Geometric processing method and unit based on multi-level unstructured grid

Abstract

The invention discloses a geometric processing method and unit based on a multi-level unstructured grid.The geometric processing method based on the multi-level unstructured grid.The geometric processing method based on the multi-level unstructured grid.The geometric processing method based on the multi-level unstructured grid.The geometric processing method based on the multi-level unstructured gridcomprises the following steps that application scenes of geometric processing of the unstructured grid.The demand parameters corresponding to the application scenes are obtained, performing grid simplification and / or local refinement on the unstructured grid according to the demand parameters; establishing geometric background data based on a multi-level unstructured grid; according to a dimension coefficient factor (DCF) expression, calculating a DCF value of each unit in the geometric background data, and optimizing a search sequence based on a dimension coefficient factor to obtain a DCF value sorting result; according to the DCF value sorting result, geometric judgment is carried out on the corresponding units according to the DCF values from high to low in sequence, and geometric processing of the unstructured grid is achieved. The accurate geometric judgment can reduce the number of times of invalid geometric judgment, and the effect is especially obvious in a local encryption grid.

Description

Technical Field

[0001] This invention relates to the field of computer graphics, specifically to a geometric processing method and unit based on multi-level unstructured meshes. Background Technology

[0002] In the field of computer graphics, unstructured meshes are widely used due to their strong adaptability to complex geometric boundaries and irregular physical domains. However, the irregularity of unstructured meshes also brings significant computational challenges. Due to the lack of a unified indexing mechanism in the geometric processing, it is difficult to complete the task directly through simple mathematical calculations. It is necessary to rely on traversal search or complex topological traversal algorithms, resulting in low efficiency in geometric processing. This has become a key bottleneck restricting real-time rendering, interactive editing, and large-scale simulation.

[0003] Therefore, it is necessary to optimize the geometric processing methods based on multi-level unstructured meshes to improve the performance and interactive experience of graphics applications. Summary of the Invention

[0004] The purpose of this invention is to provide a geometric processing method and unit based on multi-level unstructured meshes, in order to solve the technical problem in the field of computer graphics that the geometric processing of unstructured meshes lacks a unified indexing mechanism, making it difficult to complete directly through simple mathematical calculations, and requiring traversal search or complex topological traversal algorithms, resulting in low efficiency in geometric processing.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] In a first aspect, the present invention provides a geometric processing method based on a multi-level unstructured mesh, comprising the following steps:

[0007] S1. Determine the application scenario for the geometric processing of the unstructured mesh, obtain the requirement parameters for the corresponding application scenario, and simplify and / or refine the unstructured mesh according to the requirement parameters.

[0008] S2. Establish geometric baseline data based on multi-level unstructured grids;

[0009] S3. Calculate the DCF value of each unit in the geometric background data according to the DCF expression, and optimize the search order based on the dimension coefficient factor to obtain the DCF value sorting result;

[0010] S4. Based on the DCF value sorting results, perform geometric determination on the corresponding cells in descending order of DCF value to achieve geometric processing of unstructured meshes. This precise geometric determination can reduce the number of invalid geometric determinations, especially in locally refined meshes.

[0011] This solution employs an adaptive approach to geometric processing methods for unstructured meshes, establishing baseline geometric data tailored to specific scene types and characteristics. It also introduces a Dimension Factor (DCF) to optimize the search order, improving the accuracy of geometric determinations and reducing the number of invalid geometric decisions. This addresses the inefficiency of relying on traversal searches or complex topological traversal algorithms for index building. This technical solution establishes a unified indexing mechanism for geometric processing of unstructured meshes, enabling direct completion through simple mathematical calculations without relying on traversal searches or complex topological traversal algorithms, thus effectively improving the efficiency of geometric processing.

[0012] Furthermore, in step S1, after obtaining the requirement parameters for the corresponding application scenario, a multi-level LOD model with different resolutions is constructed based on the determined application scenario and the corresponding requirement parameters. The unstructured mesh is simplified and locally refined to reduce the establishment of unstructured mesh background data and accelerate the construction of geometric index.

[0013] Furthermore, in step S1, the required parameters are one or a combination of several of the following: frame rate, memory usage, boundary layer accuracy, pressure gradient capture, real-time interaction efficiency, and collision detection efficiency.

[0014] Furthermore, the mesh simplification is one or a combination of visual fidelity simplification, far-field mesh simplification, occlusion area mesh simplification, and non-contact area simplification; the local refinement includes gradient area refinement and / or stress area refinement.

[0015] Furthermore, in step S2, the geometric baseline data of the unstructured mesh is established. First, the hierarchical structure of the unstructured mesh is constructed to obtain multiple leaf nodes after division and store the corresponding original geometric data. Then, each leaf node is represented by a structure. Then, each structure is spatially indexed to obtain the geometric baseline data of the unstructured mesh.

[0016] Furthermore, the construction of the hierarchical structure of the unstructured mesh includes the following steps:

[0017] S2.1 Starting from the root node of the unstructured mesh, divide the geometric units in the current node into at least two subsets to obtain a corresponding number of new nodes;

[0018] S2.2 Determine whether the number of geometric units in each of the new nodes satisfies the termination condition;

[0019] Each geometric unit corresponding to the new node that meets the termination condition is taken as the remaining unit, and step S2.3 is continued.

[0020] The new nodes that do not meet the termination condition are successively designated as the current nodes, and the process returns to step S2.1.

[0021] S2.3 Place the remaining units into leaf nodes and store the corresponding original geometric data; realize the construction of the hierarchical structure of the unstructured mesh.

[0022] Furthermore, in step S3, the DCF expression is:

[0023] ;

[0024] In the formula, For the first The volume of each unit; For the first Surface area of ​​each unit; For the first The maximum side length of each unit; This is the average value of the corresponding attribute for all cells in the grid; For adjustable weighting coefficients, satisfying .

[0025] Secondly, the present invention also discloses a geometric processing unit based on a multi-level unstructured mesh, utilizing the geometric processing method based on a multi-level unstructured mesh as described above, comprising:

[0026] The application scenario determination module is used to determine the application scenarios for the geometric processing of unstructured meshes;

[0027] A geometric background data construction and storage module is used to build and store geometric background data of unstructured grids, wherein the geometric background data includes multiple cells;

[0028] The DCF value calculation and sorting module calculates the DCF value of each unit in the geometric background data according to the DCF expression, and optimizes the search order based on the dimension coefficient factor to obtain the DCF value sorting result.

[0029] The geometry determination module performs geometry determination on the corresponding cells in descending order of DCF value based on the DCF value sorting result, thereby realizing the geometry processing of unstructured meshes.

[0030] Thirdly, the present invention also discloses an electronic device, characterized in that it includes a processor and a memory, the processor being coupled to the memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the geometric processing method based on a multi-level unstructured mesh as described in claim 7.

[0031] Fourthly, the present invention also discloses a computer-readable storage medium for storing a computer program, characterized in that the computer program is executed by a processor to implement the geometric processing method based on a multi-level unstructured mesh as described in claim 7.

[0032] The present invention has the following beneficial effects: The geometric processing method based on multi-level unstructured meshes disclosed in this invention establishes geometric baseline data under unstructured meshes in an adaptive manner according to specific scene types and characteristics. It also introduces a dimensionality factor to optimize the search order, improving the accuracy of geometric determination under unstructured meshes and reducing the number of invalid geometric determinations. This solves the problem of low efficiency in indexing by relying on traversal search or complex topological traversal algorithms. The technical solution of this embodiment establishes a unified indexing mechanism in the geometric processing of unstructured meshes, which can be completed directly through simple mathematical calculations without relying on traversal search or complex topological traversal algorithms, effectively improving the efficiency of geometric processing. Attached Figure Description

[0033] To make the objectives, technical solutions, and advantages of the invention clearer, the invention will now be described in further detail with reference to the accompanying drawings, wherein:

[0034] Figure 1 This is a flowchart of a geometric processing method based on multi-level unstructured meshes according to an embodiment of the present invention.

[0035] Figure 2 These are multi-level LOD models (Model 1, Model 2, Model 3, Model 4) with different resolutions in embodiments of the present invention.

[0036] Figure 3 The images shown are partial screenshots of Model 1 and partial unit numbers from an embodiment of the present invention.

[0037] Figure 4 The images shown are partial screenshots of Model 2 and partial unit numbers from an embodiment of the present invention.

[0038] Figure 5 The images shown are screenshots of part of Model 3 and some unit numbers from an embodiment of the present invention.

[0039] Figure 6 The image shows a partial screenshot of Model 4 and some unit numbers from an embodiment of the present invention. Detailed Implementation

[0040] The technical solutions of some embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments disclosed in the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments disclosed in the present invention are within the scope of protection of the present invention. It should be noted that in the drawings with reference numerals, the same reference numerals and letters represent similar parts. Once a part is defined in a drawing, it will not be defined and explained again in subsequent drawings.

[0041] This invention can be applied to the field of computer graphics, especially to the geometric processing of unstructured meshes. It solves the technical problem in computer graphics where the geometric processing of unstructured meshes lacks a unified indexing mechanism, making it difficult to complete directly through simple mathematical calculations and requiring reliance on traversal searches or complex topological traversal algorithms, resulting in low efficiency. This invention constructs an index database of geometry under unstructured meshes based on different application scenarios and their characteristics. Then, it improves the efficiency of geometric processing under unstructured meshes by constructing the geometric index and calculating DCF values ​​based on multi-threaded methods.

[0042] The geometric processing method and elements based on multi-level unstructured meshes disclosed in this invention have the following technical advantages:

[0043] 1. In this invention, for geometric processing under unstructured meshes, adaptive mesh simplification and local refinement are performed based on the specific scene type and scene characteristics, reducing the computational load of mesh processing from the source. Simultaneously, geometric baseline data under unstructured meshes is established, solving the problem of inefficient index building relying on traversal search or complex topological traversal algorithms.

[0044] 2. This invention also introduces a Dimension Factor (DCF) to optimize the search order, improve the accuracy of geometric determination under unstructured grids, and reduce the number of invalid geometric determinations.

[0045] 3. The present invention also optimizes geometry processing under multi-threaded unstructured grids by concurrently executing the tasks of establishing the background data and calculating the DCF value, thereby accelerating the construction of the data index structure and improving the accuracy of geometry processing.

[0046] To further illustrate the geometric processing method and unit based on multi-level unstructured meshes provided by the present invention, the following embodiments are disclosed.

[0047] In some embodiments, this embodiment provides a geometric processing method based on multi-level unstructured meshes, including the following steps:

[0048] S1. Determine the application scenario for the geometric processing of the unstructured mesh, obtain the requirement parameters for the corresponding application scenario, and simplify and / or refine the unstructured mesh according to the requirement parameters.

[0049] S2. Establish geometric baseline data based on multi-level unstructured grids;

[0050] S3. Calculate the DCF value of each unit in the geometric background data according to the DCF expression, and optimize the search order based on the dimension coefficient factor to obtain the DCF value sorting result;

[0051] S4. Based on the DCF value sorting results, perform geometric determination on the corresponding cells in descending order of DCF value to achieve geometric processing of unstructured meshes. This precise geometric determination can reduce the number of invalid geometric determinations, especially in locally refined meshes.

[0052] In this embodiment, for the geometric processing method under unstructured meshes, the geometric baseline data under unstructured meshes is established adaptively based on the specific scene type and scene characteristics. A Dimension Factor (DCF) is also introduced to optimize the search order, improve the accuracy of geometric determination under unstructured meshes, and reduce the number of invalid geometric determinations. This solves the problem of low efficiency in indexing by relying on traversal search or complex topological traversal algorithms. The technical solution of this embodiment establishes a unified indexing mechanism in the geometric processing of unstructured meshes, which can be completed directly through simple mathematical calculations without relying on traversal search or complex topological traversal algorithms, effectively improving the efficiency of geometric processing.

[0053] Specifically, the geometric determination refers to performing high-precision, computationally expensive geometric tests on the cells with higher DCF values ​​in the sorting results to confirm their exact relationship with the target geometry (such as points, lines, and surfaces). The result also confirms the exact relationship with the target geometry (such as points, lines, and surfaces), avoiding traversing all cells, reducing the number of invalid geometric determinations, and reducing computational resource overhead.

[0054] In some embodiments, in step S1, the application scenario options include computational fluid dynamics (CFD), real-time rendering / game engine, structural mechanics simulation, or virtual reality / interactive modeling.

[0055] In some embodiments, in step S1, the required parameters are one or a combination of several of the following: frame rate, memory usage, boundary layer accuracy, pressure gradient capture, real-time interaction efficiency, and collision detection efficiency.

[0056] In some embodiments, in step S1, after obtaining the requirement parameters for the corresponding application scenario, a multi-level LOD model with different resolutions is constructed based on the determined application scenario and the corresponding requirement parameters. The unstructured mesh is simplified and locally refined to reduce the establishment of unstructured mesh background data and accelerate the construction of geometric index.

[0057] In some embodiments, the mesh simplification is one or a combination of visual fidelity simplification, far-field mesh simplification, occlusion region mesh simplification, and non-contact region simplification; the local refinement includes gradient region refinement and / or stress region refinement.

[0058] If the application scenario is determined to be a real-time rendering / game engine, the required parameters are high frame rate and low memory usage. The goal of mesh simplification and local refinement is to prioritize visual fidelity, with significant simplification of the mesh in far-field / occluded areas.

[0059] If the application scenario is computational fluid dynamics (CFD), the required parameters are boundary layer accuracy and pressure gradient capture. The goal of mesh simplification and local refinement is to refine the mesh in the high gradient region and simplify the mesh in the far field.

[0060] If the application scenario is structural mechanics simulation, the required parameters are the accuracy of the stress concentration area, and the goal of mesh simplification and local refinement is to refine the high-stress area and simplify the far-field mesh.

[0061] If the application scenario is virtual reality / interactive modeling, the required parameters are real-time interaction and collision detection efficiency. The goal of mesh simplification and local refinement is to simplify non-contact areas while retaining details of interactive hotspots.

[0062] Specifically, the mesh simplification and local refinement are achieved by constructing a Level of Detail (LOD) model. The number of levels in the LOD model is determined by the different goals of simplification in different scenarios. For example, in real-time rendering / game engines, where the core requirements of high frame rate and low memory usage are met, the goal of mesh simplification and local refinement is to prioritize visual fidelity, with significant simplification of the mesh in distant / occluded areas.

[0063] Specifically, when simplifying unstructured meshes, first process them using Open3D, MeshLab, or Simplygon, then import them into Unity, and generate LOD models using the Auto LOD Generator plugin in Unity.

[0064] Here are some specific application examples to illustrate this:

[0065] If the application scenario is a real-time rendering / game engine scenario, select to generate a 4-level LOD model, where:

[0066] LOD0 level is the original model, containing all visible details, with a face count of 100%;

[0067] LOD1 level is a moderately simplified model that retains the main outline and features, removes faces that are not visible on the back, merges small parts, smooths non-critical chamfers, and has a face count of 40-80%.

[0068] LOD2 level is a significantly simplified model that merges adjacent material regions, replaces complex structures with simple geometry, removes non-contour features, and has a face count of 40-10%.

[0069] LOD3 is a minimalist model that retains only 2-10% of the convex hulls. If an object is meaningless at a distance, it can be completely removed.

[0070] If the application scenario is computational fluid dynamics, select to generate a 3-level LOD model, where:

[0071] LOD0 level is the boundary layer core region with an extremely fine anisotropic mesh. The first mesh layer has a very small thickness and a high normal stretch ratio. The laminar flow bottom layer has at least 10 to 15 mesh layers.

[0072] LOD1 level is the boundary layer and strong pressure gradient region (such as separation initiation point, shock root, strong reverse pressure gradient region).

[0073] The transition from LOD0 is smooth, avoiding excessive mesh jumps. The isotropic or slightly stretched mesh can clearly capture structures such as separated bubbles and weak shock waves.

[0074] LOD2 level represents the flow characteristic region (such as large-scale vortex structure, wake region, main shock wave, free shear layer), which is sufficient to distinguish large-scale turbulent structure and main wave system, and the grid scale matches the scale of the main flow characteristics.

[0075] If the application scenario is a structural mechanics scenario, select to generate a level 3 LOD model, where:

[0076] LOD0 is the core area of ​​stress concentration (such as the edge of a hole, the root of a notch, a rounded corner, etc.). The extremely fine mesh cells are very small in size, which can capture changes in stress gradient. In areas with large geometric curvature, the mesh fits the geometry.

[0077] LOD1 level stress concentration transition zone, medium density mesh, smooth transition from LOD0 region, avoid mesh size jump;

[0078] In the far-field region of LOD2 level, the relatively coarse mesh cells are larger, reducing the computational load.

[0079] If the application scenario is a non-virtual reality / interactive modeling scenario, select to generate a 4-level LOD model, where:

[0080] LOD0 level is a macro interaction level that retains all visual details and precise collision geometry;

[0081] LOD1 level close-up observation level, retaining key features and outlines, simplified collider;

[0082] LOD2 level is a mid-range environment level that retains basic shape features and performs bounding box level collision detection.

[0083] LOD3 level, distant background level, minimalist design, no collision or simple trigger body.

[0084] In some embodiments, in step S2, the geometric baseline data of the unstructured mesh is established by first constructing a hierarchical structure of the unstructured mesh, obtaining multiple leaf nodes after partitioning, and storing the corresponding original geometric data; then, each leaf node is represented by a structure; and then, each structure is spatially indexed to obtain the geometric baseline data of the unstructured mesh. Specifically, all nodes used to represent the structure are stored in an array and accessed by index.

[0085] In some embodiments, constructing the hierarchical structure of the unstructured mesh includes the following steps:

[0086] S2.1 Starting from the root node of the unstructured mesh, divide the geometric units in the current node into at least two subsets to obtain a corresponding number of new nodes;

[0087] S2.2 Determine whether the number of geometric units in each of the new nodes satisfies the termination condition;

[0088] Each geometric unit corresponding to the new node that meets the termination condition is taken as the remaining unit, and step S2.3 is continued.

[0089] The new nodes that do not meet the termination condition are successively designated as the current nodes, and the process returns to step S2.1.

[0090] S2.3 Place the remaining units into leaf nodes and store the corresponding original geometric data to realize the construction of the hierarchical structure of the unstructured mesh.

[0091] This hierarchical structure division allows control over the number of leaf nodes, ensuring that each leaf node contains only a small number of units, making the subsequent geometric decision calculations manageable. It also avoids over-division, which increases tree depth and memory overhead. During traversal, once a leaf node is reached, geometric decisions can be made directly on all its units.

[0092] In some embodiments, in step S2, a spatial index of the unstructured mesh is constructed through a bounding box hierarchy (BVH). The geometric elements under the unstructured mesh are input, all elements are traversed, and their minimum bounding boxes are calculated. All bounding boxes are merged to obtain the global bounding box of the entire mesh, which serves as the bounding box of the BVH root node. Specifically, the structure includes attributes such as the current node's bounding box and the left and right child node indices.

[0093] In some embodiments, in step S2.1, the subset is partitioned using the object median (OMS) partitioning method.

[0094] In some embodiments, starting from the root node of the unstructured mesh, the geometric units in the current node are divided into two subsets to obtain a corresponding number of new nodes, with the difference in the number of geometric units in the two subsets being 1 to 2; it is determined whether the number of geometric units in the two new nodes satisfies the termination condition; each geometric unit corresponding to the new node that satisfies the termination condition is taken as the remaining unit, and the new node that does not satisfy the termination condition is taken as the current node in turn, until the geometric units in all nodes satisfy the termination condition; then the geometric units in all nodes that satisfy the termination condition are placed as the remaining units into the leaf node, and their original geometric data is stored.

[0095] In some embodiments, the termination condition is: the number of units contained in the current node is less than or equal to a preset threshold. Preferably, the preset threshold ranges from 4 to 8, and more preferably, the preset threshold is 5.

[0096] In some embodiments, in step S3, the dimension factor (DCF) of the defined unit is:

[0097] For the first unstructured mesh Unit The DCF expression is:

[0098] (1)

[0099] In equation (1), For the first The volume of each unit; For the first Surface area of ​​each unit; For the first The maximum side length of each unit; This is the average value of the corresponding attribute for all cells in the grid; For adjustable weighting coefficients, satisfying .

[0100] Specifically, It can be configured according to the application scenario, such as focusing on volume in physical simulation and focusing on surface area in rendering.

[0101] In some embodiments, step S3 involves defining the dimensionality factor (DCF) of a cell, calculating the DCF of each cell, and optimizing the search order based on the DCF, including:

[0102] Define the dimension coefficient factor of the unit;

[0103] Retrieve the candidate unit list by index;

[0104] Calculate the DCF value for each cell and sort them in descending order of DCF value to obtain the sorted DCF value results.

[0105] In some embodiments, when performing steps S2 and / or S3, the tasks of steps S2 and / or S3 are divided into multiple subtasks, each subtask corresponding to a thread. By executing each subtask simultaneously through multiple threads, the progress of geometry processing can be further accelerated.

[0106] Specifically, these subtasks can be executed using the operating system's concurrent execution mechanism, thus forming a multi-threaded system.

[0107] In some embodiments, when each of the subtasks is executed, if one of the subtasks completes step S2, the corresponding thread directly executes step S3 to calculate the DCF value of the obtained unit and sorts them in descending order of DCF value, prioritizing geometric determination of units with high DCF; if other threads have not completed step S2, they continue to execute step S2 until step S2 is completed and then continue to execute step S3.

[0108] After step S2 of a certain thread is completed, the amount of data in the current cell index list is determined. If the amount of data in the current cell index list is large, a new thread is created to execute step S3.

[0109] In some embodiments, this embodiment also provides a geometric processing unit based on a multi-level unstructured mesh, used to implement the above-described geometric processing method based on a multi-level unstructured mesh, including:

[0110] The application scenario determination module is used to determine the application scenarios for the geometric processing of unstructured meshes;

[0111] A geometric baseline data construction and storage module is used to build and store geometric baseline data of unstructured grids, wherein the geometric baseline data includes multiple cells;

[0112] The DCF value calculation and sorting module calculates the DCF value of each unit in the geometric background data according to the DCF expression, and optimizes the search order based on the dimension coefficient factor to obtain the DCF value sorting result.

[0113] The geometry determination module performs geometric determination on the corresponding cells in descending order of DCF value based on the DCF value sorting result, thereby realizing the geometric processing of unstructured meshes.

[0114] In some embodiments, please refer to Figures 1 to 6 According to the technical solution in the above embodiments, geometric processing is performed on the unstructured mesh whose application scenario is determined to be real-time rendering / game engine.

[0115] First, based on the application scenario, the required parameters are determined to be high frame rate and low memory usage. The goal of mesh simplification and local refinement is to prioritize visual fidelity, with significant simplification of the mesh in distant / occluded areas.

[0116] When simplifying and refining unstructured meshes according to application scenarios, the Auto LOD Generator plugin in Unity automatically generates 4-level LOD (Level of Detail) models. The automatic generation mechanism includes the following steps:

[0117] D1. First, perform model analysis and check the polygon composition and vertex positions of the original model;

[0118] D2. Then, using the required parameters for the corresponding application scenario obtained in step S1, as well as the target parameters for mesh simplification and local refinement, set the Simplification Ratio parameter in the plugin to specify the degree of original mesh detail retention in percentage form.

[0119] Specifically, in step D2, based on the distance from the camera to the model in the scene, the simplified scaling parameters are set to 100% → 50% → 25% → 10%. At the same time, the Preserve Borders option is checked to try to keep the hard edges, UV seams, or contours of the model from being distorted.

[0120] D3. Construct the model hierarchy, setting LOD0 to 100% polygon count, LOD1 to 50%, LOD2 to 25%, and LOD4 to 10%. Use the Unity plugin AutoLOD Generator to automatically generate the 4-level LOD (Level of Detail) model. Figure 2 As shown, these are Model 1, Model 2, Model 3, and Model 4, respectively.

[0121] After the above processing, a mesh for constructing the geometric baseline data is obtained. Next, the bounding box hierarchy (BVH) spatial index of the mesh is constructed. Constructing the BVH spatial index involves inputting the geometric elements (triangles) under the mesh, using multi-threading technology to traverse the vertex positions of all elements, and calculating the minimum and maximum x, y, and z values ​​of their coordinate axes to obtain the minimum bounding box (defined by two points: min = (x_min, y_min, z_min) and max = (x_max, y_max, z_max)). All bounding boxes are merged to obtain the global bounding box of the entire mesh, which serves as the bounding box of the BVH root node. Using the median partitioning method, starting from the root node, the geometric elements in the current node are divided into two subsets on average until a termination condition is met. In this embodiment, the termination condition is: the number of elements contained in the current node is less than or equal to a preset threshold; in this embodiment, the threshold is set to 4. At this point, the remaining elements are placed in leaf nodes, and their original geometric data is stored. Nodes are represented using structs. The struct contains properties such as the current node's bounding box and the indices of its left and right child nodes. All nodes can be stored in an array and accessed by index.

[0122] If the current thread completes the construction of the bounding box hierarchy (BVH) spatial index, the next task is to obtain the candidate cell list through the index, calculate the DCF value of each cell, and sort them in descending order of DCF value. The current scene application is model rendering, so the focus is on surface area. In the DFC formula, α is set to 0.0, β is set to 0.9, and γ is set to 0.1. Substituting these values ​​into equation (1), we get the following formula:

[0123] (2)

[0124] In equation (2), For the first Surface area of ​​each unit; For the first The maximum side length of each unit; This is the average value of the corresponding attribute for all cells in the grid; For adjustable weighting coefficients, satisfying .

[0125] Calculate the DCF value of each cell in the mesh.

[0126] The highlighted areas in Model 1 are obtained through spatial indexing, such as... Figure 3 As shown (partial screenshot of Model 1), Figure 3The "1" in the table refers to the first element, with a corresponding DCF value of DCF1, and so on. The DCF values ​​of each element are calculated as shown in the table below. Only the DCF values ​​of 12 elements are shown here; the calculation method for the DCF values ​​of the remaining elements is the same and will not be shown here.

[0127]

[0128] The areas highlighted in yellow in Model 2 represent a subset of units obtained through spatial indexing, such as... Figure 4 As shown in the table below, the DCF values ​​are calculated. Only the DCF values ​​of 10 units are shown here. The DCF values ​​of the remaining units are calculated in the same way and will not be shown here.

[0129]

[0130] The areas highlighted in yellow in Model 3 represent a subset of units obtained through spatial indexing, such as... Figure 5 As shown in the table below, the DCF values ​​are calculated. Only the DCF values ​​of 7 units are shown here. The DCF values ​​of the remaining units are calculated in the same way and will not be shown here.

[0131]

[0132] The areas highlighted in yellow in Model 4 represent some of the units obtained through spatial indexing, such as... Figure 6 As shown in the table below, the DCF values ​​are calculated. Only the DCF values ​​of 5 units are shown here. The DCF values ​​of the remaining units are calculated in the same way and will not be shown here.

[0133]

[0134] From the four tables above and Figures 3 to 6 As can be seen in the diagram, the largest DCF in the yellow area of ​​Model 1 is the DCF. 12 In Model 2, the largest DCF is DCF. 10 In Model 3, the largest DCF values ​​are DCF4 and DCF5, while in Model 4, the largest DCF values ​​are DCF3 and DCF4. Geometric determination is prioritized for the highest DCF unit in each model to reduce the number of invalid geometric determinations.

[0135] In the description disclosed in this invention, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to has a specific orientation, or is constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the disclosure of this invention.

[0136] Unless the context otherwise requires, throughout the specification and claims, the term "comprising" is interpreted as open-ended and encompassing, meaning "including, but not limited to." In the description, terms such as "one embodiment," "some embodiments," "exemplary embodiment," "exemplary," or "some examples" are intended to indicate that a particular feature, structure, material, or characteristic associated with that embodiment or example is included in at least one embodiment or example disclosed in the invention. The illustrative representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics mentioned may be included in any suitable manner in any one or more embodiments or examples.

[0137] The terms "first" and "second" are used merely to distinguish different descriptive objects and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated; that is, they do not limit the position, order, priority, quantity, or content of the described objects. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the embodiments disclosed in this invention, unless otherwise stated, "a plurality of" means two or more.

[0138] In describing some embodiments, the term "connection" and its derivative expressions may be used. For example, the term "connection" may be used in describing some embodiments to indicate that two or more components have direct physical or electrical contact with each other. The embodiments disclosed herein are not necessarily limited to the content of this document.

[0139] "At least one of A, B, and C" has the same meaning as "at least one of A, B, or C," both including the following combinations of A, B, and C: only A, only B, only C, a combination of A and B, a combination of A and C, a combination of B and C, and a combination of A, B, and C. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B; "A and / or B" includes the following three combinations: only A, only B, and a combination of A and B.

[0140] The use of “configured as” in this article implies an open and inclusive language that does not exclude the applicability to or configuration of devices to perform additional tasks or steps.

[0141] In addition, the use of “based on” implies openness and inclusivity, because processes, steps, calculations or other actions “based on” one or more of the stated conditions or values ​​may in practice be based on additional conditions or values ​​beyond those stated.

[0142] As used herein, “about” and “approximately” include the values ​​stated and the average values ​​within an acceptable range of deviation from a particular value, wherein the acceptable range of deviation is determined by a person skilled in the art taking into account the measurement under discussion and the error associated with the measurement of the particular quantity (i.e., the limitations of the measurement system).

[0143] This document describes exemplary embodiments with reference to cross-sectional views and / or plan views, which are idealized exemplary drawings. In the drawings, the thickness of layers and regions is enlarged for clarity. Therefore, variations in shape relative to the drawings are contemplated due to, for example, manufacturing techniques and / or tolerances. Thus, exemplary embodiments should not be construed as limited to the shapes of the regions shown herein, but rather include shape deviations due to, for example, manufacturing processes. For example, etched regions shown as rectangular would typically have curved features. Therefore, the regions shown in the drawings are schematic in nature, and their shapes are not intended to show the actual shapes of the regions of the device, nor are they intended to limit the scope of the exemplary embodiments.

[0144] The above description is merely a specific embodiment of the present invention, but the scope of protection disclosed in the present invention is not limited thereto. Any variations or substitutions conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection disclosed in the present invention. Therefore, the scope of protection disclosed in the present invention should be determined by the scope of the claims.

Claims

1. A geometric processing method based on multi-level unstructured meshes, characterized in that, Includes the following steps: S1. Determine the application scenario for the geometric processing of the unstructured mesh, obtain the requirement parameters for the corresponding application scenario, and simplify and / or refine the unstructured mesh according to the requirement parameters. S2. Establish geometric baseline data based on multi-level unstructured grids; S3. Calculate the DCF value of each unit in the geometric background data according to the DCF expression, and optimize the search order based on the dimension coefficient factor to obtain the DCF value sorting result; S4. Based on the DCF value sorting results, perform geometric determination on the corresponding units in descending order of DCF value to achieve geometric processing of unstructured mesh.

2. The geometric processing method based on multi-level unstructured meshes according to claim 1, characterized in that, In step S1, after obtaining the requirement parameters for the corresponding application scenario, a multi-level LOD model with different resolutions is constructed based on the determined application scenario and the corresponding requirement parameters, and the unstructured mesh is simplified and locally refined.

3. The geometric processing method based on multi-level unstructured meshes according to claim 1, characterized in that, In step S1, the required parameters are one or a combination of several of the following: frame rate, memory usage, boundary layer accuracy, pressure gradient capture, real-time interaction efficiency, and collision detection efficiency.

4. The geometric processing method based on multi-level unstructured meshes according to claim 2, characterized in that, The mesh simplification is one or a combination of visual fidelity simplification, far-field mesh simplification, occlusion area mesh simplification, and non-contact area simplification; the local refinement includes gradient area refinement and / or stress area refinement.

5. The geometric processing method based on multi-level unstructured meshes according to claim 1, characterized in that, In step S2, the geometric background data of the unstructured mesh is established. First, the hierarchical structure of the unstructured mesh is constructed to obtain multiple leaf nodes after division and store the corresponding original geometric data. Then, each leaf node is represented by a structure. Then, each structure is spatially indexed to obtain the geometric background data of the unstructured mesh.

6. The geometric processing method based on multi-level unstructured meshes according to claim 5, characterized in that, The construction of the hierarchical structure of the unstructured mesh includes the following steps: S2.1 Starting from the root node of the unstructured mesh, divide the geometric units in the current node into at least two subsets to obtain a corresponding number of new nodes; S2.2 Determine whether the number of geometric units in each of the new nodes satisfies the termination condition; Each geometric unit corresponding to the new node that meets the termination condition is taken as the remaining unit, and step S2.3 is continued. The new nodes that do not meet the termination condition are successively designated as the current nodes, and the process returns to step S2.

1. S2.3 Place the remaining units into leaf nodes and store the corresponding original geometric data; realize the construction of the hierarchical structure of the unstructured mesh.

7. The geometric processing method based on multi-level unstructured meshes according to any one of claims 1 to 6, characterized in that, In step S3, the DCF expression is: ； In the formula, For the first The volume of each unit; For the first Surface area of ​​each unit; For the first The maximum side length of each unit; This is the average value of the corresponding attribute for all cells in the grid; For adjustable weighting coefficients, satisfying .

8. A geometric processing unit based on a multi-level unstructured mesh, utilizing the geometric processing method based on a multi-level unstructured mesh as described in claim 1, characterized in that, include: The application scenario determination module is used to determine the application scenario of the geometric processing of the unstructured mesh, obtain the requirement parameters of the corresponding application scenario, and perform mesh simplification and / or local refinement on the unstructured mesh according to the requirement parameters. The geometric baseline data construction and storage module is used to build and store the geometric baseline data of unstructured grids; The DCF value calculation and sorting module calculates the DCF value of each unit in the geometric background data according to the DCF expression, and optimizes the search order based on the dimension coefficient factor to obtain the DCF value sorting result. The geometry determination module performs geometric determination on the corresponding cells in descending order of DCF value based on the DCF value sorting result, thereby realizing the geometric processing of unstructured meshes.

9. An electronic device, characterized in that, The device includes a processor and a memory, the processor being coupled to the memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to implement the geometric processing method based on a multi-level unstructured mesh as described in any one of claims 1 to 7.

10. A computer-readable storage medium for storing a computer program, characterized in that, The computer program is executed by a processor to implement the geometric processing method based on a multi-level unstructured mesh as described in any one of claims 1 to 7.

Images