Vehicle modeling parameter checking method and system

Abstract

The invention discloses a vehicle modeling parameter checking method and system, and belongs to the technical field of vehicle modeling optimizing.The method comprises the steps that vehicle three-dimensional modeling data are obtained; according to the three-dimensional modeling data of the vehicle, acquiring a planar graph of the vehicle; according to the plane graph, key points of the vehicle model and the connection relation between the key points are determined; taking the key points of the vehicle model as nodes, and constructing graph structure data of the vehicle model according to the connection relationship among the key points; extracting modeling parameter features from the graph structure data; according to the modeling parameter characteristics, modeling parameter data of the vehicle are determined; predicting the aerodynamic performance of the vehicle model according to the model parameter data of the vehicle; and checking the vehicle model according to the aerodynamic performance prediction result. Automatic extraction and checking of the vehicle modeling parameters and optimization of the vehicle modeling are realized.

Description

Technical Field

[0001] This invention relates to the field of vehicle styling optimization technology, and in particular to a method and system for verifying vehicle styling parameters. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] With the rapid development of the automotive industry, the iteration speed of car styling is gradually increasing. Every time a new vehicle styling is developed, its aerodynamic performance must be checked to evaluate whether the new vehicle styling meets the aerodynamic performance requirements.

[0004] When determining the aerodynamic performance of a vehicle's shape, the first step is to obtain the vehicle's shape parameter data. Currently, this is mainly done manually by using 3D design software or 3D simulation software to measure the constructed 3D vehicle shape data.

[0005] Because there are many vehicle styling parameters, obtaining vehicle styling parameter data through manual measurement is time-consuming. Furthermore, when manually measuring vehicle styling parameter data, it is necessary to manually locate and capture the geometric feature lines involved in the vehicle styling parameters, which has positioning and capture errors. The accuracy of the geometric feature line positioning and capture cannot be guaranteed, resulting in low accuracy of the measured vehicle styling parameter data. When using this vehicle styling parameter data for vehicle styling evaluation, the accuracy of the vehicle styling evaluation cannot be guaranteed. Summary of the Invention

[0006] To address the aforementioned problems, this invention proposes a method and system for verifying vehicle styling parameters, enabling the automatic extraction and verification of vehicle styling parameter data. The technical solution is as follows: On one hand, embodiments of the present invention provide a method for verifying vehicle styling parameters, including: Obtain vehicle 3D modeling data; Based on the vehicle's 3D modeling data, obtain the vehicle's 2D plan view; Based on the plan, determine the key points of the vehicle's design and the connections between these key points; Using the key points of vehicle styling as nodes, and based on the connection relationships between these key points, a graph structure data of vehicle styling is constructed. Extracting shape parameter features from graph structure data; Based on the styling parameter characteristics, determine the vehicle's styling parameter data; Based on the vehicle's styling parameter data, the aerodynamic performance of the vehicle's styling is predicted; The vehicle's design was verified based on the predicted aerodynamic performance.

[0007] Optionally, the vehicle plan view includes a left view, a top view, and a front view of the vehicle; Based on the left view, top view, and front view of the vehicle, determine the key points of the vehicle's shape and the connection relationships between these key points.

[0008] Optionally, edge detection is performed on the planar image to determine the geometric feature lines in the planar image; Based on geometric feature lines, determine the key points of vehicle styling; The relationship between key points is determined by whether there are geometric feature lines between them.

[0009] Optionally, the vehicle styling verification is deemed passed when the predicted aerodynamic performance meets the aerodynamic performance requirements; When the predicted aerodynamic performance does not meet the aerodynamic performance requirements, the vehicle's 3D shape is modified until the aerodynamic performance of the vehicle shape predicted based on the modified 3D shape data meets the aerodynamic performance requirements.

[0010] Optionally, the vehicle 3D modeling data is preprocessed, including: mesh repair and mesh cleaning. The scale, orientation, and centroid location of the preprocessed data are normalized. The normalized data is then projected onto a two-dimensional image to obtain a plan view of the vehicle.

[0011] Optionally, a styling parameter extraction model is used to extract styling parameter features from graph structure data; based on the styling parameter features, the styling parameter data of the vehicle is determined; wherein, the styling parameter extraction model is obtained by constructing a GCN.

[0012] On the other hand, embodiments of the present invention also provide a vehicle styling parameter verification system, including: The data acquisition unit is used to acquire three-dimensional modeling data of the vehicle. The plan view acquisition unit is used to acquire a plan view of the vehicle based on the vehicle's three-dimensional modeling data; The graph structure data determination unit is used to determine the key points of the vehicle shape and the connection relationships between the key points based on the plan view; and to construct the graph structure data of the vehicle shape based on the key points of the vehicle shape as nodes and the connection relationships between the key points. The styling parameter determination unit is used to extract styling parameter features from the graphic structure data; and to determine the styling parameter data of the vehicle based on the styling parameter features. The vehicle styling parameter verification unit is used to predict the aerodynamic performance of the vehicle styling based on the vehicle styling parameter data; and to verify the vehicle styling based on the aerodynamic performance prediction results.

[0013] In another aspect, embodiments of the present invention also provide a computer device, the device comprising: A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by the processor, implements a vehicle styling parameter verification method provided in this embodiment of the invention.

[0014] In another aspect, embodiments of the present invention also provide a computer-readable storage medium storing a computer program adapted for loading and executing by a processor a vehicle styling parameter verification method provided in embodiments of the present invention.

[0015] In another aspect, embodiments of the present invention also provide a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements a vehicle styling parameter verification method provided in embodiments of the present invention.

[0016] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: This invention provides a method and system for verifying vehicle styling parameters. The method converts three-dimensional vehicle styling data into a two-dimensional model of the vehicle. Based on the two-dimensional model, it determines the key points of the vehicle styling and the connections between these key points. Using these key points as nodes, it constructs a graph structure data representation of the vehicle styling based on the connections between them. This graph structure data accurately reflects the key points in the vehicle styling and the relationships between them. Then, by extracting styling parameter features from this graph structure data and identifying the extracted features, it achieves automatic and accurate measurement of the vehicle styling parameters, improving measurement efficiency. When this vehicle styling parameter data is used to predict vehicle aerodynamic performance for vehicle styling verification, it improves the accuracy and efficiency of vehicle styling verification, thereby shortening the vehicle styling development cycle.

[0017] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a vehicle styling parameter verification method provided by an embodiment of the present invention; Figure 2 This is a flowchart illustrating the training process of the model involved in a vehicle styling parameter verification method provided in an embodiment of the present invention. Detailed Implementation

[0020] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0021] It should be noted that the following detailed descriptions are illustrative and intended to provide optional explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0022] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0023] Before providing a detailed explanation of the vehicle styling parameter verification method provided in the embodiments of the present invention, the application scenarios and implementation environment of the vehicle styling parameter verification method provided in the embodiments of the present invention will be explained in detail first.

[0024] First, the application scenarios of a vehicle styling parameter verification method provided in the embodiments of the present invention will be explained.

[0025] With the rapid development of the automotive industry, the iteration speed of car styling is gradually increasing. Every time a new vehicle styling is developed, its aerodynamic performance must be checked to evaluate whether the new vehicle styling meets the aerodynamic performance requirements.

[0026] When determining the aerodynamic performance of a vehicle's shape, the first step is to obtain the vehicle's shape parameter data. Currently, this is mainly done manually by using 3D design software or 3D simulation software to measure the constructed 3D vehicle shape data.

[0027] Because there are many vehicle styling parameters, obtaining vehicle styling parameter data through manual measurement is time-consuming. Furthermore, when manually measuring vehicle styling parameter data, it is necessary to manually locate and capture the geometric feature lines involved in the vehicle styling parameters, which has positioning and capture errors. The accuracy of the geometric feature line positioning and capture cannot be guaranteed, resulting in low accuracy of the measured vehicle styling parameter data. When using this vehicle styling parameter data for vehicle styling evaluation, the accuracy of the vehicle styling evaluation cannot be guaranteed.

[0028] Based on this application scenario, this invention provides a vehicle styling parameter verification method that can automatically and accurately capture vehicle styling parameters. Secondly, the implementation environment of the vehicle styling parameter verification method provided in the embodiments of the present invention will be explained.

[0029] The vehicle styling parameter verification method is applied in a terminal, which is equipped with a car model building program, a styling parameter extraction program, and an aerodynamic performance prediction program. The car model building program can build a 3D vehicle model, and through this 3D vehicle model, 3D vehicle styling data can be obtained. The styling parameter extraction program can automatically extract vehicle styling parameter data based on the 3D vehicle styling data. The aerodynamic performance prediction program can predict the aerodynamic performance of the vehicle styling based on the vehicle styling parameter data, and evaluate the vehicle styling based on the aerodynamic performance prediction results.

[0030] Figure 1 This is a flowchart of a vehicle styling parameter verification method provided by an embodiment of the present invention. The vehicle styling similarity evaluation method may include the following steps: Step 101: Obtain vehicle 3D modeling data; Step 102: Obtain the vehicle's plan view based on the vehicle's 3D modeling data; Step 103: Based on the plan, determine the key points of the vehicle's design and the connection relationships between these key points; Step 104: Using the key points of the vehicle's design as nodes, construct the graph structure data of the vehicle's design based on the connection relationships between the key points; Step 105: Extract styling parameter features from the graph structure data; determine the vehicle's styling parameter data based on the styling parameter features; Step 106: Based on the vehicle's styling parameter data, predict the aerodynamic performance of the vehicle's styling; Step 107: Verify the vehicle styling based on the aerodynamic performance prediction results.

[0031] This invention provides a method for verifying vehicle styling parameters. It converts three-dimensional vehicle styling data into a two-dimensional model of the vehicle. Based on the two-dimensional model, it determines the key points of the vehicle styling and the connections between these key points. Using these key points as nodes, it constructs a graph structure data representation of the vehicle styling, accurately reflecting the key points and their relationships within the vehicle styling. Then, it extracts styling parameter features from this graph structure data and identifies these features, achieving automatic and accurate measurement of the vehicle styling parameters and improving measurement efficiency. When this vehicle styling parameter data is used to predict vehicle aerodynamic performance for styling verification, it improves the accuracy and efficiency of vehicle styling verification, thereby shortening the vehicle styling development cycle.

[0032] Based on the vehicle styling design requirements, two types of vehicle styling parameters are determined. The vehicle styling is then evaluated and verified using these parameters to determine whether it meets the requirements. These two types of vehicle styling parameters include global parameters and local parameters.

[0033] The overall parameters of the vehicle's design include length, width, height, wheelbase (tolerance ±1.5mm), track width, front overhang length, rear overhang length, ground clearance, and roof width.

[0034] The specific parameters of a vehicle's design include the angle of the front windshield, the angle of the rear windshield, the angle of the rear wing tip, the height of the front of the vehicle, the approach angle, and the departure angle.

[0035] The corresponding vehicle styling parameter features are: measuring the geometric feature lines involved in each vehicle styling parameter.

[0036] Optionally, the vehicle's 3D modeling data is obtained by acquiring CAD data of the vehicle's 3D modeling. This CAD data can be data from CATIA, AutoForm, etc.

[0037] Optionally, the vehicle 3D modeling data is preprocessed, including: mesh repair and mesh cleaning. The scale, orientation, and centroid location of the preprocessed data are normalized. The normalized data is then projected onto a two-dimensional image to obtain a plan view of the vehicle.

[0038] Mesh repair refers to fixing geometric and topological errors in the vehicle 3D modeling data mesh, such as holes, self-intersections, overlapping patches, surface voids, islands, redundant loops, singularities, inconsistent orientations, and complex edges, in order to improve mesh quality.

[0039] Mesh cleaning refers to removing or correcting errors, inconsistencies, or redundant information in the vehicle 3D modeling data mesh, such as removing non-manifold edges and filling holes.

[0040] By preprocessing the vehicle's 3D modeling data, the quality of the data can be improved. By normalizing the preprocessed data, deviations in subsequent 2D image projection can be avoided, thereby improving the accuracy of the obtained vehicle plan view.

[0041] Optionally, the vehicle plan view includes a left view, a top view, and a front view of the vehicle; Based on the left view, top view, and front view of the vehicle, determine the key points of the vehicle's shape and the connection relationships between these key points.

[0042] By recognizing the left view, top view, and front view of the vehicle, a comprehensive and accurate identification of the key points of the vehicle's shape and the connection relationships between these key points is achieved.

[0043] Key elements in vehicle styling include the feature points of the vehicle's outer contour, the turning points of curved surfaces, and the proportion control points of the body.

[0044] Optionally, edge detection is performed on the planar image to determine the geometric feature lines in the planar image; Based on geometric feature lines, determine the key points of vehicle styling; The relationship between key points is determined by whether there are geometric feature lines between them.

[0045] Edge detection is performed on the left, top, and front views using the Canny edge detection algorithm to determine the geometric feature lines in the three views. Key points are calculated and determined using regular features, enabling automatic annotation of these key points.

[0046] Optionally, a styling parameter extraction model is used to extract styling parameter features from graph structure data; based on the styling parameter features, the styling parameter data of the vehicle is determined; wherein, the styling parameter extraction model is obtained by constructing a GCN.

[0047] The intelligent verification method of Graph Convolutional Network (GCN) shows unique advantages. It trains the constructed graph convolutional network by training the training dataset and obtains the modeling parameter extraction model after training.

[0048] The process of constructing the training dataset includes: We acquired 3D modeling data of vehicles from historical projects of automakers, the publicly available dataset DriAerNet, and parametric deformation mesh data. The data covered vehicle types including sedans, SUVs, and MPVs.

[0049] The above three types of data are preprocessed, and then the three preprocessed data are projected into a two-dimensional image to obtain the left view, top view and front view of each data. Edge detection was performed on the above three views using the Canny edge detection algorithm, and geometric feature lines were extracted. Based on the extracted geometric feature lines, the key points of the three types of data and the relationships between the key points are determined, and then the graph structure data of the three types of data is constructed. The vehicle styling parameter features and styling parameter types of the three types of graph structure data are labeled to form a training dataset.

[0050] The training dataset is divided into a training set, a validation set, and a test set in a ratio of 7:1.5:1.5.

[0051] The constructed graph convolutional network is trained using the training set, the training effect of the graph convolutional network is verified using the validation set, and the effect of the model for extracting shape parameters obtained from the training is verified using the test set.

[0052] In terms of model architecture design, GCN adopts a multi-layered convolutional layer stacked structure. Each layer updates the feature representation of a node by aggregating the feature information of its neighbors. Specifically, the input layer of GCN receives graph structure data, taking the feature vectors of the nodes in the graph structure data and the topological structure (adjacency matrix) of the graph structure data as input. The adjacency matrix is ​​determined according to the relationship between each node. The hidden layer in the middle performs weighted aggregation and nonlinear transformation on the features of the node and its neighboring nodes through designed convolution operations, based on the adjacency relationship of the graph, to achieve layer-by-layer abstraction of features. The output layer integrates the extracted high-order features based on the hidden layer through a fully connected layer, and finally outputs the styling parameter features of the graph structure data. The styling parameter features are transformed into vehicle styling parameter data through feature measurement rules.

[0053] To improve the training efficiency and performance of the GCN model, several training optimization strategies are employed. First, batch normalization is introduced, normalizing the feature vectors of nodes after each convolutional layer to accelerate model convergence and improve generalization ability. Second, a dropout mechanism is used to randomly discard information from some nodes or edges during training to prevent overfitting. Furthermore, a learning rate decay strategy is employed, gradually reducing the learning rate as the number of training epochs increases to ensure stable convergence in the later stages of training. Simultaneously, to further enhance training efficiency, GPU parallel acceleration technology is used, fully utilizing the high computational power of the graphics processing unit to significantly shorten training time. This embodiment of the invention also defines the loss function of the GCN model as the MSE loss function, which measures the difference between the model's predictions and the true labels. Stochastic gradient descent (SGD) is used as the optimization algorithm, calculating the gradient of the loss function with respect to the model parameters through backpropagation and updating the parameters to minimize the loss.

[0054] A multi-dimensional evaluation method was employed to comprehensively assess the performance of the styling parameter extraction model. In terms of quantitative analysis, mean squared error (MSE), mean absolute error (MAE), and coefficient of determination (R^2) were used to quantify the deviation between the model's predicted and actual values. MSE directly reflects the magnitude of the average error, while R^2 measures the model's goodness of fit to the data. The model was tested on different vehicle model datasets to verify its generalization ability. Simultaneously, the model was tested on a test set containing data from four SUVs, three sedans, and three MPVs. The measured results for length and angle parameters met engineering application standards.

[0055] Optionally, the vehicle styling verification is deemed passed when the predicted aerodynamic performance meets the aerodynamic performance requirements; When the predicted aerodynamic performance does not meet the aerodynamic performance requirements, the vehicle's 3D shape is modified until the aerodynamic performance of the vehicle shape predicted based on the modified 3D shape data meets the aerodynamic performance requirements.

[0056] AI models or fluid dynamics (CFD) models are used to analyze vehicle styling parameter data to predict vehicle aerodynamic performance and ensure the accuracy of aerodynamic performance prediction.

[0057] By predicting aerodynamic performance, it is possible to determine whether the vehicle's design meets aerodynamic performance requirements, thus ensuring that the final vehicle design meets these requirements and guarantees the vehicle's driving stability and safety.

[0058] Optionally, the vehicle's three-dimensional shape can be modified within a set range, and the aerodynamic performance of the vehicle shape predicted by the modified three-dimensional shape data can be obtained. The vehicle shape with the best aerodynamic performance can be selected as the optimal vehicle shape, which can meet the shape improvement requirements and achieve the best aerodynamic performance.

[0059] This invention provides a vehicle styling parameter verification method. It converts three-dimensional vehicle styling data into a two-dimensional model of the vehicle. Based on the two-dimensional model, it determines the key points of the vehicle styling and the connections between these key points. Using these key points as nodes, it constructs a graph structure data representation of the vehicle styling, accurately reflecting the key points and their relationships. Then, by extracting styling parameter features from this graph structure data and identifying the extracted features, it achieves automatic and accurate measurement of vehicle styling parameters, improving measurement efficiency, effectively reducing errors from manual measurement, and possessing high scalability. The model can continuously learn and optimize to adapt to new design trends. When using this vehicle styling parameter data to predict vehicle aerodynamic performance for styling verification, it improves the accuracy and efficiency of vehicle styling verification, thereby shortening the vehicle styling development cycle. Furthermore, it can screen vehicle styling based on aerodynamic performance to determine the optimal vehicle styling that meets both styling improvement needs and achieves optimal aerodynamic performance.

[0060] On the other hand, embodiments of the present invention also provide a vehicle styling parameter verification system, including: The data acquisition unit is used to acquire three-dimensional modeling data of the vehicle. The plan view acquisition unit is used to acquire a plan view of the vehicle based on the vehicle's three-dimensional modeling data; The graph structure data determination unit is used to determine the key points of the vehicle shape and the connection relationships between the key points based on the plan view; and to construct the graph structure data of the vehicle shape based on the key points of the vehicle shape as nodes and the connection relationships between the key points. The styling parameter determination unit is used to extract styling parameter features from the graphic structure data; and to determine the styling parameter data of the vehicle based on the styling parameter features. The vehicle styling parameter verification unit is used to predict the aerodynamic performance of the vehicle styling based on the vehicle styling parameter data; and to verify the vehicle styling based on the aerodynamic performance prediction results.

[0061] It should be noted that the vehicle styling parameter verification system provided in the above embodiments is only illustrated by the division of the functional units described above when evaluating vehicle aerodynamic performance. In practical applications, the functions described above can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above. Furthermore, the vehicle styling parameter verification system and the vehicle styling parameter verification method embodiment provided in the above embodiments belong to the same concept, and their specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0062] The present invention also discloses a computer device, the device comprising: A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by the processor, implements a vehicle styling parameter verification method provided in this embodiment of the invention.

[0063] The present invention also discloses a computer-readable storage medium storing a computer program adapted for loading and execution by a processor of a vehicle styling parameter verification method provided in the embodiments of the present invention.

[0064] The present invention also discloses a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements a vehicle styling parameter verification method provided in the embodiments of the present invention.

[0065] The method provided in this invention can be directly executed by a hardware processor, or executed by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, detailed descriptions are omitted here.

[0066] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0067] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for verifying vehicle styling parameters, characterized in that, include: Obtain vehicle 3D modeling data; Based on the vehicle's 3D modeling data, obtain the vehicle's 2D plan view; Based on the plan, determine the key points of the vehicle's design and the connections between these key points; Using the key points of vehicle styling as nodes, and based on the connection relationships between these key points, a graph structure data of vehicle styling is constructed. Extracting shape parameter features from graph structure data; Based on the styling parameter characteristics, determine the vehicle's styling parameter data; Based on the vehicle's styling parameter data, the aerodynamic performance of the vehicle's styling is predicted; The vehicle's design was verified based on the predicted aerodynamic performance.

2. The vehicle styling parameter verification method as described in claim 1, characterized in that, The vehicle's plan view includes the vehicle's left view, top view, and front view; Based on the left view, top view, and front view of the vehicle, determine the key points of the vehicle's shape and the connection relationships between these key points.

3. The vehicle styling parameter verification method as described in claim 1, characterized in that, Edge detection is performed on the planar image to determine the geometric feature lines in the planar image; Based on geometric feature lines, determine the key points of vehicle styling; The relationship between key points is determined by whether there are geometric feature lines between them.

4. The vehicle styling parameter verification method as described in claim 1, characterized in that, When the predicted aerodynamic performance meets the aerodynamic performance requirements, the vehicle styling verification is deemed passed. When the predicted aerodynamic performance does not meet the aerodynamic performance requirements, the vehicle's 3D shape is modified until the aerodynamic performance of the vehicle shape predicted based on the modified 3D shape data meets the aerodynamic performance requirements.

5. The vehicle styling parameter verification method as described in claim 1, characterized in that, Preprocessing of vehicle 3D modeling data includes: mesh repair and mesh cleaning; The scale, orientation, and centroid location of the preprocessed data are normalized. The normalized data is then projected onto a two-dimensional image to obtain a plan view of the vehicle.

6. The vehicle styling parameter verification method as described in claim 1, characterized in that, A styling parameter extraction model is used to extract styling parameter features from graph structure data; based on the styling parameter features, the styling parameter data of the vehicle is determined; the styling parameter extraction model is constructed using GCN.

7. A vehicle styling parameter verification system, characterized in that, include: The data acquisition unit is used to acquire three-dimensional modeling data of the vehicle. The plan view acquisition unit is used to acquire a plan view of the vehicle based on the vehicle's three-dimensional modeling data; The graph structure data determination unit is used to determine the key points of the vehicle shape and the connection relationships between the key points based on the plan view; and to construct the graph structure data of the vehicle shape based on the key points of the vehicle shape as nodes and the connection relationships between the key points. The shaping parameter determination unit is used to extract shaping parameter features from the graphic structure data; Based on the styling parameter characteristics, determine the vehicle's styling parameter data; The vehicle styling parameter verification unit is used to predict the aerodynamic performance of the vehicle styling based on the vehicle styling parameter data. The vehicle's design was verified based on the predicted aerodynamic performance.

8. An electronic device, characterized in that, The device includes: A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by the processor, implements a vehicle styling parameter verification method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded by a processor and executed by a method for verifying vehicle styling parameters as described in any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements a vehicle styling parameter verification method according to any one of claims 1-6.

Images