0054] Specifically, step S20 includes the following steps:

[0055] S21: While satisfying that all edge lengths are smaller than the maximum folded edge length of the corresponding LOD level, delete the edge with the smallest quadratic error.

[0056] S22: Create a new point in the model grid, where the new point is the point with the minimum distance to the plane where the two endpoints of the deleted edge are located.

[0057] S23: Update the new point to the model grid according to the information parameters of the new point.

[0058] After the update, the LOD models are organized from large to small according to the maximum folded side length, that is, the LOD models are divided into several different levels according to the model's degree of refinement, from coarse to fine. This makes it convenient to directly select the maximum error model that satisfies the model's appearance (the error between the simplified model and the original model is the smallest).

[0059] The specific process of the first embodiment is as follows Figure 3 As shown, this is a flowchart for LOD model generation.

[0060] Embodiment 2

[0061] The switching of LOD levels affects the overall look and feel of the model. The amount of model data is directly proportional to the model look and feel. Ensuring the look and feel of the model at the cost of the minimum amount of model data is the focus of the LOD level switching method.

[0062] In the first embodiment, it is mentioned that the triangular mesh formed by the maximum folding side length is an inaccurate mesh. Therefore, reducing the impact of such triangular mesh on the model appearance is the key to LOD level switching. Based on this, the concept of screen space error is introduced, and the following method is provided to achieve LOD level switching, which specifically includes the following steps:

[0063] S30: Simulate the length value of the triangle mesh mapped to the screen according to the screen space error, and obtain the pixel size of the longest side mapped to the screen and the mesh error according to the length value.

[0064] The acquisition of the length value of the triangular mesh mapped to the screen depends on the following formula:

[0065]

[0066] Among them, θ represents the angle of triangle mesh mapping, ε represents the side length of triangle mesh, d represents the viewing distance from the point to the triangle mesh, and x represents the length of the front plane of the viewing cone.

[0067] The acquisition of the pixel size of the longest side mapped to the screen depends on the following formula:

[0068]

[0069] Among them, SSE(x) is the length of the horizontal coordinate mapped to the screen, SSE(y) is the length of the vertical coordinate mapped to the screen, x represents the horizontal axis of the plane where the screen is located, y represents the vertical axis of the plane where the screen is located, Weight represents the width of the plane, and Height represents the height of the plane;

[0070] In addition, ε is designated as the maximum folding side length, and the size of Δ(mesh) is calculated to represent the pixel size mapped to the screen.

[0071] If the mesh error is acceptable, the LOD level with the maximum error within the acceptable range is selected to balance the data volume and model appearance.

[0072] The grid error is the maximum pixel value mapped to the screen by the triangular grid with the maximum folded edge as the side length. The grid error is expressed as follows:

[0073]

[0074] Wherein, SSE(x) represents the length of the horizontal coordinate mapped to the screen.

[0075] S40: When switching the model accuracy, if the mesh error meets the preset value, the LOD level is generated as a rendering target.

[0076] When step S40 is executed, the pixel size of the longest side obtained in step S30 mapped to the screen is obtained in sequence from low to high LOD levels. If the pixel size is not less than the preset maximum error, a higher LOD level is entered for comparison until the pixel size is less than the preset maximum error, and the LOD level is generated as the rendering target.

[0077] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them.

Claims

1. A method for generating LOD of a BIM model, characterized in that: The steps include: S10: obtaining information parameters of all edges in the model grid, and obtaining the data volume and quadratic error of each edge according to the information parameters; S20: Delete the edge length with the smallest quadratic error, re-establish a new point with the minimum distance to all planes of the deleted edge endpoint, and update the point to the model grid; S30: simulating a length value of the triangular mesh mapped to the screen according to the screen space error, and obtaining a pixel size of the longest side mapped to the screen and a mesh error according to the length value; S40: When switching the model accuracy, if the mesh error meets the preset value, the LOD level is generated as a rendering target.

2. The LOD generation method of a BIM model according to claim 1, characterized in that: The step S10 comprises the following steps: S11: Obtain information parameters of all points in the model grid, and record the length of the edge formed by the line connecting two points in the model grid; S12: Calculate the scene bounding box, take the bounding box diameter as the maximum side length, perform interpolation calculation between the minimum side length and the maximum side length, and obtain the maximum folded side lengths of different LOD levels; S13: Obtain the Q matrix of all points, and obtain the quadratic error of the edge according to the Q matrix.

3. The LOD generation method of a BIM model according to claim 1, characterized in that: The method for obtaining the Q matrix depends on the following formula: Among them, p is the plane where point v is located, pp T is a symmetric matrix.

4. The LOD generation method of a BIM model according to claim 3, characterized in that: The step S20 comprises the following steps: S21: while satisfying that all edge lengths are less than the maximum folded edge length of the corresponding LOD level, delete the edge with the smallest quadratic error; S22: establishing a new point in the model grid, the new point being a point with a minimum distance to the plane where the two endpoints of the deleted edge are located; S23: Update the new point to the model grid according to the information parameters of the new point.

5. The LOD generation method of a BIM model according to claim 2, characterized in that: The acquisition of the length value of the triangular mesh mapped to the screen depends on the following formula: Among them, θ represents the angle of triangle mesh mapping, ε represents the side length of triangle mesh, d represents the viewing distance from the point to the triangle mesh, and x represents the length of the front plane of the viewing cone.

6. The LOD generation method of a BIM model according to claim 5, characterized in that: The acquisition of the pixel size of the longest side mapped to the screen depends on the following formula: Among them, SSE(x) is the length of the horizontal coordinate mapped to the screen, SSE(y) is the length of the vertical coordinate mapped to the screen, x represents the horizontal axis of the plane where the screen is located, y represents the vertical axis of the plane where the screen is located, Weight represents the width of the plane, and Height represents the height of the plane; The grid error is the maximum pixel value mapped to the screen by the triangular grid with the maximum folded edge as the side length. The grid error is expressed as follows: Among them, SSE(x ) Expressed as the horizontal coordinate length mapped to the screen.

7. The LOD generation method of a BIM model according to claim 6, characterized in that: If the mesh error is acceptable, the LOD level with the maximum error within the acceptable range is selected to balance the data volume and model appearance.

8. A method for generating LOD of a BIM model according to any one of claims 2 to 7, characterized in that: When step S40 is executed, the pixel size of the longest side obtained in step S30 mapped to the screen is obtained in sequence from low to high LOD levels. If the pixel size is not less than the preset maximum error, a higher LOD level is entered for comparison until the pixel size is less than the preset maximum error, and the LOD level is generated as the rendering target.

Images