Space target spectral imaging simulation method based on space-based observation scene

Abstract

The invention provides a space target spectral imaging simulation method based on a space-based observation scene. The method comprises the following steps: (1) inputting attitude and orbit parameters and simulation time parameters of a space target and a spectral imager; (2) establishing a space-based observation scene by using the parameters in the step (1), and obtaining position, attitude and speed information; (3) carrying out target visibility analysis, and firstly judging an earth shadow area; (4) judging sight shielding; (5) performing imager view field constraint judgment; (6) carrying out radiation transfer modeling, analyzing surface element visibility and calculating a spectral radiance image at the entrance pupil of the spectral imager; and (7) carrying out spectral imager modeling, and generating a simulation image by using the spectral radiance image at the entrance pupil obtained in the step (6). According to the method, dynamic simulation is carried out in combination with a track and relative motion, factors influencing target visibility are considered in detail, and the full-link simulation method under the space-based observation scene is constructed by considering the reflection characteristic in the direction of a material and visibility of different surface elements.

Description

Technical Field

[0001] The present invention relates to a method for simulating spectral imaging of space targets in a space-based observation scenario, belongs to the field of optical imaging data acquisition, simulation and recording, and is suitable for research on spectral simulation image generation in a space-based observation scenario. Background Art

[0002] Space target supervision is of great significance to the development and sustainable development of outer space resources. Building a space-based observation platform is an important means of space target detection, identification and supervision at this stage, and the application of spectral analysis technology can provide richer information for space target detection. Due to the lack of measured data at this stage, the cost of launching and operating space-based remote sensing payloads is high, and it is difficult to obtain high-quality optical observation data in deep space environments. Therefore, simulation analysis methods are often used in current research to make up for the lack of measured data. However, in my country, the space-based imaging system for space targets is still in the preliminary research stage, and the imaging results are difficult to predict.

[0003] The existing space target observation simulation and analysis work can be divided into two categories: space target detection simulation and space target observation data simulation.

[0004] Space target detection simulation refers to the simulation analysis of the scene of using detectors to detect space targets, which is the premise of space target observation simulation. First, the detector is used to search for targets within the field of view, and then the targets are distinguished from the background or other non-target objects, and the targets are tracked, and preliminary classification and feature information are cataloged. Researchers have used simulation methods to use space-based detectors placed on sun-synchronous orbits to detect and track 100 geosynchronous orbit objects with destinations and 200 geosynchronous orbit objects without destinations. At the same time, the detectability of space targets in the infrared spectrum was simulated and analyzed, and the reflection and emission characteristics of the targets were modeled respectively. The reflection sources include direct solar radiation, reflected radiation from the earth and the moon, infrared thermal radiation from the earth, and cosmic background radiation, while the emission sources include the heat of the object itself and the energy radiation of its aerodynamic heating system. The noise sources involved in the simulation system mainly include the absorption and scattering effects of the earth's atmosphere, the background radiation of the starry sky, etc. Researchers have used data-driven methods to model and simulate the future evolution of the number of targets based on spacecraft launch data and trends, and historical data on spacecraft explosions and clearances.

[0005] After the space target is detected, the space target observation data can be simulated under the corresponding observation conditions. For example, based on the Phong model, the light curve simulation analysis of geosynchronous orbit satellites of different shapes and sizes is carried out. The results show that the simulation error of the satellite light curve considering the directional reflection characteristics is about 3%, which is much more accurate than the simulation results using the diffuse reflection model.

[0006] In summary, the existing space target observation simulation methods have the following three difficulties that have not been resolved: (1) There is a lack of modeling and simulation for close-range full-link spectral imaging of space targets; (2) The model adopts a lot of simplifications, treating the target as a Lambertian body without considering its directional reflection characteristics, resulting in a large gap between the simulation results and the measured data; (3) There is no dynamic simulation combined with orbit and relative motion, and the three factors that affect the visibility of the target in the space environment, namely, the ground shadow constraint, the line of sight occlusion constraint, and the imager field of view constraint, are not considered in detail. In order to provide a theoretical and data basis for the feasibility analysis of using space-based observation platforms to detect and identify space targets, it is urgent to carry out full-link research on space target spectral imaging simulation based on space-based observations. Summary of the invention

[0007] The purpose of the present invention is to realize the full link of spectral imaging simulation of space targets based on space-based observation scenes, and provide a spectral imaging simulation method of space targets based on space-based observation scenes.

[0008] The technical solution of the present invention is as follows: a space-based observation scene is established based on the actual orbital operation data of the space target and the spectral imager, various data of the space target and the spectral imager in the dynamic scene are obtained, and the space target visualization analysis is performed based on the actual motion data; after the visualization analysis is completed, the surface element visibility analysis is performed on the three-dimensional model of the space target pixel by pixel, and the directional reflection characteristics of different materials are considered, and the observation angles of the visible surfaces are calculated in turn, and the spectral radiance image of the entrance pupil of the spectral imager is calculated by combining the radiation transmission modeling results and the surface optical characteristics modeling results of the space target; the imaging principle of the spectral imager is analyzed, and the five links of radiation response, spectral response, motion blur, random noise, and image quantization in the imaging process are modeled, and finally a simulated image of the space target based on the space-based observation scene is obtained.

[0009] The present invention is a space target spectral imaging simulation method based on a space-based observation scene, and the steps are as follows:

[0010] (1) Input the attitude and orbit parameters of the space target and the spectral imager, the simulation start and end time, and the simulation step size;

[0011] (2) Input the three parameters of the space target and the spectral imager input in step (1), the attitude and orbit parameters, the simulation start and end time, and the simulation step size into STK, establish a specific scenario for the space-based spectral imager to observe and image the space target, and obtain the six parameters of the position of the earth in the Earth-centered Earth-fixed coordinate system at each simulation moment, the position and velocity of the space target, the position and velocity of the spectral imager, and the position of the sun, as well as the attitude of the space target in the space target track coordinate system;

[0012] (3) Determine whether the space target is in the shadow area of ​​the earth at the current simulation moment. Consider the space target as a point in the space-based observation scene. First, use the position coordinates of the space target, the sun and the earth obtained in step (2) and use a cylindrical earth shadow model to describe the earth shadow interval. Calculate whether the position coordinates of the space target are in the shadow area of ​​the earth. If the space target is in the shadow area, the simulation of the current simulation moment ends, enters the next simulation moment and re-executes step (3). Otherwise, execute step (4).

[0013] (4) determining whether the line of sight of the spectral imager is blocked by the earth at the current simulation moment, using the position coordinates of the spectral imager, the space target and the earth obtained in step (2), regarding the earth as a standard ellipsoid, and determining whether the line vector connecting the spectral imager and the space target intersects with the earth; if there is no intersection or the two intersections are not located between the spectral imager and the space target, the line of sight of the spectral imager is not blocked by the earth at the current simulation moment, and executing step (5); otherwise, the simulation at the current simulation moment ends, and the next simulation moment is entered and step (3) is executed again;

[0014] (5) Perform a field of view constraint judgment on the spectral imager at the current simulation moment, use the position coordinates of the spectral imager and the space target obtained in step (2) to construct the relative position relationship between the two, calculate the boundary point set of the coordinates of the intersection of all detection light rays and the object plane where the space target is located, and determine the single-view coverage range that can be achieved by the spectral imager without posture adjustment and the panoramic coverage range that can be achieved by posture adjustment. If the space target is in both the single-view and panoramic coverage ranges at the current moment, or the space target is in the panoramic view but not in the single-view range at the current moment, and the space target is not in the ground shadow area or the line of sight obstruction area and is in the panoramic range at the next moment, it means that the spectral imager can perform observation imaging, and execute step (6). Otherwise, the simulation at the current simulation moment ends, and the next simulation moment is entered and step (3) is re-executed;

[0015] (6) Modeling the transmission process of the radiation characteristics of the space target, considering the three-dimensional detailed structure and material distribution of the space target, segmenting the space target based on the triangular face representation method, and modeling the space target in three dimensions by determining the vertices, normal vectors and corresponding material types of each triangular face. Using the three-dimensional model of the space target and the seven types of parameters obtained in step (2), two radiation transmission paths and relative position relationships of the sun-space target-spectral imager and the earth-space target-spectral imager are obtained. Based on the space radiation transmission calculation method, the radiation intensity of the sunlight received at the space target and the radiation intensity of the reflected light from the earth are obtained. Ray tracing is used to calculate the observation angles of the detection light emitted by the spectral imager and the intersecting face normal vectors in the two radiation transmission paths. The spectral reflectance of the two radiation transmission paths is calculated using the bidirectional reflectance distribution function and the calculated observation angles. The spectral radiance of the spectral imager at the entrance pupil is obtained by combining the spectral radiation intensity and the spectral reflectance under the corresponding path, and the calculation of the entrance pupil spectral radiance image is completed.

[0016] (7) Modeling the imaging process of the spectral imager is carried out, and the five links of radiation response, spectral response, motion blur, random noise, and image quantization are comprehensively analyzed to establish the imaging model of the spectral imager. The spectral radiance image at the entrance pupil calculated by step (6) is used to generate the final simulation image.

[0017] Wherein, the step (1) inputs the attitude and orbit parameters of the space target and the spectral imager, the simulation start and end time, and the simulation step length: the orbit parameters are defined in the format of six orbit elements, including the semi-major axis, eccentricity, orbit inclination, pericentric angle, ascending node longitude and true anomaly; the attitude parameters of the space target include four types: three-axis stabilization to the sun, three-axis stabilization to the earth, spin to the sun and spin to the earth; the simulation start and end time are defined in the coordinated universal time format, and the simulation step length is in seconds.

[0018] In which, the step (2) inputs the three parameters of the attitude and orbit parameters of the space target and the spectral imager input in step (1), the simulation start and end time, and the simulation step size into STK, establishes a specific scene for the space-based spectral imager to observe and image the space target, obtains six parameters of the position of the earth, the position and speed of the space target, the position and speed of the spectral imager, and the position of the sun in the geocentric earth-fixed coordinate system at each simulation moment, and the attitude of the space target in the space target track coordinate system: inputs the parameters required for establishing the space-based observation scene into STK, establishes a specific space-based spectral imager to observe and image the space target, and uses the report format preset by the STK system to obtain two types of information, one is the position and speed of the space target, the position and speed of the spectral imager, and the position of the sun based on the geocentric earth-fixed coordinate system, and the other is the attitude angle of the space target based on the space target track coordinate system, including the roll angle, the pitch angle, and the yaw angle. When the position coordinates of the spectral imager, the space target, the sun, and the earth are all known and in the same coordinate system at each simulation moment, the relative position relationship is determined accordingly.

[0019] Wherein, the step (3) determines whether the space target is in the shadow area of ​​the earth at the current simulation moment, regards the space target as a point in the space-based observation scene, first uses the position coordinates of the space target, the sun and the earth obtained by step (2), adopts a cylindrical earth shadow model to describe the earth shadow interval, and calculates whether the position coordinates of the space target are in the shadow area of ​​the earth. If the space target is in the shadow area, the simulation of the current simulation moment ends, enters the next simulation moment and re-executes step (3), otherwise executes step (4): based on the relative position relationship constructed by the space target, the imager, the sun and the earth obtained in step (2), determines whether the space target is in the shadow area of ​​the earth at the current simulation moment, approximates the sunlight received at the spatial position near the earth as parallel light, adopts a cylindrical earth shadow model to describe the earth shadow interval, and calculates the angle between the geocentric vector of the target and the geocentric vector of the sun. The calculation formula is as follows:

[0020]

[0021] In the formula is the vector obtained by the position coordinates of the target in the Earth-centered Earth-fixed coordinate system, is the vector obtained by the position coordinates of the sun in the Earth-centered Earth-fixed coordinate system, α is the angle between the two vectors, 0≤α≤π, when 0≤α≤π / 2, the target is not in the Earth's shadow and can be observed, when π / 2≤α≤π, if the target is not in the Earth's shadow, the following relationship must be satisfied:

[0022]

[0023] Where R eis a constant, defined as the average radius of the earth. Therefore, under the cylindrical earth shadow model, the constraint condition for the space target to be in the non-earth shadow area is:

[0024]

[0025] If the space target is in the earth shadow area, the simulation of the current simulation moment ends, and the next simulation moment is entered and step (3) is re-executed; otherwise, step (4) is executed.

[0026] Wherein, the step (4) determines whether the line of sight of the spectral imager is blocked by the earth at the current simulation moment. The position coordinates of the spectral imager, the space target and the earth obtained by step (2) are used to regard the earth as a standard ellipsoid, and it is solved whether the connecting vector of the spectral imager and the space target has an intersection with the earth. If there is no intersection or the two intersections are not located between the spectral imager and the space target, the line of sight of the spectral imager is not blocked by the earth at the current simulation moment, and step (5) is executed. Otherwise, the simulation at the current simulation moment ends, and the next simulation moment is entered and step (3) is re-executed: the space target and the spectral imager are regarded as a certain point in the space-based observation scene, and the earth is regarded as a standard ellipsoid. The detection ray vector is obtained by subtracting the coordinate vectors of the spectral imager and the space target obtained by step (2). The calculation method is as follows:

[0027] (kx,ky,kz)=(obj_x-sat_x,obj_y-sat_y,obj_z-sat_z)

[0028] Where kx, ky, kz are the three-dimensional coordinates of the detection ray vector, obj_x, obj_y, obj_z are the three-dimensional coordinates of the space target in the Earth-centered Earth-fixed coordinate system, sat_x, sat_y, sat_z are the three-dimensional coordinates of the spectral imager in the Earth-centered Earth-fixed coordinate system, and the coefficients of the simultaneous equations can be obtained by calculating whether a vector in three-dimensional space intersects with an ellipsoid as follows:

[0029]

[0030] Where A, B, C are the polynomial coefficients in the simultaneous equations, a, b, c are the radius lengths of the earth on the x, y, z axes respectively. If the detection light intersects the earth, the calculation method is as follows:

[0031]

[0032]

[0033] Where t 1 ,t 2 To calculate the process variable, are the coordinate vectors of the two intersection points, so the constraints that the spectral imager's line of sight is not blocked by the earth are as follows:

[0034]

[0035] In the formula and are (obj_x, obj_y, obj_z) and (sat_x, sat_y, sat_z) respectively. If the line of sight of the spectral imager is not blocked by the earth at the current simulation moment, execute step (5); otherwise, the simulation at the current simulation moment ends, enters the next simulation moment and executes step (3) again.

[0036] Among them, the step (5) determines the field of view constraint of the spectral imager at the current simulation moment, uses the position coordinates of the spectral imager and the space target obtained by step (2) to construct the relative position relationship between the two, calculates the boundary point set of the coordinates of the intersection of all detection light rays and the object plane where the space target is located, and determines the single-view coverage range that can be achieved by the spectral imager without posture adjustment and the panoramic coverage range that can be achieved by posture adjustment. If the space target is in both the single-view and panoramic coverage ranges at the current moment, or if the space target is in the panoramic view but not in the single-view range at the current moment, and the space target is not in the ground shadow area or the line of sight obstruction area and is in the panoramic range at the next moment, it means that the spectral imager can Perform observation imaging and execute step (6). Otherwise, the simulation at the current simulation moment ends, and the next simulation moment is entered and step (3) is executed again: Based on the relative position information that has been determined, the imaging geometric relationship is used to determine the single-view coverage and maximum coverage of the spectral imager. The single-view coverage is defined as the imaging area that the spectral imager can actually observe at a specific moment and a specific spectral imager attitude angle, while the maximum coverage is the union of all imaging areas that the spectral imager can observe through attitude adjustment under the assumption that the attitude adjustment time of the spectral imager is 0. The single-view coverage at a specific moment is a subset of the maximum coverage at the same moment, that is:

[0037] C cat (t,i)=f(pos(t),yaw(t,i),pitch(t,i),roll(t,i),sensortype)

[0038] C poss (t) = ∑ i C cat (t,i)

[0039]

[0040] Where pos(t) is the relative position relationship matrix of the spectral imager, the target and the earth at time t, yaw(t,i), pitch(t,i), roll(t,i) are the i-th combination of the attitude angles of the spectral imager at time t, the attitude angles include yaw angle, pitch angle and roll angle, sensortype represents the field of view of the spectral imager, C cat (t,i) represents the single scene coverage of the spectral imager at time t and the i-th attitude angle combination, C poss (t,i) represents the maximum coverage of the spectral imager at time t, which is equal to the union of all single-view coverage at that time. f represents the coordinates of the intersection of all detection rays and the object plane where the space target is located under the i-th attitude angle combination at time t. In order to save computing resources, the coverage is defined as a rectangle, and the boundary point set corresponding to the four vertices of the coverage is used to represent the coverage of the spectral imager, thereby obtaining the single-view coverage and the maximum coverage. The detection ray vector at each coverage vertex is calculated as follows:

[0041]

[0042] Where R x ,R y ,R z is the three-axis rotation matrix, e 0 is the central detection ray vector of the spectral imager at time t. The central detection vector of the spectral imager is set to point from the spectral imager to the center of the earth at the initial simulation time. The central detection vector at the non-initial simulation time has the same direction as the previous simulation time. e is the detection light vector at a vertex of the coverage range of the spectral imager at time t. θ, φ, ψ are the rotation angles of the detection light vector at a vertex of the coverage range of the spectral imager around the x, y, and z axes based on the central detection light. In the calculation process of the single-scene coverage range, θ, φ, ψ are substituted in the following way:

[0043]

[0044] Where HFOV is the horizontal field of view of the spectral imager, VFOV is the vertical field of view of the spectral imager, and θ, φ, ψ are substituted in the following way during the calculation of the panoramic coverage:

[0045]

[0046] Where Roll max is the maximum roll angle of the spectral imager, Pitch maxis the maximum pitch angle of the spectral imager. The object plane where the space target is located is set to contain the coordinates of the space target point and the plane is perpendicular to the central detection vector of the spectral imager. The above formula is used to obtain the detection light vectors at the four vertices of the single-view coverage range and the panoramic coverage range at the current simulation moment, and the intersection point set with the object plane where the space target is located is calculated. It can be judged whether the position coordinates of the space target at the current simulation moment are located within the maximum coverage range or the single-view coverage range. There are four situations. The first situation is that the space target is in both the single-view and panoramic coverage ranges at the current moment, indicating that the spectral imager can observe and image the space target at the current simulation moment, and execute step (6). The second situation is that the space target is in the panoramic view but not in the single-view range at the current moment. At the next moment, the space target is not in the ground shadow area or the line of sight obstruction area and is in the panoramic range. This indicates that the spectral imaging instrument cannot observe imaging at the current moment, but can observe imaging by adjusting the attitude at the next simulation moment. The simulation at the current simulation moment ends, and the next simulation moment is entered to calculate the adjusted attitude angle. After the attitude angle calculation is completed, step (6) is executed. The third case is that the space target is not in the panoramic coverage of the imaging spectrometer at the current simulation moment, indicating that the space target cannot be observed and imaged at the current simulation moment, and the next simulation moment is entered and step (3) is executed. The fourth case is that the space target is in the panoramic view but not in the single view range at the current moment, and the space target is in at least one of the three areas outside the panoramic range, the ground shadow area, and the line of sight obstruction area at the next simulation moment, indicating that the spectral imaging instrument cannot observe and image the space target at both the current simulation moment and the next simulation moment, and the next simulation moment is entered and step (3) is executed.

[0047] When the spectral imager needs to adjust the attitude angle, in order to ensure the correct trajectory of the imager, the yaw angle cannot be adjusted, only the pitch angle and roll angle are adjusted. In order to achieve the best observation effect, the space target should be placed at the center of the single-view coverage area, that is, the central detection light of the spectral imager should be directed to the target. The specific calculation formula is as follows:

[0048]

[0049] In the formula The operator represents the search for two vectors The angle of and Respectively The projection vectors on the xOz plane and yOz plane of the spectral imager coordinate system, Pitch represents the vector pointing from the spectral imager to the space target. cal and Roll cal They represent the pitch angle and roll angle adjustments obtained by calculation, are the x-axis and y-axis direction vectors of the spectral imager track coordinate system, respectively. The z-axis of the spectral imager track coordinate system is pointed from the spectral imager to the center of the earth, the x-axis is pointed from the spectral imager to the speed direction, and the y-axis direction of the spectral imager is determined by the right-hand rule. Since the single-view coverage area cannot exceed the maximum coverage area, it is necessary to further limit the spectral imager attitude adjustment amount. Then the actual attitude angle adjustment amount should be:

[0050]

[0051] Where Pitch is the pitch angle of the spectral imager after adjusting its attitude, and Roll is the roll angle of the spectral imager after adjusting its attitude. So far, the visibility analysis of the space target at the current simulation moment and the calculation of the attitude adjustment amount of the spectral imager are completed.

[0052] Among them, the step (6) models the transmission process of the radiation characteristics of the space target, considers the three-dimensional detailed structure and material distribution of the space target, segments the space target based on the triangular surface element representation method, and performs three-dimensional modeling of the space target by determining the vertices, normal vectors and corresponding material types of each triangular surface element. The three-dimensional model of the space target and the seven types of parameters obtained in step (2) are used to obtain two radiation transmission paths and relative position relationships of the sun-space target-spectral imager and the earth-space target-spectral imager. Based on the space radiation transmission calculation method, the solar radiation intensity received at the space target and the earth's reflected light radiation intensity are obtained. , use ray tracing to calculate the observation angle between the detection light emitted by the spectral imager and the normal vector of the intersecting facet in the two radiation transmission paths, use the bidirectional reflectance distribution function and the calculated observation angle to calculate the spectral reflectance of the two radiation transmission paths respectively, combine the spectral radiation intensity with the spectral reflectance under the corresponding path to obtain the spectral radiance at the entrance pupil of the spectral imager, and complete the calculation of the entrance pupil spectral radiance image: the first step, consider the three-dimensional detailed structure and material distribution of the space target, segment the space target based on the triangular facet representation method, and perform three-dimensional modeling of the space target by determining the vertices, normal vectors and corresponding material types of each triangular facet;

[0053] The second step is to conduct the visibility analysis of the facets of the three-dimensional model of the space target, and determine the visibility of each facet on the surface of the space target, that is, the process of determining whether the light corresponding to the pixel point of the imaging plane intersects with each facet. The method of finding the intersection of the space ray and the triangular facet set is used to calculate the intersection of the ray emitted from the imaging center and passing through each pixel point with each facet on the surface of the space target, that is, the observation point recorded corresponding to each pixel. An imager with M×N pixels is used to emit M×N rays from the observation point. The rays pass through the pixel grid center of the imaging plane. The ray set is used to find the intersection with the triangular facet set of the target surface, so as to determine the facet observed corresponding to each pixel.

[0054] The third step is to construct the space radiation transmission path. In the visible light band, the radiation sources received by the space target are mainly direct solar radiation and earth-reflected solar radiation. Direct solar radiation can use measured data. Earth-reflected radiation refers to the radiation component of sunlight reflected to the outer space after reaching the surface of the earth. It is related to the reflection characteristics of the earth's surface and the solar irradiance value received on the ground. In order to simplify the problem, the average albedo model is used to describe the reflection characteristics of the earth. The direct solar radiation received at a point in the outer space at any time is determined by the distance between the target position and the sun at that time and the solar constant. The calculation formula is as follows:

[0055]

[0056] In the formula, r so is the straight-line distance between the spatial target position and the sun at that moment, r 0 is the average distance between the sun and the earth, λ is the wavelength, E 0 (λ) is the solar constant, whose physical meaning is the light energy received per unit area of ​​the top surface of the Earth's atmosphere at the average distance between the Sun and the Earth, per unit time and per unit wavelength interval, in W / m2 / nm, which is a function of the wavelength λ. sun (λ) is the direct solar radiation component received by the space target in a specific wavelength range;

[0057] It is known that the average albedo of the earth's surface is ρ = 0.35, so the earth's reflected radiation can be expressed as:

[0058] E earth (λ)=ρE sun (λ)

[0059] Where E earth (λ) is the Earth reflected radiation component received by the space target in a specific wavelength range;

[0060] The fourth step is to calculate the observation angle of each pixel corresponding to the facet in different radiation transmission paths, including the incident angle and the exit angle. The incident angle and the exit angle include the azimuth angle and the zenith angle respectively. The calculation formula is:

[0061]

[0062] In the formula, and are the vectors pointing from the intersection of the detection light and the surface element to the radiation source and from the intersection of the detection light and the surface element to the spectral imager, respectively. is the surface element normal vector, and also the z-axis direction vector of the surface element coordinate system. is the x-axis coordinate vector of the surface coordinate system, and They are and In the surface coordinate system X f O f Y f The projection vector on the plane is calculated as follows:

[0063]

[0064]

[0065] The fifth step is to calculate the entrance pupil radiance curve pixel by pixel, and use the Davies model to calculate the amount of light energy entering the entrance pupil of the imager after being reflected by the surface material of the space target. The specific calculation formula is as follows:

[0066]

[0067] In the formula is the reflectivity of a material at a specific observation angle and wavelength, σ is the root mean square of surface roughness, and λ is α is the surface autocorrelation length, ρ dh,s is the directional hemispherical reflectivity of the smooth surface, which can be determined by the Fresnel formula. Finally, the entrance pupil radiance curve is calculated pixel by pixel. The calculation formula is as follows:

[0068] E final (λ)=E sun (λ)·f r,λ,sun +E earth (λ)·f r,λ,earth

[0069] Where E final (λ) is the spectral pupil radiance of a certain wavelength range finally obtained at the entrance pupil of the spectral imager, f r,λ,sun is the reflectivity calculated in the radiation transmission path from sun to space target to spectral imager, f r,λ,earth The reflectivity calculated for the Earth-space target-spectral imager radiation transmission path.

[0070] Among them, the step (7) models the imaging process of the spectral imager, comprehensively analyzes the five links of radiation response, spectral response, motion blur, random noise, and image quantization to establish the spectral imager imaging model, and uses the spectral radiance image at the entrance pupil calculated by step (6) to generate the final simulation image: the first step is to perform radiation response calculation to convert the light radiation signal into an electrical signal. The specific calculation formula is as follows:

[0071] S(λ)=E final (λ)·a _rc +b _rc

[0072] Where S(λ) is the signal value obtained after the spectral pupil radiance of a certain wavelength range is calibrated by radiation, and a _rc ,b _rc are the radiation calibration coefficients respectively;

[0073] The second step is to calculate the spectral response. The spectral response function R(λ) is generated according to the set central wavelength, half-width and response function type. Then, the signal obtained by the radiation response in the first step is integrated pixel by pixel as follows:

[0074]

[0075] Where DN s (λ) is the signal value of a certain wavelength interval obtained after spectral response calculation;

[0076] In the third step, considering the motion blur problem caused by the speed of the space target, the length of the point spread function along the track and vertical track is determined by using the norm of the speed of the space target along the track and vertical track, and the DN obtained pixel by pixel in the second step is converted into s (λ) is organized into a two-dimensional image and convolved with the vertical track and along-track point spread functions respectively. Then the squares of the convolution results in the two directions are added and the bi-norm is calculated to obtain the final two-dimensional image DN considering the motion blur problem. t,h,w (λ), where h and w are the height and width of the two-dimensional image;

[0077] The fourth step is to consider the noise in the system imaging, generate random Gaussian noise according to the set signal-to-noise ratio and add it to the image generated in the third step to obtain DN n,h,w (λ);

[0078] The fifth step is to quantize the final spectral image to n bits and map the image pixel values ​​to 0-2 n -1 range and store it as the final simulation image Image h,w (λ).

[0079] Compared with the prior art, the advantages of the present invention are as follows: (1) This method is a full-link modeling and simulation of close-range spectral imaging of space targets. It considers the three-dimensional structure of the space target, performs visibility analysis on different face elements, and establishes a full-link simulation model of attitude-orbit modeling-radiation transfer modeling-imager modeling; (2) Dynamic simulation is performed in combination with orbit and relative motion, and the real position, attitude, and velocity information in the dynamic scene are used to quantitatively analyze the three factors affecting the visibility of the target in the space environment, namely, the ground shadow constraint, the line of sight occlusion constraint, and the imager field of view constraint, before the simulation image is generated; (3) This method does not make too many simplifications when modeling the optical properties of the surface of the space target. It considers the directional reflection characteristics of different materials instead of treating the target as a Lambertian body. It can simulate the directional characteristics of different visible face elements in the same observation scene due to different observation angles, so that the simulation effect is more in line with the actual situation. DETAILED DESCRIPTION

[0080] In order to better illustrate the space target spectral imaging simulation method involved in the present invention, the actual equatorial orbit data is used to simulate the observation process and the spectral image is generated in accordance with the AOTF imaging mode. At each simulation moment, only a spectral image in a certain band is generated. The calculated band range is 400nm to 1000nm, and the space target is set to the GF-6 three-dimensional model. The specific implementation steps are as follows:

[0081] (1) Input the attitude and orbit parameters of the space target and the spectroscopic imager, the simulation start and end time, and the simulation step size: input the space target orbit semi-major axis of 7078137m, eccentricity 0, orbit inclination 0°, ascending node right ascension 0°, perigee angular distance 0°, true perigee 150°, the space target rotates around the sun at 0.1r / min, the spectroscopic imager orbit semi-major axis of 7082137m, eccentricity 0, orbit inclination 0°, ascending node right ascension 0°, perigee angular distance 0°, true perigee 150°, the simulation time range is 2021-10-23 08:21:00 to 2021-10-23 08:21:30, and the simulation step size is 5s.

[0082] (2) Input the three parameters of the attitude and orbit parameters of the space target and the spectral imager, the simulation start and end time, and the simulation step size input in step (1) into STK, establish a specific scene for the space-based spectral imager to observe and image the space target, obtain six parameters of the position of the earth, the position and velocity of the space target, the position and velocity of the spectral imager, and the position of the sun in the geocentric earth-fixed coordinate system at each simulation moment, and the attitude of the space target in the space target track coordinate system: Input the parameters required for establishing the space-based observation scene into STK, establish a specific observation scene for the space-based spectral imager to observe and image the space target, and use the report format preset by the STK system to obtain two types of information, one is the position and velocity of the space target, the position and velocity of the spectral imager, and the position of the sun based on the geocentric earth-fixed coordinate system, and the other is the attitude angle of the space target based on the space target track coordinate system, including the roll angle, pitch angle, and yaw angle. When the position coordinates of the spectral imager, the space target, the sun, and the earth are all known and in the same coordinate system at each simulation moment, the relative position relationship is determined accordingly.

[0083] (3) Determine whether the space target is in the shadow area of ​​the earth at the current simulation moment. Consider the space target as a point in the space-based observation scene. First, use the position coordinates of the space target, the sun and the earth obtained in step (2) and adopt a cylindrical earth shadow model to describe the earth shadow interval to calculate whether the position coordinates of the space target are in the shadow area of ​​the earth. If the space target is in the shadow area, the simulation of the current simulation moment ends, enters the next simulation moment and re-executes step (3). Otherwise, execute step (4): Based on the relative position relationship constructed by the space target, the imager, the sun and the earth obtained in step (2), determine whether the space target is in the shadow area of ​​the earth at the current simulation moment. Approximate the sunlight received at the spatial position near the earth as parallel light. Use a cylindrical earth shadow model to describe the earth shadow interval. Calculate the angle between the geocentric vector of the target and the geocentric vector of the sun. The calculation formula is as follows:

[0084]

[0085] In the formula is the vector obtained by the position coordinates of the target in the Earth-centered Earth-fixed coordinate system, is the vector obtained by the position coordinates of the sun in the Earth-centered Earth-fixed coordinate system, α is the angle between the two vectors, 0≤α≤π, when 0≤α≤π / 2, the target is not in the Earth's shadow and can be observed, when π / 2≤α≤π, if the target is not in the Earth's shadow, the following relationship must be satisfied:

[0086]

[0087] Where R eis a constant, defined as the average radius of the earth. Therefore, under the cylindrical earth shadow model, the constraint condition for the space target to be in the non-earth shadow area is:

[0088]

[0089] If the space target is in the earth's shadow area, the simulation of the current simulation moment ends, and the next simulation moment is entered and step (3) is re-executed. Otherwise, step (4) is executed. The simulation results show that the space target is not in the earth's shadow area in the six simulation moments included in this simulation.

[0090] (4) Determine whether the line of sight of the spectral imager is blocked by the earth at the current simulation moment. Use the position coordinates of the spectral imager, the space target and the earth obtained in step (2), regard the earth as a standard ellipsoid, and solve whether the line vector connecting the spectral imager and the space target intersects with the earth. If there is no intersection or the two intersections are not located between the spectral imager and the space target, the line of sight of the spectral imager is not blocked by the earth at the current simulation moment, and execute step (5). Otherwise, the simulation at the current simulation moment ends, enters the next simulation moment and re-executes step (3): regard the space target and the spectral imager as a point in the space-based observation scene, regard the earth as a standard ellipsoid, and use the coordinate vectors of the spectral imager and the space target obtained in step (2) to subtract to obtain the detection ray vector. The calculation method is as follows:

[0091] (kx,ky,kz)=(obj_x-sat_x,obj_y-sat_y,obj_z-sat_z)

[0092] Where kx, ky, kz are the three-dimensional coordinates of the detection ray vector, obj_x, obj_y, obj_z are the three-dimensional coordinates of the space target in the Earth-centered Earth-fixed coordinate system, sat_x, sat_y, sat_z are the three-dimensional coordinates of the spectral imager in the Earth-centered Earth-fixed coordinate system, and the coefficients of the simultaneous equations can be obtained by calculating whether a vector in three-dimensional space intersects with an ellipsoid as follows:

[0093]

[0094]

[0095] Where A, B, C are the polynomial coefficients in the simultaneous equations, a, b, c are the radius lengths of the earth on the x, y, z axes respectively. If the detection light intersects the earth, the calculation method is as follows:

[0096]

[0097] Where t 1 ,t 2 To calculate the process variable, are the coordinate vectors of the two intersection points, so the constraints that the spectral imager's line of sight is not blocked by the earth are as follows:

[0098]

[0099] In the formula and are (obj_x, obj_y, obj_z) and (sat_x, sat_y, sat_z) respectively. If the line of sight of the spectral imager is not blocked by the earth at the current simulation moment, execute step (5); otherwise, the simulation at the current simulation moment ends, enters the next simulation moment and executes step (3) again. The simulation results show that the line of sight of the spectral imager is not blocked in the 6 simulation moments included in this simulation.

[0100] (5) Determine the field of view constraint of the spectral imager at the current simulation moment. Use the position coordinates of the spectral imager and the space target obtained in step (2) to construct the relative position relationship between the two. Calculate the boundary point set of the intersection coordinates of all detection light rays and the object plane where the space target is located. Determine the single-view coverage range that the spectral imager can achieve without posture adjustment and the panoramic coverage range that can be achieved through posture adjustment. If the space target is in both the single-view and panoramic coverage ranges at the current moment, or if the space target is in the panoramic view but not in the single-view range at the current moment, and the space target is not in the ground shadow area or the line of sight obstruction area at the next moment but is in the panoramic range, it means that the spectral imager can observe. Imaging, execute step (6), otherwise the simulation of the current simulation moment ends, enter the next simulation moment and re-execute step (3): Based on the relative position information that has been determined, use the imaging geometric relationship to determine the single-view coverage and maximum coverage of the spectral imager. The single-view coverage is defined as the imaging area that the spectral imager can actually observe at a specific moment and a specific spectral imager attitude angle, and the maximum coverage is the union of all imaging areas that the spectral imager can observe through attitude adjustment under the premise that the attitude adjustment time of the spectral imager is 0. The single-view coverage at a specific moment is a subset of the maximum coverage at the same moment, that is:

[0101] C cat (t,i)=f(pos(t),yaw(t,i),pitch(t,i),roll(t,i),sensortype)

[0102] C poss (t)=∑ i C cat (t,i)

[0103]

[0104] Where pos(t) is the relative position relationship matrix of the spectral imager, the target and the earth at time t, yaw(t,i), pitch(t,i), roll(t,i) are the i-th combination of the attitude angles of the spectral imager at time t, the attitude angles include yaw angle, pitch angle and roll angle, sensortype represents the field of view of the spectral imager, C cat (t,i) represents the single scene coverage of the spectral imager at time t and the i-th attitude angle combination, C poss (t,i) represents the maximum coverage of the spectral imager at time t, which is equal to the union of all single-view coverage at that time. f represents the coordinates of the intersection of all detection rays and the object plane where the space target is located under the i-th attitude angle combination at time t. In order to save computing resources, the coverage is defined as a rectangle, and the boundary point set corresponding to the four vertices of the coverage is used to represent the coverage of the spectral imager, thereby obtaining the single-view coverage and the maximum coverage. The detection ray vector at each coverage vertex is calculated as follows:

[0105]

[0106] Where R x ,R y ,R z is the three-axis rotation matrix, e 0 is the central detection ray vector of the spectral imager at time t. The central detection vector of the spectral imager is set to point from the spectral imager to the center of the earth at the initial simulation time. The central detection vector at the non-initial simulation time has the same direction as the previous simulation time. e is the detection light vector at a vertex of the coverage range of the spectral imager at time t. θ, φ, ψ are the rotation angles of the detection light vector at a vertex of the coverage range of the spectral imager around the x, y, and z axes based on the central detection light. In the calculation process of the single-scene coverage range, θ, φ, ψ are substituted in the following way:

[0107]

[0108] Where HFOV is the horizontal field of view of the spectral imager, VFOV is the vertical field of view of the spectral imager, and θ, φ, ψ are substituted in the following way during the calculation of the panoramic coverage:

[0109]

[0110] Where Roll max is the maximum roll angle of the spectral imager, Pitch maxis the maximum pitch angle of the spectral imager. The object plane where the space target is located is set to contain the coordinates of the space target point and the plane is perpendicular to the central detection vector of the spectral imager. The above formula is used to obtain the detection light vectors at the four vertices of the single-view coverage range and the panoramic coverage range at the current simulation moment, and the intersection point set with the object plane where the space target is located is calculated. It can be judged whether the position coordinates of the space target at the current simulation moment are located within the maximum coverage range or the single-view coverage range. There are four situations. The first situation is that the space target is in both the single-view and panoramic coverage ranges at the current moment, indicating that the spectral imager can observe and image the space target at the current simulation moment, and execute step (6). The second situation is that the space target is in the panoramic view but not in the single-view range at the current moment. At the next moment, the space target is not in the ground shadow area or the line of sight obstruction area and is in the panoramic range. This indicates that the spectral imaging instrument cannot observe imaging at the current moment, but can observe imaging by adjusting the attitude at the next simulation moment. The simulation at the current simulation moment ends, and the next simulation moment is entered to calculate the adjusted attitude angle. After the attitude angle calculation is completed, step (6) is executed. The third case is that the space target is not in the panoramic coverage of the imaging spectrometer at the current simulation moment, indicating that the space target cannot be observed and imaged at the current simulation moment, and the next simulation moment is entered and step (3) is executed. The fourth case is that the space target is in the panoramic view but not in the single view range at the current moment, and the space target is in at least one of the three areas outside the panoramic range, the ground shadow area, and the line of sight obstruction area at the next simulation moment, indicating that the spectral imaging instrument cannot observe and image the space target at both the current simulation moment and the next simulation moment, and the next simulation moment is entered and step (3) is executed.

[0111] When the spectral imager needs to adjust the attitude angle, in order to ensure the correct trajectory of the imager, the yaw angle cannot be adjusted, only the pitch angle and roll angle are adjusted. In order to achieve the best observation effect, the space target should be placed at the center of the single-view coverage area, that is, the central detection light of the spectral imager should be directed to the target. The specific calculation formula is as follows:

[0112]

[0113] In the formula The operator represents the search for two vectors The angle of and Respectively The projection vectors on the xOz plane and yOz plane of the spectral imager coordinate system, Pitch represents the vector pointing from the spectral imager to the space target. cal and Roll cal They represent the pitch angle and roll angle adjustments obtained by calculation, are the x-axis and y-axis direction vectors of the spectral imager track coordinate system, respectively. The z-axis of the spectral imager track coordinate system is pointed from the spectral imager to the center of the earth, the x-axis is pointed from the spectral imager to the speed direction, and the y-axis direction of the spectral imager is determined by the right-hand rule. Since the single-view coverage area cannot exceed the maximum coverage area, it is necessary to further limit the spectral imager attitude adjustment amount. Then the actual attitude angle adjustment amount should be:

[0114]

[0115] Wherein, Pitch is the pitch angle after the spectral imager adjusts its attitude, and Roll is the roll angle after the spectral imager adjusts its attitude. So far, the visibility analysis of the space target at the current simulation moment and the calculation of the attitude adjustment amount of the spectral imager are completed. The simulation results show that the spectral imager can observe and image the space target at the 1st, 3rd, and 5th simulation moments in this simulation, but cannot image at the 2nd and 4th moments, and the attitude angle of the spectral imager is adjusted so that the spectral imager can observe and image at the 3rd and 5th moments. The 6th simulation moment is the last simulation moment, and the space target cannot be observed and imaged. At the same time, the simulation ends after the spectral imager adjusts its attitude angle, so the space target will not be observed and imaged at the next moment.

[0116] (6) Modeling the transmission process of the radiation characteristics of the space target, considering the three-dimensional detailed structure and material distribution of the space target, segmenting the space target based on the triangular face representation method, and modeling the space target in three dimensions by determining the vertices, normal vectors and corresponding material types of each triangular face. Using the three-dimensional model of the space target and the seven types of parameters obtained in step (2), the two radiation transmission paths and relative position relationships of the sun-space target-spectral imager and the earth-space target-spectral imager are obtained. Based on the space radiation transmission calculation method, the solar radiation intensity received at the space target and the earth's reflected light radiation intensity are obtained. Ray tracing is used to calculate the observation angle between the detection light emitted by the spectral imager and the normal vector of the intersecting facet in the two radiation transmission paths. The spectral reflectance of the two radiation transmission paths is calculated using the bidirectional reflectance distribution function and the calculated observation angle. The spectral radiance at the entrance pupil of the spectral imager is obtained by combining the spectral radiation intensity with the spectral reflectance under the corresponding path, and the calculation of the entrance pupil spectral radiance image is completed: the first step is to consider the three-dimensional detailed structure and material distribution of the space target, segment the space target based on the triangular facet representation method, and perform three-dimensional modeling of the space target by determining the vertices, normal vectors and corresponding material types of each triangular facet;

[0117] The second step is to conduct the visibility analysis of the facets of the three-dimensional model of the space target, and determine the visibility of each facet on the surface of the space target, that is, the process of determining whether the light corresponding to the pixel point of the imaging plane intersects with each facet. The method of finding the intersection of the space ray and the triangular facet set is used to calculate the intersection of the ray emitted from the imaging center and passing through each pixel point with each facet on the surface of the space target, that is, the observation point recorded corresponding to each pixel. An imager with M×N pixels is used to emit M×N rays from the observation point. The rays pass through the pixel grid center of the imaging plane. The ray set is used to find the intersection with the triangular facet set of the target surface, so as to determine the facet observed corresponding to each pixel.

[0118] The third step is to construct the space radiation transmission path. In the visible light band, the radiation sources received by the space target are mainly direct solar radiation and earth-reflected solar radiation. Direct solar radiation can use measured data. Earth-reflected radiation refers to the radiation component of sunlight reflected to the outer space after reaching the surface of the earth. It is related to the reflection characteristics of the earth's surface and the solar irradiance value received on the ground. In order to simplify the problem, the average albedo model is used to describe the reflection characteristics of the earth. The direct solar radiation received at a point in the outer space at any time is determined by the distance between the target position and the sun at that time and the solar constant. The calculation formula is as follows:

[0119]

[0120] In the formula, r so is the straight-line distance between the spatial target position and the sun at that moment, r 0 is the average distance between the sun and the earth, λ is the wavelength, E 0 (λ) is the solar constant, whose physical meaning is the light energy received per unit area of ​​the top surface of the Earth's atmosphere at the average distance between the Sun and the Earth, per unit time and per unit wavelength interval, in W / m2 / nm, which is a function of the wavelength λ. sun (λ) is the direct solar radiation component received by the space target in a specific wavelength range;

[0121] It is known that the average albedo of the earth's surface is ρ = 0.35, so the earth's reflected radiation can be expressed as:

[0122] E earth (λ)=ρE sun (λ)

[0123] Where E earth (λ) is the Earth reflected radiation component received by the space target in a specific wavelength range;

[0124] The fourth step is to calculate the observation angle of each pixel corresponding to the facet in different radiation transmission paths, including the incident angle and the exit angle. The incident angle and the exit angle include the azimuth angle and the zenith angle respectively. The calculation formula is:

[0125]

[0126]

[0127] In the formula, and are the vectors pointing from the intersection of the detection light and the surface element to the radiation source and from the intersection of the detection light and the surface element to the spectral imager, respectively. is the surface element normal vector, and also the z-axis direction vector of the surface element coordinate system. is the x-axis coordinate vector of the surface coordinate system, and They are and In the surface coordinate system X f O f Y f The projection vector on the plane is calculated as follows:

[0128]

[0129] The fifth step is to calculate the entrance pupil radiance curve pixel by pixel, and use the Davies model to calculate the amount of light energy entering the entrance pupil of the imager after being reflected by the surface material of the space target. The specific calculation formula is as follows:

[0130]

[0131] In the formula is the reflectivity of a material at a specific observation angle and wavelength, σ is the root mean square of surface roughness, and λ is α is the surface autocorrelation length, ρ dh,s is the directional hemispherical reflectivity of the smooth surface, which can be determined by the Fresnel formula. Finally, the entrance pupil radiance curve is calculated pixel by pixel. The calculation formula is as follows:

[0132] E final (λ)=E sun (λ)·f r,λ,sun +E earth (λ)·f r,λ,earth

[0133] Where E final (λ) is the spectral pupil radiance of a certain wavelength range finally obtained at the entrance pupil of the spectral imager, f r,λ,sun is the reflectivity calculated in the radiation transmission path from sun to space target to spectral imager, f r,λ,earth The reflectivity calculated for the Earth-space target-spectral imager radiation transmission path.

[0134] (7) Modeling the imaging process of the spectral imager, comprehensively analyzing the five links of radiation response, spectral response, motion blur, random noise, and image quantization to establish the spectral imager imaging model, and using the spectral radiance image at the entrance pupil calculated in step (6) to generate the final simulation image: The first step is to calculate the radiation response and convert the light radiation signal into an electrical signal. The specific calculation formula is as follows:

[0135] S(λ)=E fi x al (λ)·a _rc +b _rc

[0136] Where S(λ) is the signal value obtained after the spectral pupil radiance of a certain wavelength range is calibrated by radiation, and a _rc ,b _rc are the radiation calibration coefficients, set to 1.1 and 0.2 respectively;

[0137] The second step is to calculate the spectral response. The central wavelength range is 400nm-1000nm, and the initial value is 400nm. At the same time, as the simulation time increases, the central wavelength increases by 50nm each time, which is 400nm, 450nm, 500nm, 550nm, 600nm, and 650nm respectively. The half-height width is set to 10nm, and the response function type is set to Gaussian function. The spectral response function R(λ) is generated according to the set central wavelength, half-height width and response function type, and then the signal obtained by the radiation response in the first step is integrated and calculated pixel by pixel as follows:

[0138]

[0139] Where DN s (λ) is the signal value of a certain wavelength interval obtained after spectral response calculation;

[0140] In the third step, considering the motion blur problem caused by the speed of the space target, the length of the point spread function along the track and vertical track is determined by using the norm of the speed of the space target along the track and vertical track, and the DN obtained pixel by pixel in the second step is converted into s (λ) is organized into a two-dimensional image and convolved with the vertical track and along-track point spread functions respectively. Then the squares of the convolution results in the two directions are added and the bi-norm is calculated to obtain the final two-dimensional image DN considering the motion blur problem. t,h,w (λ), where h and w are the height and width of the two-dimensional image;

[0141] The fourth step is to consider the noise in the system imaging, set the signal-to-noise ratio to 30, generate random Gaussian noise based on the set signal-to-noise ratio and add it to the image generated in the third step to obtain DN n,h,w (λ);

[0142] The fifth step is to quantize the final spectral image to 10 bits, map the image pixel values ​​to the range of 0-1023 and store them as the final simulated image Image h,w (λ), and finally generate three GF-6 simulation images based on the space-based observation scene at the 1st, 3rd, and 5th simulation moments.

[0143] The embodiments of the present invention are described in detail above, but the contents are only preferred implementation cases of the present invention and cannot be considered to limit the scope of implementation of the present invention. The implementation methods of each step can be changed, and all equal changes and improvements made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A method for simulating spectral imaging of space targets based on space-based observation scenes, characterized in that: It contains the following steps: (1) Input the attitude and orbit parameters of the space target and the spectral imager, the simulation start and end time, and the simulation step size; (2) Input the three parameters of the space target and the spectral imager input in step (1), the attitude and orbit parameters, the simulation start and end time, and the simulation step size into STK, establish a specific scenario for the space-based spectral imager to observe and image the space target, and obtain the six parameters of the position of the earth in the Earth-centered Earth-fixed coordinate system at each simulation moment, the position and velocity of the space target, the position and velocity of the spectral imager, and the position of the sun, as well as the attitude of the space target in the space target track coordinate system; (3) Determine whether the space target is in the shadow area of ​​the earth at the current simulation moment. Consider the space target as a point in the space-based observation scene. First, use the position coordinates of the space target, the sun and the earth obtained in step (2) and use a cylindrical earth shadow model to describe the earth shadow interval. Calculate whether the position coordinates of the space target are in the shadow area of ​​the earth. If the space target is in the shadow area, the simulation of the current simulation moment ends, enters the next simulation moment and re-executes step (3). Otherwise, execute step (4). (4) determining whether the line of sight of the spectral imager is blocked by the earth at the current simulation moment, using the position coordinates of the spectral imager, the space target and the earth obtained in step (2), regarding the earth as a standard ellipsoid, and determining whether the line vector connecting the spectral imager and the space target intersects with the earth; if there is no intersection or the two intersections are not located between the spectral imager and the space target, the line of sight of the spectral imager is not blocked by the earth at the current simulation moment, and executing step (5); otherwise, the simulation at the current simulation moment ends, and the next simulation moment is entered and step (3) is executed again; (5) Perform a field of view constraint judgment on the spectral imager at the current simulation moment, use the position coordinates of the spectral imager and the space target obtained in step (2) to construct the relative position relationship between the two, calculate the boundary point set of the coordinates of the intersection of all detection light rays and the object plane where the space target is located, and determine the single-view coverage range that can be achieved by the spectral imager without posture adjustment and the panoramic coverage range that can be achieved by posture adjustment. If the space target is in both the single-view and panoramic coverage ranges at the current moment, or the space target is in the panoramic view but not in the single-view range at the current moment, and the space target is not in the ground shadow area or the line of sight obstruction area and is in the panoramic range at the next moment, it means that the spectral imager can perform observation imaging, and execute step (6). Otherwise, the simulation at the current simulation moment ends, and the next simulation moment is entered and step (3) is re-executed; (6) Modeling the transmission process of the radiation characteristics of the space target, considering the three-dimensional detailed structure and material distribution of the space target, segmenting the space target based on the triangular face representation method, and modeling the space target in three dimensions by determining the vertices, normal vectors and corresponding material types of each triangular face. Using the three-dimensional model of the space target and the seven types of parameters obtained in step (2), two radiation transmission paths and relative position relationships of the sun-space target-spectral imager and the earth-space target-spectral imager are obtained. Based on the space radiation transmission calculation method, the radiation intensity of the sunlight received at the space target and the radiation intensity of the reflected light from the earth are obtained. Ray tracing is used to calculate the observation angles of the detection light emitted by the spectral imager and the intersecting face normal vectors in the two radiation transmission paths. The spectral reflectance of the two radiation transmission paths is calculated using the bidirectional reflectance distribution function and the calculated observation angles. The spectral radiance of the spectral imager at the entrance pupil is obtained by combining the spectral radiation intensity and the spectral reflectance under the corresponding path, and the calculation of the entrance pupil spectral radiance image is completed. (7) Modeling the imaging process of the spectral imager is carried out, and the five links of radiation response, spectral response, motion blur, random noise, and image quantization are comprehensively analyzed to establish the imaging model of the spectral imager. The spectral radiance image at the entrance pupil calculated by step (6) is used to generate the final simulation image.

2. The method for simulating spectral imaging of space targets based on space-based observation scenes according to claim 1, characterized in that: The step (1) inputs the attitude and orbit parameters of the space target and the spectral imager, the simulation start and end time, and the simulation step length: the orbit parameters are defined in the format of six orbit elements, including the semi-major axis, eccentricity, orbit inclination, pericentric angle, ascending node longitude, and true anomaly; the attitude parameters of the space target include four types: three-axis stabilization toward the sun, three-axis stabilization toward the earth, spin toward the sun, and spin toward the earth; the simulation start and end time are defined in the coordinated universal time format, and the simulation step length is in seconds.

3. The method for simulating spectral imaging of space targets based on space-based observation scenes according to claim 1, characterized in that: The step (2) inputs the three parameters of the attitude and orbit parameters of the space target and the spectral imager input in step (1), the simulation start and end time, and the simulation step size into STK, establishes a specific scene for the space-based spectral imager to observe and image the space target, obtains six parameters of the position of the earth, the position and speed of the space target, the position and speed of the spectral imager, and the position of the sun in the geocentric earth-fixed coordinate system at each simulation moment, and the attitude of the space target in the space target track coordinate system: the parameters required for establishing the space-based observation scene are input into STK, and a specific space-based spectral imager is established to observe and image the space target. Two types of information are obtained using the report format preset by the STK system, one type is the position and speed of the space target, the position and speed of the spectral imager, and the position of the sun based on the geocentric earth-fixed coordinate system, and the other type is the attitude angle of the space target based on the space target track coordinate system, including the roll angle, the pitch angle, and the yaw angle. When the position coordinates of the spectral imager, the space target, the sun, and the earth are all known and are in the same coordinate system at each simulation moment, the relative position relationship is determined accordingly.

4. The method for simulating spectral imaging of space targets based on space-based observation scenes according to claim 1, characterized in that: The step (3) determines whether the space target is in the shadow area of ​​the earth at the current simulation moment. The space target is regarded as a point in the space-based observation scene. First, the position coordinates of the space target, the sun and the earth obtained by step (2) are used to describe the shadow interval of the earth using a cylindrical earth shadow model to calculate whether the position coordinates of the space target are in the shadow area of ​​the earth. If the space target is in the shadow area, the simulation of the current simulation moment ends, and the next simulation moment is entered and step (3) is re-executed. Otherwise, step (4) is executed: based on the relative position relationship constructed by the space target, the imager, the sun and the earth obtained in step (2), it is determined whether the space target is in the shadow area of ​​the earth at the current simulation moment. The sunlight received at the spatial position near the earth is approximated as parallel light, and the cylindrical earth shadow model is used to describe the shadow interval of the earth. The angle between the geocentric vector of the target and the geocentric vector of the sun is calculated. The calculation formula is as follows: In the formula is the vector obtained by the position coordinates of the target in the Earth-centered Earth-fixed coordinate system, is the vector obtained by the position coordinates of the sun in the Earth-centered Earth-fixed coordinate system, α is the angle between the two vectors, 0≤α≤π, when 0≤α≤π / 2, the target is not in the Earth's shadow and can be observed, when π / 2≤α≤π, if the target is not in the Earth's shadow, the following relationship must be satisfied: Where R e is a constant, defined as the average radius of the earth. Therefore, under the cylindrical earth shadow model, the constraint condition for the space target to be in the non-earth shadow area is: If the space target is in the earth shadow area, the simulation of the current simulation moment ends, and the next simulation moment is entered and step (3) is re-executed; otherwise, step (4) is executed.

5. The method for simulating spectral imaging of space targets based on space-based observation scenes according to claim 1, characterized in that: The step (4) determines whether the line of sight of the spectral imager is blocked by the earth at the current simulation moment. The position coordinates of the spectral imager, the space target and the earth obtained in step (2) are used to regard the earth as a standard ellipsoid, and it is solved whether the connecting vector of the spectral imager and the space target has an intersection with the earth. If there is no intersection or the two intersections are not located between the spectral imager and the space target, the line of sight of the spectral imager is not blocked by the earth at the current simulation moment, and step (5) is executed. Otherwise, the simulation at the current simulation moment ends, and the next simulation moment is entered and step (3) is re-executed: the space target and the spectral imager are regarded as a certain point in the space-based observation scene, and the earth is regarded as a standard ellipsoid. The detection ray vector is obtained by subtracting the coordinate vectors of the spectral imager and the space target obtained in step (2). The calculation method is as follows: (kx,ky,kz)=(obj_x-sat_x,obj_y-sat_y,obj_z-sat_z) Where kx, ky, kz are the three-dimensional coordinates of the detection ray vector, obj_x, obj_y, obj_z are the three-dimensional coordinates of the space target in the Earth-centered Earth-fixed coordinate system, sat_x, sat_y, sat_z are the three-dimensional coordinates of the spectral imager in the Earth-centered Earth-fixed coordinate system, and the coefficients of the simultaneous equations can be obtained by calculating whether a vector in three-dimensional space intersects with an ellipsoid as follows: Where A, B, C are the polynomial coefficients in the simultaneous equations, a, b, c are the radius lengths of the earth on the x, y, z axes respectively. If the detection light intersects the earth, the calculation method is as follows: Where t1, t2 are calculated process variables, are the coordinate vectors of the two intersection points, so the constraints that the spectral imager's line of sight is not blocked by the earth are as follows: In the formula and are (obj_x, obj_y, obj_z) and (sat_x, sat_y, sat_z) respectively. If the line of sight of the spectral imager is not blocked by the earth at the current simulation moment, execute step (5); otherwise, the simulation at the current simulation moment ends, enters the next simulation moment and executes step (3) again.

6. The method for simulating spectral imaging of space targets based on space-based observation scenes according to claim 1, characterized in that: The step (5) determines the field of view constraint of the spectral imager at the current simulation moment, uses the position coordinates of the spectral imager and the space target obtained in step (2) to construct the relative position relationship between the two, calculates the boundary point set of the intersection coordinates of all detection light rays and the object plane where the space target is located, and determines the single-view coverage range that can be achieved by the spectral imager without posture adjustment and the panoramic coverage range that can be achieved by posture adjustment. If the space target is in both the single-view and panoramic coverage ranges at the current moment, or if the space target is in the panoramic view but not in the single-view range at the current moment, and the space target is not in the ground shadow area or the line of sight obstruction area and is in the panoramic range at the next moment, it means that the spectral imager can be Observe the imaging and execute step (6). Otherwise, the simulation at the current simulation moment ends, and the next simulation moment is entered and step (3) is executed again: Based on the relative position information that has been determined, the imaging geometric relationship is used to determine the single-view coverage and maximum coverage of the spectral imager. The single-view coverage is defined as the imaging area that the spectral imager can actually observe at a specific moment and a specific spectral imager attitude angle, while the maximum coverage is the union of all imaging areas that the spectral imager can observe through attitude adjustment under the assumption that the attitude adjustment time of the spectral imager is 0. The single-view coverage at a specific moment is a subset of the maximum coverage at the same moment, that is: C cat (t,i)=f(pos(t),yaw(t,i),pitch(t,i),roll(t,i),sensortype) C poss (t)=∑ i C cat (t,i) Where pos(t) is the relative position relationship matrix of the spectral imager, the target and the earth at time t, yaw(t,i), pitch(t,i), roll(t,i) are the i-th combination of the attitude angles of the spectral imager at time t, the attitude angles include yaw angle, pitch angle and roll angle, sensortype represents the field of view angle of the spectral imager, C cat (t,i) represents the single scene coverage of the spectral imager at time t and the i-th attitude angle combination, C poss (t,i) represents the maximum coverage of the spectral imager at time t, which is equal to the union of all single-view coverage at that time. f represents the coordinates of the intersection of all detection rays and the object plane where the space target is located under the i-th attitude angle combination at time t. In order to save computing resources, the coverage is defined as a rectangle, and the boundary point set corresponding to the four vertices of the coverage is used to represent the coverage of the spectral imager, thereby obtaining the single-view coverage and the maximum coverage. The detection ray vector at each coverage vertex is calculated as follows: Where R x ,R y ,R z is the three-axis rotation matrix, e0 is the central detection ray vector of the spectral imager at time t, the central detection vector of the spectral imager is set to point from the spectral imager to the center of the earth at the initial simulation time, and the central detection vector at the non-initial simulation time is the same as the direction of the previous simulation time, e is the detection light vector at a vertex of the coverage range of the spectral imager at time t, θ, φ, ψ are the rotation angles of the detection light vector at a vertex of the coverage range of the spectral imager around the x, y, z axes with the central detection light as the reference, and θ, φ, ψ are substituted in the following way during the calculation of the single scene coverage range: Where HFOV is the horizontal field of view of the spectral imager, VFOV is the vertical field of view of the spectral imager, and θ, φ, ψ are substituted in the following way during the calculation of the panoramic coverage: Where Roll max is the maximum roll angle of the spectral imager, Pitch max is the maximum pitch angle of the spectral imager. The object plane where the space target is located is set to contain the coordinates of the space target point and the plane is perpendicular to the central detection vector of the spectral imager. The above formula is used to obtain the detection light vectors at the four vertices of the single-view coverage range and the panoramic coverage range at the current simulation moment, and the intersection point set with the object plane where the space target is located is calculated. It can be judged whether the position coordinates of the space target at the current simulation moment are located within the maximum coverage range or the single-view coverage range. There are four situations. The first situation is that the space target is in both the single-view and panoramic coverage ranges at the current moment, indicating that the spectral imager can observe and image the space target at the current simulation moment, and execute step (6). The second situation is that the space target is in the panoramic view but not in the single-view range at the current moment. At the next moment, the space target is not in the ground shadow area or the line of sight obstruction area and is in the panoramic range. This indicates that the spectral imaging instrument cannot observe imaging at the current moment, but can observe imaging by adjusting the attitude at the next simulation moment. The simulation at the current simulation moment ends, and the next simulation moment is entered to calculate the adjusted attitude angle. After the attitude angle calculation is completed, step (6) is executed. The third case is that the space target is not in the panoramic coverage of the imaging spectrometer at the current simulation moment, indicating that the space target cannot be observed and imaged at the current simulation moment, and the next simulation moment is entered and step (3) is executed. The fourth case is that the space target is in the panoramic view but not in the single view range at the current moment, and the space target is in at least one of the three areas outside the panoramic range, the ground shadow area, and the line of sight obstruction area at the next simulation moment, indicating that the spectral imaging instrument cannot observe and image the space target at both the current simulation moment and the next simulation moment, and the next simulation moment is entered and step (3) is executed. When the spectral imager needs to adjust the attitude angle, in order to ensure the correct trajectory of the imager, the yaw angle cannot be adjusted, only the pitch angle and roll angle are adjusted. In order to achieve the best observation effect, the space target should be placed at the center of the single-view coverage area, that is, the central detection light of the spectral imager should be directed to the target. The specific calculation formula is as follows: In the formula The operator represents the search for two vectors The angle of and Respectively The projection vectors on the xOz plane and yOz plane of the spectral imager coordinate system, Pitch represents the vector pointing from the spectral imager to the space target. cal and Roll cal They represent the pitch angle and roll angle adjustments obtained by calculation, are the x-axis and y-axis direction vectors of the spectral imager track coordinate system, respectively. The z-axis of the spectral imager track coordinate system is pointed from the spectral imager to the center of the earth, the x-axis is pointed from the spectral imager to the speed direction, and the y-axis direction of the spectral imager is determined by the right-hand rule. Since the single-view coverage area cannot exceed the maximum coverage area, it is necessary to further limit the spectral imager attitude adjustment amount. Then the actual attitude angle adjustment amount should be: Where Pitch is the pitch angle of the spectral imager after adjusting its attitude, and Roll is the roll angle of the spectral imager after adjusting its attitude. So far, the visibility analysis of the space target at the current simulation moment and the calculation of the attitude adjustment amount of the spectral imager are completed.

7. The method for simulating spectral imaging of space targets based on space-based observation scenes according to claim 1, characterized in that: The step (6) models the transmission process of the radiation characteristics of the space target, considers the three-dimensional detailed structure and material distribution of the space target, segments the space target based on the triangular face representation method, and performs three-dimensional modeling of the space target by determining the vertices, normal vectors and corresponding material types of each triangular face. The three-dimensional model of the space target and the seven types of parameters obtained in step (2) are used to obtain two radiation transmission paths and relative position relationships of the sun-space target-spectral imager and the earth-space target-spectral imager. The radiation intensity of the sunlight received at the space target and the radiation intensity of the earth's reflected light are obtained based on the space radiation transmission calculation method. Ray tracing is used to calculate the observation angle between the detection light emitted by the spectral imager and the normal vector of the intersecting facet in the two radiation transmission paths. The spectral reflectance of the two radiation transmission paths is calculated using the bidirectional reflectance distribution function and the calculated observation angle. The spectral radiance at the entrance pupil of the spectral imager is obtained by combining the spectral radiation intensity with the spectral reflectance under the corresponding path, and the calculation of the entrance pupil spectral radiance image is completed: the first step is to consider the three-dimensional detailed structure and material distribution of the space target, segment the space target based on the triangular facet representation method, and perform three-dimensional modeling of the space target by determining the vertices, normal vectors and corresponding material types of each triangular facet; The second step is to conduct the visibility analysis of the facets of the three-dimensional model of the space target, and determine the visibility of each facet on the surface of the space target, that is, the process of determining whether the light corresponding to the pixel point of the imaging plane intersects with each facet. The method of finding the intersection of the space ray and the triangular facet set is used to calculate the intersection of the ray emitted from the imaging center and passing through each pixel point with each facet on the surface of the space target, that is, the observation point recorded corresponding to each pixel. An imager with M×N pixels is used to emit M×N rays from the observation point. The rays pass through the pixel grid center of the imaging plane. The ray set is used to find the intersection with the triangular facet set of the target surface, so as to determine the facet observed corresponding to each pixel. The third step is to construct the space radiation transmission path. In the visible light band, the radiation sources received by the space target are mainly direct solar radiation and earth-reflected solar radiation. Direct solar radiation can use measured data. Earth-reflected radiation refers to the radiation component of sunlight reflected to the outer space after reaching the surface of the earth. It is related to the reflection characteristics of the earth's surface and the solar irradiance value received on the ground. In order to simplify the problem, the average albedo model is used to describe the reflection characteristics of the earth. The direct solar radiation received at a point in the outer space at any time is determined by the distance between the target position and the sun at that time and the solar constant. The calculation formula is as follows: In the formula, r so is the straight-line distance between the spatial target position and the sun at that moment, r0 is the average distance between the sun and the earth, λ represents the wavelength, and E0(λ) is the solar constant. Its physical meaning is the light energy received per unit area on the top surface of the earth's atmosphere at the average distance between the sun and the earth in unit time and per unit wavelength interval. The unit is W / m2 / nm, which is a function of the wavelength λ. E sun (λ) is the direct solar radiation component received by the space target in a specific wavelength range; It is known that the average albedo of the earth's surface is ρ = 0.35, so the earth's reflected radiation can be expressed as: E earth (λ)=ρE sun (l) Where E earth (λ) is the Earth reflected radiation component received by the space target in a specific wavelength range; The fourth step is to calculate the observation angle of each pixel corresponding to the facet in different radiation transmission paths, including the incident angle and the exit angle. The incident angle and the exit angle include the azimuth angle and the zenith angle respectively. The calculation formula is: In the formula, and are the vectors pointing from the intersection of the detection light and the surface element to the radiation source and from the intersection of the detection light and the surface element to the spectral imager, respectively. is the surface element normal vector, and also the z-axis direction vector of the surface element coordinate system. is the x-axis coordinate vector of the surface coordinate system, and They are and In the surface coordinate system X f O f Y f The projection vector on the plane is calculated as follows: The fifth step is to calculate the entrance pupil radiance curve pixel by pixel, and use the Davies model to calculate the amount of light energy entering the entrance pupil of the imager after being reflected by the surface material of the space target. The specific calculation formula is as follows: In the formula is the reflectivity of a material at a specific observation angle and wavelength, σ is the root mean square of surface roughness, and λ is α is the surface autocorrelation length, ρ dh,s is the directional hemispherical reflectivity of the smooth surface, which can be determined by the Fresnel formula. Finally, the entrance pupil radiance curve is calculated pixel by pixel. The calculation formula is as follows: E final (λ)=E sun (λ)·f r,λ,sun +E earth (λ)·f r,λ,earth Where E final (λ) is the spectral pupil radiance of a certain wavelength range finally obtained at the entrance pupil of the spectral imager, f r,λ,sun is the reflectivity calculated in the radiation transmission path from sun to space target to spectral imager, f r,λ,earth The reflectivity calculated for the Earth-space target-spectral imager radiation transmission path.

8. The method for simulating spectral imaging of space targets based on space-based observation scenes according to claim 1, characterized in that: The step (7) models the imaging process of the spectral imager, comprehensively analyzes the five links of radiation response, spectral response, motion blur, random noise, and image quantization to establish a spectral imager imaging model, and uses the spectral radiance image at the entrance pupil calculated by step (6) to generate the final simulation image: the first step is to perform radiation response calculation to convert the light radiation signal into an electrical signal. The specific calculation formula is as follows: S(λ)=E final (λ)·a _rc +b _rc Where S(λ) is the signal value obtained after the spectral pupil radiance of a certain wavelength range is calibrated by radiation, and a _rc ,b _rc are the radiation calibration coefficients respectively; The second step is to calculate the spectral response. The spectral response function R(λ) is generated according to the set central wavelength, half-width and response function type. Then, the signal obtained by the radiation response in the first step is integrated pixel by pixel as follows: Where DN s (λ) is the signal value of a certain wavelength interval obtained after spectral response calculation; In the third step, considering the motion blur problem caused by the speed of the space target, the length of the point spread function along the track and vertical track is determined by using the norm of the speed of the space target along the track and vertical track, and the DN obtained pixel by pixel in the second step is converted into s (λ) is organized into a two-dimensional image and convolved with the vertical track and along-track point spread functions respectively. Then the squares of the convolution results in the two directions are added and the bi-norm is calculated to obtain the final two-dimensional image DN considering the motion blur problem. t,h,w (λ), where h and w are the height and width of the two-dimensional image; The fourth step is to consider the noise in the system imaging, generate random Gaussian noise according to the set signal-to-noise ratio and add it to the image generated in the third step to obtain DN n,h,w (λ); The fifth step is to quantize the final spectral image to n bits and map the image pixel values ​​to 0-2 n -1 range and store it as the final simulation image Image h,w (λ).

Images