A method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data

Abstract

The present invention provides a method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data, belonging to the fields of natural cloud visual scene simulation and radar image simulation. This method uses the new generation radar basic reflectivity image of the National Meteorological Information Center, designs an image processing algorithm to extract the radar reflectivity map in the image, and uses this data to generate a visual scene cloud data map as the basis for representing the basic shape and concentration of the visual scene cloud; based on the radar reflectivity map and the visual scene cloud data map, a radar cloud data map required for radar image simulation is generated as the basis for representing the statistical information of the electromagnetic characteristics of cloud and rain particles, which is used for radar echo generation, so that the visual scene cloud picture seen by the operator in flight simulation and the cloud imaging picture presented by the radar are consistent, providing support for training the operator to interpret meteorological targets.

Description

Technical Field

[0001] The present invention belongs to the fields of natural cloud visual scene simulation and radar image simulation, and particularly relates to a method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data. Background Art

[0002] Flight simulation training plays an important role in military training. The development of advanced avionics system simulation technology and visual scene simulation technology enables flight personnel to improve their operation and decision-making abilities under the simulation fidelity of equipment close to the actual installation, and also enables them to obtain a sense of immersion in training in a realistic natural environment, enhancing the functions and effects of simulation training.

[0003] As an important sensor device on an aircraft, a radar has various functions such as target reconnaissance, terrain mapping, and meteorological detection. It is one of the main simulation devices for flight simulation training. The simulation of these functions requires the radar to interact with the environmental targets in the visual scene simulation. The detectable targets for realizing certain training functions in the simulation environment should be converted into radar detection information. If these targets will appear in the pilot's field of vision, they should also be realistically displayed in the visual scene simulation. This requires the radar detection information and the visual scene simulation information to be consistent, so as to ensure the consistency between the radar image screen and the visual scene image screen.

[0004] Currently, the function of the airborne radar simulation system is complete, and it maintains a high consistency with the visual scene in terms of platform target detection and imaging, and terrain mapping. However, there are the following deficiencies in the radar imaging simulation of cloud and rain meteorological targets and the natural cloud simulation in the visual scene:

[0005] 1) Existing simulation methods mostly focus on the realization of single functions, and the established models can only achieve meteorological target detection simulation or visual scene simulation, and cannot meet the requirements for the consistency of visual scene cloud and meteorological radar cloud imaging simulation;

[0006] 2) In the simulation modeling of airborne meteorological radar imaging driven by visual scene data, emphasis is placed on the manifestation of the radar image effect, and a simulation model is not established based on the principle of meteorological target radar imaging to simulate the attenuation characteristics of meteorological targets to radar electromagnetic waves. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide a method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data in view of the above deficiencies of the existing technology, enabling flight personnel to judge the correlation between the two through visual scene clouds and radar images and training their radar image interpretation ability.

[0008] A method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data according to the present invention is characterized by including the following steps:

[0009] Step S1 extracts corresponding regions through image processing based on the radar reflectivity values, roads, and text objects identified by different colors in the radar basic reflectivity image.

[0010] Step S2 generates a radar reflectivity image based on the regions obtained in Step S1 and their object types.

[0011] Step S3 generates a visual scene cloud data map with the same pixel size through the radar reflectivity image generated in Step S2, and calculates the values of each point in the visual scene cloud data map.

[0012] Step S4 uses the visual scene cloud data map generated in Step S3 to design a method for representing the visual scene cloud data, and calculates the cloud density in the three-dimensional cloud field.

[0013] Step S5 generates a radar cloud data map through the radar reflectivity image generated in Step S2 and the visual scene cloud data map generated in Step S3, and calculates the values of each point in the radar cloud data map.

[0014] Step S6 uses the radar cloud data map generated in Step S5 and the cloud density calculation method in Step S4 to design a method for representing the radar cloud data.

[0015] The specific steps of Step S1 are as follows:

[0016] Step S11: According to the color value definition of different basic reflectivity values in the radar basic reflectivity image (image product of the National Meteorological Science Data Center), use color thresholds to segment the image and extract different radar basic reflectivity regions.

[0017] Step S12: Merge different radar basic reflectivity regions and obtain a closed region through the closing operation of the regions.

[0018] Step S13: According to the color definition of place names and ground features in the radar basic reflectivity image, use color thresholds to extract the corresponding regions, and find the intersection of this region and the closed region in S12 to obtain the place name and ground feature regions within the closed region.

[0019] The specific steps of Step S2 are as follows:

[0020] Step S21: Generate a grayscale image with the same size as the radar basic reflectivity image. According to the radar basic reflectivity regions in S11, transform the radar basic reflectivity value z′ of each pixel point in different radar basic reflectivity regions into the grayscale value g of the corresponding pixel point in the grayscale image, then:

[0021] g = [16 - (z′ / 5)] * 10

[0022] z' ∈ {5, 15, 20, 25, 35, 40, 45, 50, 55, 60, 65, 70, 75}. The grayscale values of other pixel points are set to 255;

[0023] Step S22: Calculate the pixel coordinates of the place names and ground object areas obtained in S13, and use the filling algorithm to set the grayscale values of each pixel of the grayscale image in S21 corresponding to the pixel coordinates, so as to obtain the filled grayscale image.

[0024] Step S23: For the grayscale image obtained in step S22, use the grayscale threshold to extract the area with radar reflectivity, obtain the minimum circumscribed rectangle of this area, the width M of the rectangle, the height N of the rectangle, and intercept the image within the circumscribed rectangle range from the image as the radar reflectivity image.

[0025] Preferably, in step S22, the filling algorithm used when setting the grayscale value of the grayscale image includes the following steps:

[0026] Step S221: Make a copy of the grayscale image in step S21 as a copy image;

[0027] Step S223: For the pixel coordinates of each place name and ground object area, search in the copy image for the pixel point with the closest grayscale value less than 255 to this coordinate point, and assign the grayscale value of this point to the corresponding coordinate point of the grayscale image in S21.

[0028] The specific steps of the said step S3 are as follows:

[0029] According to the radar reflectivity image in step S2, generate a visual scene cloud data map with the same pixel size. For any pixel point in the radar reflectivity image:

[0030] a. If the pixel value of this point is less than 255, calculate the value of the corresponding coordinate point of the visual scene cloud data map according to this pixel value;

[0031] b. If the pixel value of this point is equal to 255, with this point as the center, set the area width w, and use the window search method to calculate the corresponding value of the visual scene cloud data map.

[0032] For any coordinate point P(m, n), where m = 1, 2,..., M, n = 1, 2,..., N are the coordinates in the image, the pixel value of the radar reflectivity image at this place is g(m, n), and calculate the value of the visual scene cloud data map at this place as c(m, n):

[0033] If g(m, n) < 255, c(m, n) = [160 - g(m, n)] / 510.0 + rand * 0.015686, where rand is a random number with a value range of [-1, 0);

[0034] If g(m,n) = 255, set a rectangular search area of w×w centered at this point, and find the pixel point P(m′,n′) closest to the pixel point P(m,n) with a pixel value less than 255 and the pixel value g(m′,n′). Then the value of c(m,n) is:

[0035]

[0036] δ is the distance attenuation factor, and its value is greater than 0. Through δ, the edge of the generated 3D visual scene cloud can have a sloped and smoothly transitioning appearance.

[0037] The specific steps of step S4 are as follows:

[0038] Step S41: The cloud density in the 3D visual scene cloud field consists of two parts. One is the radar-detectable cloud generated from the visual scene cloud data map, and the other is the cloud composed of fine cloud particles that cannot form a certain intensity radar echo. First, calculate the cloud density in the 3D visual scene cloud field according to the visual scene cloud data map in step S3.

[0039] When the distance of the cloud field range (left-handed coordinate system) in the XZ direction is W, L, and the thickness and cloud base height of the cloud field are H, b h , as Figure 1 shown, the position PC(x c ,y c ,z c ) of the center of the visual scene cloud data map in the cloud field, the distance size Δd represented by a single data point of the visual scene cloud data map, then the length of the coverage range of the visual scene cloud data map is ΔdM, and the width is ΔdN. The texture coordinates of the visual scene cloud data map are defined as: the lower left corner is (0,0), and the upper right corner is (1,1). For any point PC(x,y,z) in the 3D cloud field space, its coordinates (x′,z′) projected onto the coordinate system of the visual scene cloud data map are:

[0040]

[0041] If x′ ∈ [0, 1] and z′ ∈ [0, 1], then this point is located in the cloud field area where the visual scene cloud data map is located, and obtain the visual scene cloud map data value c at this place. Otherwise, let the obtained visual scene cloud map data value c = 0. The normalized height range of the main cloud body of the cloud field here is [y min ,y max , y min ∈(0, 1), y max ∈(0, 1), y min is a constant for adjusting the height of the cloud base, and y max is used to adjust the height of the cloud top, and the value is:

[0042]

[0043] c max is the maximum value of the visual scene cloud data map.

[0044] Then the cloud density at this location is:

[0045]

[0046] is the ratio of the distance from this position to the cloud base, c min is the minimum value of the visual scene cloud data map.

[0047] Step S42: Use the noise-based volume cloud generation method to calculate the visual scene cloud density at which the entire cloud field cannot form a radar echo of a certain intensity, and add it to the cloud density calculated in Step S41.

[0048] Use a four-channel three-dimensional Worley noise as the basic shape, and calculate the cloud density Pc2 at Pos(x, y, z) through three-dimensional noise sampling. The cloud density Pc(x, y, z) at Pos(x, y, z) is synthesized by Pc1 and Pc2:

[0049]

[0050] ψ is the density coefficient used to adjust the concentration and distribution of clouds in the visual scene. In order to make the edge of the visual scene cloud more natural visually, generate a three-channel three-dimensional Worley noise as the detail noise to perform detail processing on the density of the three-dimensional cloud field.

[0051] The specific steps of Step S5 are as follows:

[0052] Generate a 2-channel radar cloud data map with the same pixel size according to the radar reflectivity image in Step S2. For any coordinate point P(m, n) in the radar reflectivity image, if g(m, n) < 255 in the radar reflectivity image, set the value of the first channel of the radar cloud data map at this coordinate point to 1, otherwise set it to 0. At the same time, according to the visual scene cloud data map in Step S4, assign the value c(m, n) of the visual scene cloud data map to the second channel of the radar cloud data map at this coordinate point.

[0053] The specific steps of Step S6 are as follows:

[0054] S61: Calculate the radar cloud reflectivity coefficient of PC(x, y, z) at any position in the cloud field according to the radar cloud data map.

[0055] For any point PC(x, y, z) in the three-dimensional cloud field space, its projected coordinates (x′, z′) in the XZ direction. If x′ ∈ [0, 1] and z′ ∈ [0, 1], then this point is located in the cloud field area where the radar cloud data map is located, and the data value at this location is obtained. When the value of the first channel is 1, it means that there is a radar cloud in the XY direction. In order to represent the reflectivity coefficient of the radar cloud at different heights, according to the value c of the second channel, the cloud density Pc1 at this location is calculated. A visual cloud density threshold η is set. When Pc1 ≥ η, the radar reflectivity at PC(x, y, z) of the visual cloud is:

[0056] Z = lg(Pc1 × c × 25.5)

[0057] S62: Calculate the statistical information of the radar electromagnetic characteristics of the meteorological target according to the basic radar reflectivity Z

[0058] The magnitude of the basic radar reflectivity Z value represents the intensity of the meteorological target and can be directly used for calculating the radar echo energy. However, when the electromagnetic wave emitted by the radar passes through the meteorological target, it will be attenuated due to the scattering and absorption of cloud and rain particles in the meteorological target. The calculation of the attenuation coefficient needs to be indirectly calculated through Z. In order to simulate the attenuation effect of the radar echo intensity of the meteorological target, the formula K = K2I γ represents the attenuation coefficient of the meteorological target, where K2 and γ are functions of the radar wavelength, and I is the rainfall rate. From the Z-I relationship Z = AI b we get:

[0059]

[0060] where A and b are empirical constants. Therefore, the attenuation coefficient at PC(x, y, z) can be expressed as:

[0061]

[0062] The beneficial effects of the present invention are as follows:

[0063] 1. The present invention designs a method for consistent representation of visual cloud data and radar cloud data for simulated flight training, enabling the visual clouds seen by flight personnel in simulated flight training to correspond to the radar images of clouds, which is beneficial for flight personnel to correctly interpret radar images and assist flight personnel in making corresponding decisions.

[0064] 2. Calculate the radar echo energy based on the radar statistical characteristics of cloud particles, consider the attenuation effect of clouds on electromagnetic waves, truly simulate the generation process of radar echoes, and improve the fidelity of radar system simulation. Description of the Drawings

[0065] Figure 1 It is a schematic diagram of the cloud field area;

[0066] Figure 2 It is the reflectivity grayscale image after filling;

[0067] Figure 3 It is the radar reflectivity image;

[0068] Figure 4 It is the 3D visual scene cloud effect diagram;

[0069] Figure 5 It is the non-attenuated echo image;

[0070] Figure 6 It is the echo image considering the attenuation effect;

[0071] Figure 7 It is the echo image with increased beam height;

[0072] Figure 8 It is the flowchart of the present invention. Detailed implementation manners

[0073] Embodiment 1

[0074] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present application.

[0075] As Figure 1-8 shown, a method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data provided by an embodiment of the present invention includes:

[0076] Step S1: Select a basic radar reflectivity image as the basic data source, and extract the basic radar reflectivity data. Steps:

[0077] Step S11: According to the color value definition used for different basic reflectivities in the basic radar reflectivity image, use a color threshold to perform regional segmentation on the image, and extract different basic radar reflectivity regions;

[0078] a. Convert the RGB format image to the YUV format, and extract the YUV three-channel image data; convert the RGB values of the colors used for different reflectivities into YUV values;

[0079] b. Utilize the YUV three-channel image data, and select a suitable color channel to extract the corresponding region Region lv , where lv is an integer in [1,..., 15], and the corresponding basic radar reflectivity value z' = lv * 5.

[0080] Step S12 merges different radar reflectivity regions and obtains closed regions through region closing operations;

[0081] In step S13, according to the definition of place names and ground object colors in the basic reflectivity image, corresponding regions are extracted from the YUV image using color thresholds, and the intersection of this region and the closed region in step S12 is obtained to get the place name and ground object regions within the closed region;

[0082] Step S2 generates a radar reflectivity image based on the regions and their object types obtained in S1, with steps:

[0083] Step S21 generates a grayscale image with the same pixel size as the radar basic reflectivity image. For different radar basic reflectivity regions Region lv the grayscale value of the corresponding grayscale image region is g = [16 - (z' / 5)] * 10, and the grayscale values of other regions are set to 255.

[0084] Step S22 calculates the pixel coordinates of the place name and ground object regions in step S13, and makes a copy of the grayscale image in S21. For the pixel coordinates of each place name and ground object region, the pixel value of the pixel point closest to this coordinate point with a grayscale value less than 255 is searched in the copy image, and this value is assigned to the corresponding coordinate point in the grayscale image in S21 (attached Figure 2 );

[0085] For the grayscale image in step S22, the threshold range is set to 0 - 254, and the region with radar reflectivity is extracted. The dimensions of the minimum bounding rectangle are M = 224 and N = 273. According to the rectangle coordinates, the radar reflectivity image is intercepted from the image (attached Figure 3 ).

[0086] Step S3 generates a visual scene cloud data map with the same pixel size based on the radar reflectivity image in step S2. For any coordinate point P(m, n) in the radar reflectivity image, where m = 1, 2,..., M and n = 1, 2,..., N are the coordinates in the image, and the pixel value g(m, n) of the radar reflectivity image at this point, the value c(m, n) of the visual scene cloud data map at this point is calculated.

[0087] If g(m, n) < 255, c(m, n) = [160 - g(m, n)] / 510.0 + rand * 0.015686, where rand is a random number with a value range of [-1, 0).

[0088] If g(m,n) = 255, a search area of 71×71 is set centered at this point, and the pixel point P(m′,n′) that is closest to the pixel point P(m,n) and has a pixel value less than 255 is found. Its pixel value is g(m′,n′), then the value of c(m,n) is:

[0089]

[0090] Take the distance decay factor δ = 10.0 to make the edge of the generated three-dimensional visual scene cloud have a slope and a smooth transition appearance. According to the calculated value of c(m,n), the maximum value c of the visual scene cloud data map is obtained max = 0.4392495, and the minimum value c min = 0.005886275.

[0091] In step S41, set W = 256000m, L = 256000m, H = 3000m, b z = 1500m for the cloud field range. The cloud field center coordinates in the XZ direction are (0,0), and the coordinates of the lower left corner are (-128000, -128000). The position of the center of the visual scene cloud data map in the cloud field is PC(35500m, 3000m, 51500m). The distance size Δd represented by a single data point of the visual scene cloud data map is 500m. Then the length of the coverage range of the visual scene cloud data map is ΔdM = 103500m, and the width is ΔdN = 120500m. The texture coordinates of the visual scene cloud data map are defined as: the lower left corner is (0,0), and the upper right corner is (1,1). For any point PC(x,y,z) in the three-dimensional cloud field space, its coordinates (x′,z′) projected onto the coordinate system of the visual scene cloud data map are:

[0092]

[0093] If x′ ∈ [0, 1] and z′ ∈ [0, 1], then this point is located in the cloud field area where the visual scene cloud data map is located, and the visual scene cloud map data value c at this place is obtained. Otherwise, set the obtained visual scene cloud map data value c = 0. Set y min = 0.001, then:

[0094]

[0095] The ratio of this position to the cloud bottom is

[0096] Then the cloud density at this place is:

[0097]

[0098] Step S42 pre-generates a four-channel three-dimensional Worley noise as the basic shape, calculates the cloud density Pc2 at Pos(x, y, z) through three-dimensional noise sampling, and sets the non-radar cloud density coefficient ψ = 0.31. Synthesize the cloud density Pc(x, y, z) at Pos(x, y, z) from Pc1 and Pc2:

[0099]

[0100] To make the edge of the visual scene cloud more natural visually, generate a three-channel three-dimensional Worley noise as the detail noise, perform detail processing on the density of the three-dimensional cloud field, and use the ray marching algorithm to draw the volume cloud with a step length of 10m (attached Figure 4 ).

[0101] Step S5 generates a two-channel radar cloud data map with the same pixel size according to the radar reflectivity image in Step S2. For any coordinate point (m, n), if g(m, n) < 255 in the radar reflectivity image, set the value of the first channel of the radar cloud data map at this coordinate point to 1, otherwise set it to 0; according to the visual scene cloud data map in Step S4, assign the value c(m, n) of the visual scene cloud data map to the second channel of the radar cloud data map at this coordinate point;

[0102] Step S61 For any point PC(x, y, z) in the three-dimensional cloud field space, and its projected coordinates (x′, y′) in the XY direction, if x′ ∈ [0, 1] and y′ ∈ [0, 1], then this point is located in the cloud field area where the radar cloud data map is located, and obtain the radar cloud data at this place. When the value of the first channel is 1, it means that there is a radar cloud in the XY direction at this place. To represent the reflectivity coefficient of the radar cloud at different heights, calculate the cloud density Pc1 at this place according to the value c of the second channel, and set the visual scene cloud density threshold η = 0.3. When Pc1 ≥ η, the basic radar reflectivity of the visual scene cloud at PC(x, y, z) is Z = lg(Pc1 × c × 25.5).

[0103] Step S62 Calculate the statistical information of the radar electromagnetic characteristics of the meteorological target at PC(x, y, z) according to the basic radar reflectivity Z.

[0104] The magnitude of the basic radar reflectivity Z value represents the intensity of the meteorological target and can be directly used for radar echo energy calculation. However, when the electromagnetic wave emitted by the radar passes through the meteorological target, it will be attenuated due to the scattering and absorption of the cloud and rain particles in the meteorological target. The calculation of the attenuation coefficient needs to be indirectly calculated through Z. Let K2 = 0.04, γ = 1.13, A = 200, b = 1.6, the attenuation coefficient at PC(x, y, z) can be expressed as:

[0105]

[0106] Radar meteorological equation under attenuation factor:

[0107]

[0108] In the equation, C is a constant related to radar parameters, ψ is the beam filling coefficient indicating that the radar beam is filled by meteorological targets, R is the distance between the radar and the meteorological target. Radar echo energy calculation is performed to generate a radar echo image. Assume that the scanning center of the radar antenna is 0°, the scanning range is ±60°, and the detection distance is 150000 m. Figure 5 , 6 are the radar echo images without attenuation and considering the attenuation effect at the radar position coordinates (40000 m, 2500 m, -25000 m) respectively. The attenuation effect will weaken the echo energy on the radar beam path; Figure 7 is the radar image considering attenuation at the radar coordinate position (40000 m, 4500 m, -25000 m). As the radar beam increases, the effective volume of the meteorological targets irradiated by the beam decreases, and the echo energy weakens.

[0109] The above are only the embodiments of the present application and are not used to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data, characterized in that It includes the following steps: Step S1: According to the radar reflectivity values, roads, and text objects identified by different colors in the radar basic reflectivity image, extract the corresponding regions through image processing. Step S2: Generate a radar reflectivity image based on the regions obtained in Step S1 and their object types. Step S3: Generate a visual scene cloud data map with the same pixel size through the radar reflectivity image generated in Step S2, and calculate the values of each point in the visual scene cloud data map. Step S4: Use the visual scene cloud data map generated in Step S3 to design a method for representing the visual scene cloud data, and calculate the cloud density in the three-dimensional cloud field. Step S5: Generate a radar cloud data map through the radar reflectivity image generated in Step S2 and the visual scene cloud data map generated in Step S3, and calculate the values of each point in the radar cloud data map. Step S6: Use the radar cloud data map generated in Step S5 and the cloud density calculation method in Step S4 to design a method for representing the radar cloud data.

2. A method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data according to claim 1, characterized in that The said Step S1 includes the following steps: Step S11: According to the color value definitions of different basic reflectivities in the radar basic reflectivity image, use color thresholds to segment the image regions and extract different radar basic reflectivity regions. Step S12: Merge different radar basic reflectivity regions and obtain a closed region through the closing operation of the regions. Step S13: According to the color definitions of place names and ground objects in the radar basic reflectivity image, use color thresholds to extract the corresponding regions, and find the intersection of this region and the closed region in Step S12 to obtain the place name and ground object regions within the closed region.

3. A method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data according to claim 2, characterized in that The said Step S2 includes the following steps: Step S21: Generate a grayscale image with the same size as the radar basic reflectivity image. According to the radar basic reflectivity regions in Step S11, transform the radar basic reflectivity value z′ of each pixel point in different radar basic reflectivity regions into the grayscale value g of the corresponding pixel point in the grayscale image: g = [16 - (z′ / 5)] * 10 z′ ∈ {5, 15, 20, 25, 35, 40, 45, 50, 55, 60, 65, 70, 75}, and the grayscale values of other pixel points are set to 255. Step S22: Calculate the pixel point coordinates of the place name and ground object regions obtained in Step S13, and use the filling algorithm to set the grayscale values of each pixel point in the grayscale image corresponding to the pixel point coordinates in Step S21 to obtain the filled grayscale image. Step S23: For the grayscale image obtained in Step S22, use the grayscale threshold to extract the region with radar reflectivity, obtain the minimum circumscribed rectangle of this region, with the width M of the rectangle and the height N of the rectangle, and intercept the image within the range of the circumscribed rectangle from the image as the radar reflectivity image.

4. A method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data according to claim 3, characterized in that: In the said Step S22, the filling algorithm used when setting the grayscale values of the grayscale image in Step S21 includes the following steps: Step S221: Make a copy of the grayscale image in Step S21 as a copy image. Step S223: For the pixel coordinates of each place name and feature area, search in the copy image for the pixel point with the gray value less than 255 that is closest to this coordinate point, and assign the gray value of this point to the corresponding coordinate point in the gray image in Step S21.

5. A method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data according to claim 1, characterized in that The said Step S3 includes the following steps: Based on the radar reflectivity image in Step S2, generate a visual cloud data map with the same pixel size. For any coordinate point P(m,n) in the radar reflectivity image, where m = 1, 2,..., M and n = 1, 2,..., N are the horizontal and vertical coordinate values of this point in the image, and the pixel value of the radar reflectivity image at this coordinate point is g(m,n), the calculation method of the value c(m,n) of the visual cloud data map at this place is: (1) When g(m,n) ≠ 255, c(m,n) = [160 - g(m,n)] / 510.0 + rand * 0.015686, rand is a random number with the value range [-1, 0); (2) When g(m,n) = 255, adopt the window search method. Set a rectangular search area with both the horizontal width and the vertical width being w centered on this point, and find the pixel point P(m′,n′) within this area that is closest to the pixel point P(m,n) and has a pixel value less than 255. Its pixel value is g(m′,n′), then the value of c(m,n) is: δ is the distance attenuation factor, and its value is greater than 0. Through δ, the edge of the generated three-dimensional visual cloud can have a slope and a smooth transition appearance.

6. A method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data according to claim 1, characterized in that The said Step S4 includes the following steps: Step S41: Calculate the cloud density in the three-dimensional visual cloud field according to the visual cloud data map in Step S3; When the distance of the cloud field range (left - hand coordinate system) in the XZ direction is W, L, the thickness of the cloud field and the cloud base height are H, b respectively h , the position PC(x c ,y c ,z c ) of the center of the visual - scene cloud data map in the cloud field, and the distance size Δd represented by a single data point of the visual - scene cloud data map, then the length of the coverage range of the visual - scene cloud data map is ΔdM, the width is ΔdN, and the texture coordinates of the visual - scene cloud data map are defined as: the lower - left corner is (0,0), the upper - right corner is (1,1). For any point PC(x,y,z) in the three - dimensional cloud field space, its coordinates (x′,z′) projected onto the visual - scene cloud data map coordinate system are: If x′ ∈ [0, 1] and z′ ∈ [0, 1], then the point is located in the cloud field area where the visual scene cloud data map is located, and the visual scene cloud map data value c at this location is obtained. Otherwise, let the obtained visual scene cloud map data value c = 0. The normalized height range of the main cloud body of the cloud field here is [y min , y max , y min ∈ (0, 1), y max ∈ (0, 1), y min is a constant for adjusting the height of the cloud bottom, y max is used to adjust the height of the cloud top, and the value is: c max is the maximum value of the visual scene cloud data graph; Then the cloud density at this place is: is the ratio of the position to the cloud base, c min is the minimum value of the visual cloud data map; Step S42: Use the volume cloud generation method based on noise to calculate the visual cloud density where the entire cloud field cannot form a certain intensity radar echo, and add it to the cloud density calculated in Step S41; Generate a four-channel three-dimensional Worley noise as the basic shape, and calculate the cloud density Pc2 at Pos(x,y,z) through three-dimensional noise sampling. The cloud density Pc(x,y,z) at Pos(x,y,z) is synthesized by Pc1 and Pc2: ψ is the density coefficient, which adjusts the concentration and distribution of the clouds in the visual scene. In order to make the edge of the visual cloud more natural visually, generate a three-channel three-dimensional Worley noise as the detail noise to perform detail processing on the density of the three-dimensional cloud field.

7. A method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data according to claim 1, characterized in that The said Step S5 includes the following steps: Based on the radar reflectivity image in Step S2, generate a two-channel radar cloud data map with the same pixel size. For any coordinate point P(m,n) in the radar reflectivity image, if g(m,n) ≠ 255 in the radar reflectivity image, set the value of the first channel of the radar cloud data map at this coordinate point to 1, otherwise set it to 0. At the same time, according to the visual cloud data map in Step S4, assign the value c(m,n) of the visual cloud data map to the second channel of the radar cloud data map at this coordinate point.

8. A method for consistent representation of virtual simulation visual scene cloud data and airborne radar meteorological cloud data according to claim 1, characterized in that The said Step S6 includes the following steps: Step S61: Calculate the reflectivity coefficient of the radar cloud at any position PC(x,y,z) in the cloud field according to the radar cloud data map For any point PC(x, y, z) in the three-dimensional cloud field space, with its projected coordinates (x′, z′) in the XZ direction, if x′ ∈ [0, 1] and z′ ∈ [0, 1], then this point is located in the cloud field area where the radar cloud data map is located, and the data value at this location is obtained. When the value of the first channel is 1, it indicates that there is a radar cloud in the XY direction. To represent the reflectivity coefficient of the radar cloud at different heights, according to the value c of the second channel, the cloud density Pc1 at this location is calculated, and a visual scene cloud density threshold η is set. When Pc1 ≥ η, the radar reflectivity at PC(x, y, z) of the visual scene cloud is: Z = lg(Pc1 × c × 25.5) Step S62: Calculate the statistical information of the radar electromagnetic characteristics of the meteorological target according to the basic radar reflectivity Z; The magnitude of the basic radar reflectivity Z value represents the intensity of meteorological targets and is directly used for radar echo energy calculation. However, when the electromagnetic wave emitted by the radar passes through meteorological targets, it will be attenuated due to the scattering and absorption of cloud and rain particles in the meteorological targets. The calculation of the attenuation coefficient needs to be indirectly calculated through Z. To simulate the attenuation effect of the radar echo intensity of meteorological targets, the formula K = K2I is adopted. γ represents the attenuation coefficient of meteorological targets, where K2 and γ are functions of the radar wavelength, I is the rainfall rate, and from the Z-I relationship Z = AI b it is obtained that: Where A and b are empirical constants. Therefore, the attenuation coefficient at PC(x, y, z) can be expressed as:

Images