Three-dimensional model dynamic deformation method responding to real-time data

Abstract

The invention discloses a three-dimensional model dynamic deformation method responding to real-time data, which comprises the following steps: S1, reading three-dimensional model grid data in Unity, and extracting model vertex coordinates; s2, determining the position of an actual monitoring deformation point on the three-dimensional model; s3, screening all grid vertexes with deformation displacement in the deformation area according to the actual deformation area influenced by the monitored deformation points; s4, determining the deformation displacement direction of the vertex of the grid in the deformation area range; and S5, determining the deformation displacement value of the vertex of the grid in the deformation area range, and controlling the deformation displacement speed of the vertex of the grid according to the deformation rate. According to the method, the real-time dynamic deformation capability is remarkably improved, the deformation effect is more natural and accurate, the applicability is wide, the universality is high, and the data processing flow is simplified to improve the data processing efficiency.

Description

Technical Field

[0001] The present invention belongs to the technical field of three-dimensional model deformation, and in particular to a three-dimensional model dynamic deformation method that responds to real-time data. Background Art

[0002] With the rapid development of computer graphics, virtual reality and augmented reality technologies, 3D models are increasingly used in various fields. From architectural design and civil engineering to film and television special effects and game development, 3D models have become an important tool for expressing complex spatial structures and dynamic scenes. However, in the actual application of 3D models, how to achieve the dynamic response of the model to the deformation monitoring data of the physical structure has always been a difficult problem that has troubled researchers and engineers.

[0003] From the perspective of 3D scene construction, the application of 3D models has evolved from the initial visualization of construction results to management and control applications based on visualization scenes, and then to the visualization of the scene throughout its life cycle, and integrated management and control of operation and maintenance. However, existing research on 3D model scene construction mainly focuses on the construction of static scene models and the unidirectional effect of dynamic data on static models. Static 3D models cannot achieve dynamic response to the deformation monitoring data of the physical structure, and cannot dynamically and intuitively visualize the structural changes of the model.

[0004] Domestic and foreign scholars' research in the field of dynamic deformation of three-dimensional models mainly focuses on the following aspects: First, simulation technology based on physical engines, which realizes dynamic deformation of models by simulating the physical properties of materials; second, deformation methods based on data-driven, which directly drives model deformation by inputting deformation data; third, predictive deformation combined with machine learning algorithms, which predicts future deformation trends by training models. However, most of these methods have problems such as high computational complexity, poor real-time performance, and unnatural deformation effects, which make it difficult to meet the needs of practical applications. Most of the existing research results focus on the dynamic construction and high-precision simulation of three-dimensional models.

[0005] The problems existing in the prior art are as follows:

[0006] (1) Insufficient dynamic response capability of static 3D models: Current 3D model technologies mostly focus on the construction and display of static scenes, and are unable to respond to deformation monitoring data of physical structures in real time and dynamically. This technical limitation results in the inability to intuitively reflect the actual dynamic changes of structures in application scenarios that require real-time monitoring of structures, such as bridge health monitoring and building structure deformation analysis, affecting the effective use of data and the timeliness of decision-making.

[0007] (2) Unnatural deformation and low accuracy: Traditional 3D model deformation methods usually rely on preset animations or scripts, which cannot accurately simulate the real deformation of physical structures caused by external factors (such as gravity, wind load, temperature change, etc.). In addition, these methods often result in stiff and unnatural deformation effects, which are difficult to meet the needs of high-precision monitoring and display.

[0008] (3) Lag in real-time data processing: When processing real-time deformation monitoring data, existing technologies often experience delays due to high algorithm complexity or limited data processing capabilities, resulting in untimely deformation feedback. This lag limits the application of 3D models in areas that require rapid response to deformation changes (such as earthquake early warning, geological disaster monitoring, etc.).

[0009] (4) Limited application scenarios due to technical limitations: Due to the above technical difficulties, static 3D models are limited in many application scenarios that require high precision, real-time performance, and dynamic response. These fields include but are not limited to engineering construction, transportation, geological exploration, environmental protection, etc. They all require 3D models to be able to reflect the changes in actual structures in real time and accurately to support effective monitoring, early warning, and decision-making. Summary of the invention

[0010] The present invention aims to provide a three-dimensional model dynamic deformation method responding to real-time data, so as to solve the above problems.

[0011] The technical solution of the present invention is: a three-dimensional model dynamic deformation method responding to real-time data, comprising:

[0012] S1: Read the 3D model mesh data in Unity and extract the model vertex coordinates;

[0013] S2: Determine the actual monitoring deformation point position on the three-dimensional model;

[0014] S3: According to the actual deformation area affected by the monitoring deformation point, all mesh vertices that have deformation displacement in the deformation area are screened;

[0015] S4: determining the deformation displacement direction of the mesh vertices within the deformation area;

[0016] S5: Determine the mesh vertex deformation displacement value within the deformation area, and control the mesh vertex deformation displacement speed according to the deformation rate.

[0017] Preferably, S1 specifically includes: importing the three-dimensional model into Unity, obtaining the triangular mesh data of the three-dimensional model based on the Mesh class, extracting the local coordinates of the model mesh vertices, and multiplying the local coordinates with the transformation matrix to obtain the world coordinates of the model mesh vertices;

[0018] Unity's coordinate system includes two expressions: local coordinates and world coordinates. The local coordinate system is the coordinate system relative to its parent object, which is used to describe the position relationship within the object. The world coordinate system corresponds to the coordinate system of the entire Unity world space and is not affected by the object hierarchy. To convert local coordinates to world coordinates to determine the specific position of the model in Unity space, the following calculation formula can be used:

[0019] P w =M parent ·P l +P parentw

[0020] P w and P l Represent the world coordinates and local coordinates of the model, M parent and P parent They represent the coordinate transformation matrix and world coordinates of the model's parent object respectively.

[0021] Preferably, S2 is specifically: determining the actual monitoring deformation point position in the three-dimensional model, outputting the local coordinates of the deformation monitoring point through the collision of rays and the model, and multiplying the local coordinates of the actual monitoring deformation point with the transformation matrix to obtain the world coordinates of the actual monitoring deformation point; wherein, the actual monitoring deformation point position is determined by converting the corresponding position of the real-world monitoring point to the corresponding coordinate position after the three-dimensional model; that is, a monitoring point is set at each interval of the real tunnel slope, and the deformation data transmitted by the monitoring point must accurately correspond to the three-dimensional model.

[0022] Preferably, S3 is specifically: traversing the mesh vertex coordinate array, calculating the Euclidean distance between the actual deformation monitoring point and the mesh vertex, and screening all mesh vertices within the deformation area according to the actual deformation area affected by each actual monitoring deformation point; wherein, the three-dimensional model will eventually be imported into the rendering engine in the form of a mesh, and the mesh is composed of a series of vertices and triangular faces connecting different vertices; the mesh vertices have coordinates.

[0023] Preferably, S4 specifically includes:

[0024] S41: Determine the position of the triangle mesh where the deformation point is located: traverse all triangle mesh vertices, record the coordinates of the monitored deformation point as P, the coordinates of any triangle mesh vertex as A, B, C, calculate the cross product results of vectors PA, PB, PC, and when the results have the same sign, the deformation point is located in the triangle mesh;

[0025] S42: Calculate the vertical vector of the triangular mesh: calculate two vectors in a clockwise direction according to the coordinates of the three vertices of the triangular mesh where the monitoring deformation point is located, and obtain the vertical vector of the triangular mesh by cross-product of the two vectors, with the direction pointing to the outside of the model;

[0026] S43: Determine the deformation displacement direction of each mesh vertex in the deformation area: normalize the vertical vector of the triangular mesh; when the deformation value of the monitoring deformation point is negative, the mesh vertex is deformed and displaced along the normalized vertical vector, and the model deformation form is stretching; when the deformation value of the monitoring deformation point is positive, the deformation displacement direction of the mesh vertex is the opposite of the normalized vertical vector, and the model deformation form is compression.

[0027] Preferably, S5 specifically includes: the deformation value of the deformation monitoring point is given based on the actual deformation monitoring value, and the deformation displacement value of each mesh vertex in the deformation range decreases as the distance between the deformation monitoring point and the mesh vertex increases; the deformation displacement value of the mesh vertex farthest from the monitoring deformation point in the deformation range is zero, and the deformation displacement values ​​of other mesh vertices are linearly interpolated according to their distance from the monitoring deformation point; the deformation rate is given based on the actual deformation monitoring value, and the deformation rate is added to the mesh vertex that has undergone deformation displacement, and the deformation rate controls the moving speed of the mesh vertex along the deformation displacement direction. When the deformation displacement of the mesh vertex reaches the set value, a three-dimensional model structure deformation is completed.

[0028] Preferably, in S42, when the mesh vertices are deformed and displaced, the direction of the vertical vector of the triangular mesh changes accordingly; during the process of the mesh vertices being deformed and displaced, the mesh vertex coordinates are updated in real time and the vertical vector of the triangular mesh is calculated.

[0029] The beneficial effects of the present invention are:

[0030] (1) Significant improvement in real-time dynamic deformation capability: Compared with the method for dynamically creating a BIM model for a highway tunnel project based on the Unity platform (existing technology 1), the present invention has significantly improved real-time dynamic deformation capability. The present invention can not only dynamically create a model, but more importantly, it can realize the real-time response and dynamic deformation of the three-dimensional model to the deformation monitoring data of the physical structure, and support high-precision monitoring and timely decision-making in complex environments (such as bridge health monitoring, building structure deformation analysis, etc.).

[0031] (2) The deformation effect is more natural and accurate: Traditional methods and existing technologies mostly rely on preset animations or complex physical calculations, resulting in stiff, unnatural deformation effects and low precision. The present invention precisely controls the deformation direction and displacement value of the mesh vertices, combines physical simulation and data-driven technology, and makes the deformation effect more natural and realistic, conforming to actual physical laws, and meeting the needs of high-precision monitoring and display.

[0032] (3) Wide applicability and high versatility: Existing technologies are often limited to specific application scenarios (such as highway tunnel engineering), while the technical solution of the present invention has wide applicability and can be applied to dynamic deformation of three-dimensional models in various fields such as bridges, buildings, and geology. At the same time, the method of the present invention does not rely on specific algorithm implementation and has higher versatility and scalability.

[0033] (4) Simplify the data processing process and improve data processing efficiency: The present invention integrates a data processing module in Unity, simplifies the integration process of deformation monitoring data and 3D model mesh data, and realizes real-time data transmission and processing. Compared with the complex data processing algorithms and processes in the prior art, the present invention can significantly improve the efficiency and accuracy of data processing, and provide strong support for rapid response to deformation changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic flow chart of a method for dynamic deformation of a three-dimensional model in response to real-time data provided by an embodiment of the present invention;

[0035] Figure 2 A cube mesh vertex displacement effect diagram provided by an embodiment of the present invention;

[0036] Figure 3 A code implementation diagram of mesh vertex displacement provided by an embodiment of the present invention;

[0037] Figure 4 A tunnel and real-scene model grid visualization diagram provided by an embodiment of the present invention

[0038] Figure 5 A monitoring point location and instantiation diagram provided for an embodiment of the present invention;

[0039] Figure 6 A monitoring point control deformation area division diagram provided in an embodiment of the present invention;

[0040] Figure 7 This is a real-time deformation effect diagram provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it. The implementation methods of the present invention are not limited thereto.

[0042] Example 1

[0043] like Figure 1As shown, the present invention includes acquiring three-dimensional model mesh data and extracting vertex coordinates, determining the coordinates of actual detection deformation points on the three-dimensional model, screening mesh vertices that have undergone deformation and displacement, determining the deformation displacement direction of the mesh vertices, and determining the deformation value and deformation rate of the mesh vertices; and finally measuring whether the deformation of the model structure is completed by judging whether the deformation displacement of the mesh vertices reaches a set value.

[0044] like Figure 2 As shown in the figure, in three-dimensional space, the deformation of a cube is taken as an example: it has a total of 6 faces, each of which contains two triangular faces. For two triangular faces on the same square face, their normals are the same, so the overlapping mesh vertices do not need to be stored repeatedly. However, the normals of different square faces are different. When deformation rules are only applied to the four vertices of a single face, the model may have problems such as cracks or openings, affecting the integrity of the model. Therefore, overlapping vertices need to be recorded repeatedly; to ensure the continuity and closure of the model, when applying deformation rules for specific vertices, the vertices that overlap with it need to be given the same displacement rules. Taking a cube as an example, it contains a total of twenty-four mesh vertices. Figure 2 The example shown shows the result of selecting the vertices on the right face and displacing them along the X, Y, and Z axes by (0.5, 0.5, 0.5).

[0045] Figure 3 The code implementation for the above operations.

[0046] like Figure 4 As shown, a three-dimensional tunnel model that realizes dynamic deformation based on settlement monitoring data is taken as an example. The method of the present invention can also be used for other types of models such as bridges, and is not limited to tunnels.

[0047] like Figure 5 As shown, a point on the tunnel outline is selected as the reference point and designated as the parent object of all monitoring points. At this time, the reference point is the coordinate origin relative to the monitoring point. On this basis, according to the location data of the measurement point layout, the monitoring point is translated along the X and Y axes for a fixed distance on the model to determine the location of the monitoring point. Taking the IV-level surrounding rock as an example, the red ball model is instantiated at the coordinate position of the monitoring point for visualization, and the positioning effect is shown in the figure.

[0048] in, Figure 5 The yellow triangle frame in the middle is the reference point.

[0049] The reason for the above translation is to locate the five monitoring points 1, 2, 3, 4 and A in the figure. Assuming that the reference point is in the plane coordinate (0, 0), point 2 starts from the reference point, translates 1m to the right, and translates 2m upward (if right and upward are positive directions), then the coordinate representation of monitoring point 2 is (1m, 2m); for issues related to the Z axis, refer to the "layout length" part of the claims. Assuming that the constructed layout length is 3m, the next reference point of the straight tunnel is at the same horizontal position 3m adjacent to the current reference point.

[0050] like Figure 6 As shown in the figure, in actual construction measurement, the value recorded by the monitoring point only represents the measurement value of the specific location, and cannot fully reflect the deformation of the overall cross-sectional profile of the tunnel. For the cross section, the deformation area controlled by each measuring point is bounded by the midpoint of the arc between the measuring point and the adjacent measuring point.

[0051] Taking the tunnel as an example, the deformation area of ​​the plane control of each measuring point is bounded by the midpoint of the arc between the measuring point and the adjacent measuring point, as shown in the figure; the longitudinal control area is the length of the construction and lofting of this section of the tunnel. Different accuracy requirements can be set in Unity. The setting on the left side of the figure is 3m.

[0052] like Figure 7 As shown, according to the actual monitoring data, within the deformation area of ​​the monitoring point, the mesh vertices show a cumulative deformation effect (real-time dynamic deformation).

[0053] In addition, in S4, in 3D computer graphics, triangles are the most basic geometric units for constructing and representing complex 3D shapes. The vertex deformation calculation of triangular meshes is relatively simple, because each vertex is only connected to three adjacent vertices, and the deformation calculation is more efficient in real-time dynamic deformation scenes. The effects and numerical stability of simulating various deformations such as stretching, compression, and bending are better.

[0054] According to the cross product of vectors and the right-hand screw rule, the direction of AB^AM (^ represents the cross product, and the arrow symbol above the letter is omitted here) is pointing outward from the screen, and AB^AN is also pointing outward from the screen, but the direction of AB^AO is pointing inward from the screen, so M and N are on the same side of the straight line AB, and M and O are on both sides of the straight line AB. In actual calculations, only the positive and negative values ​​of the cross product need to be considered. Assuming the coordinates of the above points are A(0,0), B(4,0), M(1,2), N(3,4), O(3,-4), then:

[0055] AB^AM=(4,0)^(1,2)=4*2-0*1=8

[0056] AB^AN=(4,0)^(3,4)=4*4–0*3=16

[0057] AB^AO=(4,0)^(3,-4)=4*-4–0*3=–16

[0058] From the above values, we can know that the direction of the vector after the cross product can be determined based on the sign of the value. That is, if the results of the cross product AB^AM and AB^AN have the same sign, then points M and N are on the same side of the line, otherwise they are not on the same side. In particular, if point M is on line AB, the value of AB^AM is 0. (If it is in a three-dimensional coordinate system, the cross product is a vector, and the sign of the dot product of the two vectors can be used to determine whether the two vectors point to the same side). Therefore, to determine whether a point is inside the triangle: If P is inside triangle ABC, the following three conditions are met: P and A are on the same side of BC, P and B are on the same side of AC, and PC is on the same side of AB. If any one of them is not met, it means that P is not inside the triangle.

[0059] Normalization means: mapping the "absolute value" of the actual deformation data value range to 0~1. The absolute value is used here because deformation can be divided into stretching and contraction (positive and negative deformation). Therefore, when the deformation is negative, we give a combination of deformation value and deformation direction, such as (3mm, +) means positive deformation of 3mm, (3mm, -) means negative deformation of 3mm; normalization assumes that the deformation range is (-8mm, 6mm), and the maximum absolute value is 8mm. After normalization, the absolute value range of deformation is (0, 1), where the maximum positive deformation is 0.75=6 / 8.

[0060] In S5, in the actual deformation process, the points closer to the deformation source (ie, the deformation monitoring point) are usually more affected, and the deformation displacement is also larger; while the points farther from the deformation source are less affected, and the deformation displacement is correspondingly reduced.

[0061] For the monitoring deformation point, pay attention to its deformation. Within the deformation range, the deformation displacement of the mesh vertex farthest from the monitoring deformation point is set to zero, which can be regarded as a boundary condition. For other mesh vertices, their deformation displacement is not given directly, but is calculated by linear interpolation of the distance from the monitoring deformation point.

[0062] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for dynamic deformation of a three-dimensional model in response to real-time data, characterized in that: include: S1: Read the 3D model mesh data in Unity and extract the model vertex coordinates; S2: Determine the actual monitoring deformation point position on the three-dimensional model; S3: According to the actual deformation area affected by the monitoring deformation point, all mesh vertices that have deformation displacement in the deformation area are screened; S4: determining the deformation displacement direction of the mesh vertices within the deformation area; S5: Determine the mesh vertex deformation displacement value within the deformation area, and control the mesh vertex deformation displacement speed according to the deformation rate.

2. A three-dimensional model dynamic deformation method in response to real-time data according to claim 1, characterized in that: S1 specifically includes: importing the 3D model into Unity, obtaining the triangular mesh data of the 3D model based on the Mesh class, extracting the local coordinates of the model mesh vertices, and multiplying the local coordinates with the transformation matrix to obtain the world coordinates of the model mesh vertices; Unity's coordinate system includes two expressions: local coordinates and world coordinates. The local coordinate system is the coordinate system relative to its parent object, which is used to describe the position relationship within the object. The world coordinate system corresponds to the coordinate system of the entire Unity world space and is not affected by the object hierarchy. To convert local coordinates to world coordinates to determine the specific position of the model in Unity space, the following calculation formula can be used: P w =M parent ·P l +P parentw P w and P l Represent the world coordinates and local coordinates of the model, M parent and P parent They represent the coordinate transformation matrix and world coordinates of the model's parent object respectively.

3. A three-dimensional model dynamic deformation method in response to real-time data according to claim 1, characterized in that: S2 is specifically as follows: determining the actual monitoring deformation point position in the three-dimensional model, outputting the local coordinates of the deformation monitoring point through the collision of rays and the model, and multiplying the local coordinates of the actual monitoring deformation point with the transformation matrix to obtain the world coordinates of the actual monitoring deformation point; wherein, determining the actual monitoring deformation point position is to convert the corresponding position of the real-world monitoring point to the corresponding coordinate position after the three-dimensional model; that is, setting a monitoring point at each interval of the real tunnel slope, and the deformation data transmitted by the monitoring point must accurately correspond to the three-dimensional model.

4. A three-dimensional model dynamic deformation method in response to real-time data according to claim 1, characterized in that: S3 is specifically as follows: traverse the mesh vertex coordinate array, calculate the Euclidean distance between the actual deformation monitoring point and the mesh vertex, and filter all mesh vertices within the deformation area according to the actual deformation area affected by each actual monitoring deformation point; the three-dimensional model will eventually be imported into the rendering engine in the form of a mesh, and the mesh is composed of a series of vertices and triangular faces connecting different vertices; the mesh vertices have coordinates.

5. The method for dynamic deformation of a three-dimensional model in response to real-time data according to claim 1, characterized in that: S4 specifically includes: S41: Determine the position of the triangle mesh where the deformation point is located: traverse all triangle mesh vertices, record the coordinates of the monitored deformation point as P, the coordinates of any triangle mesh vertex as A, B, C, calculate the cross product results of vectors PA, PB, PC, and when the results have the same sign, the deformation point is located in the triangle mesh; S42: Calculate the vertical vector of the triangular mesh: calculate two vectors in a clockwise direction according to the coordinates of the three vertices of the triangular mesh where the monitoring deformation point is located, and obtain the vertical vector of the triangular mesh by cross-product of the two vectors, with the direction pointing to the outside of the model; S43: Determine the deformation displacement direction of each mesh vertex in the deformation area: normalize the vertical vector of the triangular mesh; when the deformation value of the monitoring deformation point is negative, the mesh vertex is deformed and displaced along the normalized vertical vector, and the model deformation form is stretching; when the deformation value of the monitoring deformation point is positive, the deformation displacement direction of the mesh vertex is the opposite of the normalized vertical vector, and the model deformation form is compression.

6. A three-dimensional model dynamic deformation method in response to real-time data according to claim 1, characterized in that: S5 specifically includes: the deformation value of the deformation monitoring point is given based on the actual deformation monitoring value, and the deformation displacement value of each mesh vertex within the deformation range decreases as the distance between the deformation monitoring point and the mesh vertex increases; the deformation displacement value of the mesh vertex farthest from the monitoring deformation point within the deformation range is zero, and the deformation displacement values ​​of other mesh vertices are linearly interpolated according to their distances from the monitoring deformation point; the deformation rate is given based on the actual deformation monitoring value, and the deformation rate is added to the mesh vertex that undergoes deformation displacement, and the deformation rate controls the moving speed of the mesh vertex along the deformation displacement direction; when the deformation displacement of the mesh vertex reaches the set value, a three-dimensional model structure deformation is completed.

7. A three-dimensional model dynamic deformation method in response to real-time data according to claim 5, characterized in that: In S42, when the mesh vertices are deformed and displaced, the direction of the vertical vector of the triangular mesh changes accordingly; during the process of the mesh vertices being deformed and displaced, the mesh vertex coordinates are updated in real time and the vertical vector of the triangular mesh is calculated.

Images