A texture-based three-dimensional cloud infrared calculation method

Abstract

The application discloses a three-dimensional cloud infrared calculation method based on texture, comprising the following steps: generating normal vectors of each spherical surface in a football model according to the football model; generating texture maps of a three-dimensional cloud which are consistent with the normal direction of each spherical surface respectively according to preset parameters of the three-dimensional cloud and the normal vectors of each spherical surface; placing the texture maps of the three-dimensional cloud which are consistent with the normal direction of each spherical surface respectively in a preset center position of the three-dimensional cloud in a scene editor; selecting one of the texture maps as an infrared cloud picture and selecting another one of the texture maps as an infrared cloud shadow; obtaining the gray value of the intersection point of the ray of a simulated detector and the infrared cloud picture; and calculating the radiation brightness of the intersection point according to a preset temperature and the gray value of the intersection point. The method can effectively improve the calculation efficiency of the infrared cloud picture while ensuring that the infrared cloud picture is consistent with the form of the cloud in the scene editor.

Description

Technical Field

[0001] This invention belongs to the field of infrared radiation calculation technology, specifically relating to a texture-based three-dimensional cloud infrared calculation method. Background Technology

[0002] In simulated infrared scene calculations, clouds are an indispensable element, and their radiation effect affects the infrared radiation effect of the entire field of view and related elements. In order to simulate the infrared radiation effect of three-dimensional clouds, researchers have proposed several simulation methods, including cloud modeling based on fractal methods, cloud modeling based on particle systems, and cloud modeling based on cellular automata.

[0003] For cloud modeling based on fractal methods, see the paper "Dynamic Cloud Simulation Based on Fractal Geometry" published in the journal Computer Simulation by Shi Jiandi et al. in April 2006. It uses the Diamond-Square algorithm to create a surface with relatively complex textures, and then generates cloud textures. However, the values ​​of each pixel in the generated cloud texture map are assigned based on random numbers generated by the computer system. This results in different cloud textures being calculated and generated each time, and the cloud textures generated at different angles are completely unrelated, making it difficult to form a three-dimensional cloud texture. It is difficult to use this method to form three-dimensional clouds in the scene editor. If this method is used to generate cloud textures in the scene editor, it will be difficult to keep the shape of the cloud consistent with that in the scene editor if the method used in the scene editor is different. For cloud modeling based on particle systems and cloud modeling based on cellular automata, see Li Ping's 2020 dissertation "Research on Key Technologies of 3D Cloud Modeling and Rendering" published at Nanjing University of Information Science and Technology. In the cloud modeling based on particle systems, it is pointed out that the number of cloud particles has a significant impact on the simulation results. Moreover, the shape of 3D clouds is dynamically changing, and it is difficult to capture and store all the positions and states of all particles at the current moment in the scene editor. Furthermore, the complexity of calculation and rendering is high, and it is not convenient to export data for infrared scene calculation. As a result, cloud textures with the same shape as those in the scene editor cannot be used in infrared scenes. In cloud modeling based on cellular automata, regular 3D meshes are used to simulate the average division of space. Each mesh represents a single cell, and each cell represents a single water vapor particle. During the modeling process, parameter values ​​need to be obtained through a large number of experiments to achieve ideal values. Moreover, different parameter values ​​are used for different cloud shapes. At the same time, due to the large scene and the existence of various cloud shapes and types, the computational load for generating multiple cloud textures is very large, which is not conducive to infrared scene simulation calculations. Furthermore, the initial values ​​in each cell are set using a random value method, which cannot guarantee the consistency of the morphology of each generated 3D cloud, and thus cannot guarantee that the morphology of the 3D cloud remains consistent between the infrared scene and the scene editor.

[0004] Because the infrared clouds generated by existing technologies and their shadow shapes on the ground cannot be consistent with the 3D cloud effects in the scene editor in terms of shape, the simulation effect is not ideal, which reduces the user experience. Summary of the Invention

[0005] To address the aforementioned problems in the prior art, the present invention is achieved through the following technical solution: This invention provides a texture-based three-dimensional cloud infrared calculation method, comprising: Step S1: Based on the soccer model, generate the normal vector of each sphere in the soccer model; Step S2: Based on the preset parameters of the 3D cloud and the normal vector of each sphere, generate texture maps of the 3D cloud that are consistent with the normal direction of each sphere. Step S3: Place the texture map of the 3D cloud, which is consistent with the normal direction of each sphere, at the preset center position of the 3D cloud in the scene editor. Step S4: Select one texture map from all texture maps and mark it as an infrared cloud map. The infrared cloud map serves as the texture map of the three-dimensional cloud in the infrared scene. Select one texture map from all texture maps that is different from the infrared cloud map and mark it as an infrared cloud shadow. The infrared cloud shadow serves as the texture map of the shadow cast by the three-dimensional cloud relative to the ground in the infrared scene. Step S5: Obtain the grayscale value of the intersection point of the infrared cloud image and the ray, wherein the ray comes from the simulated detector preset in the scene editor; Step S6: Calculate the radiance of the intersection point based on the preset temperature and the gray value of the intersection point. The radiance is used as the radiance of the intersection point in the infrared scene. Step S7: Based on the field of view angle of the simulated detector and the number of pixels divided by the field of view range, repeat steps S5 to S6 to obtain the radiance of other points in the infrared cloud image.

[0006] In one embodiment of the present invention, generating the normal vector of each sphere in the soccer model according to the soccer model includes: connecting the center point of the soccer model with the center point of each sphere, and calculating the normal vector of the sphere containing the center point connection line according to the center point connection line.

[0007] In one embodiment of the present invention, the texture maps of the three-dimensional cloud, which are consistent with the normal directions of each sphere, are placed at preset center positions of the three-dimensional cloud in the scene editor, including: aligning the center points of all texture maps with the preset center positions of the three-dimensional cloud in the scene editor, and ensuring that the normal direction of each texture map is the normal direction of the corresponding sphere; and moving each texture map along its normal direction according to the preset size of the three-dimensional cloud, so that all texture maps do not overlap in the space of the scene editor.

[0008] In one embodiment of the present invention, selecting one texture map from all texture maps and marking it as an infrared cloud map, and selecting one texture map different from the infrared cloud map from all texture maps and marking it as an infrared cloud shadow, includes: In all texture maps, the texture map whose normal direction is consistent with the reverse direction of the ray is marked as the first cloud map. In all first cloud maps, the texture map with the smallest angle to the reverse direction of the ray is marked as the infrared cloud map, wherein the ray comes from the simulated detector preset in the scene editor. In all texture maps, the texture map whose normal direction is consistent with the direction of the sun is marked as the second cloud map. In all second cloud maps, the texture map of the 3D cloud with the smallest angle to the direction of the sun is marked as the infrared cloud shadow, wherein the direction of the sun is the direction from the sun to the earth, the sun is the simulated sun component preset in the scene editor, and the earth is the simulated earth component preset in the scene editor.

[0009] In one embodiment of the present invention, obtaining the grayscale value of the intersection point of the infrared cloud image and the ray includes: When ray tracing is performed on a ray from a pre-set simulated detector in the scene editor, the geometric coordinates of the intersection point of the ray and the infrared cloud map are obtained, the texture coordinates of the intersection point are calculated based on the geometric coordinates, the pixel position of the intersection point is calculated based on the texture coordinates and the size of the infrared cloud map, and the gray value of the intersection point is obtained based on the pixel position.

[0010] In one embodiment of the present invention, calculating the radiance of the intersection point based on a preset temperature and the grayscale value of the intersection point includes: According to the formula Calculate the median radiance value at the intersection point, where L represents the median radiance value at the intersection point, in units of W / (sr·m²). ~ This indicates the wavelength range of the preset band, in meters (m). Wavelength is indicated, with the unit being meters (m). Indicates to Integral This represents the average emissivity of the preset band. This represents the first radiation constant, with units of W·m. 2 , This represents the second radiation constant, in m·K, where T is the preset temperature, in K; according to the formula... Calculate the radiance at the intersection point, where, The radiance at the intersection point is expressed in W / (sr·m²), and rho represents the modulation coefficient at the intersection point, according to the formula... Calculate the modulation coefficient, where gray represents the gray value of the intersection point.

[0011] In one embodiment of the present invention, obtaining the radiance of other points in the infrared cloud image includes: obtaining the normal direction of the ray at each pixel position based on the number of pixels divided by the field of view angle and field range of the simulated detector; repeating steps S5 and S6 based on the normal direction of the ray at each pixel position until all other intersection points of the ray at each pixel position with the infrared cloud image are calculated, and the radiance of all other intersection points is obtained.

[0012] In one embodiment of the present invention, the soccer ball model includes thirty-two spherical surfaces, including twelve regular pentagonal spheres and twenty regular hexagonal spheres.

[0013] In one embodiment of the present invention, the soccer ball model includes ninety-two spherical surfaces, the ninety-two spherical surfaces including: keeping the twelve regular pentagonal spherical surfaces in the soccer ball structure unchanged, and decomposing each of the twenty regular hexagonal spherical surfaces in the soccer ball structure into four identical triangular spherical surfaces.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The present invention provides a texture-based three-dimensional cloud infrared calculation method, which is based on a soccer ball model. The soccer ball model ensures that the angle between the normal directions of any two adjacent spheres in the same soccer ball model remains consistent. Texture maps of three-dimensional clouds with different shapes under the normal direction of each sphere in the soccer ball model are generated respectively. That is, it can simulate the texture maps of the three-dimensional cloud uniformly distributed in multiple directions under the full view in the infrared scene. Then, in the scene editor, simulated detectors, simulated solar components and simulated earth components are preset according to the requirements of the infrared scene. The ray direction of the simulated detector is modeled as the observer's perspective. Among the texture maps whose normal direction is consistent with the reverse direction of the ray, the texture map with the smallest angle to the reverse direction of the ray is selected and marked as the infrared cloud map. This can reconstruct the texture map that is closest to the three-dimensional cloud from the observer's perspective. Among the texture maps whose normal direction is consistent with the direction of the sun, the texture map with the smallest angle to the direction of the sun is selected and marked as the infrared cloud shadow. This can reconstruct the cloud shadow formed by the three-dimensional cloud on the ground under sunlight in the infrared scene. Based on the infrared cloud map, the radiance of each point in the infrared cloud map is further calculated, thereby obtaining an infrared image of the cloud that is consistent with the three-dimensional cloud shape in the scene editor.

[0015] 2. Compared to traditional methods where the parameter values ​​of each cell in the infrared scene of modeling the 3D cloud require extensive experimentation to obtain ideal values, resulting in a large computational burden when generating multiple cloud textures, this invention reconstructs the radiance calculation formula of the infrared cloud image and controls the modulation coefficient, giving the infrared radiance of the entire cloud texture a gradient effect. This makes the 3D cloud shape more realistic from the observer's perspective, further improving the consistency with the 3D cloud shape in the scene editor, enhancing the user experience, effectively reducing the cost of determining the infrared image of the 3D cloud in the infrared scene, and increasing the computational speed of 3D cloud infrared simulation.

[0016] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a texture-based three-dimensional cloud infrared calculation method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the texture maps of 3D clouds under different normal directions of 32 spherical surfaces generated by the texture-based 3D cloud infrared calculation method in Embodiment 1 of the present invention. Figure 3 This is a schematic diagram comparing the infrared cloud image of the three-dimensional cloud generated in the scene editor provided in this embodiment of the invention with that generated by the infrared cloud image of the three-dimensional cloud based on the texture-based three-dimensional cloud infrared calculation method. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve the intended purpose, the following detailed description of the solution according to the present invention is provided in conjunction with the accompanying drawings and specific embodiments.

[0019] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations are intended to cover non-exclusive inclusion, such that an article or apparatus comprising a list of elements includes not only those elements but also other elements not expressly listed. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or apparatus that includes said element.

[0020] Example 1 Please see Figure 1 , Figure 1 This is a flowchart illustrating a texture-based three-dimensional cloud infrared calculation method provided in an embodiment of the present invention. The three-dimensional cloud infrared calculation method includes: Step S1: Based on the soccer ball model, generate the normal vector of each sphere in the soccer ball model; Specifically, the center point of the soccer ball model is connected to the center point of each sphere, and the normal vector of the sphere containing the center point connection is calculated based on the center point connection.

[0021] A traditional soccer ball consists of thirty-two spherical surfaces: twenty regular hexagonal surfaces and twelve regular pentagonal surfaces. Therefore, in this embodiment, the coordinates of the center point of the soccer ball model are set to (0, 0, 0). These coordinates represent the X, Y, and Z axes in a three-dimensional coordinate system. The distance from the center point of the soccer ball model to the center point of each spherical surface is 1, a dimensionless physical quantity. The center point of one spherical surface in the soccer ball model is taken as (1, 0, 0), and the center point of the adjacent spherical surface is taken as (0.934172, 0.356822, 0). Once the coordinates of the center point of the soccer ball model are determined, the center points of the two adjacent spherical surfaces are also determined, thus determining the position and angle of the soccer ball model in the three-dimensional coordinate system. Connecting the center point of the soccer ball model to the center points of the thirty-two spherical surfaces allows for the calculation of the normal vector of each of the thirty-two spherical surfaces.

[0022] In this embodiment, among the normal vectors of each of the thirty-two spheres, the normal vector of the first sphere is (0.525731, 0, 0.850651), the normal vector of the second sphere is (-0.525731, 0, 0.850651), the normal vector of the third sphere is (0.577350, -0.577350, -0.577350), the normal vector of the fourth sphere is (0.934172, 0, -0.356822), and the normal vector of the fifth sphere is (0.934172, 0, 0.356822). These normal vectors of the spheres are all dimensionless unit vectors. The normal vectors of the other spheres will not be described further.

[0023] Step S2: Based on the preset parameters of the 3D cloud and the normal vector of each sphere, generate texture maps of the 3D cloud that are consistent with the normal direction of each sphere. Please see Figure 2 , Figure 2 This is a schematic diagram of the texture maps of 3D clouds under different normal directions of the 3D clouds generated by the texture-based 3D cloud infrared calculation method in Embodiment 1 of the present invention. In this embodiment, a new scene is created using the complex battlefield environment infrared scene generation software (XD_CAE-IR). The latitude and longitude of the scene are set to (120°E, 40°N). The 3D coordinate system in the scene editor is constructed as follows: In the scene editor, the latitude and longitude point at the lower left corner is taken as the origin O. The directions of the three axes are defined based on the scene coordinate system. Specifically, the positive direction of the x-axis points east (unit: mm), the positive direction of the y-axis points north (unit: mm), and the positive direction of the z-axis points to the sky (unit: mm). The z-axis points vertically outward from the origin O, and the z-axis conforms to the right-hand rule with the x-axis and y-axis. Using ray Marching (RM) technology, based on the preset parameters of the 3D cloud and the normal vectors of the thirty-two spherical surfaces in the soccer model, texture maps of the 3D cloud are generated from the perspective of the normal direction of each spherical surface in the soccer model. In other words, texture maps of the 3D cloud are generated one-to-one with the normal direction of the corresponding spherical surface. The preset parameters of the 3D cloud include: the three-axis dimensions of the ellipsoid, the texture image resolution, and the reference matrix of the 3D cloud. Specifically, the ellipsoidal dimensions of the 3D cloud are set to (2, 4, 2), the texture image resolution of the 3D cloud is 256*256, and the reference matrix of the 3D cloud is set to [0, 0.8, 0.6, -0.8, 0.36, -0.48, -0.6, -0.48, 0.64]. Generating texture maps of the 3D cloud under different normal directions based on preset parameters and the normal vector of each sphere in the soccer ball model is a common technique in this field and will not be elaborated further here. Furthermore, this invention does not limit the model or type of scene editor.

[0024] Step S3: Place the texture map of the 3D cloud, which is consistent with the normal direction of each sphere, at the preset center position of the 3D cloud in the scene editor. Specifically, the center point of all texture maps is aligned with the preset center position of the 3D cloud in the scene editor, and the normal direction of each texture map is the normal direction of the corresponding sphere. According to the preset size of the 3D cloud, each texture map is moved along its normal direction, and each texture map is offset relative to its initial position, which is the preset center position of the 3D cloud in the scene editor. This ensures that all texture maps do not overlap in the space of the scene editor, thus avoiding overlap between texture maps in space. When the offset distance is positive, it means that the texture map is moved along its normal direction; when the offset distance is negative, it means that the texture map is moved in the opposite direction of its normal direction.

[0025] Optionally, for ease of operation, different bulletin boards can be placed at the preset center position of the 3D cloud in the scene editor to represent different texture maps. Each bulletin board corresponds to one of the texture maps mentioned above. In the scene editor, the bulletin board is used to replace the corresponding texture map. The method of operating the bulletin board in the scene editor is the same as the method of processing texture maps in the scene editor. That is, according to the preset center position and size of the 3D cloud in the scene editor, the coverage area of ​​the 3D cloud in the scene editor is determined. Within the coverage area, the bulletin board corresponding to the normal of each sphere is placed, and the orientation of each bulletin board is consistent with the normal direction of the sphere it is located on. The corresponding bulletin board is moved in the orientation direction of each bulletin board so that all bulletin boards do not overlap in space. In subsequent operations, the operation of processing the bulletin board is similar to the operation of processing the texture map represented by the bulletin board. The operation of the bulletin board will not be described again hereafter.

[0026] In this embodiment, the length coordinates of all texture maps are set on the x-axis, the width coordinates are set on the y-axis, and the normal coordinates are set on the z-axis. The origin of the coordinate system of all texture maps (the center point of the texture map) is the preset center position of the 3D cloud in the scene editor, namely (205058.0mm, 906876.4mm, 2545032.6mm). The preset size of the cloud is 20000mm*20000mm, that is, the preset length of the 3D cloud is 20000mm and the preset width of the 3D cloud is 20000mm. At the preset center position of the 3D cloud, texture maps with the normal directions of the thirty-two spheres in the soccer ball model are placed. Each texture map is moved +0.5mm along its normal direction, so that each texture map is offset relative to its initial position, so as to achieve the effect that all texture maps do not overlap in the space of the scene editor.

[0027] Step S4: Select one texture map from all texture maps and mark it as an infrared cloud map. This infrared cloud map serves as the texture map of the three-dimensional cloud in the infrared scene. Select one texture map from all texture maps that is different from the infrared cloud map and mark it as an infrared cloud shadow. This infrared cloud shadow serves as the texture map of the shadow cast by the three-dimensional cloud relative to the ground in the infrared scene. Specifically, among all texture maps, the texture map whose normal direction is consistent with the reverse direction of the ray is marked as the first cloud map. Among all the first cloud maps, the texture map with the smallest angle with the reverse direction of the ray is marked as the infrared cloud map. This ray comes from the simulated detector preset in the scene editor. Among all the texture maps, the texture map whose normal direction is consistent with the direction of the sun is marked as the second cloud map. Among all the second cloud maps, the texture map of the three-dimensional cloud with the smallest angle with the direction of the sun is marked as the infrared cloud shadow. Here, the direction of the sun is the direction in which the sun points to the earth. The sun is the simulated sun component preset in the scene editor, and the earth is the simulated earth component preset in the scene editor.

[0028] This invention proposes a model based on a soccer ball with thirty-two spherical surfaces. By using the soccer ball model, the angle between any two adjacent directional vectors in any of the thirty-two directions can be kept consistent. Texture maps of three-dimensional clouds with different shapes under the normal direction of the thirty-two spherical surfaces can be generated respectively. That is, it is possible to simulate the texture maps of the three-dimensional clouds uniformly distributed in the thirty-two directions under the full view in an infrared scene. Then, in the scene editor, according to the requirements of the infrared scene, simulated detectors, simulated solar components, and simulated Earth components are preset. The ray direction of the simulated detector is modeled as the observer's perspective. Therefore, the texture map in the first cloud map, whose normal direction is consistent with the reverse direction of the ray, is the texture map of the three-dimensional cloud as seen from the observer's perspective in the infrared scene. In the texture map of the first cloud map, the texture map with the smallest angle to the reverse direction of the ray is marked as the infrared cloud map, which can restore the texture map that is closest to the three-dimensional cloud as seen from the observer's perspective. Similarly, the texture map in the second cloud map, whose normal direction is consistent with the direction of the sun, is the cloud shadow of the three-dimensional cloud under the sun in the infrared scene. In the texture map of the second cloud map, the texture map with the smallest angle to the direction of the sun is marked as the infrared cloud shadow, which can restore the cloud shadow formed on the ground by the three-dimensional cloud under sunlight in the infrared scene.

[0029] In this embodiment, the simulated detector is located at (847690, -730410, 278351), its normal vector is (-0.0396405, 0.465429, -0.884197), its field of view is set to (30°*30°), and the simulated solar component is located at (120.805859°E, 23.012907°N). Based on the average distance between the Earth and the Sun and the three-dimensional coordinate system in the scene editor, the solar direction vector (0.01313473, -0.89009951, 0.45382296) can be calculated. When calculating the solar direction vector using the three-dimensional coordinate system in the scene editor, the solar direction vector in the scene coordinate system at the current moment is calculated based on the time (Beijing time) and latitude and longitude information defined in the scene, thereby calculating the texture map with the smallest angle to the solar direction.

[0030] Step S5: Obtain the grayscale value of the intersection point between the infrared cloud image and the ray; Specifically, when ray tracing is performed on the ray from the simulated detector preset in the scene editor, the geometric coordinates of the intersection point of the ray and the infrared cloud map are obtained, that is, the three-dimensional coordinates of the intersection point in the scene editor. The texture coordinates of the intersection point are calculated based on the geometric coordinates of the intersection point. The pixel position of the intersection point is calculated based on the texture coordinates of the intersection point and the size of the infrared cloud map. The gray value of the intersection point is obtained based on the pixel position of the intersection point.

[0031] In this embodiment, the three-dimensional coordinates of the intersection point of the ray and the infrared cloud map in the scene editor are (206501.7mm, 913284.4mm, 2545032.6mm). Based on the geometric coordinates of the intersection point, the coordinates of the intersection point in the coordinate system of the infrared cloud map are calculated to be (95.3, 684.2, 0). Then, the texture coordinates of the intersection point are calculated to be (0.50953, 0.56482). Subsequently, based on the texture coordinates (0.50953, 0.56482) and the size of the infrared cloud map (i.e., the width and height of the texture map), the pixel position of the intersection point is calculated and the corresponding gray value is taken. Specifically, the pixel row and column number corresponding to the intersection point is (130, 145), and its corresponding gray value is 200. Typically, the actual grayscale value range of a 3D cloud texture map is (0~255). When the grayscale value is greater than zero, it indicates that the current grayscale value is a valid grayscale value of the 3D cloud. When the grayscale value of the intersection point is greater than a preset threshold, it indicates that the current intersection point is a point on the 3D cloud. When the grayscale value of the intersection point is less than or equal to the preset threshold, it indicates that the current intersection point is a point around the 3D cloud. In this embodiment, the preset threshold is zero.

[0032] Step S6: Calculate the radiance of the intersection point based on the preset temperature and the gray value of the intersection point.

[0033] First, according to the formula Calculate the median radiance value at the intersection point, where L represents the median radiance value at the intersection point, in units of W / (sr·m²). ~ This indicates the wavelength range of the preset band, in meters (m). Wavelength is indicated, with the unit being meters (m). Represents the 5th power of the wavelength. Indicates to Integral This represents the average emissivity of the preset band. This represents the first radiation constant, with units of W·m. 2 The value is 3.7417749 × 10 -16 W·m 2 , The second radiation constant is represented by m·K and has a value of 0.01438769 m·K. T is the preset temperature, in K.

[0034] Subsequently, according to the formula Calculate the radiance at the intersection point, where, The radiance at the intersection point is expressed in W / (sr·m²), and rho represents the modulation coefficient at the intersection point, according to the formula... Calculate the modulation coefficient, where gray represents the gray value of the intersection point.

[0035] In this embodiment, the wavelength range is 3µm to 5µm, the preset temperature of the three-dimensional cloud is 295.675931K, i.e., T=295.675931K, and the average emissivity is... It is 0.8, that is =0.8, and thus the median radiance at the intersection point is calculated to be 1.3667*10. 12 W / (sr·m²), since the gray value at the intersection point is 200, i.e., gray=200, the modulation coefficient at this intersection point is calculated to be 0.78125, i.e., rho=0.78125, and the radiance at the intersection point is 1.0677*10. 12 W / (sr·m²).

[0036] Step S7: Based on the field of view angle of the simulated detector and the number of pixels divided by the field of view range, repeat steps S5 to S6 to obtain the radiance of other points in the infrared cloud image.

[0037] Specifically, based on the field of view angle of the simulated detector preset in the scene editor, for example, 30°*30°, and the number of pixels divided by the field of view of the simulated detector, for example, the number of pixels divided is 512*512, the normal direction of the ray at each pixel position is obtained. Similar to obtaining the radiance of the intersection point in steps S5 and S6, steps S5 and S6 are repeated based on the normal direction of the ray at each pixel position until all other intersection points of the ray at each pixel position with the infrared cloud map are calculated, and the radiance of all other intersection points is obtained, thereby obtaining all the radiance of the three-dimensional cloud in the infrared scene.

[0038] This invention reconstructs the radiance calculation formula of infrared cloud images and controls the modulation coefficient to create a gradient effect in the infrared radiance of the entire cloud texture. This results in a more realistic 3D cloud shape as seen from the observer's perspective in an infrared scene, further improving the consistency with the 3D cloud shape in the scene editor and enhancing the user experience. Furthermore, in traditional methods, when modeling the texture map of the 3D cloud in an infrared scene, the parameter values ​​of each cell require extensive experimentation to obtain ideal values, leading to a high computational load when generating multiple cloud textures. This invention effectively reduces the cost of determining the infrared image of the 3D cloud in an infrared scene and improves the computational speed of 3D cloud infrared simulation.

[0039] Example 2 In another embodiment of the present invention, the soccer model includes ninety-two spherical surfaces. Each of the twenty regular hexagonal spherical surfaces in the thirty-two spherical surfaces of the soccer model in Embodiment 1 is decomposed into four identical triangular spherical surfaces, while keeping the other twelve regular pentagonal spherical surfaces unchanged, thereby obtaining a soccer model with ninety-two spherical surfaces, which has a different number of spherical surfaces than in Embodiment 1.

[0040] Subsequently, using the same method as in Example 1, namely the corresponding methods in steps S1 to S7, the texture map of the 3D cloud in the infrared scene that most closely resembles the 3D cloud in the scene editor, as well as the texture map of the 3D cloud's shadow relative to the ground, are obtained using the 92-spherical soccer ball model. Compared to the 32-spherical soccer ball model, the 92-spherical soccer ball model has a higher and more uniform spherical density in space, thus allowing for the acquisition of spherical normal directions with a higher spatial density. Therefore, compared to the 32-spherical soccer ball model, a texture map that more closely resembles the 3D cloud in the infrared scene can be obtained. The radiance of the 3D cloud is calculated using this texture map in the infrared scene, thereby obtaining the infrared image of the 3D cloud that most closely resembles the cloud in the scene editor. This effectively reduces the cost of determining the infrared image of the 3D cloud in the infrared scene while maintaining consistency with the morphology of the 3D cloud in the scene editor, improving the user experience. The above steps are the same as in Example 1 and will not be repeated here.

[0041] In this embodiment, a soccer ball model with 92 spherical surfaces is used. This ensures that the angle between the normal directions of any two adjacent spherical surfaces remains consistent, simulating the texture map of the 3D cloud uniformly distributed in 92 directions across the entire infrared scene. This results in a texture map of the 3D cloud most closely resembling the observer's viewpoint, as well as a texture map of the cloud's projection onto the ground. In other embodiments, a texture map of the 3D cloud that maintains consistency with the cloud's shape in the scene editor can be obtained. In other embodiments, the soccer ball model can also be a soccer ball model with other numbers of spherical surfaces.

[0042] Please see Figure 3 , Figure 3 This is a comparative illustration of the infrared cloud image of a 3D cloud generated in the scene editor provided in this embodiment of the invention and generated by the infrared cloud image of the same 3D cloud using a texture-based 3D cloud infrared calculation method. Specifically, Figure 3 (a) in the image is a 3D cloud generated in the scene editor. Figure 3 (b) in the figure is a texture-based three-dimensional cloud infrared calculation method according to an embodiment of the present invention. Figure 3 The infrared cloud image generated from the 3D cloud in (a) shows that the infrared cloud image of the 3D cloud finally generated using this application is consistent with the 3D cloud shape in the scene editor.

[0043] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A texture-based three-dimensional cloud infrared calculation method, characterized in that, include: Step S1: Based on the soccer model, generate the normal vector of each sphere in the soccer model; Step S2: Based on the preset parameters of the 3D cloud and the normal vector of each sphere, generate texture maps of the 3D cloud that are consistent with the normal direction of each sphere. Step S3: Place the texture map of the 3D cloud, which is consistent with the normal direction of each sphere, at the preset center position of the 3D cloud in the scene editor. Step S4: Select one texture map from all texture maps and mark it as an infrared cloud map. The infrared cloud map serves as the texture map of the three-dimensional cloud in the infrared scene. Select one texture map from all texture maps that is different from the infrared cloud map and mark it as an infrared cloud shadow. The infrared cloud shadow serves as the texture map of the shadow cast by the three-dimensional cloud relative to the ground in the infrared scene. Step S5: Obtain the grayscale value of the intersection point of the infrared cloud image and the ray, wherein the ray comes from the simulated detector preset in the scene editor; Step S6: Calculate the radiance of the intersection point based on the preset temperature and the gray value of the intersection point. The radiance is used as the radiance of the intersection point in the infrared scene. Step S7: Based on the field of view angle of the simulated detector and the number of pixels divided by the field of view range, repeat steps S5 to S6 to obtain the radiance of other points in the infrared cloud image.

2. The texture-based three-dimensional cloud infrared calculation method according to claim 1, characterized in that, Based on the soccer ball model, generate the normal vector for each sphere in the soccer ball model, including: Connect the center point of the soccer ball model to the center point of each sphere, and calculate the normal vector of the sphere containing the center point connection line.

3. The texture-based three-dimensional cloud infrared calculation method according to claim 2, characterized in that, The texture maps of the 3D cloud, which are aligned with the normal directions of each sphere, are placed at preset center positions in the scene editor, including: Make the center point of all texture maps coincide with the preset center position of the 3D cloud in the scene editor, and make the normal direction of each texture map the normal direction of the corresponding sphere. Based on the preset size of the 3D cloud, each texture map is moved along the normal direction of each texture map so that all texture maps do not overlap in the space of the scene editor.

4. The texture-based three-dimensional cloud infrared calculation method according to claim 3, characterized in that, Select one texture map from all texture maps and mark it as an infrared cloud map. Select one texture map from all texture maps that is different from the infrared cloud map and mark it as an infrared cloud shadow, including: In all texture maps, the texture map whose normal direction is consistent with the reverse direction of the ray is marked as the first cloud map. In all first cloud maps, the texture map with the smallest angle with the reverse direction of the ray is marked as the infrared cloud map, wherein the ray comes from the simulated detector preset in the scene editor. In all texture maps, the texture map whose normal direction is consistent with the direction of the sun is marked as the second cloud map. In all second cloud maps, the texture map of the three-dimensional cloud with the smallest angle to the direction of the sun is marked as the infrared cloud shadow. The direction of the sun is the direction in which the sun points to the earth. The sun is the simulated sun component preset in the scene editor, and the earth is the simulated earth component preset in the scene editor.

5. The texture-based three-dimensional cloud infrared calculation method according to claim 4, characterized in that, Obtaining the grayscale value of the intersection point between the infrared cloud image and the ray includes: When ray tracing is performed on a ray from a pre-set simulated detector in the scene editor, the geometric coordinates of the intersection point of the ray and the infrared cloud map are obtained, the texture coordinates of the intersection point are calculated based on the geometric coordinates, the pixel position of the intersection point is calculated based on the texture coordinates and the size of the infrared cloud map, and the gray value of the intersection point is obtained based on the pixel position.

6. The texture-based three-dimensional cloud infrared calculation method according to claim 5, characterized in that, Calculating the radiance of the intersection point based on the preset temperature and the gray value of the intersection point includes: According to the formula Calculate the median radiance value at the intersection point, where L represents the median radiance value at the intersection point, in units of W / (sr·m²). ~ This indicates the wavelength range of the preset band, in meters (m). Wavelength is indicated, with the unit being meters (m). Indicates to Integral This represents the average emissivity of the preset band. This represents the first radiation constant, with units of W·m. 2 , The second radiation constant is represented by m·K, and T is the preset temperature, in K. According to the formula Calculate the radiance at the intersection point, where, The radiance at the intersection point is expressed in W / (sr·m²), and rho represents the modulation coefficient at the intersection point, according to the formula... Calculate the modulation coefficient, where gray represents the gray value of the intersection point.

7. The texture-based three-dimensional cloud infrared calculation method according to claim 6, characterized in that, Obtaining the radiance of other points in the infrared cloud image includes: Based on the field of view angle and the number of pixels divided by the field of view range of the simulated detector, the normal direction of the ray at each pixel position is obtained. Based on the normal direction of the ray at each pixel position, steps S5 and S6 are repeated until all other intersection points of the ray at each pixel position with the infrared cloud map are calculated, and the radiance of all other intersection points is obtained.

8. The texture-based three-dimensional cloud infrared calculation method according to any one of claims 1-7, characterized in that, The soccer ball model comprises thirty-two spherical surfaces, including twelve regular pentagonal spheres and twenty regular hexagonal spheres.

9. The texture-based three-dimensional cloud infrared calculation method according to any one of claims 1-7, characterized in that, The soccer ball model comprises ninety-two spherical surfaces, which include: keeping the twelve regular pentagonal spherical surfaces in the soccer ball structure unchanged, and decomposing each of the twenty regular hexagonal spherical surfaces in the soccer ball structure into four identical triangular spherical surfaces.

Images