Infrared flame three-dimensional rendering method, device and medium based on GPU acceleration

Abstract

The application relates to a GPU acceleration-based infrared flame three-dimensional rendering method, device and medium. The method comprises the following steps: firstly, analyzing an aircraft whole-machine grid model, combining with a LOD technology for lightweight processing, then constructing different working condition flow field distribution templates based on jet flow field temperature data, calling a multi-line group wideband k distribution model to generate multi-view brightness distribution, finally, constructing infrared rendering content and sending the infrared rendering content into a GPU, using a Stream Output feature of DirectX 10.0 to realize GPU end vertex data caching and updating, and completing cyclic rendering. The application improves infrared sequence image generation efficiency and fidelity, saves particle resources, and meets the real-time and precision requirements of aviation simulation.

Description

Technical Field

[0001] This application relates to the field of infrared 3D rendering technology, and in particular to a GPU-accelerated infrared flame 3D rendering method, apparatus and medium. Background Technology

[0002] Infrared 3D rendering technology is a core supporting technology in fields such as aerospace target simulation and infrared characteristic analysis. It can present the infrared radiation characteristics of a target in a visual form, providing important simulation data and visual basis for target detection, identification, and anti-interference design. For aerial targets such as aircraft, their infrared radiation mainly comes from two parts: the solid surface of the fuselage and the engine exhaust. The fuselage skin is a solid target, and its grayscale changes slowly with the observation angle. Traditional infrared rendering technology relies on 3D solid modeling and rendering logic of solid surfaces to achieve fast and stable rendering of such solid targets, meeting the needs of conventional simulation.

[0003] However, the exhaust plume of aircraft turbofan engines is a high-temperature gas medium, mainly composed of radiation-participating components such as high-temperature water vapor, carbon dioxide, and soot particles. This differs fundamentally from the infrared radiation characteristics and morphological distribution of solid surfaces, making it impossible to directly apply traditional solid-surface infrared rendering techniques for efficient and accurate rendering. Currently, the mainstream approach to infrared rendering of aircraft engine exhaust plumes is to use particle system effects to visualize the plume. While this technique can initially simulate the dynamic morphology of the plume, it suffers from numerous insurmountable technical defects and cannot meet the core requirements of modern aerospace simulation systems for high fidelity and real-time performance. Specifically, this manifests in insufficient rendering fidelity, poor real-time performance, and excessive resource consumption. Summary of the Invention

[0004] Therefore, it is necessary to provide a GPU-accelerated infrared flame 3D rendering method, device, and medium that can balance rendering realism, real-time performance, and resource utilization efficiency to address the aforementioned technical problems.

[0005] A GPU-accelerated 3D rendering method for infrared flame projection, the method comprising:

[0006] A three-dimensional digital scene of the aircraft infrared simulation is obtained, which includes the whole aircraft mesh model and the calculated data of the three-dimensional flow field temperature distribution of the engine exhaust flame; The mesh model component classes of the aircraft whole mesh model are extracted by semantic analysis and geometric analysis. The general body mesh model and the flame mesh model are parameterized at the algorithm end. While instantiating the mesh model, a progressive adaptive rendering method based on LOD multi-level of detail technology is used to lightweight the aircraft whole mesh model to obtain a lightweight aircraft whole mesh model. Based on the calculated temperature distribution data of the three-dimensional flow field of the engine exhaust flame, corresponding exhaust flame flow field distribution templates are constructed for different engine operating conditions. The brightness variation law in the exhaust flame flow field distribution template is extracted as the basic data for rendering. The exhaust flame radiation multi-line group broadband k-distribution model is called to generate the exhaust flame brightness distribution under different perspectives. Based on the lightweight aircraft mesh model and the flame brightness distribution from different perspectives, infrared flame rendering content is constructed. The infrared flame rendering content is sent to the GPU for 3D rendering. The StreamOutput feature of DirectX 10.0 is used to cache and update the vertex data on the GPU side. The updated vertex data is then re-entered into the vertex shader to complete the loop rendering, thus realizing GPU-accelerated 3D rendering of infrared flames.

[0007] In one embodiment, the general shape mesh model is a mesh model corresponding to the solid wall of the aircraft, and the flame mesh model is a dedicated mesh model corresponding to the flame of the aircraft turbofan engine.

[0008] In one embodiment, by reducing the total data volume and the number of triangles in the aircraft mesh model, the initial loading time of the scene is accelerated and the frame rate of the overall scene loading is improved, thereby achieving the lightweighting of the aircraft mesh model.

[0009] In one embodiment, the different operating conditions of the engine include at least cruise and afterburner conditions, and the flame flow field distribution template constructed for each operating condition includes the physical morphological parameters and basic brightness distribution characteristics of the flame under the corresponding operating condition.

[0010] In one embodiment, when the wideband k-distribution model of the jet radiation multi-line group is invoked to generate the jet brightness distribution under different viewing angles: Based on the brightness variation law of the aforementioned flame flow field distribution template, and by adapting the flame radiation multi-line group broadband k-distribution model to the three radiation-participating components of water vapor, carbon dioxide, and soot particles in the flame, the flame radiation brightness under a specified observation angle is accurately calculated. The calculation result is used as the direct basis for generating the flame brightness distribution under different perspectives.

[0011] In one embodiment, the infrared flame rendering content is constructed step-by-step based on the lightweight aircraft mesh model and the flame brightness at different viewing angles, including: Based on the flame brightness distribution under different viewpoints, a texture map containing thermal energy information adapted to infrared radiation effects is obtained. The texture map is mapped to the flame mesh model in the lightweight aircraft's overall mesh model; Based on the optical characteristics of the infrared environment, the corresponding lighting and shadow parameters and post-processing effects are obtained and superimposed on the flame mesh model. Then, the particle system is fused to obtain the infrared flame rendering content.

[0012] In one embodiment, the caching and updating of GPU-side vertex data is implemented using the Stream Output feature of DirectX 10.0, including: After the vertex shader or geometry shader has finished processing the vertex data, the processed vertex data is directly cached into a dedicated data structure inside the GPU through the Stream Output feature of DirectX 10.0; The vertex information required for flame rendering is updated in real time based on the cached vertex data, and the dedicated data structure is deployed in the GPU throughout the process without needing to be sent back to the CPU.

[0013] This application also provides a GPU-accelerated infrared flame 3D rendering device, the device comprising: The scene import module is used to acquire the three-dimensional digital scene of the aircraft infrared simulation. The three-dimensional digital scene includes the whole aircraft mesh model and the calculated data of the three-dimensional flow field temperature distribution of the engine exhaust flame. The lightweight mesh model module is used to extract the mesh model component classes of the aircraft whole mesh model through semantic analysis and geometric analysis. It performs parameterization processing on the general body mesh model and the flame mesh model at the algorithm end. While instantiating the mesh model, it adopts a progressive adaptive rendering method based on LOD multi-level of detail technology to lightweight the aircraft whole mesh model and obtain a lightweight aircraft whole mesh model. The flame template construction and brightness calculation module is used to construct corresponding flame flow field distribution templates for different engine operating conditions based on the three-dimensional flow field temperature distribution calculation data of the engine flame, extract the brightness variation law in the flame flow field distribution template as the basic data for rendering, call the flame radiation multi-line group broadband k-distribution model, and generate the flame brightness distribution under different perspectives. The rendering content construction module is used to construct infrared flame rendering content based on the lightweight aircraft mesh model and the flame brightness distribution from different perspectives. The GPU-accelerated rendering module is used to send the infrared flame rendering content to the GPU for 3D rendering. It utilizes the Stream Output feature of DirectX 10.0 to cache and update the vertex data on the GPU side, and then re-enters the updated vertex data into the vertex shader to complete the loop rendering, thereby realizing GPU-accelerated 3D rendering of the infrared flame.

[0014] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps in the above-described GPU-accelerated infrared flame 3D rendering method.

[0015] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described GPU-accelerated infrared flame 3D rendering method.

[0016] The aforementioned GPU-accelerated infrared flame 3D rendering method, device, and medium extract the mesh model component classes of the aircraft's overall mesh model through semantic and geometric analysis. The general body mesh model and the flame mesh model are parameterized at the algorithm level. While instantiating the mesh model, a progressive adaptive rendering method based on LOD (Level of Detail) technology is used to lightweight the aircraft's overall mesh model, resulting in a lightweight aircraft mesh model. Based on the calculated temperature distribution data of the engine flame's 3D flow field, corresponding flame flow field distribution templates are constructed for different engine operating conditions. The brightness variation patterns in the flame flow field distribution templates are extracted as the basic rendering data. The wideband k-distribution model of the flame radiation multi-line group is called to generate the flame brightness distribution from different viewpoints. Based on the lightweight aircraft mesh model and the flame brightness distribution from different viewpoints, infrared flame rendering content is constructed. This infrared flame rendering content is sent to the GPU for 3D rendering. The Stream Output feature of DirectX 10.0 is used to cache and update the vertex data on the GPU side. The updated vertex data is then re-entered into the vertex shader to complete the loop rendering, achieving GPU-accelerated 3D rendering of infrared flames.

[0017] This method achieves lightweight mesh modeling by performing semantic and geometric analysis on the aircraft's overall mesh model, combined with LOD (Level of Detail) technology. This effectively reduces the model's data volume and the number of triangles, significantly improving scene loading efficiency and frame rate, providing an efficient model foundation for infrared rendering of complex exhaust flame meshes. Simultaneously, based on the calculated temperature distribution data of the engine exhaust's 3D flow field, it constructs exhaust flow field distribution templates for different operating conditions. Combined with a wideband k-distribution model of exhaust radiation multi-line groups, it accurately calculates exhaust brightness from multiple perspectives, upgrading exhaust rendering from simple visual effects simulation to precise data rendering based on physical laws. This perfectly adapts to the exhaust radiation characteristics of different engine operating conditions such as cruise and afterburner, significantly improving the realism of infrared exhaust rendering. Furthermore, leveraging the StreamOutput feature of DirectX 10.0, it enables local caching and autonomous updating of vertex data on the GPU, allowing the entire vertex data update and loop rendering process of exhaust rendering to be completed within the GPU, eliminating the need for the CPU to repeatedly generate and transmit vertex data, thus drastically reducing CPU and GPU overhead. The data interaction cost between them is reduced, the particle resource consumption of the particle system is greatly reduced, and the technical pain point of the traditional particle system in rendering flames is that it is difficult to balance realism and real-time performance. In the end, it realizes high-fidelity and high-efficiency GPU-accelerated 3D rendering of infrared flames of aircraft turbofan engines, effectively improves the generation efficiency and display effect of infrared sequence images of aircraft targets, and meets the real-time and accuracy requirements of aviation infrared simulation system for flame rendering. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a GPU-accelerated infrared flame 3D rendering method in one embodiment. Figure 2 This is a schematic diagram of the model lightweighting process in one embodiment; Figure 3 This is a schematic diagram illustrating the trend of flame brightness during engine cruise in one embodiment; Figure 4 This is a schematic diagram of the central axis brightness variation curve in one embodiment; Figure 5 This is a schematic diagram illustrating the trend of flame brightness under the applied force state in one embodiment; Figure 6 This is a schematic diagram of molecular vibrational-rotational energy level transitions in one embodiment; Figure 7 This is a schematic diagram of the energy levels of the basic transitions of carbon dioxide, water vapor, and soot particles in one embodiment; Figure 8 This is a schematic diagram of a multiple rendering process in one embodiment; Figure 9This is a structural block diagram of a GPU-accelerated infrared flame 3D rendering device in one embodiment; Figure 10 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0020] In response to the limitations of existing technologies, such as the inability to rapidly render the infrared exhaust plumes of turbofan engines targeting aircraft using traditional solid-wall rendering techniques, and the difficulty in simultaneously ensuring the realism and real-time performance of existing particle system-based plume rendering methods, this application addresses these issues. Figure 1 As shown, a GPU-accelerated 3D rendering method for infrared flames is provided, which specifically includes the following steps: Step S100: Obtain the three-dimensional digital scene of the aircraft infrared simulation. The three-dimensional digital scene includes the whole aircraft mesh model and the calculated data of the three-dimensional flow field temperature distribution of the engine exhaust flame.

[0021] Step S110: Extract the mesh model component classes of the aircraft whole mesh model through semantic analysis and geometric analysis. Perform parameterization processing on the general body mesh model and the flame mesh model at the algorithm end. While instantiating the mesh model, adopt a progressive adaptive rendering method based on LOD multi-level of detail technology to lighten the aircraft whole mesh model and obtain a lightweight aircraft whole mesh model.

[0022] Step S120: Based on the calculated temperature distribution data of the three-dimensional flow field of the engine exhaust flame, construct corresponding exhaust flame flow field distribution templates for different engine operating conditions, extract the brightness variation law in the exhaust flame flow field distribution templates as the basic data for rendering, call the exhaust flame radiation multi-line group broadband k-distribution model, and generate exhaust flame brightness distributions from different perspectives.

[0023] Step S130: Based on the lightweight aircraft mesh model and the flame brightness distribution from different perspectives, construct the infrared flame rendering content.

[0024] In step S140, the infrared flame rendering content is sent to the GPU for 3D rendering. The Stream Output feature of DirectX 10.0 is used to cache and update the vertex data on the GPU side. The updated vertex data is then sent back to the vertex shader to complete the loop rendering, thus realizing GPU-accelerated 3D rendering of the infrared flame.

[0025] In step S100, the aircraft's whole-aircraft mesh model is the structured mesh model required for aircraft infrared simulation. It includes general shape mesh models corresponding to solid walls such as the aircraft fuselage skin, wings, and engine body, as well as the exhaust mesh model corresponding to the high-temperature gas medium of the aircraft turbofan engine exhaust. It can be directly used for subsequent mesh analysis and parameterization. The engine exhaust three-dimensional flow field temperature distribution calculation data is the full-condition temperature distribution data obtained from exhaust flow field simulation. It includes core physical parameters such as temperature gradient and component concentration of the exhaust flow field under different engine conditions such as cruise and afterburner. It is the basic data support for subsequent construction of exhaust flow field distribution template and calculation of exhaust radiation brightness, and can be directly retrieved from the flow field calculation database of aircraft infrared simulation.

[0026] In step S110, the mesh model component classes of the aircraft's overall mesh model are extracted through semantic and geometric analysis, clearly distinguishing two core mesh models: one is the general shape mesh model corresponding to the aircraft's solid walls, covering the meshes corresponding to solid structures such as the aircraft fuselage skin, wings, and engine casing; the other is the dedicated exhaust mesh model corresponding to the aircraft's turbofan engine exhaust, adapted to the rendering characteristics of high-temperature gas media. For the different structural features and rendering requirements of these two types of mesh models, parameterization is performed at the algorithm level. The general shape mesh model focuses on preserving parameters for structural integrity, while the exhaust mesh model focuses on parameter design for dynamic morphological adaptation. While instantiating the two types of mesh models mentioned above, a progressive adaptive rendering method based on LOD (Levels of Detail) technology is adopted. This method reduces the total data volume and the number of triangles in the aircraft mesh model, achieving lightweighting of the entire aircraft mesh model. The resulting lightweight aircraft mesh model accelerates the initial loading time of the infrared simulation scene and significantly improves the overall frame rate of scene loading, providing efficient model support for the subsequent complex infrared rendering of the flame mesh model. The entire lightweighting process is as follows: Figure 2 As shown.

[0027] In this embodiment, in order to achieve accurate rendering and dynamic shape adaptation of the flame mesh model, step S110 also combines Mesh rendering technology to build an efficient rendering foundation, whose core logic and mesh lightweight processing complement each other.

[0028] The core principle of Mesh rendering technology is to decompose all 3D graphics into a combination of several triangular facets. For example, a planar rectangle can be formed by splicing two triangles together, and complex 3D shapes are restored through the orderly arrangement of triangles. In this embodiment, a dedicated Mesh class is provided, which supports the creation and modification of meshes via scripts. With the help of this Mesh class, the mesh structure of objects can be flexibly generated or adjusted, thereby achieving highly realistic dynamic deformation effects of flames. At the same time, a Mesh filter and a Mesh renderer are configured. When importing the whole aircraft mesh model, the algorithm will automatically create a Mesh filter component. This component extracts mesh data from the resources and passes it to the MeshRenderer for subsequent drawing. The Mesh object instantiated by the Mesh filter will store the core mesh data in an optimized way. That is, it first stores a vertex array (Vector3 type) containing the x, y, and z coordinates of each vertex, and then records the combination relationship of the triangular facets. By utilizing the characteristic that the facets share vertices, memory usage is greatly saved, providing efficient and flexible rendering data support for the lightweight flame mesh model.

[0029] In step S120, the engine operates under at least two conditions: cruise and afterburner. The flame flow field distribution template constructed for each condition includes the physical morphology parameters and basic brightness distribution characteristics of the flame under that condition. The physical morphology and temperature gradient of the flame differ significantly between the two conditions. Specifically, under cruise condition, the flame is subsonic, with a maximum diameter of 0.6 m and a maximum length of 14 m. The temperature decreases gradually from the engine outlet to the far end, with the core high-temperature region concentrated near the nozzle. The brightness distribution trend is as follows: Figure 3 As shown: The left side is the engine outlet. The curve showing the change (normalized) in brightness of the central axis with nozzle distance is as follows. Figure 4 As shown. Under afterburner conditions, the flame exhibits supersonic speeds, with a maximum diameter of 1.2m and a maximum length of 26m. The overall temperature level is higher, and the high-temperature zone extends over a wider area. Its brightness trend is as follows. Figure 5 As shown.

[0030] Furthermore, it is necessary to construct dedicated jet flow field distribution templates for the two types of operating conditions. When constructing the templates, the three-dimensional flow field temperature distribution calculation data is used as the core basis. The jet is divided into several units according to spatial regions. The temperature value, component concentration (mainly water vapor, carbon dioxide and soot particles), and spatial coordinates of each unit are recorded to form a structured template containing jet physical morphology parameters and temperature distribution characteristics. At the same time, the brightness variation law of each spatial unit in the template is extracted. Taking the engine outlet as the reference point, along the jet central axis and radial direction, the corresponding relationship curve between temperature and brightness is established (such as the central axis brightness under cruise conditions showing an exponential decay trend with the increase of nozzle distance). This law is used as the basis data for subsequent rendering to ensure that the brightness distribution conforms to the real jet radiation characteristics.

[0031] In this embodiment, when generating the flame brightness distribution from different perspectives, a wideband k-distribution model of multi-line group of flame radiation adapted to the characteristics of the flame is invoked. This model is specifically optimized for the three radiation-participating components in the flame: water vapor, carbon dioxide, and soot particles. It takes into account the multi-line component segmentation under large temperature gradients and the radiation absorption overlap effect between different components, and can accurately calculate the flame radiation brightness at specific temperatures, pressures, and component concentrations.

[0032] This paper further introduces a broadband k-distribution model of multi-line groups in flame radiation. Under certain conditions of quantum mechanics, the transitions in the vibrational and rotational energy levels of the products of high-temperature gas combustion alter the energy levels of molecules. These transitions are accompanied by photon absorption and radiation, resulting in the typical vibrational-rotational waveband characteristics (i.e., spectral line groups) of the thermal radiation of molecular gases. According to the "selection rule" of quantum mechanics, simultaneous changes in rotational energy levels are also permissible or required. Therefore, the thermal radiation of molecular gases exhibits typical vibrational-rotational waveband characteristics (i.e., spectral line groups), with each spectral line corresponding to a specific transition, denoted by the vibrational and rotational quantum numbers V and J. Figure 6 and 7 The diagrams illustrate molecular vibrational-rotational energy level transitions and vibrational modes, respectively.

[0033] Neglecting scattering, the radiation transfer equation can be simplified to: (1) In formula (1), For wave number, For spectral radiance, Right now Let be the blackbody radiation function (based on Prandtl's law of blackbody radiation). The absorption coefficient is... This represents the set of all local variables that affect the gas absorption coefficient, including temperature. ,pressure and component concentration , It represents the direction of radiation transmission.

[0034] Furthermore, based on the idea of ​​the k-distribution and the related k assumption, both sides of the radiative transfer equation are multiplied by the Dirac function. Then Integrate over the range of (narrowband wavenumber width), and divide the integral by... ,get: (2) In formula (2), ,in, .

[0035] To further simplify formula (2), the k-distribution function Integrating yields the cumulative k-distribution function: Divide both sides of formula (2) by ,get: (3) In formula (3), Compared to formula (2), the cumulative k distribution function in formula (3) has a range of 0 to 1, which can be conveniently calculated numerically using Gaussian integrals, as shown in the following formula: (4) Formula (4) is the narrowband k-distribution model (NB). Applying the ideas of the k-distribution and the associated k-assumption to the full spectrum range yields the broadband model: (5) In formula (5), Given a stretching function, the jet flame contains three radiation-participating components. Where subscript 1 represents water vapor, subscript 2 represents carbon dioxide, and subscript 3 represents soot particles. Formula (1) can be decomposed into: (6) From a physical perspective, formula (6) can be interpreted as the radiation emitted by any gas being absorbed by all other gases during transmission. Considering that the correlation k assumption no longer holds under large temperature gradients, based on the idea of ​​multi-line groups, the absorption / emission spectrum of each component can be decomposed into M uncorrelated sub-spectrums, and the corresponding radiative transfer formula (6) can also be decomposed into M uncorrelated radiative transfer equations: (7) In formula (7), the subscript The group number representing the absorption coefficient of the gas is used to introduce the idea of ​​k-distribution into formula (7). The method is similar to the derivation of formula (2), and we can obtain: (8) In formula (8), , , , It is the overlap coefficient of the absorption coefficient of the nth component in the mth group with the absorption coefficients of all other components at that wavenumber, specifically expressed as: (9) Equation (9) is the broadband k-distribution gas radiation model MSMGWB. The gray-body solid-wall boundary condition of Equation (8) can be written as: (10) In formula (10), It is solid-wall emissivity. and These are the unit vector normal to the wall and the direction of radiative transmission, respectively. yes Corresponding spatial solid angle, This represents the solid wall temperature and pressure, and the gas composition distribution of the adjacent solid wall grid cells.

[0036] Furthermore, formula (10) is obtained based on the fact that when the radiation-participating medium... The radiance of a gas layer of infinite thickness composed of this medium is equal to the radiance of a blackbody at the same temperature, expressed as: (10) At the same time, it is not difficult to deduce that: (11) Finally, based on the improved convolution theory, the overlap coefficient that guarantees the calculated radiation characteristics of a homogeneous gas mixture (with arbitrary state and optical thickness) can be derived. Implicit expression: (12) The aforementioned multi-line group broadband k-distribution model for exhaust radiation (MSMGWB) is not a general gas radiation calculation model, but rather a model specifically adapted and optimized for the high-temperature, multi-component characteristics of aircraft turbofan engine exhaust. On one hand, the model accurately covers the three core radiation-involved components in the exhaust: water vapor (H2O), carbon dioxide (CO2), and soot particles. Through group processing and overlap coefficient calculation, it solves the problem of traditional broadband models neglecting radiation absorption interference between components, ensuring the accuracy of exhaust radiation characteristic calculations under different operating conditions. On the other hand, considering the large temperature gradient in the exhaust during engine afterburner / cruise switching, the model overcomes the applicability limitations of the associated k-assump in scenarios with drastic temperature changes through multi-line group spectral decomposition, ensuring that the brightness distribution calculation of the exhaust from the core high-temperature region to the far-end low-temperature region conforms to actual physical laws.

[0037] In step S130, the infrared flame rendering content is constructed step by step according to the lightweight aircraft mesh model and the flame brightness under different viewing angles. This includes: obtaining a texture map containing thermal energy information that adapts to the infrared radiation effect based on the flame brightness distribution under different viewing angles; mapping the texture map to the flame mesh model in the lightweight aircraft mesh model; obtaining the corresponding lighting and shadow parameters and post-processing effects based on the optical characteristics of the infrared environment and superimposing them onto the flame mesh model; and then fusing the particle system to obtain the infrared flame rendering content.

[0038] Specifically, the first step is to create the material system and shaders needed to achieve the infrared effect of the object. Custom shaders are created, taking into account the object's surface reflection and heat emission characteristics, as well as the representation of thermal infrared images. Infrared calculation results for the model are imported, and corresponding textures containing simulated thermal energy information are rendered. These texture maps are used to render the object's surface, giving it a special effect under infrared vision. During rendering, the texture maps are applied to the object's materials, and the thermal infrared image is correctly projected onto the object through texture coordinate mapping. Furthermore, lighting and shadows are also crucial factors in showcasing the object's infrared effect. By adjusting lighting and shadow settings, the object can appear more realistic under infrared vision. During model rendering, the properties and types of lights, as well as the intensity and hardness of shadows, are adjusted to simulate lighting conditions in a thermal infrared environment. Post-processing effects are used to blur and distort the infrared image to simulate heat conduction and optical effects. Simultaneously, particle effects are combined to add more realism and atmosphere to the infrared scene. Furthermore, considering that the above-mentioned 3D rendering draw calls to the GPU will invoke a draw call, but the particle consumption will be very high, and the number of particles rendered is very limited. In step S140, the Stream Output feature of DirectX 10.0 is used to implement the caching and updating of vertex data on the GPU side. This includes: after the vertex shader or geometry shader has processed the vertex data, the processed vertex data is directly cached into a dedicated data structure inside the GPU through the Stream Output feature of DirectX 10.0. Based on the cached vertex data, the vertex information required for flame rendering is updated in real time, and the dedicated data structure is deployed entirely in the GPU without needing to be sent back to the CPU. The process is as follows: Figure 8 As shown.

[0039] In the aforementioned GPU-accelerated 3D rendering method for infrared flames, the component classes of the mesh model are extracted through semantic and geometric analysis. General volumetric mesh models and flame mesh models are classified and parameterized. LOD (Level of Detail) technology is used to achieve model lightweighting. Mesh rendering technology is employed to optimize mesh creation, modification, and data storage. Based on the calculated temperature distribution data of the engine flame's 3D flow field, flame flow field distribution templates for different operating conditions such as cruise and afterburner are constructed, and brightness variation patterns are extracted. A wideband k-distribution model of flame radiation multi-line group, adapted to the three components of water vapor, carbon dioxide, and soot particles, is used to accurately calculate the flame brightness distribution from different perspectives. Finally, a material system and shaders adapted to infrared effects are created. Infrared calculation data is imported to render texture maps containing thermal information. Combined with lighting and shadow adjustments, post-processing effects, and a particle system, the infrared flame rendering content is constructed and sent to the GPU via the 3D rendering's draw function. Utilizing the Stream Output feature of DirectX 10.0 (corresponding to Transform Feedback in OpenGL 3.0), vertex data is cached on the GPU after processing by the vertex shader or geometry shader. The internal structure enables autonomous vertex data updates and iterative rendering without requiring CPU backhaul, ultimately achieving a visual representation of infrared targets with flame plumes. This method not only ensures computational efficiency and realism through flame plume templates and a multi-line group broadband k-distribution model, but also saves particle consumption in the particle system and accelerates the rendering process by leveraging GPU-side data caching, significantly improving the generation efficiency and realism of target infrared sequence images. This method effectively solves the limitations of traditional target infrared rendering techniques, which are suitable for solid walls but cannot quickly render high-temperature gas flame plumes, and the shortcomings of commonly used particle system effects rendering, such as insufficient realism and inability to meet real-time requirements.

[0040] It should be understood that, although Figure 1The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0041] In one embodiment, such as Figure 9 As shown, a GPU-accelerated infrared flame 3D rendering device is provided, including: a scene import module 200, a mesh model lightweighting module 210, a flame template construction and brightness calculation module 220, a rendering content construction module 230, and a GPU-accelerated rendering module 240, wherein: The scene import module 200 is used to acquire a three-dimensional digital scene of the aircraft infrared simulation. The three-dimensional digital scene includes the whole aircraft mesh model and the calculated data of the three-dimensional flow field temperature distribution of the engine exhaust flame. The lightweight mesh model module 210 is used to extract the mesh model component classes of the aircraft whole mesh model through semantic analysis and geometric analysis, and to perform parameterization processing on the general body mesh model and the flame mesh model at the algorithm end. While instantiating the mesh model, a progressive adaptive rendering method based on LOD multi-level of detail technology is used to lightweight the aircraft whole mesh model to obtain a lightweight aircraft whole mesh model. The flame template construction and brightness calculation module 220 is used to construct corresponding flame flow field distribution templates for different engine operating conditions based on the three-dimensional flow field temperature distribution calculation data of the engine flame, extract the brightness change law in the flame flow field distribution template as the basic data for rendering, call the flame radiation multi-line group broadband k-distribution model, and generate the flame brightness distribution under different perspectives. The rendering content construction module 230 is used to construct infrared flame rendering content based on the lightweight aircraft mesh model and the flame brightness distribution from different perspectives. The GPU-accelerated rendering module 240 is used to send the infrared flame rendering content to the GPU for 3D rendering. It utilizes the Stream Output feature of DirectX 10.0 to cache and update the vertex data on the GPU side, and then re-enters the updated vertex data into the vertex shader to complete the loop rendering, thereby realizing GPU-accelerated 3D rendering of the infrared flame.

[0042] Specific limitations regarding the GPU-accelerated infrared flame 3D rendering device can be found in the limitations of the GPU-accelerated infrared flame 3D rendering method described above, and will not be repeated here. Each module in the aforementioned GPU-accelerated infrared flame 3D rendering device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0043] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a GPU-accelerated infrared flame 3D rendering method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0044] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0045] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps: A three-dimensional digital scene of the aircraft infrared simulation is obtained, which includes the whole aircraft mesh model and the calculated data of the three-dimensional flow field temperature distribution of the engine exhaust flame; The mesh model component classes of the aircraft whole mesh model are extracted by semantic analysis and geometric analysis. The general body mesh model and the flame mesh model are parameterized at the algorithm end. While instantiating the mesh model, a progressive adaptive rendering method based on LOD multi-level of detail technology is used to lightweight the aircraft whole mesh model to obtain a lightweight aircraft whole mesh model. Based on the calculated temperature distribution data of the three-dimensional flow field of the engine exhaust flame, corresponding exhaust flame flow field distribution templates are constructed for different engine operating conditions. The brightness variation law in the exhaust flame flow field distribution template is extracted as the basic data for rendering. The exhaust flame radiation multi-line group broadband k-distribution model is called to generate the exhaust flame brightness distribution under different perspectives. Based on the lightweight aircraft mesh model and the flame brightness distribution from different perspectives, infrared flame rendering content is constructed. The infrared flame rendering content is sent to the GPU for 3D rendering. The StreamOutput feature of DirectX 10.0 is used to cache and update the vertex data on the GPU side. The updated vertex data is then re-entered into the vertex shader to complete the loop rendering, thus realizing GPU-accelerated 3D rendering of infrared flames.

[0046] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor: A three-dimensional digital scene of the aircraft infrared simulation is obtained, which includes the whole aircraft mesh model and the calculated data of the three-dimensional flow field temperature distribution of the engine exhaust flame; The mesh model component classes of the aircraft whole mesh model are extracted by semantic analysis and geometric analysis. The general body mesh model and the flame mesh model are parameterized at the algorithm end. While instantiating the mesh model, a progressive adaptive rendering method based on LOD multi-level of detail technology is used to lightweight the aircraft whole mesh model to obtain a lightweight aircraft whole mesh model. Based on the calculated temperature distribution data of the three-dimensional flow field of the engine exhaust flame, corresponding exhaust flame flow field distribution templates are constructed for different engine operating conditions. The brightness variation law in the exhaust flame flow field distribution template is extracted as the basic data for rendering. The exhaust flame radiation multi-line group broadband k-distribution model is called to generate the exhaust flame brightness distribution under different perspectives. Based on the lightweight aircraft mesh model and the flame brightness distribution from different perspectives, infrared flame rendering content is constructed. The infrared flame rendering content is sent to the GPU for 3D rendering. The StreamOutput feature of DirectX 10.0 is used to cache and update the vertex data on the GPU side. The updated vertex data is then re-entered into the vertex shader to complete the loop rendering, thus realizing GPU-accelerated 3D rendering of infrared flames.

[0047] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0048] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0049] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A GPU-accelerated 3D rendering method for infrared flame projection, characterized in that, The method includes: A three-dimensional digital scene of the aircraft infrared simulation is obtained, which includes the whole aircraft mesh model and the calculated data of the three-dimensional flow field temperature distribution of the engine exhaust flame; The mesh model component classes of the aircraft whole mesh model are extracted by semantic analysis and geometric analysis. The general body mesh model and the flame mesh model are parameterized at the algorithm end. While instantiating the mesh model, a progressive adaptive rendering method based on LOD multi-level of detail technology is used to lightweight the aircraft whole mesh model to obtain a lightweight aircraft whole mesh model. Based on the calculated temperature distribution data of the three-dimensional flow field of the engine exhaust flame, corresponding exhaust flame flow field distribution templates are constructed for different engine operating conditions. The brightness variation law in the exhaust flame flow field distribution template is extracted as the basic data for rendering. The exhaust flame radiation multi-line group broadband k-distribution model is called to generate the exhaust flame brightness distribution under different perspectives. Based on the lightweight aircraft mesh model and the flame brightness distribution from different perspectives, infrared flame rendering content is constructed. The infrared flame rendering content is sent to the GPU for 3D rendering. The StreamOutput feature of DirectX 10.0 is used to cache and update the vertex data on the GPU side. The updated vertex data is then re-entered into the vertex shader to complete the loop rendering, thus realizing GPU-accelerated 3D rendering of infrared flames.

2. The GPU-accelerated infrared flame 3D rendering method according to claim 1, characterized in that, The general shape mesh model is the mesh model corresponding to the solid wall of the aircraft, and the flame mesh model is the dedicated mesh model corresponding to the flame of the aircraft turbofan engine.

3. The GPU-accelerated infrared flame 3D rendering method according to claim 1, characterized in that, By reducing the total data volume and the number of triangles in the aircraft mesh model, the initial scene loading time is accelerated and the overall scene loading frame rate is improved, thereby achieving lightweighting of the aircraft mesh model.

4. The GPU-accelerated infrared flame 3D rendering method according to claim 1, characterized in that, The engine's different operating conditions include at least cruise and afterburner conditions. The flame flow field distribution template constructed for each operating condition includes the physical morphological parameters and basic brightness distribution characteristics of the flame under the corresponding operating condition.

5. The GPU-accelerated infrared flame 3D rendering method according to claim 1, characterized in that, When using the aforementioned wideband k-distribution model of multi-line jet radiation to generate jet brightness distributions at different viewpoints: Based on the brightness variation law of the aforementioned flame flow field distribution template, and by adapting the flame radiation multi-line group broadband k-distribution model to the three radiation-participating components of water vapor, carbon dioxide, and soot particles in the flame, the flame radiation brightness under a specified observation angle is accurately calculated. The calculation result is used as the direct basis for generating the flame brightness distribution under different perspectives.

6. The GPU-accelerated infrared flame 3D rendering method according to claim 1, characterized in that, Based on the lightweight aircraft mesh model and the step-by-step rendering of the flames from different perspectives, the infrared flame rendering content is constructed as follows: Based on the flame brightness distribution under different viewpoints, a texture map containing thermal energy information adapted to infrared radiation effects is obtained. The texture map is mapped to the flame mesh model in the lightweight aircraft overall mesh model; Based on the optical characteristics of the infrared environment, the corresponding lighting and shadow parameters and post-processing effects are obtained and superimposed on the flame mesh model. Then, the particle system is fused to obtain the infrared flame rendering content.

7. The GPU-accelerated infrared flame 3D rendering method according to claim 1, characterized in that, This section describes how to implement GPU-side vertex data caching and updating using DirectX 10.0's Stream Output feature, including: After the vertex shader or geometry shader has finished processing the vertex data, the processed vertex data is directly cached into a dedicated data structure inside the GPU through the StreamOutput feature of DirectX 10.0; The vertex information required for flame rendering is updated in real time based on the cached vertex data, and the dedicated data structure is deployed in the GPU throughout the process without needing to be sent back to the CPU.

8. A GPU-accelerated infrared flame 3D rendering device, characterized in that, The device includes: The scene import module is used to acquire the three-dimensional digital scene of the aircraft infrared simulation. The three-dimensional digital scene includes the whole aircraft mesh model and the calculated data of the three-dimensional flow field temperature distribution of the engine exhaust flame. The lightweight mesh model module is used to extract the mesh model component classes of the aircraft whole mesh model through semantic analysis and geometric analysis. It performs parameterization processing on the general body mesh model and the flame mesh model at the algorithm end. While instantiating the mesh model, it adopts a progressive adaptive rendering method based on LOD multi-level of detail technology to lightweight the aircraft whole mesh model and obtain a lightweight aircraft whole mesh model. The flame template construction and brightness calculation module is used to construct corresponding flame flow field distribution templates for different engine operating conditions based on the three-dimensional flow field temperature distribution calculation data of the engine flame, extract the brightness variation law in the flame flow field distribution template as the basic data for rendering, call the flame radiation multi-line group broadband k-distribution model, and generate the flame brightness distribution under different perspectives. The rendering content construction module is used to construct infrared flame rendering content based on the lightweight aircraft mesh model and the flame brightness distribution from different perspectives. The GPU-accelerated rendering module is used to send the infrared flame rendering content to the GPU for 3D rendering. It utilizes the Stream Output feature of DirectX 10.0 to cache and update the vertex data on the GPU side, and then re-enters the updated vertex data into the vertex shader to complete the loop rendering, thereby realizing GPU-accelerated 3D rendering of the infrared flame.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images