A method and system for generating an extraterrestrial body surface illumination simulation image

Abstract

The present invention relates to a method and system for generating an extraterrestrial surface illumination simulation image, wherein the method comprises: firstly, based on the rapid calculation of the geometrically occluded Boolean light-dark map, the preparation of a horizon database and the interpolation of the Boolean light-dark map; then, based on the rapid calculation of the albedo of the terrain, including solving the terrain normal vector, estimating the brightness value of the simulation image, that is, combining the light source vector and the normal vector to render a realistic image effect. Compared with the prior art, the present invention realizes rapid, high-resolution and high-precision simulation of the surface illumination image of an extraterrestrial body.

Description

Technical Field

[0001] The present invention relates to the field of image simulation, and in particular to a method and system for generating an extraterrestrial body surface illumination simulation image. Background Art

[0002] Image simulation of the surface of extraterrestrial bodies through terrain and ephemeris data is of great significance for future deep space exploration activities. Taking the moon as an example, the resolution of the optical images of the lunar surface currently obtained by humans has reached 0.5 to 1.5 meters per pixel (LROC NAC images), and these images contain extremely rich morphological details of the lunar surface. However, each image has different solar illumination conditions when it is taken, which means that the inconsistency of the direction and range of shadows in the image seriously interferes with the application of remote sensing images in mapping and spatial analysis. Reverse simulation of images through terrain and ephemeris data has positive significance in the fields of image calibration, fusion of multi-source remote sensing data, and high-precision mapping. The classic image map simulation scheme uses the principle of ray tracing to calculate light occlusion and reflection. Although it has high accuracy, the calculation is complex and is not suitable for large-scale image simulation tasks. The corresponding simplified scheme usually shortens the maximum boundary that needs to be considered in the process of calculating light occlusion. In essence, it abandons accuracy in order to improve calculation efficiency. At the moment when the solar altitude angle is low, this approximation will introduce serious errors. Considering that the range of solar altitude angle in the lunar polar region is limited to about ±1.5°, the demand for accurate image simulation of such scenes cannot be ignored.

[0003] In summary, a lot of work has been done on image simulation, but there is no suitable method for high-precision, high-resolution, large-scale extraterrestrial lighting simulation tasks. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a method and system for generating fast, high-resolution and high-precision extraterrestrial surface lighting simulation images based on terrain and ephemeris data.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A method for generating an extraterrestrial body surface illumination simulation image comprises the following steps:

[0007] S1. Convert terrain data from terrain grid space to Cartesian space, and establish topology in point cloud vector form for terrain grid; search for the maximum terrain occlusion elevation angle for all terrain units at fixed azimuth intervals, and record the maximum terrain occlusion elevation angle values ​​of all terrain units at each azimuth as a horizon map according to the terrain grid space topology; a set of horizon maps uniformly covering 360° azimuth range is regarded as a sampling of continuous horizons, which is defined as a horizon database;

[0008] S2. For a specific moment, the solar altitude and azimuth at the center of the target terrain are queried through the ephemeris; for a specific solar azimuth, the two nearest horizon maps are queried in the horizon database, and then at each grid, the corresponding values ​​in the two horizon maps are linearly interpolated according to the azimuth difference to obtain a new horizon map under the current azimuth; a new Boolean grid data is defined in the terrain grid space, called a Boolean brightness map, and the maximum terrain occlusion elevation angle recorded in the horizon map under the current azimuth is compared with the current solar altitude angle. If the solar altitude angle value is larger, the corresponding grid of the Boolean brightness map is assigned a value of true, otherwise it is assigned a value of false;

[0009] S3. Define a spherical area with a query point as the center and r as the radius as the neighborhood range of the terrain point; calculate the vector coordinates from the query point to the neighborhood point, perform covariance analysis, calculate the eigenvalue of the covariance matrix, and use the eigenvector corresponding to the minimum eigenvalue as the normal vector of the query point, and the normal vector is positive in the opposite direction of the center of the extraterrestrial body;

[0010] S4. Calculate the dot product between the normal vector and the unit vector representing the sun's position grid by grid, record the result in the corresponding position of the new grid, and define the new grid as the reflectance map; multiply the reflectance map, the Boolean brightness map, and the preset albedo map grid by grid to obtain a simulated image.

[0011] Furthermore, in step S1, the origin of the Cartesian space is located at the center point of the terrain grid, and the xOy plane coincides with the section of the extraterrestrial body at the origin.

[0012] Furthermore, in step S1, the neighbor relationship of points in the topological space is described by a Kd tree.

[0013] Furthermore, in step S3, the value of the radius r is 1.5 times the length of the maximum grid side in the terrain grid.

[0014] Furthermore, in step S4, the preset albedo map is initialized to 1.

[0015] A system for generating an extraterrestrial body surface illumination simulation image comprises a processor and a memory, wherein the processor calls a program stored in the memory and performs the following steps:

[0016] S1. Convert terrain data from terrain grid space to Cartesian space, and establish topology in point cloud vector form for terrain grid; search for the maximum terrain occlusion elevation angle for all terrain units at fixed azimuth intervals, and record the maximum terrain occlusion elevation angle values ​​of all terrain units at each azimuth as a horizon map according to the terrain grid space topology; a set of horizon maps uniformly covering 360° azimuth range is regarded as a sampling of continuous horizons, which is defined as a horizon database;

[0017] S2. For a specific moment, the solar altitude and azimuth at the center of the target terrain are queried through the ephemeris; for a specific solar azimuth, the two nearest horizon maps are queried in the horizon database, and then at each grid, the corresponding values ​​in the two horizon maps are linearly interpolated according to the azimuth difference to obtain a new horizon map under the current azimuth; a new Boolean grid data is defined in the terrain grid space, called a Boolean brightness map, and the maximum terrain occlusion elevation angle recorded in the horizon map under the current azimuth is compared with the current solar altitude angle. If the solar altitude angle value is larger, the corresponding grid of the Boolean brightness map is assigned a value of true, otherwise it is assigned a value of false;

[0018] S3. Define a spherical area with a query point as the center and r as the radius as the neighborhood range of the terrain point; calculate the vector coordinates from the query point to the neighborhood point, perform covariance analysis, calculate the eigenvalue of the covariance matrix, and use the eigenvector corresponding to the minimum eigenvalue as the normal vector of the query point, and the normal vector is positive in the opposite direction of the center of the extraterrestrial body;

[0019] S4. Calculate the dot product between the normal vector and the unit vector representing the sun's position grid by grid, record the result in the corresponding position of the new grid, and define the new grid as the reflectance map; multiply the reflectance map, the Boolean brightness map, and the preset albedo map grid by grid to obtain a simulated image.

[0020] Furthermore, in step S1, the origin of the Cartesian space is located at the center point of the terrain grid, and the xOy plane coincides with the section of the extraterrestrial body at the origin.

[0021] Furthermore, in step S1, the neighbor relationship of points in the topological space is described by a Kd tree.

[0022] Furthermore, in step S3, the value of the radius r is 1.5 times the length of the maximum grid side in the terrain grid.

[0023] Furthermore, in step S4, the preset albedo map is initialized to 1. Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention realizes the rapid calculation of light and dark information by pre-establishing a horizon database, and then realizes the accurate simulation of reflection intensity under the same spatial topology by combining the Lambertian reflection principle, thereby taking into account both the simulation accuracy and the rendering time, and realizing fast, high-resolution and high-precision simulation of the surface illumination image of extraterrestrial bodies. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the process of the present invention.

[0026] Figure 2a This is a 0.5-meter resolution LROC image of the edge of Shackleton Crater at the South Pole of the Moon.

[0027] Figure 2b This is a simulated image of the edge of Shackleton Crater at the South Pole of the Moon with a resolution of 5 meters. DETAILED DESCRIPTION

[0028] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0029] This embodiment provides a method for generating a simulated image of the surface illumination of an extraterrestrial body, which mainly includes the following aspects:

[0030] 1. Fast calculation of Boolean shading map based on geometric occlusion

[0031] A: Creation of horizon database: including the transformation from geographic grid space to Cartesian space, topological reconstruction of point cloud vectors; including the discretization and traversal of light source azimuth, and the search and recording of horizons;

[0032] B: Interpolation of Boolean brightness and darkness map: including finding the horizon map under the azimuth of the light source in the horizon database, and judging the shadow area in combination with the altitude angle of the light source.

[0033] 2. Rapid calculation of albedo based on terrain

[0034] A: Solving terrain normal vector: In vector topology, the normal vector is estimated based on the neighborhood;

[0035] B: Estimate the brightness value of the simulated image: combine the light source vector and the normal vector to render a realistic image effect.

[0036] like Figure 1 As shown, the steps of this embodiment are specifically expanded as follows:

[0037] Step S1: Creation of horizon database

[0038] Terrain data is usually stored in the form of geographic longitude and latitude grids or projected grids. In order to better calculate occlusion, the coordinates of the center point of each grid are described in Cartesian space, and then another set of topology is established for the terrain grid in the form of a three-coordinate vector. The origin of the Cartesian space is located at the center point of the terrain grid, and the xOy plane coincides with the tangent plane of the planet at the origin. In the new topological space, the neighbor relationship of the point is described by a Kd tree. Then, the horizon is searched for all terrain points along the xOy plane at fixed azimuth intervals. Therefore, for a certain azimuth, a horizon (elevation) map with the original grid topology can be obtained. A set of such horizon maps that evenly cover a 360° azimuth range can be regarded as a sampling of a continuous horizon, defined as a horizon database.

[0039] Step S2: Interpolation of Boolean brightness map

[0040] For a specific moment, first query the solar altitude and azimuth at the midpoint of the target terrain through the ephemeris. Then, for a specific solar azimuth, query the two nearest horizon maps in the horizon database, and then at each grid, linearly interpolate the corresponding values ​​in the two horizon maps for each grid on the horizon map according to the azimuth difference to obtain a new horizon map at the current azimuth; define a new Boolean grid data in the terrain grid space, called the Boolean brightness map, compare the maximum terrain occlusion elevation angle recorded in the horizon map at the current azimuth and the current solar altitude angle, if the solar altitude angle value is larger, assign the corresponding grid of the Boolean brightness map to true, otherwise assign it to false; in this way, the Boolean brightness map at the current moment is obtained.

[0041] Step S3: Solve the terrain normal vector

[0042] Since the real topology of terrain data is recorded in vector form, it is necessary to first determine the neighborhood influence range. Define a spherical area with a radius of r and a query point as the neighborhood range of the terrain point (r is recommended to be 1.5 times the maximum grid side length). Then, calculate the vector coordinates from the query point to the neighborhood point, perform covariance analysis, calculate the eigenvalue of the covariance matrix, and use the eigenvector corresponding to the minimum eigenvalue as the normal vector of the query point. Next, define the normal vector direction, with the opposite direction of the center of the moon as the positive direction.

[0043] Step S4: Estimating the brightness value of the simulated image and performing simulated image rendering

[0044] After independently obtaining the terrain normal vector for any point in the terrain grid, the vector is normalized to a unit vector, and then dot product with the unit vector representing the sun's position. The result is recorded at the corresponding grid position, and the new grid is defined as the reflectance map (Reflectance). According to the image intensity formula (1), the reflectance map is multiplied grid by grid with the Boolean shadow map (BS) at the corresponding moment and the preset albedo map, and a simulated image (Intensity) at the real moment can be obtained. Among them, if the albedo map (Albedo) is not defined, it will be initialized to 1.

[0045] Intensity=BS×Reflectance×Albedo (1)

[0046] Experiment and analysis

[0047] The platform of this experiment is to select the LOLA DEM with a resolution of 5 meters near the south pole of the moon as the terrain data source to calculate the simulated image map. At the same time, the simulated illumination map produced by the method of the present invention is compared with the actual LROC image. Figure 2a and Figure 2b It can be proved that the present invention can accurately simulate a high-resolution image map of the lunar surface at any time. Figure 2b The data volume is 100,000 pixels (1800×400), and the calculation and rendering time is only 0.79 seconds.

[0048] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for generating a simulated image of the surface illumination of an extraterrestrial body, characterized in that: The following steps are involved: S1. Convert terrain data from terrain grid space to Cartesian space, and establish topology in point cloud vector form for terrain grid; search for the maximum terrain occlusion elevation angle for all terrain units at fixed azimuth intervals, and record the maximum terrain occlusion elevation angle values ​​of all terrain units at each azimuth as a horizon map according to the terrain grid space topology; a set of horizon maps uniformly covering 360° azimuth range is regarded as a sampling of continuous horizons, which is defined as a horizon database; S2. For a specific moment, the solar altitude and azimuth at the center of the target terrain are queried through the ephemeris; for a specific solar azimuth, the two nearest horizon maps are queried in the horizon database, and then at each grid, the corresponding values ​​in the two horizon maps are linearly interpolated according to the azimuth difference to obtain a new horizon map under the current azimuth; a new Boolean grid data is defined in the terrain grid space, called a Boolean brightness map, and the maximum terrain occlusion elevation angle recorded in the horizon map under the current azimuth is compared with the current solar altitude angle. If the solar altitude angle value is larger, the corresponding grid of the Boolean brightness map is assigned a value of true, otherwise it is assigned a value of false; S3. Define a spherical area with a query point as the center and r as the radius as the neighborhood range of the terrain point; calculate the vector coordinates from the query point to the neighborhood point, perform covariance analysis, calculate the eigenvalue of the covariance matrix, and use the eigenvector corresponding to the minimum eigenvalue as the normal vector of the query point, and the normal vector is positive in the opposite direction of the center of the extraterrestrial body; S4, calculating the dot product between the normal vector and the unit vector representing the sun's position grid by grid, recording the result at the corresponding position of the new grid, and defining the new grid as a reflectance map; The reflectance map, Boolean brightness map and preset albedo map are multiplied grid by grid to obtain the simulated image.

2. The method for generating a simulated image of an extraterrestrial body surface illumination according to claim 1, characterized in that: In step S1, the origin of the Cartesian space is located at the center point of the terrain grid, and the xOy plane coincides with the section of the extraterrestrial body at the origin.

3. The method for generating a simulated image of an extraterrestrial body surface illumination according to claim 1, characterized in that: In step S1, the neighbor relationship of points in the topological space is described by a Kd tree.

4. The method for generating a simulated image of an extraterrestrial body surface illumination according to claim 1, characterized in that: In step S3, the value of the radius r is 1.5 times the length of the maximum grid side in the terrain grid.

5. The method for generating a simulated image of an extraterrestrial body surface illumination according to claim 1, characterized in that: In step S4, the preset albedo map is initialized to 1.

6. A system for generating an extraterrestrial surface illumination simulation image, comprising a processor and a memory, characterized in that: The processor calls the program stored in the memory to perform the following steps: S1. Convert terrain data from terrain grid space to Cartesian space, and establish topology in point cloud vector form for terrain grid; search for the maximum terrain occlusion elevation angle for all terrain units at fixed azimuth intervals, and record the maximum terrain occlusion elevation angle values ​​of all terrain units at each azimuth as a horizon map according to the terrain grid space topology; a set of horizon maps uniformly covering 360° azimuth range is regarded as a sampling of continuous horizons, which is defined as a horizon database; S2. For a specific moment, the solar altitude and azimuth at the center of the target terrain are queried through the ephemeris; for a specific solar azimuth, the two nearest horizon maps are queried in the horizon database, and then at each grid, the corresponding values ​​in the two horizon maps are linearly interpolated according to the azimuth difference to obtain a new horizon map under the current azimuth; a new Boolean grid data is defined in the terrain grid space, called a Boolean brightness map, and the maximum terrain occlusion elevation angle recorded in the horizon map under the current azimuth is compared with the current solar altitude angle. If the solar altitude angle value is larger, the corresponding grid of the Boolean brightness map is assigned a value of true, otherwise it is assigned a value of false; S3. Define a spherical area with a query point as the center and r as the radius as the neighborhood range of the terrain point; calculate the vector coordinates from the query point to the neighborhood point, perform covariance analysis, calculate the eigenvalue of the covariance matrix, and use the eigenvector corresponding to the minimum eigenvalue as the normal vector of the query point, and the normal vector is positive in the opposite direction of the center of the extraterrestrial body; S4, calculating the dot product between the normal vector and the unit vector representing the sun's position grid by grid, recording the result at the corresponding position of the new grid, and defining the new grid as a reflectance map; The reflectance map, Boolean brightness map and preset albedo map are multiplied grid by grid to obtain the simulated image.

7. The system for generating an extraterrestrial body surface illumination simulation image according to claim 6, characterized in that: In step S1, the origin of the Cartesian space is located at the center point of the terrain grid, and the xOy plane coincides with the section of the extraterrestrial body at the origin.

8. The system for generating an extraterrestrial body surface illumination simulation image according to claim 6, characterized in that: In step S1, the neighbor relationship of points in the topological space is described by a Kd tree.

9. The system for generating an extraterrestrial body surface illumination simulation image according to claim 6, characterized in that: In step S3, the value of the radius r is 1.5 times the length of the maximum grid side in the terrain grid.

10. The system for generating an extraterrestrial body surface illumination simulation image according to claim 6, characterized in that: In step S4, the preset albedo map is initialized to 1.

Images